ESMO Supporter 2018

Displaying One Session

Hall A3 - Poster Area Networking Hub Poster Display session
Date
20.10.2018
Time
12:30 - 13:30
Location
Hall A3 - Poster Area Networking Hub
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

Biomarkers

Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

66P - TCR beta chain convergence defines the tumor infiltrating T cell repertoire of melanoma and non-small cell lung carcinoma.

Presentation Number
66P
Lecture Time
12:30 - 12:30
Speakers
  • Timothy Looney (Singapore, SG)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

T cell convergence refers to the process whereby antigen-driven selection enriches for T cell receptors having a shared antigen specificity but different amino acid or nucleotide sequence. T cell recruitment and expansion within the tumor microenvironment (TME) may be directed by responses to tumor neoantigen, suggesting that elevated T cell convergence could be a general feature of the tumor infiltrating T cell repertoire. Here we evaluate evidence for T cell convergence in tumor biopsy from research subjects with melanoma and non-small cell lung carcinoma (NSCLC) and peripheral blood leukocytes (PBL) from healthy donors.

Methods

Total RNA from 63 melanoma and 19 NSCLC tumor biopsy research samples (non-FFPE) was extracted for use in long-amplicon TCRB chain sequencing (mean amplicon of 330bp covering CDR1, 2 and 3) via the Oncomine TCR Beta-LR Research Assay. To evaluate T cell convergence, we searched for instances where TCRB chains were identical in amino acid space but had distinct nucleotide sequences owing to N-addition and exonucleotide chewback within the V-D and D-J junctions of the CDR3. To provide context, we evaluated evidence for T cell convergence in PBL T cell repertoires derived from 16 healthy donors.

Results

Sequencing of melanoma biopsy research samples typically yielded within the range of 2000 to 8000 clones per sample. Convergent T cell receptors were identified in the great majority of melanoma and NSCLC tumor infiltrating T cell repertoires having greater than 100 detected clones (92% and 100%, respectively). The frequency of convergent T cell rearrangements was significantly greater in melanoma and NSCLC tumor biopsies than T cell repertoires derived from healthy PBL research samples.

Conclusions

These data suggest that T cell convergence may be a common feature of the melanoma and NSCLC infiltrating T cell repertoire. Convergence was more frequently observed within the TME than T cell repertoires derived from healthy PBL, consistent with elevated antigen-driven T cell selection within the TME. The finding of elevated T cell convergence in melanoma and NSCLC suggests that convergence may be a hallmark of immunogenic tumors. For research use only.

Legal entity responsible for the study

Thermo Fisher Scientific.

Funding

Thermo Fisher Scientific.

Disclosure

T. Looney, G. Lowman, L. Miller, E. Linch: Employee: Thermo Fisher Scientific.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

67P - PI3K inhibition and modulation of immune and tumor microenvironment markers by copanlisib in patients with non-Hodgkin's lymphoma or advanced solid tumors

Presentation Number
67P
Lecture Time
12:30 - 12:30
Speakers
  • Ahmad H. Awada (Brussels, BE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Copanlisib, a pan-class I phosphatidylinositol 3-kinase (PI3K) inhibitor with predominant activity against α and δ isoforms, shows immune stimulating effects and anti-tumor activity in combination with immune checkpoint blockers in tumor models. Here we explore tumor PI3K signaling and the effects of copanlisib on immune and tumor microenvironment modulation in subjects with NHL and solid tumors (ST) in a phase I pharmacodynamic study (NCT02155582).

Methods

Patients received 0.4 or 0.8 mg/kg (equivalent to a flat dose of 60 mg) copanlisib on an intermittent schedule (QW, IV, 3 wks on/1 wk off). Tumor biopsies were performed at baseline and Day 15. Tumor PI3K isoforms, PTEN, and CD3, CD4 and CD8 tumor infiltrating lymphocytes (TILs) were assessed by IHC. Plasma protein markers were measured using multiplexed immunoassay. Tumor and lymphoma responses were based on RECIST 1.1 and modified Cheson 2007 criteria, respectively.

Results

A total of 61 patients were treated: 33 NHL (20 at 0.4 mg/kg and 13 at 0.8 mg/kg) and 28 ST (14 at 0.4 mg/kg and 14 at 0.8 mg/kg). Among patients treated at 0.8 mg/kg, 2 had CR (PTCL and DLBCL) and 5 had PR (MCL, FL, 2 DLBCL, and endometrial adenocarcinoma); 1 additional PR (DLBCL) occurred at 0.4 mg/kg. Tumor PI3K α was detected in most NHL (18/19) and ST (22/25) samples with comparable intensity. PI3K δ was predominantly present in NHL (18/19). PI3K ß and γ were present in a subset of NHL and ST. PTEN loss was more frequent in ST than NHL. CD3, CD4 and CD8 TIL numbers were higher in NHL than in ST. At 0.8 mg/kg copanlisib decreased tumor pAKT and pS6 in both NHL and ST, and reduced CD4 TILs in NHL (mean -81%, n = 7), with little effect on CD8 TILs in both NHL and ST. In plasma, copanlisib decreased cytokine/chemokines (e.g. CCL2, CCL5, CCL17), and factors associated with macrophages (e.g. CD163, CCL4, CCL22) and Treg cells (IL-2Ra). High baseline levels of CD27 and IL2Ra were associated with tumor reduction in NHL (p < 0.05, unadjusted).

Conclusions

The high prevalence of the PI3K isoforms – especially α in both NHL and ST and δ in NHL – is consistent with a role for PI3K signaling in immune suppression. The immune modulation profile for copanlisib supports combination studies with immunotherapy.

Clinical trial identification

NCT02155582.

Legal entity responsible for the study

Bayer AG.

Funding

Bayer AG.

Disclosure

I. Genvresse, K. Koechert, G. Cisternas, C. Granvil, C. Pena, L. Liu: Employee: Bayer. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

68P - Pan-squamous genomic profiling stratified by anatomic tumor site and viral association

Presentation Number
68P
Lecture Time
12:30 - 12:30
Speakers
  • Meagan Montesion (Cambridge, MA, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Squamous cell carcinomas (SCC) have diverse anatomic etiologies but may share common genomic biomarkers. We profiled 7,871 unique SCCs across nine anatomic sites to investigate commonality in genomic alterations (GA), tumor mutational burden (TMB), human papillomavirus (HPV) association, and mutational signatures.

Methods

Tissue from over 8,100 unique SCC samples originating from nine anatomic sites (anogenital (anus, cervix, penis, vagina, vulva), esophagus, head and neck, lung, and skin) were sequenced by hybrid capture-based comprehensive genomic profiling to evaluate GA and TMB. About 3% of non-cutaneous SCC samples had UV signatures, indicative of potential primary site misdiagnoses, and were filtered from the analysis. Detection of HPV, including high-risk strains 16, 18, 31, 33, and 45, was implemented through de novo assembly of non-human sequencing reads and BLASTn comparison against all viral nucleotide sequences in the NCBI database.

Results

The proportion of HPV+ patients by anatomic site varied, with the highest being anal (91%) and cervical (83%). The mutational landscape of each cohort was similar, regardless of anatomic origin, but clustered based on HPV status. The largest differences in GA frequency as stratified by HPV- vs. HPV+ were TP53 (87% vs. 12%), CDKN2A (45% vs. 6%), and PIK3CA (22% vs. 33%). The median TMB in cases originating from HPV-associated sites was similar, regardless of HPV status. Higher median TMB was observed in lung and skin cases, which exhibited significant enrichment of mutational signatures indicative of tobacco- and UV-induced DNA damage, respectively.

Conclusions

HPV+ and HPV- SCC populations have distinct genomic profiles and, for the latter, anatomic site is correlated with TMB distribution, secondary to associated carcinogen exposure. As such, biomarkers such as TMB and UV signature can provide unexpected insight into site of origin misdiagnoses and may correlate with benefit from immune checkpoint inhibitors.

Tumor Site% HPV+Median TMB (Interquartile Range)% TMB  > = 10% TMB  > = 20
Anogenital (n = 1213)765 (6)175
Head and Neck (n = 1843)364 (5)155
Esophageal (n = 416)65 (4)132
Lung (n = 3977)59 (8)439
Skin (n = 422)840 (69)6862

Legal entity responsible for the study

Foundation Medicine, Inc.

Funding

Foundation Medicine, Inc.

Disclosure

M. Montesion, E.S. Sokol: Employee: Foundation Medicine, Inc. L.A. Albacker, G.M. Frampton, V.A. Miller, J.S. Ross, S.M. Ali: Employee and stock ownership: Foundation Medicine, Inc. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

69P - Toward the Standardization of Bioinformatics Methods for the Accurate Assessment of Tumor Mutational Burden (TMB)

Presentation Number
69P
Lecture Time
12:30 - 12:30
Speakers
  • Han Chang (Princeton, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

TMB has emerged as a predictive biomarker of response to immune checkpoint inhibitors. CheckMate 227 demonstrated that patients with non-small cell lung cancer (NSCLC) with TMB ≥10 mutations/megabase derived enhanced benefit from first-line treatment with nivolumab + ipilimumab vs chemotherapy (Hellmann et al. NEJM 2018). Standardized approaches for the measurement and reporting of TMB are essential for the real-world implementation of TMB. This study aimed to refine a bioinformatic pipeline for mutation calling and annotation of whole exome sequencing (WES) data for TMB assessment.

Methods

In CheckMate 026, TMB was assessed by WES on formalin-fixed, paraffin-embedded tumor samples and matched blood from 312 patients with NSCLC (Carbone et al. NEJM 2017). Data from each sample were aligned to a reference human genome and somatic mutations were called by comparing matched tumor and blood samples using the TNsnv and Strelka algorithms. The somatic mutations were additionally filtered for germline variants in public databases. TMB scores were compared with data from 710 NSCLC samples in The Cancer Genome Atlas (TCGA) dataset. We examined the concordance of TMB estimates using several mutation filtering schemes, with and without matched germline controls.

Results

TMB scores including synonymous, indel, frameshift, and nonsense mutations (all mutations) were ∼3-fold higher than matched data filtered for missense mutations only, but values were highly correlated (Spearman’s r = 0.99). Scores including missense mutations only were similar to those generated from TCGA, but those including all mutations were on average higher. Using public databases for germline subtraction showed a trend for race-dependent increases in TMB scores.

Conclusions

Standardization of bioinformatic analyses is critical to the clinical implementation of TMB assessment. TMB assessment is sensitive to variations in bioinformatic parameters (eg, which type of mutation to include), which may affect the identification of patients likely to respond to immunotherapy. These results show that data from different pipelines are highly correlated, suggesting that reliable assessment of TMB across different centers and platforms is achievable.

Clinical trial identification

NCT02041533.

Legal entity responsible for the study

Bristol-Myers Squibb.

Funding

Bristol-Myers Squibb.

Editorial Acknowledgement

Medical writing assistance was provided by Stuart Rulten, PhD, of Spark Medica Inc. (US), funded by Bristol-Myers Squibb.

Disclosure

H. Chang, S. Srinivasan, A. Sasson, R. Golhar, D. Greenawalt, J.D. Szustakowski: Employment and stock ownership: Bristol-Myers Squibb. S. Kirov: Stock ownership and compensation: Bristol-Myers Squibb.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

70P - Prediction of primary resistance to anti-PD1 therapy (APD1) in 2nd line NSCLC

Presentation Number
70P
Lecture Time
12:30 - 12:30
Speakers
  • Egbert F. Smit (Amsterdam, NL)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

APD1, while capable of restoring immunity, does not benefit all patients. While molecular tests like PD-L1 expression and TMB help in enriching response in respective subsets, a test identifying patients showing primary resistance to APD1 which does not require tissue samples could help in optimizing treatment regimens.

Methods

Outcome data (PFS/OS) were correlated with protein profiles from mass spectrometry of the circulating proteome of pretreatment serum from 116 2nd line NSCLC patients treated with nivolumab (development set S) using multivariate machine learning methods related to deep learning. The resulting test stratified patients into three groups: group A having very poor outcomes, group B having intermediate outcomes, and group C having very good outcomes. Development results were obtained using out-of-bag estimators. Two additional patient cohorts treated with nivolumab, V1(N = 58) and V2(N = 75), were used for validation.

Results

The proportions of patients in A, B, and C were 41:43:32 in S, 23:18:17 in V1, and 32:19:24 in V2. Median PFS/OS in the poor prognosis group A was 43/132 days in S, 105/189 days in V1, 90/278 days in V2, and in the good prognosis group C 276/528 days in S, 192/459 days in V1, and 155/not reached days in V2. In a comparison with historical controls treated with single agent chemotherapy and analyzed with the same technique, nivolumab appeared substantially superior in the good prognosis group C, while there was no evidence of superiority in the poor prognosis group A. In multivariate analysis including performance status, smoking history, and histology, the test remained an independent predictor of outcome. The patterns of protein expression related to poor prognosis in group A patients were associated with elevated complement, wound healing, and acute phase reactants.

Conclusions

We developed and validated a test stratifying patients into three groups with significantly different outcomes on nivolumab. The poor prognosis group showed little benefit from nivolumab, and other treatments may be needed, while in the good prognosis group outcomes were very good for a 2nd line population. These results emphasize the importance of the host immune response in the prediction of APD1 efficacy.

Legal entity responsible for the study

Biodesix.

Funding

Has not received any funding.

Disclosure

J.G. Aerts: Advisory boards: BMS, MSD, Roche, AstraZeneca, Eli-Lilly, Boehringer Ingelheim, Amphera; Stock owner: Amphera. H. Roder: Officer, salary, stock options: Biodesix. C. Oliveira, J. Roder: Employment, stock options: Biodesix. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

71P - Immunomodulatory germline variation impacts the development of multiple primary melanoma (MPM).

Presentation Number
71P
Lecture Time
12:30 - 12:30
Speakers
  • Robert Ferguson (New York, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

During their lifetime about 8% of patients with single primary cutaneous melanoma (SPM) will develop multiple primary melanomas (MPM), which are associated with significantly higher mortality compared to patients with SPM. Based on the evidence that the immune system plays a role in regulating melanoma progression we explored whether germline genetic variants controlling the expression of immunomodulatory genes (immunomodulatory quantitative trait loci, eQTLs) discern risk of MPM compared to patients with SPM or healthy controls.

Methods

Previously, we identified 50 eQTLs significantly associated with the expression of 265 immunomodulatory genes using the MuTHer twin cohort. These 50 SNPs were genotyped in 837 SPM and 104 MPM individuals using MassARRAY system. 1047 healthy controls were obtained from a publically available GWAS on CM ascertained at MD Anderson (phs000187.v1.p1). We employed multivariate logistic regression to test the association of SNPs with MPM vs cancer-free controls and MPM vs SPM.

Results

When comparing MPM vs SPM, rs2071304, previously linked to expression of SPI1 in MuTHer data, showed a strong association with reduction of MPM risk (OR = 0.60; 95% CI = 0.45-0.81; p = 0.0007). Intriguingly, this variant also trended toward significance when comparing MPM vs controls (OR = 0.61; 95% CI: 0.44-0.85; p = 0.003). Finally, our most significant association when comparing MPM to controls was for rs2276645 (OR = 0.60; 95% CI = 0.45-0.81; p = 0.0008), an eQTL associated with Zap-70 expression.

Conclusions

Our data, for the first time, indicate that the inherited host immunity impacts risk of MPM in individuals with SPM, highlighting an importance of immune involvement in melanoma progression. The MPM risk-predicting genetic variants identified here or in expanded efforts, currently underway, may eventually lead to a diagnostic tool allowing for enhanced screening and clinical management of patients at risk of MPM, hence reducing elevated MPM-associated mortality. Additionally, our results further support that MPM and SPM may have different genetics underpinnings and should be treated as separate clinical entities.

Legal entity responsible for the study

Tomas Kirchhoff.

Funding

NIH.

Disclosure

D. Polsky: Research grant: BioRad-laboratory reagents: Novartis. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

72P - Leukocyte telomere length and recurrence risk after EGFR-TKIs therapy in patients with advanced lung adenocarcinoma

Presentation Number
72P
Lecture Time
12:30 - 12:30
Speakers
  • Ming Yang (Jinan, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Gefitinib is currently one of the mostly used epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) recommended for treating non-small cell lung cancer. However, the factors that predict treatment prognosis and drug resistance to EGFR-TKIs remain elusive. The objective of this study is to exam the association between leukocyte relative telomere length (RTL) and prognosis or drug resistance of advanced lung adenocarcinoma to gefitinib treatment.

Methods

In this study, three hundred and sixty-nine patients with stage IIIB or IV lung adenocarcinoma were recruited between January 2009 and June 2013. All patients were treated with gefitinib orally at a daily dose of 250 mg as first-line monotherapy. Leukocyte RTL of each patient was measured using quantitative polymerase chain reaction (qPCR) protocol on an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems) and calculated according to Cawthon’s formula. Differences in patients’ characteristics were calculated by Pearson’s χ2 tests or Student’s t test. Cox proportional hazard regression analyses were used to calculate univariate and multivariate hazard ratios (HRs). Survival differences were examined using the log-rank test. Two-sided P < 0.05 indicated a significant difference.

Results

Among 369 patients, EGFR mutations were positive in 181 patients (49.1%). Compared to long RTL, short leukocyte RTL was significantly associated with poor prognosis in all patients after gefitinib treatment (overall survival: 12.9 months vs. 17.8 months, P = 1.2 × 10-4; progression free survival: 7.8 months vs. 13.0 months, P = 0.043). Additionally, statistically significant association between short leukocyte RTL and shorten OS still existed among the EGFR mutant patients with gefitinib treatment (HR = 1.65, 95% CI = 1.28-2.12; P = 0.006). Besides EGFR mutation status, short RTL also contributed to significantly elevated risk of gefitinib primary resistance (HR = 1.50, 95% CI = 1.05-2.15, P = 0.027).

Conclusions

Our results highlight the potential of leukocyte RTL as a novel biomarker in advanced lung adenocarcinoma treated with EGFR-TKIs and the possibility of patient-tailored decisions based on leukocyte RTL.

Legal entity responsible for the study

Ming Yang.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

73P - Gene embedding: a novel machine learning approach to identify gene candidates related to immunotherapy responsiveness

Presentation Number
73P
Lecture Time
12:30 - 12:30
Speakers
  • Chi Tung Choy (Shatin, HK)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Apart from PD-L1 and mutational load, there are no genetic predictive biomarkers for checkpoint inhibitors treatment. In this study, gene embedding, a machine learning technique, was used to single out related genes of immune checkpoint proteins (i.e. PD-1, PD-L1, CTLA-4) as new potential predictors for such responders.

Methods

TCGA RNASeqV2 level 3 RSEM normalized read counts (January 2016) were downloaded from the Broad Institute TCGA GDAC Firehose. A shallow neural network, aka embedding layers for samples and genes, were trained using log2 transformed data. Neighbors closeness were evaluated by euclidean distance. The model was kept blind from any additional information, including cancer types, protein-protein interactions and gene ontologies.

Results

Gene expressions of 13045 samples from 36 cancer types were embedded into 50-dimension space, while cancer types were learnt by the model without supervision. Immunotherapy responders and non-responders were stimulated from melanoma (SKMC) and lung squamous cell carcinoma (LUSC) data, and hepatocellular carcinoma and prostate cancer data respectively. 9 genes (TNFRSF8, CLEC10A, FCN1, CD8B, SLA2, IL2RA, CTLA4, GZMH), 3 (CD101, LOC154761, RNF152) genes, and 6 (SH2D1A, MEI1, PDCD1, GFI1, SIT1, SIRPG) genes were found to be closely related neighbors with PD-1, PD-L1, and CTLA-4 respectively in responders but not in non-responders. All neighbors were neither co-expressed in SKMC/LUSC dataset nor indicated as interacting partners on existing databases (BioGRID, MINT, iRefWeb, STRING, HPRD and Reactome). 88.8% genes were evidenced as either directly related to checkpoint proteins and/or T cells activation in literature. Further evaluation of the role of identified targets in immune checkpoint blockade therapy would be warranted.

Conclusions

We identified potential biomarker candidates for immune checkpoint blockade therapy by TCGA data mining and demonstrated the utility of gene embedding learned from big gene expression dataset as a powerful tool to uncover gene relationships that may not be discovered otherwise without prior knowledge on functional interactions.

Legal entity responsible for the study

Department of Clinical Oncology, Faculty of Medicine, The Chinese University of Hong Kong.

Funding

Department of Clinical Oncology, Faculty of Medicine, The Chinese University of Hong Kong.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

74P - Common and rare DPYD variants are predictive for 5FU/capecitabine (5FU) toxicity: The MRC COIN and COIN-B trials

Presentation Number
74P
Lecture Time
12:30 - 12:30
Speakers
  • Ayman Madi (Wirral, GB)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Rare genetic variants in DPYP increase toxicity and screening for them prevents serious complications by upfront reduction in 5FU dose; however, most patients with severe toxicities do not have a rare mutation. We have previously shown that 2 common DPYD variants were associated with toxicity in patients with advanced colorectal cancer treated on COIN & COIN-B (abstract 3509, ASCO 2013): Cys29Arg [rs1801265] (Minor Allele Frequency (MAF) 0.21) and Val732Ile [rs1801160] (MAF 0.04). We have now genotyped 4 rare variants using the same cohort.

Methods

Blood samples were available from 2183 patients treated with first line oxaliplatin-5FU ± cetuximab. We assayed IVS14 + 1G>A [rs3918290], Asp949Val [rs67376798], Lys259Glu [rs45589337] and Ser534Asn [rs1801158] using KASPar. Primary endpoint was dose reduction or delay in chemotherapy in the first 12 weeks of treatment due to any toxicity except neuropathy. Secondary endpoints were grade ≥2 versus grade <2 for neutropenia, lethargy, Nausea & Vomiting (N&V), diarrhoea, stomatitis, Hand-Foot Syndrome (HFS) and infection.

Results

Two rare variants were associated with toxicity (OR (95% CI)): Asp949Val with neutropenia 3.2 (1.2-8.2) P = 0.019, N&V 3.4 (1.5-7.3) P = 0.002, diarrhoea 4.6 (2.1-10.1) P < 0.001 and infection 5.5 (1.3-24.2) P = 0.024; IVS14 + 1G>A with lethargy 5.3 (1.9-14.9) P = 0.002, diarrhoea 4.4 (1.7-11.0) P = 0.002, stomatitis 4.6 (1.7-12.6) P = 0.003, HFS 3.8 (1.2- 11.8) P = 0.021 and infection 19.2 (5.0-73.8) P < 0.001. MAF was 0.007 and 0.005, respectively. The effect on toxicity for our 2 common variants was not as marked (OR (95% CI)): Cys29Arg 0.8 (0.7-1.0) P = 0.008 (protective) and Val732Ile 1.6 (1.1-2.1) P = 0.006 for the primary endpoint.

Conclusions

We have validated 2 mutations, Asp949Val and IVS14 + 1G>A, as predictors for 5FU toxicity in a large cohort of patients and recommend they should be screened for. Our data suggest that common DPYD variants are also associated with toxicity but not to the same level seen with rare ones. While the presence of a single common variant is not an indication for dose modification, the presence of multiple variants in a patient might be. Further work is needed to establish what combinations of common DPYD variants would necessitate 5FU dose alteration.

Legal entity responsible for the study

Cardiff University and MRC CTU.

Funding

This work was supported by The Bobby Moore Fund from CRUK, Cancer Research Wales, Tenovus, the Wales Gene Park, and an unrestricted research grant from Merck Serono.

Disclosure

T.S. Maughan, J.P. Cheadle: Part funded by an unrestricted research grant from Merck Serono. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

75P - The landscape of NTRK fusions in Chinese solid tumor patients

Presentation Number
75P
Lecture Time
12:30 - 12:30
Speakers
  • Qi Ling (Hangzhou, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

NTRK gene fusions resulting in the elevated expression of TRK kinases were discovered in a wide variety of tumor types but generally at a low frequency. TRK inhibitors such as LOXO-101 and entrectinib had remarkable and durable antitumor activities in patients (pts) with TRK fusion-positive cancers, regardless of age or tumor type.

Methods

FFPE tumor samples of over 3700 Chinese solid tumor pts were collected for NGS-based assay. We measured the gene fusions, mutations, and copy number alterations in tumor tissue against matched blood. Pan-Trk IHC testing was performed on five NTRK1 fusions.

Results

12 pts were idendified as NTRK fusion positive, which accounted to approximately 0.3% of the Chinese solid tumor pts in our cohort. Seven out of 12 pts harbored NTRK1 fusions with 6 partners, and the remaining ones were NTRK3 fusions with 2 partners. The half fusions with novel partner genes or breakpoints were defined as likely fusions. The tumor types of pts with NTRK fusions included NSCLC, colorectal cancer (CRC), prostate cancer and fibrosarcoma. NTRK fusions were more likely t occur in NSCLC and CRC, which accouted for 0.3% and 1.4%, respectively. Three predicted likely fusions and two known fusions were selected to perform pan-Trk IHC assay, and four were IHC positive. One known TPR-NTRK1 fusion not detected by IHC, which highlighted the necessity to use NGS to detect NTRK fusions due to higher sensitivity and capability. NTRK fusions were not always mutually exclusive with driver mutations. Three lung cancer pts with NTRK fusions also harbored EGFR-sensitive mutations.

Gene fusionCancer typeIHCPathogenic
NTRK1
TPM3 exon10-NTRK1 exon8Colorectal cancerPositiveLikely
IRF2BP2 exon1-NTRK1 exon8Prostate cancerPositiveLikely
PRDX1 exon5-NTRK1 exon12Lung adenocarcinomaPositiveLikely
LMNA exon2-NTRK1 exon11FibrosarcomaPositiveKnown
TPR exon21-NTRK1 exon 9Lung adenocarcinomaNegativeKnown
TPM3 exon10-NTRK1 exon8Colorectal cancerNALikely
AMOTL2 exon6-NTRK1 exon12Lung adenocarcinomaNALikely
NTRK3
ETV6 exon5 -NTRK3 exon15Colorectal cancerNAKnown
ETV6 exon5 -NTRK3 exon15Colorectal cancerNAKnown
ETV6 exon5 -NTRK3 exon15Squamous cell lung cancerNAKnown
ETV6 exon4 -NTRK3 exon14Small cell lung cancerNAKnown
AKAP13 exon3-NTRK3 exon14Lung adenocarcinomaNALikely

Conclusions

This study revealed NTRK fuisons in approximately 0.3% of Chinese solid tumor pts for the first time. The NTRK gene fusions more commonly occurred in NSCLC (0.3%) and CRC (1.4%), but may occur with other targetable alterations such as EGFR-activating mutations. NGS panel sequencing showed the advantage of detecting NTRK fusion and providing structure information of partners which could potentially guide more precise treatment options.

Legal entity responsible for the study

OrigiMed.

Funding

Has not received any funding.

Disclosure

R. Ma, D. Chen, H. Chen, X. Dong, W. Wang, M. Yao: Employee: OrigiMed. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

76P - Correlative analyses of serum biomarkers and efficacy outcomes in the randomized phase 2 trial of lenvatinib (LEN), everolimus (EVE), or LEN+EVE in patients with metastatic renal cell carcinoma

Presentation Number
76P
Lecture Time
12:30 - 12:30
Speakers
  • Chung-Han Lee (New York, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

In a randomized, 3-arm, phase 2 trial of patients (pts) with metastatic renal cell carcinoma (mRCC) following 1 VEGF-targeted therapy, LEN+EVE improved median progression-free survival (PFS) compared with EVE (hazard ratio [HR] 0.40; 95% confidence interval [CI] 0.24–0.68; P < 0.001) or LEN (HR 0.66; 95% CI 0.39–1.10; P = 0.121).(Motzer et al. Lancet Oncol 2015) We present biomarker analyses from this study.

Methods

Serum samples collected at baseline were analyzed by Luminex-based xMAP® assays for 40 candidate biomarkers. Baseline biomarker levels were correlated with PFS using Cox regression analysis. Biomarkers with the strongest associations (top 5 ranked by log-rank P-value and HR) with PFS in the LEN+EVE arm were integrated into composite biomarker scores (CBS)(Voss et al. Br J Cancer 2016) All P-values are nominal.

Results

Serum samples from 145 pts (LEN+EVE, n = 49; LEN, n = 50; EVE, n = 46) were analyzed. HGF, MIG, IL-18BP, IL-18, and ANG-2 concentrations demonstrated the strongest correlation with PFS and were selected for the CBS analysis. Associations with PFS are summarized in the table. In the LEN+EVE arm, median PFS for pts with high (3–5) vs low (0–2) CBS was 20.1 vs 5.6 months, respectively (HR 0.28; P = 0.002), whereas no significant difference between high vs low CBS was seen in the EVE arm (3.6 vs 5.5 months; HR 1.02; P = 0.951). Median PFS differed significantly between treatment arms for pts with high CBS (LEN+EVE vs EVE, 20.1 vs 3.6 months; P < 0.001), but not for pts with low CBS (LEN+EVE vs EVE, 5.6 vs 5.5 months; P = 0.329).

LEN+EVE
EVE
LEN+EVE vs EVE HR (95% CI)
nMedian PFS (months)nMedian PFS (months)
Total5114.6505.50.40 (CI 0.24–0.68) P < 0.001
High CBS2820.1203.60.19 (0.09–0.41) P < 0.001
Low CBS205.6245.50.70 (0.34–1.43) P = 0.329
High vs low HR (95% CI)0.28 (0.12–0.63) P = 0.0021.02 (0.52–1.99) P = 0.951

Conclusions

In pts treated with LEN+EVE, high CBS was correlated with PFS benefit; further research is needed to determine if the score can be used to identify pts who may benefit from combination therapy. Altogether, these biomarkers may be predictors of response to LEN+EVE therapy in pts with mRCC.

Clinical trial identification

NCT01136733.

Legal entity responsible for the study

Eisai Inc.

Funding

Eisai Inc.

Editorial Acknowledgement

Editorial assistance was provided by Oxford PharmaGenesis of Newtown, PA, USA, which was funded by Eisai Inc.

Disclosure

C-H. Lee: Consulting/advisory role: Exelixis; Research funding: Bristol-Myers Squibb, Eisai, Pfizer. R.J. Motzer: Consulting/advisory role: Eisai, Exelixis, Novartis, Pfizer; Research funding: Bristol-Myers Squibb, Eisai, Genentech/Roche, GlaxoSmithKline, Novartis, Pfizer; Travel, accomodations, expenses: Bristol-Myers Squibb. H. Glen: Consulting/advisory role: Astellas Pharma, Eisai, and Janssen; Research funding: Eisai. M.D. Michaelson: Consulting/advisory role: Exelixis, Novartis, Pfizer; Research funding: Argos Therapeutics, Bristol-Myers Squibb, Eisai, Exelixis, Genentech, Merck, Pfizer, Takeda. J. Larkin: Research funding: Pfizer (institution), Novartis (institution), Bristol-Myers Squibb, Merck Sharp & Dohme; Travel, accomodations, expenses: Bristol-Myers Squibb, Merck Sharp & Dohme, Pfizer, Novartis, GlaxoSmithKline, Genentech, Eisai. Y. Minoshima, M. Kanekiyo, R. Dairiki, Y. Funahashi: Employee: Eisai Co., Ltd. P. Sachdev, C.E. Dutcus: Employee: Eisai Inc. M.H. Voss: Honoraria: Novartis; Consulting/advisory role: Alexion Pharmaceuticals, Calithera Biosciences, Exelixis, GlaxoSmithKline, Natera, Novartis, Pfizer; Research funding: Bristol-Myers Squibb, Genentech/Roche, Pfizer; Travel, accommodations, expenses: Novartis, Takeda.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

77P - End-to-end learning to predict survival in patients with gastric cancer using convolutional neural networks

Presentation Number
77P
Lecture Time
12:30 - 12:30
Speakers
  • Armin Meier (Munich, DE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

While established deep learning approaches for histopathology usually consist of a two-step process, a cell or region segmentation and subsequent feature calculation, end-to-end learning has been used to predict patient survival directly from digital tissue sections. We aimed to apply a deep learning approach in a series of gastric cancer (GC) tissue microarrays (TMAs) in order to identify regions in the tissue related to a high-risk of poor survival, and subsequently stratify patients into two risk groups.

Methods

Image patches (size 80µm) were extracted from 469 TMA cores constructed from 248 GC resection specimens which were scanned after immunohistochemistry for CD8 and KI67. For each stain, a survival convolutional neural network (CNN) was trained to maximize a log partial likelihood derived from the Cox proportional hazards model [Mobadersany, PNAS, 2018] and to predict patch-based risks for cancer-specific death in a 10-fold pre-validation procedure, creating risk heatmaps for each core. Aggregation from patch to patient level was done by averaging the risks from all patches of each patient.

Results

We generated risk heatmaps comprising on median 1300 image patches per patient for the CD8 and KI67 stained tissue sections. Stratifying patients into low- and high-risk groups by taking the cohort median as threshold led to a significant log-rank test p-value (<0.01). Regarding the Lauren classification, the diffuse type was associated with higher risks than the intestinal type (T-test p-value < 0.015). Visual assessment of the risk heatmaps revealed an association of low-risk regions in CD8-stained sections with clusters of CD8(+) cells and presence of CD8(+) cells in stroma, whereas tumor epithelium and stroma regions with a low density of CD8(+) cells are associated with higher risks.

Conclusions

We applied survival CNNs to IHC stained gastric cancer tissue samples to directly associate image regions with cancer-specific death risks. This information may be used to deepen our knowledge on how tissue morphology relates to survival risk, and to stratify patients into high and low risk groups. Our results will be extended to other biomarkers and will be validated using data from another clinical site.

Legal entity responsible for the study

Heike I. Grabsch.

Funding

Pathological Society of Great Britain and Ireland, Kanagawa Cancer Center.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

78P - Detection of Targetable Kinase Fusions in 7260 patients in an integrated Cancer System

Presentation Number
78P
Lecture Time
12:30 - 12:30
Speakers
  • Ankur Parikh (Philadelphia, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Kinase fusions (KF), such as those involving ALK, are eminently targetable genomic alterations (GA) in lung and other cancers, the latter suggested by early clinical evidence (PMID: 29079636). We undertook a review of 7260 patient samples from a tertiary cancer care-focused network of five hospitals assayed with comprehensive genomic profiling (CGP).

Methods

Hybrid capture based CGP was performed on 7260 advanced cancer cases (12/2012-2/2018), with assessment of at least 186 genes (intronic baiting for at least 14) in tissue, and 62 genes (intronic baiting for 6) in circulating tumor DNA samples. Tumor mutational burden (TMB) was determined up to 1.2 Mbp of sequenced DNA.

Results

77/7260 (1%) samples in this series harbored KF. Patients (pts) with KF+ tumors had a median age of 53 years vs. 56 years in the overall population. The TMB in KF+ cases was 3.51 mut/Mb vs. 4.39 mut/Mb for all cases. KF were found in 55 lung (71%) and 22 (29%) non-lung samples. Of KF+ cases, 71% were non-small cell lung cancer, and the remainder were sarcoma (5%), breast cancer (4%), thyroid (4%), cancer of unknown primary (4%), pancreatic (3%), colorectal (3%) and others (1% each). Of KF+ non-lung cases, 39% had BRAF fusions, 30% had ALK fusions, 26% had RET fusions, and 4% had ROS1 fusions. One KF+ sarcoma pt received matched targeted therapy with ALK inhibitors including ceritinib and crizotinib. More recently, in 2017 samples alone, 42% (10/24) of KF+ cases were non-lung.

Conclusions

Greater access to CGP has led to increased detection of advanced cancer patients with tumors harboring KF, particularly those with non-lung cancers. The low frequency of the latter is a challenge for clinical investigation. As such, innovative solutions such as basket trial for kinase inhibitors are needed, which may be feasible in an integrated cancer care system with high patient volume.

Legal entity responsible for the study

Ankur Parikh.

Funding

Has not received any funding.

Disclosure

A. Parikh: Consultant: Foundation Medicine, Inc (FMI). S.M. Ali, A.B. Schrock, P. Reddy, V.A. Miller, J.S. Ross: Employee and equity interest: FMI. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

79P - Clinical and analytical validation of an FDA approved comprehensive genomic profiling (CGP) assay incorporating multiple companion diagnostics for targeted and immunotherapies

Presentation Number
79P
Lecture Time
12:30 - 12:30
Speakers
  • Yali Li (Cambridge, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Due to the compelling predictive value of companion diagnostic (CDx) biomarkers tied to targeted and immune-based therapies, well-characterized robust analytic and clinical validation of genomic assays has become mandatory. An NGS-based CGP (comprehensive genomic profiling) platform was developed in compliance with FDA guidelines for CDx indications.

Methods

DNA extracted from FFPE tumor tissue underwent whole-genome shotgun library construction and hybridization-based capture, followed by sequencing using Illumina HiSeq 4000. Sequence data were processed using a proprietary analysis pipeline designed to identify sub substitutions, indels, copy number alterations, genomic rearrangements, microsatellite instability (MSI), and tumor mutational burden (TMB) in 324 genes.

Results

Clinical validity was demonstrated by establishing statistical non-inferiority between CGP and the respective approved CDx, e.g. cobas EGFR and BRAF mutational testing, ALK rearrangements with FISH and IHC, ERBB2 amplification with FISH, and others. For analytical validity, concordance with an orthogonal NGS platform was 94.6% for substitutions and indels, and within-assay reproducibility had positive percent agreement (PPA) of 99.4%. TMB was analytically validated via concordance with whole-exome sequencing. For the first 616 patients (25% non-small cell lung cancer) assayed in clinical care, 6.8% of cases had TMB exceeding 20 mut/Mb, with 25% of these also harboring microsatellite instability. For 143 NSCLC cases, >50% harbored 10 mut/Mb. Of 354 cases with CDx findings possible, 25.6% had such findings, which were split nearly evenly between indications to benefit from and contraindications to targeted therapies.

Conclusions

We developed a CGP assay and demonstrated clinical and analytical validity for CDx biomarkers for targeted therapy, with clinical validation for TMB in progress via correlation with prospective immunotherapy trials. Initial oncologist feedback indicates impact of assay results on course of treatment decisions in patient care.

Legal entity responsible for the study

Foundation Medicine, Inc.

Funding

Foundation Medicine, Inc.

Disclosure

Y. Li: Employee of and stockholder: Foundation Medicine Inc. J.X. Sun: Stockholder: Foundation Medicine Inc. J. Skoletsky, C. Milbury, C. Burns, W-K. Yip, N. Dewal, J. He, J. Tuesdell, J.A. Elvin, G. Otto, D. Lipson, J.S. Ross, V.A. Miller, M. Doherty, C. Vietz: Employee and stockholder: Foundation Medicine Inc. E. Peters, E. Schleifman, J. Noe: Employee and stockholder: Genentech Inc. S. Jenkins: Employee and stockholder: AstraZeneca.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

80P - SET overexpression promotes colorectal cancer progression and determines poor outcome in patients with localized disease.

Presentation Number
80P
Lecture Time
12:30 - 12:30
Speakers
  • Blanca Torrejón (Madrid, ES)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

SET deregulation is an alteration that determines poor outcome in metastatic colorectal cancer (CRC) patients promoting cell growth and decreasing sensitivity to standard chemotherapeutic agents such as oxaliplatin and 5-fluorouracil. Moreover, this alteration represents a key event to inhibit the tumor suppressor PP2A in metastatic CRC. However, the role of SET in CRC progression and its potential clinical impact in early-stage CRC patients still remain to be investigated.

Methods

In this work, we studied the biological effects of SET on migration using wound-healing and transwell migration assays, and cell invasion ability was determined by colony-forming assays after SET silencing or overexpression. Moreover, we analyzed SET expression by immunostaining in a cohort of 231 CRC patients without metastatic disease at diagnosis. We also quantified the expression of the negative SET regulator miR-199b in a set of CRC patient samples.

Results

We observed that SET deregulation promotes cell migration and markedly affects invasion ability of CRC cells. At the clinical level, SET overexpression was detected in 14.7% of cases. We found this alteration associated with worse ECOG performance status, and with relapse in the subgroup of stage II CRC patients. Moreover, SET overexpression determined significantly shorter overall survival and time to metastasis. Interestingly, its prognostic value was particularly evident in patients older than 70 years. We also identified miR-199b downregulation as a molecular mechanism to deregulate SET in CRC patients with localized disease.

Conclusions

Of importance, our results indicate that SET could serve to anticipate undesirable relapses in stage II CRC patients and define a subgroup of early stage CRC patients that could benefit by the use of SET antagonists or PP2A-activating drugs such as FTY720 in anticancer protocols.

Legal entity responsible for the study

Fundación Jiménez Díaz.

Funding

This work was supported by PI15/00934 and PI16/01468 grants from “Instituto de Salud Carlos III FEDER”. BT is supported by Fundación Conchita Rábago de Jiménez Díaz.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

81P - The functional MDM4 genetic variant in advanced lung adenocarcinoma patients treated with EGFR-TKIs

Presentation Number
81P
Lecture Time
12:30 - 12:30
Speakers
  • Nasha Zhang (Ji'nan, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

As a mostly used epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitor (TKI), gefitinib significantly prolongs survival of lung adenocarcinoma patients with EGFR mutations. However, more than 10% of EGFR mutation-positive patients do not respond and a substantial fraction of responded patients progress after 8-12 months’ treatment. Identification of new biomarkers for EGFR-TKIs prognosis is vital. The objective of this study is to explore associations between MDM4 genetic variant and survival of lung adenocarcinoma patients treated with gefitinib.

Methods

384 patients with stage IIIB or IV lung adenocarcinoma were recruited between January 2009 and June 2013. Patients were treated with gefitinib orally at a daily dose of 250 mg as 1st-line monotherapy. MDM4 rs4245739 A>C genotypes were determined using MassArray system. Dual luciferase reporter gene assays evaluated the function of MDM4 rs4245739 genetic variant in lung adenocarcinoma cell lines A549 and H1299. The differences of patient clinical characteristics were calculated by student's t test or χ2 test. Survival differences were examined by log-rank test. Multivariate Cox regression analysis assessed prognostic factors for PFS or OS. Two-sided P < 0.05 indicated a significant difference.

Results

Among 384 patients, EGFR mutations were positive in 181 patients (47.1%). Median progression-free survival (PFS) and overall survival (OS) for all patients with the rs4245739AC genotype were significantly longer than that of the AA carriers (PFS: 22.9vs.10.9 months, P < 0.001; OS: 27.3vs.16.5 months, P = 0.003). Notably, in the EGFR mutation-positive subgroup, individuals with MDM4 rs4245739AC genotype showed 14.1 months prolonged PFS (28.8 months vs. 14.7 months; P = 0.022) and 12.2 months prolonged OS (31.4 months vs. 19.2 months; P = 0.047) compared to the AA group. Reporter gene assays showed that the rs4245739A allele leads to significantly increased MDM4 expression in lung adenocarcinoma cells compared to the C allele (P < 0.05).

Conclusions

MDM4 rs4245739 genotypes may act as prognostic biomarker for patients’ survival to gefitinib therapy and offer help to individualized treatment in lung adenocarcinoma patients with EGFR mutations.

Legal entity responsible for the study

Nasha Zhang.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

82P - PD-L1 expression on pre-treatment circulating tumour cells, but not serum VEGF, is predictive of response to pembrolizumab in melanoma

Presentation Number
82P
Lecture Time
12:30 - 12:30
Speakers
  • Muhammad A. Khattak (Perth, AU)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Immune checkpoint inhibitors including pembrolizumab and nivolumab have revolutionised treatment of melanoma with a small proportion of patients deriving durable disease control lasting up to 5 years. However, majority of patients do not respond to these drugs that are costly and can lead to substantial toxicity. Therefore, there is an urgent need for biomarkers that can identify patients that will respond to these therapies.

Methods

We used multi-parametric flow cytometry to identify circulating tumour cell (CTC) subpopulations based on the expression of melanoma markers MCAM, MCSP, ABCB5, CD271 and RANK in metastatic melanoma patients prior to commencing treatment with pembrolizumab (n = 40) or with ipilimumab alone or in combination with nivolumab (n = 14). In particular, we evaluated the expression of PD-L1 on CTCs in relation with response to treatment and progression free survival (PFS). Serum vascular endothelial growth factor (VEGF) concentrations were also evaluated.

Results

Pre-treatment serum VEGF concentrations were significantly higher in patients not responding to ipilimumab treatment (alone or in combination with nivolumab) (p = 0.0094). In contrast, serum VEGF was not predictive of response to pembrolizumab. Pre-treatment CTC positivity was not associated with response or PFS in either cohorts. However, PD-L1 expression on CTCs was associated with response to therapy. PD-L1 expression was found in 13 of 16 responders with detectable CTCs, while only 4 of 10 non-responders had PD-L1 detectable on their CTCs (p = 0.0425). Expression of PD-L1 on CTCs was also associated with longer PFS (p = 0.0117).

Conclusions

Our results provide evidence for the first time in melanoma, that detection of PD-L1 on CTCs is predictive of response to pembrolizumab and longer PFS.

Legal entity responsible for the study

Research Governance Office Sir Charles Gairdner Hospital and Fiona Stanley Hospital.

Funding

MSD.

Disclosure

M.A. Khattak, M. Ziman: Research grant: MSD. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

83P - Olaparib plus paclitaxel sensitivity in biomarker subgroups of gastric cancer

Presentation Number
83P
Lecture Time
12:30 - 12:30
Speakers
  • Yu-Zhen Liu (Macclesfield, GB)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Study 39 (NCT01063517) showed a significant improvement in overall survival (OS) following olaparib plus paclitaxel (OP) vs paclitaxel (P) alone in advanced gastric cancer, with improvements greatest in patients (pts) with low or undetectable tumour ATM protein levels. The Phase III GOLD study (NCT01924533) also showed a survival benefit trend for OP vs P. We investigated biomarker subgroups, candidate genes and homologous recombination repair (HRR) deficiency using pre-defined and post-hoc exploratory analyses, to determine if a predictive relationship exists between such biomarkers and clinical outcome in gastric cancer pts treated with OP.

Methods

Candidate genes, HRR deficiencies, loss of heterozygosity (LoH) and microsatellite insufficiency (MSI) were assessed in formalin-fixed, paraffin-embedded gastric tumour samples from GOLD by next-generation sequencing. HRR deficiencies were identified as carrying pathogenic mutations in any 15 HRR genes and LoH using allele-specific copy number information coupled with assessed tumour purity. ATM protein level was assessed by immunohistochemistry (IHC). Clinical outcomes analyzed were OS, PFS and ORR.

Results

Efficacy in the genetics evaluable population (n = 400) was broadly consistent with the overall GOLD population (n = 525) for each outcome investigated. ATM-negative patients by IHC had better prognosis independent of treatment. No statistically significant associations with clinical outcomes were identified (Table). Post-hoc exploratory analyses indicated good prognosis (OS) in pts with an ATM mutation, and poor prognosis in pts with CDH1, FGFR2 or KRAS mutations.

Pre-specified subgroupPatients/events in OS, PFS, ORR, respectivelyOSPFSORR
Hazard ratioHazard ratioOdds ratio
[<1 favours OP][<1 favours OP][>1 favours OP]
(P value)(P value)(P value)
Overall population525/381, 449, 720.80 (0.026)*0.84 (0.065)1.68 (0.057)
Evaluable for genetics400/284, 342, 570.80 (0.065)*0.87 (0.214)1.40 (0.256)
ATM IHC status+ve324/233, 281, 420.79 (0.076)0.96 (0.737)1.06 (0.870)
-ve76/51, 61, 150.72 (0.292)0.61 (0.081)3.41 (0.082)
ATM IHC nullNot null381/273, 327, 520.82 (0.094)0.89 (0.315)1.25 (0.551)
Null19/11, 15, 50.55 (0.426)0.28 (0.090)6.46 (0.141)
ATMWt378/274, 326, 550.80 (0.062)0.86 (0.175)1.43 (0.244)
Mut22/10, 16, 20.83 (0.803)0.91 (0.875)0.83 (1.00)
ATM/BRCAWt369/266, 317, 540.81 (0.090)0.85 (0.160)1.52 (0.185)
Mut31/18, 25, 30.94 (0.908)1.11 (0.795)0.39 (0.578)
HRRWt354/254, 302, 520.79 (0.066)0.83 (0.122)1.54 (0.177)
Mut46/30, 40, 51.05 (0.903)0.96 (0.921)0.58 (0.659)
ARID1aWt332/232, 287, 460.82 (0.133)0.86 (0.216)1.28 (0.526)
Mut68/52, 55, 110.65 (0.326)0.91 (0.741)2.06 (0.335)
TP53Wt136/90, 113, 170.80 (0.322)0.98 (0.918)2.03 (0.197)
Mut264/194, 229, 400.77 (0.074)0.76 (0.042)1.16 (0.732)
MSIStable381/269, 325, 530.81 (0.081)0.87 (0.216)1.32 (0.377)
High19/15, 17, 40.41 (0.198)0.83 (0.784)3.22 (0.582)
LoH scoreLOH evaluable ≤6198/139, 175, 32 137/101,119, 210.73 (0.064) 0.73 (0.122)0.75 (0.062) 0.80 (0.222)1.11 (0.847) 1.32 (0.638)
>661/38, 56, 110.62 (0.175)0.71 (0.246)0.84 (1.00)

ATM IHC null, 0% tumour cells expressed ATM; HRR, homologous recombination repair; LoH, loss of heterozygosity; MSI, microsatellite instability status; Mut, mutation; ORR, objective response rate; OS, overall survival; PFS, progression free survival; Wt, wild type Analyses represent the HR for olaparib vs placebo in the stated subgroups *Significant difference for OP vs P was set at P < 0.025 Assessed using the Ventana ATM (Y170) assay (ATM negative defined as < 25% nuclei staining)

Conclusions

None of the pre-specified molecular subgroups had a better outcome from adding olaparib to paclitaxel than the overall GOLD population. Additional studies are required to understand the OS signal observed with OP treatment in pts with advanced gastric cancer in the GOLD study. Yu-Zhen Liu and Darren Hodgson are joint first authors.

Clinical trial identification

NCT01924533.

Legal entity responsible for the study

AstraZeneca.

Funding

AstraZeneca.

Editorial Acknowledgement

Claire Routley, Mudskipper Business Ltd.

Disclosure

D. Hodgson, Z. Lai, D. Balcerzak, A. Sharpe, J.C. Barrett, M. Orr, T.S. Gutjahr, B. Dougherty, X. Shi: Employee and stockholder: AstraZeneca. G. Locker: Employee: AstraZeneca. M.P. Roudier: Employee and shareholder: AstraZeneca. R. Miller: Employee and share/stockholder: Roche. W.H. Kim: Grant: Seoul National University Hospital. S-A. Im: Preclinical research grant: AstraZeneca. N. Boku: Grant and personal fees: Taiho, Ono; Personal fees: Chugai, Eli Lilly. Y-J. Bang: Advisory/consulting role, honorarium, research funding: AstraZeneca. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

84P - A diagnostic model for hepatitis B virus-related hepatocellular carcinoma in China: a large-scale, multi-center study

Presentation Number
84P
Lecture Time
12:30 - 12:30
Speakers
  • Tian Yang (Shanghai, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Almost one third of carriers of hepatitis B virus (HBV) world-wide are in China and more than 80% hepatocellular carcinoma (HCC) in China are associated with HBV infection. So early detection of HCC in HBV-infected patients is necessary. In the present study, we aimed to develop a diagnostic model by combining protein induced by Vitamin K absence or antagonist-II (PIVKA-II) and α-fetoprotein (AFP) for HBV-related HCC.

Methods

We recruited consecutive patients with HBV-related HCC, chronic hepatitis B, HBV-related cirrhosis and healthy controls at 11 hospitals in China from June 2016 to May 2017 for a training cohort. A validation cohort was enrolled at the same sites from FebruaryJune 2017 to September 2017. HCC was defined on the basis of ultrasound, CT, or MRI characteristics and confirmed by histopathology. Serum PIVKA-II level was measured by ARCHITECT immunoassay and AFP was measured with commercially available ELISA. Receiver operating characteristics (ROC) were used to calculate diagnostic accuracy.

Results

The training cohort consisted of 2019 participants, 908 with HBV-related HCC, 289 with chronic hepatitis B, 314 with HBV-related cirrhosis, and 508 healthy controls. The validation cohort comprised 655 participants, 289 with HBV-related HCC, 113 with chronic hepatitis B, 98 with HBV-related cirrhosis, and 155 healthy controls. Levels of PIVKA-II in serum were significantly higher in HBV-related HCC than all controls. ROC curves showed the optimum diagnostic cutoff for PIVKA-II was 44.18 mAU/mL (area under curve [AUC], 0.907 [95% CI 0.892-0.922], sensitivity 81.13%, and specificity 94.97% in the training cohort; 0.909 [0.883-0.934], 79.02%, and 95.46% in the validation cohort). PIVKA-II maintained diagnostic accuracy for patients with HBV-related HCC who were AFP negative. A model combined PIVKA-II, AFP, age, gender and liver cirrhosis improved diagnostic accuracy for HBV-related HCC versus all controls compared with either test alone (0.951 [0.929-0.973] in the training cohort; 0.954 [0.945-0.962] in the validation cohort).

Conclusions

PIVKA-II could complement measurement of AFP in the diagnostic of HBV-related HCC and distinguish HCC from non-malignant chronic liver disease.

Clinical trial identification

NCT03047603.

Legal entity responsible for the study

Eastern Hepatobiliary Surgery Hospital, Second Military Medical University.

Funding

The National Natural Science Foundation of China.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

85P - Dissecting Gastric Cancer biology and how and when to use immunotherapy

Presentation Number
85P
Lecture Time
12:30 - 12:30
Speakers
  • Meghna Das Thakur (South San Francisco, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Gastric cancer (GC), is a leading cause of cancer-related death, is a heterogeneous disease where survival depends on factors such as biological differences, MSI status, EBV status, region, ethnicity and patterns of care. The biological context for immune responsiveness and resistance in the clinical are only now starting emerge. Although GC has similar levels of PDL1 IC and TC expression to lung cancer, the approvals and success of immunotherapy in GC to date has been mixed.

Methods

We explored 2 randomized MetMAb trials (Ph3 Study 1 YO28322 and Ph2 Study 2 YO28252) in combination with mFOLFOX6 in metastatic HER2-negative and MET-positive GC. We used Study 1 (n = 146) to uncover novel GC biology using Nanostring based gene expression analyses and Study 2 (n = 70) to confirm the findings. Late Stage GC (Stage IV biopsies) GC from the Study 1 and 2 were compared against early stage GC (Stage I, II and III) in resections from the ACRG GC dataset.

Results

Retrospective analysis revealed key biological differences between early and late stage GC. In 1L mGC with stage IV tissue, unbiased prognostic analyses uncovered genes that grouped into: Immune/Effector T cell genes (Teff), Stromal genes and Differentiation/Proliferation genes as significantly associated with OS. Patients with high Teff genes had poor prognosis (Low/High HR: 0.43, p- value 0.01) and the worst prognosis was seen when both Teff and stromal genes were high. Furthermore, we confirmed these findings in Study 2. We found that EMT, Notch and TGFb pathways interacted with the Teff genes and were all associated with the poor Teff prognosis. Importantly immune/Teff genes have good prognosis in early stage GC and this is largely driven by MSI-H patients. Finally, when assessing what genes changed going from early to late stage, EMT, Notch, Wnt genes played a role in the transition to a more aggressive and metastatic disease.

Conclusions

Although GC has been challenging to treat, it may be possible to increase the success of immunotherapy with carefully tailored combination therapies in the Stage IV setting with molecules that inhibit pathways such as Notch, Wnt and TGFb. Furthermore, it may make a lot of sense to take immunotherapies into Stage I, II and III GC where the immune and Teff gene prognosis is good and the disease is less convoluted.

Clinical trial identification

YO28322: NCT01662869 YO28252: NCT01590719.

Legal entity responsible for the study

Hoffmann-La Roche.

Funding

Hoffmann-La Roche, Genentech.

Disclosure

M. Das Thakur, K. Okrah, D. Shames, P. Hegde, C. Bais: Employee: Hoffmann-La Roche.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

86P - Longitudinal Assessment of Multiplex Patient-Specific ctDNA Biomarkers in Bladder Cancer for Diagnosis, Surveillance, and Recurrence

Presentation Number
86P
Lecture Time
12:30 - 12:30
Speakers
  • Karin Birkenkamp-Demtröder (Aarhus, DK)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The use of circulating tumor DNA (ctDNA) as a biomarker for disease staging at diagnosis (DX), treatment response, and recurrence monitoring is an emerging field in many cancer types. In bladder cancer, the utility of ctDNA has shown promising results. Here we present a highly sensitive and specific NGS-based approach to ctDNA monitoring.

Methods

A cohort of 50 patients with locally advanced muscle-invasive bladder cancer treated with neoadjuvant chemotherapy were included prospectively. For each patient, a panel of 16 tumor-specific mutations was designed (SignateraTM RUO) based on whole-exome sequencing of tumor and germline DNA. In total, we analyzed ctDNA from longitudinally collected plasma samples from 386 time points procured at diagnosis, during treatment, at cystectomy (Cx), and during monitoring until disease recurrence or up to 2 years follow-up. Results of ctDNA analyses were compared to radiographic imaging and clinical outcomes. ctDNA from longitudinally-collected urine samples will also be analyzed for treatment response and disease recurrence.

Results

At DX, plasma ctDNA status was strongly prognostic of recurrence-free survival. Specifically, 62% (8/13) of the ctDNA+ patients at DX recurred after neoadjuvant treatment and Cx; conversely, none (0/22) of the ctDNA- patients recurred (log-rank; p < 0.0001). In addition, a strong correlation was also observed between presence of ctDNA after CX and disease relapse. Specifically, relapse after Cx was detected in 100% (10/10) of ctDNA+ patients ∼120 days (0–245 days) prior to radiographic imaging, while 0% (0/38) of ctDNA- patients relapsed (log-rank; p < 0.0001).

Conclusions

We demonstrate a strong prognostic potential of ctDNA in bladder cancer at time of DX, suggesting a potential role for ctDNA in the staging of bladder cancer. Furthermore, we show ctDNA is detected in all patients with disease recurrence after Cx. Incorporation of ctDNA analysis into routine follow-up for early detection of relapse may allow earlier initiation of alternate treatment modalities.

Legal entity responsible for the study

The National Committee on Health Research Ethics (#1302183), Denmark.

Funding

Novo Nordisk Foundation, Danish Cancer Society, Natera Inc San Carlos USA.

Disclosure

H. Sethi, S. Sharma, H-T. Wu, R. Swenerton, R. Salari, D. Hafez, R. Srinivasan, M. Balcioglu, S. Navarro, Z. Assaf, B. Zimmermann, J. Lin: Employee, stockownership or options to stock: Natera, Inc. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

87P - Distinct functional consequences of HER2 gene amplification in colorectal and lung adenocarcinomas

Presentation Number
87P
Lecture Time
12:30 - 12:30
Speakers
  • Evgeny N. Imyanitov (Saint-Petersburg, RU)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

HER2 oncogene amplification, being accompanied by its overexpression, is an established driver event in breast and gastric carcinomas. The functional role of HER2 in pathogenesis of other tumor types is less defined.

Methods

2401 archival samples of lung carcinomas (LC) and 1969 samples of colorectal carcinomas (CRC) were subjected to HER2 copy number analysis. Selected tumors with amplification of this oncogene were further subjected to HER2 immunohistochemistry and mRNA quantitation. In addition, the expression levels of some neighbouring genes located in 17q12-21 amplicon were analyzed.

Results

Frequency of HER2 amplification was similar in both groups, being 100/2401 (4.2%) in LC and 84/1969 (4.3%) in CRC, respectively. 10 (82%) out of 12 analyzed HER2-amplified CRCs demonstrated clear evidence for HER2 protein and mRNA overexpression, while this estimate approached to only 3 (27%) out of 11 for LCs. Expression analysis of GRB7, STARD3, and LASP1 revealed a statistically significant correlation between HER2 and STARD3 levels [r = 0.571, Spearman test]. High STARD3 expression was observed in HER2-amplified CRCs but not LCs [p = 0.03].

Conclusions

HER2 amplification is frequently accompanied by gene overexpression in colorectal but not lung adenocarcinomas. STARD3 gene belonging to 17q12-21 amplicon demonstrates evidence for activation in HER2-amplified colorectal neoplasms and therefore deserves further analysis.

Legal entity responsible for the study

Evgeny Imyanitov.

Funding

Russian Science Foundation.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

88P - Role of AR-V7 and AR-FL in resistance to hormonal therapy in mCRPC: independent actors or reciprocal drivers? A translational study by Meet-Uro group

Presentation Number
88P
Lecture Time
12:30 - 12:30
Speakers
  • Marzia Del Re (Pisa, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The androgen receptor splice variant 7 (AR-V7) is strongly associated with resistance to hormonal therapy (HT) in castration-resistant prostate cancer (CRPC), although it is not implemented in clinical practice as a biomarker. The AR-full length (AR-FL) is also overexpressed in CRPC but its role has yet to be clarified. The aim of the present work was to investigate the role of AR-V7 and AR-FL as predictors of resistance to HT in plasma-derived exosomal RNA.

Methods

6 ml of blood were collected in EDTA tubes before the start of abiraterone/enzalutamide; blood was centrifuged and plasma stored at -80 °C until analysis. Exosomes isolation and RNA extraction were performed using the exoRNeasy kit (Qiagen) as per manufacturer instructions. The analysis of AR-FL and AR-V7 were performed by digital droplet PCR using the One-Step RT-ddPCR kit (BioRad). The absolute target concentration as copies/ml in samples was calculated by ddPCR QuantaSoft and statistical analyses were performed by SPSS v.24.

Results

52 patients (pts) were enrolled; AR-FL was detected in all pts (median: 700 copies/ml), while 15 subjects (28.8%) were AR-V7 + (median: 310 copies/ml) at baseline. The amount of AR-FL was significantly higher in pts AR-V7+ vs AR-V7- (6700 vs 490 copies/ml, p < 0.0001). Median PFS and OS were longer in AR-V7- vs AR-V7+ pts (median PFS 25 vs 4 mo, p < 0.0001; median OS 38 vs 9 mo, p < 0.0001). A ROC curve was calculated for AR-FL in the overall population and 950 copies/ml was identified as cut-off value. Pts were then stratified across this value and it was found that PFS was 22 mo in pts with <950 AR-FL copies/ml vs 4 mo in pts with ≥950 copies/ml (p = 0.0003). In 12/15 AR-V7+ pts the AR-FL expression was ≥950 copies/ml while in 3/15 AR-V7+ pts, AR-FL expression was <950 copies/ml; however, their PFS reflected the AR-V7 better than AR-FL status, being, respectively 6, 10, 4 mo. No other clinical variables were correlated with worse PFS at the univariate analysis (i.e. Gleason score ≤7 vs > 7, age).

Conclusions

This study demonstrates that resistance to HT may be predicted by AR-V7, making it a clinically relevant biomarker. AR-FL over-expression may contribute to hormone resistance although AR-V7 plays a primary role.

Legal entity responsible for the study

Romano Danesi.

Funding

University of Pisa.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

89P - Validation of a 90-gene assay for tissue origin diagnosis of brain metastases

Presentation Number
89P
Lecture Time
12:30 - 12:30
Speakers
  • Yulong Zheng (Hangzhou, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Brain metastases (BM) are the most common intracranial tumors affecting about 8-10% of all cancer patients. Morphology and immunohistochemical staining are two common approaches used to identify the primary sites of BM samples, but morphology fails to identify poorly differentiated tumors and IHC markers usually lack specificity. About 2% to 14% of BM patients still present with unknown primary sites. A 90-gene assay, proposed in our previous study, is an RNA-based gene expression test to identify the tissue of origin in poorly differentiated and undifferentiated tumors. This study aims to evaluate the performance of the 90-gene assay in determining the primary sites for BM samples.

Methods

The sequence-based gene expression profiles of 708 primary brain tumors (PBT) collected from The Cancer Genome Atlas database were performed by a 90-gene expression signature, with a similarity score for each of 21 tumor types. We used Optimal Binning algorithm to generate a threshold for separating PBT from BM. Eighteen PBT samples from Fudan University Shanghai Cancer Center were analyzed to substantiate reliability of the threshold. In addition, the performance of the 90-gene assay for identifying the tissue of origin was validated in a cohort of 48 BM samples with known origin from The First Affiliated Hospital, Zhejiang University. For each BM sample, the tumor type with the highest similarity score was considered tissue of origin. When a sample was diagnosed as PBT but the similarity score below the threshold, the second prediction was considered as primary site.

Results

A threshold of the similarity score, 70, was identified to discriminate PBT from BM (PBT: ≥ 70, BM: < 70) with an accuracy of 99% (703/708). Eighteen PBT and 44 BM were performed by the 90-gene assay. The results of 18 PBT samples matched reference diagnosis with a concordance rate of 100% and all similarity scores were above 70. Of 44 BM samples, the 90-gene assay accurately predicted primary sites in 89% (39/44, 95%CI: 0.75-0.96) of the cases.

Conclusions

The 90-gene assay showed promising discriminatory ability to separate PBT from BM and identify the primary site of BM. Our findings demonstrated the potential that 90-gene assay can serve as a powerful tool for accurately identifying the tissue of origin for BM samples.

Legal entity responsible for the study

Yulong Zheng.

Funding

Has not received any funding.

Disclosure

Y. Sun, C. Chen, L. Chen, J. Zhu: Employment: Canhelp Genomics Co. Ltd. Q. Xu: Employment, stock ownership: Canhelp Genomics Co. Ltd. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

90P - Development of a Pan-Cancer Biomarker Panel for Improved Detection of MSI Across all Cancer Types

Presentation Number
90P
Lecture Time
12:30 - 12:30
Speakers
  • Jeff Bacher (Madison, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

A new multiplexed biomarker panel is being developed for detection of microsatellite instability (MSI) that is more sensitive than currently available systems. Preliminary research data shows increased MSI sensitivity for colon polyps and endometrial (EC), skin and prostate cancers. The sensitivity of this Pan-Cancer MSI System is being further verified on 14 different cancer types.

Methods

Selection of the new microsatellite biomarkers was done by screening 160 patients ≤55 years with ≥1 polyp and 100 EC patients ≤ 50 years for MSI. The expanded study uses samples from 100 Lynch syndrome colorectal cancers (CRC), 100 sporadic MSI-High CRC, 100 sporadic MSI stable CRC and 219 extra-colonic cancers obtained from the Colon Cancer Family Registry. DNA samples are being tested for MSI using two pan-cancer systems: Promega’s MSI Analysis System version 1.2 and the improved prototype Pan-Cancer MSI System. Mutations in mismatch repair (MMR) and BRAF genes were tested, as well as MMR expression by IHC.

Results

2.3% of colon polyps were MSI-High for the MSI Analysis System compared to 5.4% with the new prototype Pan-Cancer MSI System. Sensitivity and specificity of the new biomarker panel for detection of MMR deficient lesions was 100% and 96%. Similarly, sensitivity of the new biomarker panel for EC was about 2-fold higher. Allele size changes for MSI-High samples were significantly larger with the new biomarkers making MSI classification highly accurate and robost. The MSI and IHC results were highly correlated. Evaluation of the new biomarker panel is being performed on over 500 cancer samples from 14 different cancer types.

Conclusions

Research results indicate that MSI sensitivity for colonic polyps and many extra-colonic cancers can be increased by at least 2-fold over current MSI systems using the new MSI biomarker panel. The improved sensitivity of the Pan-Cancer MSI System should improve detection of MSI in an expanded number of cancer types and facilitate identification of individuals with both sporadic and hereditary MSI-High cancers.

Legal entity responsible for the study

Promega Corporation.

Funding

Promega Corporation.

Disclosure

J. Bacher: Employee: Promega Corporation. R. Halberg, P. Ward, K. Murphy, J. Eshleman: Corporate sponsored research funds: Promega. E. Udho, M. Uhr: Employee: Promega. D. Storts: Employee and stock ownership: Promega. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

91P - bMSI better predicts the responses to immune checkpoint inhibitors (ICI) than MMR/MSI from historical tissue specimens in metastatic gastrointestinal cancer patients

Presentation Number
91P
Lecture Time
12:30 - 12:30
Speakers
  • Zhenghang Wang (Beijing, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Microsatellite instability (MSI) has been approved as the first pan-cancer biomarker in immune checkpoint inhibitors (ICI) therapies. The tumor tissues of most metastatic cancer patients receiving ICI therapies are usually unavailable. However, polymerase chain reaction (PCR) or immunohistochemistry (IHC), the two conventional MSI evaluation methods, could only be applied to the tumor tissues. Hence, we aimed to develop a next-generation sequencing based method to detect MSI from blood circulating tumor DNA (bMSI).

Methods

A training cohort of 40 metastatic cancers patients before first-line treatments were collected to train a linear-based detection model. Then, a validation cohort of 47 metastatic gastrointestinal cancer patients before ICI therapies were collected. The prediction to the responses of ICI by bMSI was compared with that by the mismatch repair (MMR) or MSI from historical tissue specimens.

Results

bMSI showed 87.5% accuracy to predict the MMR/MSI status from tissue specimens in the training cohort, and 95.2% sensitivity in the validation cohort. bMSI-H patients had 31.4% objective response rate (ORR) and 45.7% disease control rate (DCR), which were comparable to the dMMR of historical FFPE specimens (33.3% and 47.6% respectively). However, 57.7% pMMR patients were classified as bMSI-H and showed similar ORR (27%) , DCR (40%) and progress free survival to those of dMMR patients. Furthermore, 17% bMSI-H patients with high bMSI scores (larger or equal to 28) showed 66.7% ORR and 100% DCR. Finally, 91.7% patients with controlled diseases over 6 months showed decreasing bMSI scores, and 60% patients with progressive diseases showed increasing bMSI scores during therapies.

Conclusions

A significant proportion of pMMR metastatic gastrointestinal cancer patients could be rescued by bMSI and get benefits from ICI. bMSI could further classify the patients to three groups and more precisely predict the response of ICI. The level of bMSI is dynamically related to the response during the therapies. bMSI could potentially improve clinical practices in the future.

Legal entity responsible for the study

Peking University Cancer Hospital & Institute.

Funding

National Key Research and Development Program of China (No. 2016YFC0905302).

Disclosure

H. Qin, D. Wang, S. Cai, Y. Bai, Z. Xie: Employee: 3D Medicines Inc. L. Xiong: Employee, stock holder, chairman: 3D Medicines Inc. F. Li: Employee and stock holder: 3D Medicines Inc. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

92P - Vall d’Hebron Institute of Oncology (VHIO) Immuno-Oncology prognostic index (VIO): a new tool for improved patient (pt) selection in Phase 1 (Ph1) trials with Immune Checkpoint Inhibitors (ICI).

Presentation Number
92P
Lecture Time
12:30 - 12:30
Speakers
  • Cinta Hierro (Barcelona, ES)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The Royal Marsden Hospital score (RMHs) (albumin <35 g/L, lactate dehydrogenase [LDH]>upper limit of normal [ULN], and >two sites of metastases [met]) is a validated prognostic index for Ph1 pt selection. Recently, a lung immune prognostic index (LIPI) (derived neutrophil/(leukocytes minus neutrophils) ratio [dNLR]>3, and LDH>ULN) proved to be useful for identifying pts with different outcomes under ICI. We aimed to improve pt selection for ICI Ph1 trials by developing a composite VIO that included all clinical-laboratory (CL) variables linked with worse median Overall Survival (mOS).

Methods

Retrospective analysis of pts treated with ICI at VHIO Ph1 Unit from Jan’12 to Oct’17. VIO includes four CL factors previously described (albumin<35g/L, LDH>ULN, >two sites of met, dNLR>3) and a fifth variable (liver met) as per univariate Cox modeling. The following VIO clusters were defined based on Kaplan Meier OS estimates: low risk (0 and 1), intermediate risk (2 and 3) and high risk (4 and 5).

Results

In total, 174 out of 214 pts (81%) treated with ICI (antiPD1/PDL1 ICI in 93%, combination regimens in 53%) had complete CL data for modeling. Most common tumor types were melanoma (22%) and lung (14%). Overall, best response was PD 47%, SD 38%, PR 12%, CR 2% and mOS 9.8 (95% CI 7.3-12.7) months (m). Concordance index of OS models including LIPI, RMHs or VIO scores were 0.62, 0.66 and 0.69, respectively. Estimated mOS in low risk (40.2% of all pts), intermediate risk (50.3%) and high risk (9.2%) were 22.0 m (10.5.4-33.4), 6.7 m (4.1-9.3) and 3.8 m (2.5-5.1), respectively (log rank test, p < 0.001). PD as best response was higher in high risk VIO group (81%) as compared to intermediate (50%) and low risk (34%, Chi-square p = 0.002). 6m OS rates were 85% (77%-94%), 55% (44%-67%) and 30% (14%-65%) in low, intermediate and high risk VIO groups (log rank test, p < 0.001).

Conclusions

Our results suggest that the VIO is a better predictor of OS on ICI in Ph1 trials as compared to existing prognostic scores. The VIO is a helpful tool for identifying Ph1 candidates unlikely to benefit from ICI and with higher chances of death within 6 m of trial recruitment.

Legal entity responsible for the study

VHIO.

Funding

Has not received any funding.

Disclosure

A. Oaknin: Advisory boards: Roche, AstraZeneca, PharmaMar, Clovis Oncology, Tesaro; Travel or accommodation support: Roche, AstraZeneca, PharmaMar. J. Tabernero: Advisory Boards: Bayer, Boehringer Ingelheim, Genentech/Roche, Lilly, MSD, Merck Serono, Merrimack, Novartis, Peptomyc, Roche, Sanofi, Symphogen, Taiho. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

93P - Augmenting TNM Staging with Machine Learning-based Immune Profiling for Improved Prognosis Prediction in Muscle-Invasive Bladder Cancer Patients

Presentation Number
93P
Lecture Time
12:30 - 12:30
Speakers
  • Nicolas Brieu (Munich, DE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Muscle-invasive bladder cancer (MIBC) is a highly aggressive disease whose clinical reporting is based on TNM staging. A more accurate and personalized prognosis could be achieved by profiling the immune contexture alongside clinical TNM staging. This study reports on features captured from the labelling of cells for CD3, CD8 and PD-L1 expression, within both the tumor and stroma of MIBC patients. This data is distilled to identify a novel immune-based prognostic signature which augments TNM staging.

Methods

An image analysis solution was developed to quantify the density of cell populations across immunofluorescence (IF) labelled whole slide images from 105 MIBC patients with known survival data. A regression random forest was used for the detection of nuclei using the Hoechst channel [Brieu et al., ISBI2017] and a convolutional neural network employed for the segmentation of the tumor from the stroma using the pan-cytokeratin channel [Brieu et al., SPIE2018]. The CD3, CD8 and PD-L1 channels were used to classify the cells and calculate their proportion within either the tumor or stroma regions. A decision tree of depth two was trained on these proportions as well as cross-validated. More explicitly, the patients were recursively partitioned during training to maximize at each node their survival difference, leaves showing low survival difference (p-value>0.5) being finally merged.

Results

The method yielded a decision tree which stratified patients into three groups utilizing only two parameters: the proportion of CD8(+) cells in the stroma and the proportion of PD-L1(+) cells across the whole tissue, which were negatively and positively correlated with cancer-specific death respectively. This method was used to replace TNM stages 1 to 3 while retaining the original stage 4 stratification. Testing for survival curve differences showed that this combined system yielded a higher prognostic value (Chisq=41.5, p-value=5.0x10-9) than the standalone TNM staging system (Chisq=34.9, p-value=1.25x10-7).

Conclusions

Our results suggest that immune profiling derived from image analysis provides additional prognostic value to TNM scoring for MIBC.

Legal entity responsible for the study

University of St Andrews.

Funding

Definiens AG.

Disclosure

N. Brieu, G. Schmidt: Fully employed: Definiens AG, a subsidary of Medimmune and Astrazeneca. C.G. Gavriel: PhD student at the University of St. Andrews; PhD work is funded by Definiens AG. D.J. Harrison: Professor of pathology at the University of St Andrews; Secondary PhD supervisor of C. Gavriel, funded by Definiens AG. P.D. Caie: Senior research fellow at the University of St Andrews; Primary PhD supervisor of C. Gavriel, funded by Definiens AG.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

94P - HPV circulating tumor DNA as predictive biomarker of sustained response to chemotherapy in advanced anal carcinoma

Presentation Number
94P
Lecture Time
12:30 - 12:30
Speakers
  • Alice Bernard-Tessier (Villejuif, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The Epitopes-HPV02 single arm phase II study (NCT02402842) demonstrated the efficacy of Docetaxel, Cisplatin and 5FU as first line chemotherapy (CT) for advanced squamous cell carcinoma of the anal canal (SCCA), with a 1-year progression-free survival rate (PFS) of 47% (Kim, Lancet Oncol 2018 in press). We previously reported the validity of HPV ctDNA detection and its prognostic value in localized SCCA (Cabel, Clin Cancer Res 2018 in press). This ancillary study reports the impact of HPV ctDNA detection in patients enrolled in the Epitopes-HPV02 trial.

Methods

Per protocol, serum samples (1 ml) were collected twice: before CT and, in non-progressive patients, at CT discontinuation which occurred after 5 months on CT. HPV16 ctDNA was quantified by ddPCR at both time points and correlated with prospectively registered patient characteristics and outcomes; for post-CT survival analyses, a landmark was set at the time of CT discontinuation.

Results

Among 59 patients with HPV16+ advanced SCCA, 52 (88%) had ctDNA detected at baseline (sensitivity: 91.1%; 95%CI[81.1;96.2]) with a median level of 7,148 copies/ml (range: 8.3- 3,147,000). Baseline ctDNA levels were not associated with any of the patient characteristics; ctDNA level below the median at baseline was correlated with a longer PFS (HR = 2.6, 95%CI[1.1;5.9], p = 0.02). Among 38 patients who completed the 5 months CT, residual ctDNA after CT was detected in 14 patients (36.8%; 95%CI[23.4;52.7]) with a median level of 2,662 copies/ml (range: 31.5–211,950). Residual ctDNA detected at CT discontinuation was strongly associated with shorter post-CT PFS (measured from the time of CT discontinuation, median PFS: 5.4 months vs not reached; HR = 6.2, 95%CI[2.3;16.3], p < 0.001) and OS (HR = 9.6, 95%CI[2.0;46.9], p = 0.005).

Conclusions

In this prospective study in advanced SCCA, we observed a strong prognostic impact of HPV ctDNA before 1st line CT and, in non-progressive patients, after 6 months of DCF. With a limited cost and short turnaround, this quantitative assay is a promising tool to optimize the therapeutic management of SCCA with e.g. maintenance anti PD-1/PD-L1 therapy in ctDNA-positive patients after the completion of first line CT.

Clinical trial identification

NCT02402842.

Legal entity responsible for the study

Institut Curie.

Funding

Fondation ARC.

Disclosure

T. André: Honoraria and advisory role: Roche, BMS, Merck & Co. E. Samalin: Honoraria: Sanofi, Bayer, Servier, Roche; Study support: Merck, Bayer; Congress support: Roche, Ipsen. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

95P - Predicting toxicity and response to pembrolizumab (P) through germline genomic HLA class I analysis

Presentation Number
95P
Lecture Time
12:30 - 12:30
Speakers
  • Marco Iafolla (Hamilton, CA)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

HLA class I-dependent immune activity is linked to autoimmune diseases, and HLA class I-dependent CD8+ T cells are required for immune checkpoint blockade (ICB) anti-tumor activity. It is unknown if HLA class I is predictive of toxicity to ICB.

Methods

100 patients (pts) with mixed solid tumors received single agent P (anti-PD-1) 200 mg IV Q 3 weeks in the investigator-initiated Phase II trial (INSPIRE study, NCT02644369). Germline whole exome sequencing (WES) of peripheral blood mononuclear cells was analyzed using the Illumina HiSeq2500 platform. Consensus HLA class I alleles were predicted from WES using HLAminer and HLAVBSeq. Using univariate Fisher’s exact test and logistic models adjusting for HLA features, heterozygosity of HLA-A, -B and -C, individual HLA alleles and HLA haplotype dimorphism at positions −21M and -21T of the HLA-A and -B leader sequence were analyzed as predictors of: 1) toxicity defined as ≥ Gr 2 immune-related adverse events (irAE) with at least possible attribution to P; and 2) clinical benefit (CBR) defined as either partial response or stable disease lasting ≥ 6 cycles of P.

Results

In the overall cohort of 100 pts, the frequency of irAE and CBR from P was 21% and 25%, respectively. Thus far, 99 patients had their HLA class I genotype determined. Univariate analysis showed heterozygosity of HLA-A, -B and -C, compared to homozygosity of at least one HLA locus, was not predictive of toxicity (≥ Gr 2 irAE 16.7% vs 27.3%, p = 0.29) but did trend to less response (CBR 19.7% vs 36.4%, p = 0.088). Individual heterozygosity of HLA-A, -B or -C, and HLA-A and -B haplotype dimorphism was not predictive of either toxicity or response. HLA-A*02 allele showed a trend to toxicity (≥ Gr 2 irAE 26.8% vs 11.6%, p = 0.079). A pertinent exploratory toxicity model is summarized in the table.

Toxicity
A*01 and A*02 (n = 7)4(57%)
A*01 and no A*02 (n = 13)2(15%)
A*02 and no A*01 (n = 49)11(22%)
Neither A*01 nor A*02 (n = 30)3(10%)

Table Fisher exact p-value=0.0503

Conclusions

This study is the first to assess the association between HLA class I genotype and toxicity to P. There is a possible association of HLA-A*01 and HLA-A*02 with toxicity to P.

Clinical trial identification

Trial protocol number: NCT02644369.

Legal entity responsible for the study

Lillian Siu.

Funding

Merck.

Disclosure

P. Bedard: Research funding: Bristol-Myers Squibb, Sanofi, AstraZeneca, Genentech/Roche, Servier, GlaxoSmithKline, Novartis, SignalChem, PTC Therapeutics, Nektar, Merck, Seattle Genetics. A. Spreafico: Support for clinical trials: Merck, Novartis, BMS. A. Razak: Research funding: Merck, BMS, Pfizer, Karyopharm, Deciphera, Eli Lilly, Boehringer, Boston Biomedicals, Roche, Novartis and Genentech; Consultancy: Eli Lilly, Eisai, Boeringher Ingelheim, Merck. P. Ohashi: Consultancy/advisory: Symphogen Inc., Providence Pharmaceuticals, Inc., Baxalta US. Inc., Lion Biotechnologies, Inc. T. Pugh: Honoraria: Merck, Prosigna, Chrysalis Biomedical Advisors; Consulting: DynaCare Research. Funding: Boehringer Ingelheim; Patents, royalties; other intellectual property: Hybrid-capture sequencing for determining immune cell clonality, combined hybrid-capture DNA sequencing for disease detection. L.L. Siu: Advisory board and funding for clinical trials: Merck. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

96P - AXL has a prognostic role in metastatic colorectal cancer (mCRC) and is a predictive biomarker of lack of efficacy of chemotherapy (CT) + cetuximab in RAS wild type (WT) patients (pts)

Presentation Number
96P
Lecture Time
12:30 - 12:30
Speakers
  • Claudia Cardone (Napoli, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

AXL expression promotes tumour growth, angiogenesis, epithelial to mesenchymal transition (EMT), resistance to CT and targeted agents. AXL is overexpressed in CRC. We aimed to evaluate AXL expression in mCRC pts and to correlate it with clinical outcomes.

Methods

AXL expression was assessed by immunohistochemistry in tumor samples of a consecutive series of 109 mCRC pts (75 RAS mutant and 34 RAS WT) treated at our Institution and 68 mCRC RAS WT pts enrolled in CAPRI-GOIM trial. Pts received a first line treatment according to RAS status: RAS mutant pts (n = 75) received CT + anti-angiogenic drugs, RAS WT pts (n = 102) CT + cetuximab.

Results

AXL stained positively in 20/177 samples with different intensity: 13 weak, 5 moderate, 2 intense. In RAS WT cohort 9/102 cases (9%) were positive while in RAS mutant 11/75 (15%). Tumor stroma was assessable in 166 samples. AXL expression was high (moderate + intense) in 47/96 (49%) RAS WT and in 28/70 (40%) RAS mutant cases. No significant correlation was found between AXL expression and clinico-patological features. In RAS WT cohort, AXL positive pts had a significantly worse median PFS [4.3 m (CI95% 3.2-5.5) vs 12.1 m (CI95% 11.0-13.3) p = 0.001], in RAS mutant no impact on PFS was observed. AXL expression in tumor was a negative prognostic factor in both cohorts although statistical significance was reached only in RAS mutant [median OS: 30.2 m (CI95% 18.4-42.0) vs 20.1 m (CI95% 10.6-29.6) p = 0.007]. Intriguingly, high AXL expression in stroma correlated with lower median OS in both cohorts (Table).

CohortMedian OS (months - CI95%) AXL expression in tumor
Median PFS (months - CI95%) AXL expression in tumor
Median OS (months - CI95%) AXL expression in stroma
Median PFS (months - CI95%) AXL expression in stroma
N (tumor) / N (stroma)AXL positiveAXL negativeAXL positiveAXL negativeAXL highAXL lowAXL highAXL low
Overall population N = 177 / N = 16620.1 (12.8-27.4)36.5 (30.6-42.3 ) p = 0.02--25.3 (21.4-29.3)46.4 (34.6-58.2) p = 0.003--
RAS WT (CT + cetuximab) N = 102 / N = 9623.0 (0.0- 63.3)39.8 (30.2– 49.4) p = 0.664.3 (3.2- 5.5)12.1 (11.0– 13.3) p = 0.00128.8 (17.4- 40.1)47.7 (29.7– 65.7) p = 0.02110.7 (8.4- 13.0)12.4 (9.6– 15.2) p = 0.06
RAS mutant (CT + anti-angiogenic) N = 75 / N = 7020.1 (10.6- 29.6)30.2 (18.4- 42.0) p = 0.0078.9 (5.4- 12.4)9.1 (7.6- 10.7) p = 0.44424.2 (18.2- 30.1)37.7 (16.8- 58.6) p = 0.0268.9 (6.1- 11.8)8.6 (7.3- 10.0) p = 0.53

Conclusions

AXL, marker of EMT phenotype, might represent an additional predictive biomarker of lack of efficacy in RAS WT mCRC pts treated with CT + cetuximab. Moreover, its expression in tumor and stroma might have a negative prognostic relevance in mCRC. Targeting AXL could overcome resistance to anti-epidermal growth factor receptor and represent a novel therapeutic strategy in mCRC.

Clinical trial identification

CAPRI-GOIM Trial = EudraCT 2009-014041-81.

Legal entity responsible for the study

Department of Precision Medicine, Università degli Studi della Campania "Luigi Vanvitelli".

Funding

Grant by AIRC = MFAG-2015-ID: 7778.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

97P - Multimodal detection of homologous recombination repair gene mutations (HRRm) in a Phase II trial of olaparib plus abiraterone in metastatic castrate resistant prostate cancer (mCRPC)

Presentation Number
97P
Lecture Time
12:30 - 12:30
Speakers
  • Thomas H. Carr (Cambridge, GB)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Study 8 [NCT01972217] was a randomized Phase II trial that tested the hypothesis that the combination of PARP inhibition plus abiraterone benefits unselected patients (pts) with mCRPC. The primary endpoint was progression-free survival. A key secondary objective was to understand the relationship between HRRm status and outcome, which was challenged by low-tissue acquisition and high-test failure. The primary biomarker analysis focused on testing plasma when tumour data were not available and germline mutations were not evident. Here we describe additional analyses of circulating tumour DNA (ctDNA) to further characterize HRRm status and evaluate concordance between different testing modalities.

Methods

Tumour specimens were sequenced via Foundation Medicine. Germline analysis was performed via Color Genomics. An inhouse (RUO) sequencing assay was used for baseline ctDNA analysis. A subset of plasma samples was analyzed via GuardantOMNI™ and a custom assay (Resolution Bioscience). CtDNA libraries were also subjected to shallow whole genome sequencing (∼4–5x). From 142 enrolled pts, we obtained HRRm data for 136 (from any source).

Results

Previous tumour/germline analyses identified 8 HRRm pts: 1 somatic, 7 germline (tumour success rate 38/68 [56%]; germline success rate 102/102). CtDNA analyses yielded a success rate of 93% (127/136 pts with plasma analyzed), with tumour variants detectable with high confidence in 79% (100/127). Plasma sequencing identified additional HRRm pts, including homozygous deletions, approximately tripling the number known to have a HRRm. Plasma testing in pts with tumour data revealed high concordance between tumour and ctDNA.

Conclusions

Comprehensive, sensitive sequencing of ctDNA for HRRm is feasible in mCRPC pts with a high success rate. Both targeted and whole genome approaches add value. There was good concordance across testing modalities where gene coverage overlapped, highlighting the considerable value of ctDNA testing in mCRPC where access to tissue of sufficient quality for molecular analysis is challenging, and somatic alterations are common. The first two authors contributed equally.

Clinical trial identification

NCT01972217.

Legal entity responsible for the study

AstraZeneca.

Funding

AstraZeneca.

Editorial Acknowledgement

n/a

Disclosure

T.H. Carr, C. Adelman, A. Barnicle, I. Kozarewa, S. Luke, Z. Lai, S. Menon, B. Dougherty, E.A. Harrington, J.C. Barrett, C. Goessl, D. Hodgson: Employee and stockholder: AstraZeneca. S. Hollis: Contracted: AstraZeneca and own stock. F. Saad: Grants and personal fees: AstraZeneca, Janssen, Astellas, Sanofi, Bayer. N. Sala: Personal fees: Astellas, Janssen, Bristol-Myers Squibb. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

98P - Identification of a highly suppressive Treg subset associated to immunotherapy response

Presentation Number
98P
Lecture Time
12:30 - 12:30
Speakers
  • Emilio F. Giunta (Napoli, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Cancer immunotherapy, particularly monoclonal antibodies against immune checkpoint inhibitors, has shown surprising efficacy in several types of advanced incurable tumors, including malignant melanoma. Tregs, a subset of lymphocytes involved in immune-surveillance and self-tolerance, are usually increased in melanoma patients. Lymphocytes are particularly rich in FKBP51, the intracellular receptor for FK506 and rapamycin. Melanoma aberrantly expresses this immunophilin, which supports cancer resistance and invasion. Recently, our group has shown that melanoma interaction with immune cells, through PD-L1/PD1, generated the splicing of FKBP5 gene inducing a lower molecular weight form (FKBP51s), in both melanoma and lymphocyte. Aim of this study is to assess the role of Treg FKBP51s+ as potential biomarker of response to anti-PD1 drugs.

Methods

Treg FKBP51s+ were measured in peripheral blood by flow cytometry. To date, we have outcomes of 11 patients. For 6 patients, we have collected from 4 up to 16 blood samples, before each anti-PD1 administration, with a total of 80 sample analysis. iTregs were generated by purified CD4+ T lymphocytes from normal donor, stimulated with CD3+CD28+beads. The suppressive capacity was assessed according to the parameters CD25high, Ki67high and p70S6khigh.

Results

In 5 responder patients, Treg FKBP51s+ was 1.2-4.8%; in 5 non-responders, the count was 0.04-0.8%. Interestingly, a patient with count 0.72% developed autoimmune side effects that led to drug discontinuation. Resolution of side effects was accompanied by an increase in Treg FKBP51s+ value to 9.9%. In vitro iTreg generation suggested that FKBP51s was induced in Treg CD25high Ki67high p70S6khigh, corresponding to a highly metabolically active profile associated with strong suppressive capability. Use of a siRNA for FKBP51s silencing resulted in reduction of this subtype of iTreg.

Conclusions

Our data reinforce the hypothesis that melanoma patients that benefit from immunotherapy are recognizable by an expansion of a Treg subset which plays a central role in tumor immune evasion. This Treg subset is marked by FKBP51s, a splicing protein isoform generated by triggering of surface antigens (PD-L1, PD1) abundantly expressed on highly suppressive Tregs.

Legal entity responsible for the study

Università degli Studi della Campania Luigi Vanvitelli.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

99P - A Novel Framework for Evaluating Biomarker Response Relationships in Immuno-oncology (IO)

Presentation Number
99P
Lecture Time
12:30 - 12:30
Speakers
  • Jeffrey L. Evelhoch (New York, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Unlike biomarkers of dichotomous genetic mutations/fusions required for response, biomarkers for checkpoint inhibitors are continuous biologic variables with context specific cutpoints. Selecting the cutpoint of a continuous biomarker for higher response rate in a given therapy decreases the number of biomarker positive patients (prevalence). To facilitate interpretation of biomarkers in IO, we introduce a framework for understanding how cutpoints, response rate and prevalence are interrelated.

Methods

Objective response rate (ORR) in biomarker positive patients is the product of ORR in all patients and the fraction of responding patients who are biomarker positive (FR+), divided by the prevalence. FR+ depends on the difference in biomarker distributions between responding and nonresponding patients. Biomarker [PD-L1 IHC, tumor mutation burden (TMB), T-cell activated gene expression profile (GEP)] and response data were pooled from 595 patients in 7 clinical trials of pembrolizumab monotherapy across 16 tumor types. A Bayesian model was used to estimate biomarker distributions in responders and nonresponders for each biomarker assuming normality.

Results

ORR prevalence data generated by varying the cutpoint were fit well by the biomarker distribution model for all 3 biomarkers. Individual biomarker ORR prevalence curves and 95% credible intervals overlapped substantially with each other, consistent with indistinguishable areas under the receiver operating characteristics curve (AUROC) for PD-L1, TMB and GEP in this pan tumor population. Thus, although PD-L1 or GEP identify populations only partially overlapping with that of TMB, the predictive ability is similar for all 3 biomarkers.

Value (95% CI)
BiomarkerORR @ 60% PrevalenceORR @ 30% PrevalenceORR @ 10% PrevalenceAUROC
PD-L115.5 (12.0, 19.0)21.5 (15.8, 27.4)33.6 (22.3, 45.7)0.69 (0.63, 0.76)
GEP17.5 (14.1, 21.2)24.8 (19.2, 30.1)33.5 (24.2, 44.1)0.76 (0.70, 0.82)
TMB15.0 (11.7, 18.6)21.5 (15.7, 27.7)35.9 (24.0, 49.0)0.67 (0.59, 0.74)

CI = Credible interval for ORR and confidence interval for AUROC

Conclusions

A model using biomarker distributions in responding and nonresponding patients accounts for the relationships among cutpoints, response rate and prevalence, and may provide a framework for interpretation of biomarker response data in IO.

Legal entity responsible for the study

Merck Sharp & Dohme, Corp, a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA.

Funding

Merck Sharp & Dohme, Corp, a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA.

Editorial Acknowledgement

Joanne Tomassini, Merck & Co., Inc., Kenilworth, NJ, USA.

Disclosure

J.L. Evelhoch: Employee and stock owner: Merck Sharp & Dohme Corp. R. Mogg, R. Cristescu, D. Aurora-Garg, C.H. Poehlein, A. Joe, S.M. Keefe, P. Kang, V. Karantza, J. Cheng, E.H. Rubin: Employee and stock owner/stock options: Merck Sharp & Dohme Corp. L.Q. Chow: Advisory board honoraria: Merck & Co., Inc., Sanofi-Genzyme; Institutional grant funding: Merck & Co., Inc., Bristol Myers Squibb, Lily/Imclone; Consulting honoraria: Amgen, Incyte, VentiRx; Institutional research grant and advisory board honoraria: Novartis, Genentech, AstraZeneca, Pfizer, Seattle Genetics. S. Loi: Grants: Merck Sharp & Dohme, Puma Biotechnology, Bristol-Myers Squibb, Novartis, Pfizer; Grants and non-financial support: Roche-Genentech. D.V.T. Catenacci: Consultant: Eli Lilly, Roche/Genentech, Amgen, Taiho, Five Prime Therapeutics; Speaker bureau: Eli Lilly; Institutional research funding: Amgen, Genentech. U.A. Matulonis: Funding: Merck & Co., Inc., Novartis; Consulting/advisor fees: Merck & Co., Inc., Lilly, Geneos, 2X Oncology, Myriad Genetics, Clovis Oncology, Fujifilm. P.A. Ott: Consulting or advisory role: Amgen, Bristol-Myers Squibb, Alexion, CytomX Therapeutics, Celldex, Genentech, Novartis, Pfizer, Neon Therapeutics; Speaking fees: Merck & Co., Inc.; Research funding: Bristol-Myers Squibb, Merck & Co., Inc., Astra Zeneca/MedImmune, Celldex, Neon Therapeutics, Pfizer, CytomX, ARMO BioSciences (to institution). E.S. Antonarakis: Consulting or advisory role: Sanofi, Dendreon, Medivation, Janssen Biotech, ESSA, Astellas Pharma, Merck & Co., Inc., AstraZeneca, Clovis Oncology; Travel fees: Sanofi, Dendreon, Medivation; Co-inventor of a biomarker technology that has been licensed to Qiagen; Honoraria: Sanofi, Dendreon, Medivation, Janssen Biotech, ESSA, Asstellas Pharma, Merck & Co., Inc., AstraZeneca, Clovis Oncology; Research funding: Janssen Biotech, Johnson & Johnson, Sanofi, Dendreon, Aragon Pharma, Exelixis, Millennium, Genentech, Novartis, Astellas Pharma, Tokai Pharmaceuticals, Merck & Co., Inc., AstraZeneca, Clovis Oncology, Constellation Pharmaceuticals (received by institution).

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

100P - Comparison of OncoBEAM and NGS methods to detect plasma EGFR T790M mutations at progression of NSCLC

Presentation Number
100P
Lecture Time
12:30 - 12:30
Speakers
  • Jessica Garcia (Lyon, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Various methods have been employed to detect plasma EGFR mutations in patients with non-small cell lung cancer patients (NSCLC). Therefore, we evaluated the performance of digital PCR and next generation sequencing (NGS) to detect the p.T790M EGFR mutation in prospectively collected patient samples. Paired plasma samples from patients (CIRCAN cohort) that progressed on first-line EGFR TKI therapy were compared using two platforms: OncoBEAMTM-EGFR (Sysmex Inostics) and NGS (Illumina), utilizing the 56G oncology panel (Swift Biosciences).

Methods

196 stage 4 NSCLC patients with EGFR alteration under TKI were included from various center. Blood was collected in a routine setting, when physician noted changes in CT scans that were suspicious of progression. CfDNA analysis is recommended in front line in this setting in France. Replicate plasma samples were analysed using OncoBEAM and NGS. The thresholds for calling EGFR plasma mutations were 0.5% and 0.02% for NGS and OncoBEAM, respectively and were validated using cfDNA reference standards (Horizon Discovery).

Results

OncoBEAM detected the p.T790M mutation in 36/196 patients (18.3%), whereas NGS detected T790M in 20/196 patients (10.2%). The agreement of NGS vs OncoBEAM for T790M detection was 55.6%. The p.T790M-positive samples detected by OncoBEAM but missed by NGS were all found to have low mutant allelic fractions (under 0.35%). With regard to sensitizing EGFR mutations, 28/36 OncoBEAM T790M+ patients had accompanying EGFR mutations, whereas all 20/20 NGS T790M+ samples showed presence of sensitizing mutations. In contrast to OncoBEAM, NGS testing revealed other somatic alterations including ERBB2 amplification, and mutations in TP53.

Conclusions

In conclusion, these findings highlight the value of OncoBEAMTM-EGFR and NGS for detecting T790M at early progression. While less sensitive, NGS provided broader genomic coverage which may reveal diverse mechanisms of resistance. In contrast, OncoBEAM delivers superior sensitivity for focused detection of known resistance alterations such as EGFR T790M. Thus, OncoBEAM may provide the sensitivity required to monitor the kinetics of circulating tumor DNA and correlations with therapeutic response.

Legal entity responsible for the study

Hospices Civils of Lyon.

Funding

Sysmex Innostics / AstraZeneca.

Disclosure

C. Tissot, P-J. Souquet: Membership on pharma advisory board. F.S. Jones, D. Edelstein: Sysmex Innostics. S. Couraud: AstraZeneca, Roche, MSD, Novartis, BMS, Boehringher Ingelheim, Chugai, Pfizer, Lilly, Merck. L.F. Payen: Membership on BMS; Advisory board: AstraZeneca. Z. Xu: Sophia Genetics. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

101P - Differential expression of PD-L1 and immune biomarkers by age: Decreased expression in pediatric/AYA patients with advanced cancer

Presentation Number
101P
Lecture Time
12:30 - 12:30
Speakers
  • Omid Hamid (Los Angeles, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The activity of immune checkpoint inhibitors (ICIs) varies substantially at the extremes of age. We interrogated our tissue database (n = 1,467) to determine if expression of checkpoint molecules or variations in tumor mutational burden (TMB) could explain this phenomenon.

Methods

Whole transcriptomic sequencing (RNA-Seq; ∼200x106 reads/tumor) was performed across 1,467 unselected clinical cases (NantHealth; Culver City, CA), with breast, colon, lung and sarcoma reflecting the most common tumor types assessed. To reflect the extremes of age, patients age < 25 and ≥ 80 were compared to the remainder of the cohort. PD-L1 expression was compared across these age-based subsets, along with CTLA4, TIGIT, FOXP3, LAG3, OX40, TIM3 and IDO expression. Putative markers of ICI resistance (e.g, VEGF-A/B/C) were also explored. Tumor mutational burden (TMB; defined as exonic nonsynonymous mutations/megabase [muts/Mb]) was characterized in each subset.

Results

Median age of the cohort was 59 (range, 2-97). Of 1,467 patients, 84 and 65 were age < 25 and ≥ 80, respectively. In patients < 25, significantly lower PD-L1, CTLA4, FOXP3, OX40, LAG3 and TIGIT levels were observed (P < 0.001 for each). No significant differences in IDO, LAG3 or TIM3 were observed in this younger cohort. Older patients had no significant differences in checkpoint molecule expression; curiously, a nonsignificant trend towards increased expression of PD-L1, FOXP3 and LAG3 was observed in the small subset of patients age ≥ 85. No differences in TMB were observed by age. Expression and TMB in each decile of age will be reported.

Conclusions

In pediatric and adolescent and young adult (AYA) patients, lower expression of multiple immune checkpoint molecules may have implications for immune combinatorial strategies. An opposing trend was seen in octagenarians and nonagenarians in our cohort. A detailed further breakdown by histologic subtype will be presented.

Legal entity responsible for the study

Omid Hamid.

Funding

Has not received any funding.

Disclosure

O. Hamid: Consultant: Amgen, Novartis, Roche, BMS, Merck; Speaker: BMS, Genentech, Novartis, Amgen; Contracted research (for institution): AstraZeneca, BMS, Celldex, Genentech, Immunocore, Incyte, Merck, Merck-Serono, MedImmune, Novartis, Pfizer, Rinat, Roche. C. Szeto, S. Reddy: Employee and stockholder: NantOmics LLC. S.K. Pal: Consultant: Pfizer, Inc., Novartis, Aveo Pharmaceuticals, Inc., Myriad Genetics, Genentech, Inc., Exelixis, Bristol-Myers Squibb Company, Astellas Pharma, Inc.; Research/grant support: Medivation, Inc.; Honorarium: Novartis, Medivation, Inc., and Astellas Pharma, Inc. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

102P - Potential resistance mechanisms revealed primary resistance to crizotinib in ROS1+ non-small-cell lung cancer using next generation sequencing: A multicenter study

Presentation Number
102P
Lecture Time
12:30 - 12:30
Speakers
  • Quxia Zhang (Fuzhou, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Crizotinib have greatly improved the prognosis of ROS1+ lung adenocarcinoma. However, approximately 5% to 10% of patients with ROS1+ non-small-cell lung cancer (NSCLC) have primary resistance to crizotinib treatment. The underlying mechanism is unknown.

Methods

We screened 2617 patients with NSCLC for ROS1 fusion. Among them, 23 patients received crizotinib treatment, and a total of 20 patients with stage IIIb-IV ROS1+ NSCLC were undergoing tumor biopsies or blood withdrawing by the time of primary or acquiring resistance to crizotinib, in including 4 formalin-fixed paraffin-embedded (FFPE) samples, 13 serum samples and 3 serous effusions. We used targeted NGS to detect genes status of patients.

Results

Among 23 patients treated with crizotinib, 73.9% (17/23) developed acquired resistance, and 13.04% (3/23) had primary resistance. Using the specimens at the baseline, there was 1 (33.3%) patient with BCL2L11 loss (BIM deletion polymorphism), 1 (33.3%) patient with PTEN mutation, and 1 (33.3%) patient with KIT mutation. Median PFS was significantly shorter in patients with primary resistance than those with acquired resistance (2.3 vs. 14.5 months, P < 0.001).

Conclusions

BCL2L11 loss, PTEN mutation, and KIT mutation might contribute to molecular mechanisms of primary resistance to crizotinib in ROS1+ NSCLC. Further investigations are warranted to overcome these primary resistances.

Legal entity responsible for the study

Quxia Zhang.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

103P - Prevalence of CLDN18.2, HER2, and PD-L1 in Gastric Cancer Samples

Presentation Number
103P
Lecture Time
12:30 - 12:30
Speakers
  • Diarmuid Moran (Chicago, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

In gastric cancer (GC) there is a need for therapeutic targets/biomarkers beyond HER2 and PD-L1; Claudin 18.2 (CLDN18.2) is a promising target. In healthy tissue, CLDN18.2 is confined to gastric mucosa tight junctions; however, upon malignant transformation, perturbations in cell polarity lead to exposure of CLDN18.2 on the surface of GC cells. In a randomized clinical study (FAST; NCT01630083), patients with CLDN18.2-positive advanced GC and gastroesophageal junction (GEJ) cancers treated with EOX and zolbetuximab (an anti-CLDN18.2 monoclonal antibody) had prolonged survival compared with EOX alone. The CLDN18.2, HER2, and PD-L1 prevalence in global GC/GEJ tissue samples were assessed in this study.

Methods

FFPE GC/GEJ tissue samples were stained using antibodies against CLDN18.2, PD-L1, and HER2. IHC assays were run on an automated platform; HER2 amplification was determined by HER2 CISH. Stained samples were evaluated by a trained pathologist using established scoring criteria.

Results

A total of 298 GC/GEJ tissue samples (North America, n = 100; Asia, n = 100; Europe, n = 98) were assessed; 148 (50%) were histologically classified as intestinal, 123 (41%) diffuse, 18 (6%) mixed, and 9 (3%) other. In American samples, intestinal histology was the most prevalent; diffuse and intestinal were similar within Asian and European samples. Of the 286 evaluable samples, 30% (n = 86/286) were CLDN18.2high (moderate-to-strong CLDN18.2 membrane staining in ≥75% of tumor cells). CLDN18.2high prevalence ranged from 24% (n = 22/92) in Asian samples to 34% (n = 33/97) in American samples. CLDN18.2high prevalence was 30% (n = 35/115) in diffuse and 28% (n = 40/145) in intestinal subtypes. HER2+ and PD-L1+ (≥1% membrane-stained tumor cells) occurred in 10% (n = 29/291) and 37% (n = 107/289) of the evaluable samples, respectively. Of CLDN 18.2high samples with evaluable status for HER2, CLDN18.2 overlapped with HER2 in 12% (n = 10/83) of cases.

Conclusions

CLDN18.2 was found globally to be a high prevalence target in GC/GEJ cancer with limited overlap with HER2. In light of the clinical activity observed for zolbetuximab, CLDN18.2 may serve as a therapeutic target for a large subgroup of patients with GC/GEJ cancer.

Legal entity responsible for the study

Astellas Pharma, Inc.

Funding

Astellas Pharma, Inc.

Editorial Acknowledgement

Medical writing and editorial assistance provided by Amlan RayChaudhury, PhD (SuccinctChoice Medical Communications Chicago, IL).

Disclosure

D. Moran, A. Arozullah: Employee: Astellas. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

104P - Epigenetic markers in circulating cell-free DNA for detection of early stage colorectal cancer

Presentation Number
104P
Lecture Time
12:30 - 12:30
Speakers
  • Yanqun Liu (Singapore, SG)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The detection of early stage colorectal cancer (CRC) significantly improves chances of a cure and is a key factor in reducing CRC mortality rates. Colonoscopy is currently the gold standard for CRC diagnosis, but a somewhat troublesome and invasive procedure makes its acceptance not high in the general public as a screening tool. Epigenetic silencing of tumor-related genes by promoter methylation is common in CRC, but no biomarker has been proven to be individually of sufficient sensitivity or specificity in routine clinical practice. Objective: To identify tumor-derived methylated genes in the serum of stage IIA CRC and assessed their diagnostic potentials for early stage of colorectal cancer.

Methods

In this prospective study, DNA methylation levels were measured by quantitative methylation-specific PCR. Seven genes were screened in an exploratory set of case-control serum samples. Promising methylation markers were selected and verified in the serum of a test set compromising 60 stage IIA CRC and 60 age-gender-matched healthy controls. Receiver operating characteristic curve (ROC) was constructed for assessment of assay performance.

Results

Serum methylation levels of TAC1, EYA4 and SST were significantly higher in stage IIA patients as compared to healthy controls (all P < 0.001, Mann-Whitney U test). Area under the receiver operating curve (AUC) using serum methylation of TAC1 and EYA4 was 0.76 [95% confidence interval (CI), 0.68-0.85] and 0.73 (95% CI, 0.64-0.82), respectively. At a specificity of 85%, the assay sensitivity of TAC1 and EYA4 was 58.3% and 43.3%, respectively. Combination of serum methylation levels of EYA4 and SST improved the assay sensitivity to 52.5%. With TAC1 and SST being investigated in tumor DNA as well, we noticed that methylation of both genes in the serum DNA always mirrored that of tumor DNA, exhibiting 100% concordance.

Conclusions

Serum methylation levels of TAC1, EYA4 and SST might be useful for minimally invasive detection of early stage of colorectal cancer. Validation study in larger and independent cohorts is necessary.

Legal entity responsible for the study

Singapore General Hospital.

Funding

National Medical Research Council, Singapore.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

105P - Detection and clearance of RET variants in plasma cell free DNA (cfDNA) from patients (pts) treated with LOXO-292

Presentation Number
105P
Lecture Time
12:30 - 12:30
Speakers
  • Benjamin Besse (Villejuif, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

LOXO-292 is a novel, highly-selective, small molecule inhibitor of RET currently in clinical development (Phase 1, NCT03157128) for pts with advanced cancers harboring oncogenic RET alterations (e.g. non-small cell lung cancer [NSCLC], medullary thyroid cancer [MTC], papillary thyroid cancer [PTC], etc.). Here, we update data previously presented at ASCO 2018 on modulation of RET variant allele frequencies (AF) in plasma cfDNA with LOXO-292 therapy.

Methods

Blood was collected pretreatment, after 15 days of treatment, and at each restaging for cfDNA analysis by next-generation sequencing (NGS, Guardant).

Results

As of 4/2/18, 82 pts were enrolled (38 RET fusion NSCLC, 29 RET mutated MTC, 9 RET fusion PTC, 2 RET fusion pancreatic cancer and 4 others) to 8 dose cohorts (20mg QDà240mg BID), and 343 plasma samples were collected. Here we report on 65 pts with plasma NGS results available. Of 62 pts enrolled based on a RET variant detected in a tumor sample, concordant RET alterations were detected in 41 (66%) of the corresponding pre-treatment plasma samples, including 19/30 (63%) pts with RET-fusion NSCLC and 16/21 (76%) pts with RET-mutant MTC. Median AF was higher for MTC (7.03%) than NSCLC (0.51%). In RET alteration-negative pre-treatment samples, peak AF for other detected alterations was generally low (0.28% median), suggesting low tumor DNA shed into plasma. Of 34 pts with a detectable pre-treatment plasma RET alteration and day 15 plasma NGS, RET alteration AF decreased by a median of 96%, with complete clearance in 15 pts (44%). Day 15 plasma clearance was observed at multiple doses, and was more common in RET fusion-positive (67%) than RET-mutant (8%) pts. Data for additional pts will be updated at the time of presentation.

Conclusions

The rapid clearance of RET variants from plasma cfDNA on LOXO-292 supports its observed clinical activity across a range of doses, tumor types and RET alterations. NGS of plasma cfDNA can detect a range of targetable RET variants, though tumor genotyping remains critical if the initial plasma NGS is negative. Serial plasma genotyping warrants continued study as an early pharmacodynamic marker for novel targeted therapies.

Clinical trial identification

NCT03157128.

Legal entity responsible for the study

Loxo Oncology Inc.

Funding

Loxo Oncology Inc.

Disclosure

B. Besse: Grant funding: AstraZeneca, BMS, Boehringer Ingelheim, Lilly, Pfizer, Roche-Genentech, Sanofi-Aventis, Servier, Onxeo, OncoMed, Inivata, OSE Pharma, Loxo. A. Drilon: Honoraria: Foundation Medicine; Pfizer; Advisory board Honoraria from Roche/Genentech, Takeda/Ariad, Loxo Oncology, Ignyta. L.J. Wirth: Personal fees: Eisai Inc., Novartis. B.J. Solomon: Honoraria: Pfizer and Novartis; Consulting or advisory role: AstraZeneca, Roche, Merck, BMS, Novartis. E. Zhu: Employee: Loxo. K. Gordon: Employee and stockholder: Loxo Oncology. K. Ebata: Personal fees and other support: Loxo Oncology, Inc during the conduct of the study; Personal fees and other support: Loxo Oncology, Inc., outside the submitted work. B. Tuch: Employee and share holder: Loxo Oncology. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

106P - A prediction panel with DNA methylation biomarkers for lung adenocarcinoma

Presentation Number
106P
Lecture Time
12:30 - 12:30
Speakers
  • Nan Shen (Shanghai, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Lung adenocarcinoma accounts for more than 40% of lung cancer incidence. Thus, it is urgent to identify early-stage related markers. In this study, the effectiveness of CpG methylation on predicting lung adenocarcinoma was investigated.

Methods

In total, 1,170 patients with lung adenocarcinoma from four independent databases and one medical center were sorted by three phases. In the discovery phase, 338 lung adenocarcinomas and nonmalignant samples were collected from the GEO databases which used Illumina Infinium HumanMethylation27K BeadChip for the methylation analysis. The K-Means Clustering algorithm was used to select significant CpGs. In the training phase, recursive feature elimination was performed to evaluate the importance of selected CpGs to classification model. In the validation phase, four candidate CpGs were validated using two cohorts (n = 832 and n = 10). To explore the potential biological function of selected CpGs, GO enrichment analysis was performed using the Database for DAVID version 6.8.

Results

After the selection of CpGs by the K-Means Clustering algorithm, 62 CpGs showed great different methylation profiles between lung adenocarcinomas and adjacent nonmalignant lung tissue (p < 0.05). Among these selected CpGs, 95.16% were hypermethylated in the malignant samples comparing to only 4.84% were hypomethylated. With the evaluation of recursive feature elimination, four CpGs corresponding to HOXA9, KRTAP8-1, CCND1, and TULP2 were highlighted as candidate predictors in the training phase. The performance of these four candidate CpGs were validated in two validation cohorts (p < 0.01). These disparate hypermethylated genes were significantly enriched in GO biological processes including negative regulation of transcription from RNA polymerase II promoter, DNA-templated transcription, while the hypomethylated gene was obviously enriched with the terms including adenylate cyclase-activating G-protein coupled receptor signaling pathway. The direction of methylation did not affect the enrichments for Out-CpG sites.

Conclusions

A four-CpG-based signature, including HOXA9, KRTAP8-1, CCND1 and TULP2, is useful for the prediction of lung adenocarcinoma.

Legal entity responsible for the study

Xinhua Hospital, School of Medicine, Shanghai Jiao Tong University.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

107P - Novel genomic classifier for early stage colorectal cancer patients

Presentation Number
107P
Lecture Time
12:30 - 12:30
Speakers
  • Elisabeth Letellier (Esch sur Alzette, LU)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Identifying patients at risk of relapse in early colorectal cancer (CRC) stages is an unmet clinical need. Due to the limitations of clinicopathological variables in predicting individual risk of recurrence in CRC patients, genomic information has increasingly gained prominence as a potential method for patient stratification. We have previously shown that Myosin Vb (MYO5B) expression alone or in combination with the expression of its adapter protein RAB8A shows strong prognostic value in early CRC patients (PMID 29024942).

Methods

We are currently setting up a meta-analysis including multiple CRC datasets, which allows to further validate the prognostic value of the genomic classifier. Additionally, this meta-analysis will determine the predictive value of the classifier on chemotherapy efficiency. Furthermore, pre-analytical and analytical assays will assess the reproducibility, sensitivity and specificity of the biomarker. Finally, we will prospectively collect stage II CRC tumor samples to clinically validate our classifier.

Results

In the follow-up study, we have now validated the prognostic value of the classifier in independent datasets. By multivariate analysis, we show that the gene expression signature is independent of clinicopathological features currently used in the clinics (stage, grading, T3, MSI status among others). Importantly, the identified molecular classifier outperformed the other three molecular tests (Oncotype DX, Coloprint and Oncodefender) that are commercially available but not FDA approved for predicting patient relapse. We will report on the predictive value of our molecular classifier on chemotherapy efficiency. In addition, first results on the pre-analytical and analytical analysis of the classifier will be presented.

Conclusions

Altogether, MYO5B together with RAB8A might allow delineating a high-risk population in early CRC stages. This stratification could potentially help oncologists to choose the best treatment plan, especially for stage II patients, where adjuvant chemotherapy may not always lead to beneficial results, but still results in significant side-effects.

Legal entity responsible for the study

Molecular Disease Mechanisms Group.

Funding

Fonds National de la Recherche, Luxembourg.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

108P - Baseline Blood Immune Profiling to Predict Response to antiPD-1 in Patients with Advanced Non-Small Cell Lung Cancer

Presentation Number
108P
Lecture Time
12:30 - 12:30
Speakers
  • Emanuela Romano (Paris, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Immune checkpoint inhibitors have remarkably improved the natural history of patients (pts) with non-small cell lung cancer (NSCLC), with improved clinical responses and overall survival compared to standard therapy. However, over 80% of unselected NSCLC pts do not respond, highlighting the need of theranostic biomarker discovery. In a cohort of NCSLC pts treated with antiPD-1, we investigated blood immune parameters at baseline and pts characteristics as potential theranostic biomarkers.

Methods

Thirty-four pts with locally advanced/metastatic NSCLC received antiPD-1 therapy as ≥ 2 line treatment in a prospective biomarker study (NCT02866149). Peripheral blood mononuclear cells and plasma were analyzed at baseline by multiparametric flow cytometry and Luminex technology, respectively. Primary endpoint was to correlate cellular and soluble immune parameters with clinical outcome based on RECIST criteria.

Results

Baseline CD3+/CD14+ ratio was a robust predictive biomarker with pts achieving progression free survival (PFS) ≥4 months showing an average ratio of 1.91 vs 1.11 in pts with PFS <4 months (p = 0.003). Furthermore, we found a strong positive correlation between the proportion of HLA-DRhiCD14+ monocytes and the PFS (r = 0.471), with objective responders showing higher CD86 expression, suggesting an improved antigen-presenting capacity. In addition, pts with a PFS ≥4 months, displayed higher proportions of CCR7-CD45RA- effector memory CD8+T cells and regulatory CD4+T cells suggesting pre-existing adaptive immune responses. In line with previous reports, our results confirmed the association of baseline plasma albumin with clinical outcome, with levels ≥ 3.9 g/dL associated with improved PFS (p = 0.026), likely due to the lower plasma levels of the pro-inflammatory mediator IL-6. Finally, soluble CD40 ligand was elevated in pts with reduced PFS, probably in relation with an elevated platelet activation, further supported by a 2-fold increase in plasma concentration of baseline platelet-derived growth factors (PDGFs) in these pts.

Conclusions

Our study identifies promising, predictive, immune-related biomarkers in NSCLC pts treated with PD-1 blockade.

Clinical trial identification

NCT02866149.

Legal entity responsible for the study

Institut Curie.

Funding

Institut Curie.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

109P - Evaluation of clinicopathological and molecular criterias for screening of exonucleasic domain POLE (edPOLE) mutated patients in proficient Mismatch Repair (pMMR) colorectal and endometrial cancers.

Presentation Number
109P
Lecture Time
12:30 - 12:30
Speakers
  • Justine Cohen (Créteil, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Mismatch Repair Deficiency (dMMR) and edPOLE mutations (mt) are responsible for hypermutated tumoral phenotype. Immunotherapy have shown efficacy in dMMR/high mutation burden patients (pts) and could be also active in edPOLEmt tumors. These mt occur in 6-12% of endometrial cancers (EC) and in 1-2% of Colorectal Cancer (CRC) and are likely infrequent in advanced setting. We aimed to define the most relevant clinicopathogical and molecular criterias to facilitate screening for edPOLE mt pts.

Methods

EdPOLE mutational status was evaluated in cohorts of pMMR CRC and EC using High Resolution Melting PCR on the hotspots described in literature (codons 286, 411 and 459). CRC subcohorts were enriched for BRAF mt, RAS mt, unusual BRAF/RAS mt and young pts (≤40 yrs). Pts harboring a mutated profile were confirmed by Sanger sequencing.

Results

245 pts were screened: 49 EC and 196 CRC. Among CRC, 41 were BRAF mt (30 V600E, 11 non V600E), 79 were KRAS mt (30 on codon 12/13, 49 other mt), 20 were selected because of the presence of ≥ 2 simultaneous BRAF/KRAS/NRAS mt, 30 were BRAF/KRAS/NRAS wild type (wt) and 30 were ≤40 yrs. Using our method, 9 edPOLE mt tumors were identified (Table): 4 among EC and 5 among CRC. edPOLE mt EC were all endometrioid ADK. edPOLEmt CRC were all localized in the left colon or rectum with unusual molecular alterations: 1 BRAF (p.D594G), 3 KRAS mt (p.A59T p.A146T et p.N116H) and 1 with two NRAS mt (p.Q61R et p.T58A).

Tumor subcohort% of edPOLE mt (n/N)Clinical and molecular profile of edPOLE mt pts
Whole cohort3.6 (9/245)pMMR: 100% (9/9)
EC Endometrioid adenocarcinomas Other histologies8 (4/49) 13 (4/31) 0 (0/18)3 codon 286 POLE 0 codon 411 POLE 1 codon 459 POLE
CRC BRAF/KRAS/NRAS wt p.V600E BRAF mt Non p.V600E BRAF mt Codons 12 or 13 KRAS mt Non codons 12 or 13 KRAS mt Multiple BRAF/KRAS/NRAS mt2.6 (5/196) 0 (0/30) 0 (0/30) 9 (1/11) 0 (0/30) 6 (3/49) 5 (1/20)Left colon or rectum: 100% (5/5) Age ≤ 40 years old: 20% (1/5) 2 codon 286 POLE 1 codon 411 POLE 2 codon 459 POLE

Conclusions

Our screening strategy identified edPOLEmt in 13% of endometrioid ADK and 2.6% of CRC. Pts selection on clinicopathological (histology for EC, young age, left colon or rectum), and molecular criterias (pMMR, unusual BRAF/KRAS/NRAS mt) seem to increase the proportion of edPOLEmt. The use of these criteria in practice could help select patients for edPOLE screening. Additional clinicopathological and molecular data will be shown.

Legal entity responsible for the study

Anaïs Pujals.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

110P - Immune prognostic index (IPI) and Hyper-Progressive disease (HPD) in patients (pts) exposed to targeted agents (TAs) in Phase 1 trials (Ph1T): can lessons from immune checkpoint inhibitors (ICIs) be translated to other scenarios?

Presentation Number
110P
Lecture Time
12:30 - 12:30
Speakers
  • Ignacio Matos Garcia (Barcelona, ES)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

IPI score (derived neutrophil/(leukocytes minus neutrophils)ratio [dNLR]>3 plus LDH>upper limit normal) is a emerging tool to predict overall survival (OS) with ICIs in different tumor types. HPD has been reported in 15% of pts on ICIs using different criteria, including ours (PD at first restaging with ≥40% increase in sum of target lesions or ≥ 20% with appearance of multiple new lesions; Matos et al, ASCO 2018). We aimed to assess the prognostic value of IPI and the relevance of HPD in pts enrolled in Ph1T with TAs.

Methods

Retrospective analysis of a consecutive cohort of pts treated with experimental TAs at VHIO Ph1T Unit over the last 3 years (we excluded pts in the first dose escalation cohorts of each trial as well as TAs matched to validated biomarkers). Overall survival (OS) was correlated with VHIO HPD criteria and IPI.

Results

In total, 180 pts were treated with TAs (34% FGFR, 26% PI3K, 19% MET, 13% NOTCH, 5% IDH, 3% RAF). In 39% of the cases, TAs were matched to an emerging enrichment molecular alteration. Median age 59y, 58% female, most common tumor types: 20% colorectal, 17% breast, 9% gynecological, 8% biliary tract. Best response was PD in 55%, SD in 41%, PR in 4%. Our HPD criteria was met in 10% of the cases, across all tumor types and targets (highest prevalence in colorectal pts treated with PI3K inhibitors, 5/18, 28%). Median progression-free survival was 1.8 months (m) [95% CI 1.7-2.2], not affected target selected, molecular match or IPI score (p > 0.2). Median OS was 7.9 m [6.7-10], significantly different as per IPI score (IPI0 18 m [7.8-NA]; IPI1 7.7 m [6.5-9.9]; IPI2 2.6 m [1.7-NA]; p < 0.001). Importantly, in pts with PD as best response while on TAs, HPD did not negatively affect OS (p = 0.43).

Conclusions

Our results show that a prognostic score developed in cohorts treated with ICIs also predicts long-term outcome with TAs in Ph1T. HPD criteria can be met with TAs treatment, but the lack of survival impact (different from internal and external cohorts exposed to ICIs) suggests that it is not a relevant clinical finding.

Legal entity responsible for the study

Vall d´Hebron Institute Oncology.

Funding

Has not received any funding.

Disclosure

J. Tabernero: Advisory boards: Bayer, Boehringer Ingelheim, Genentech/Roche, Lilly, MSD, Merck Serono, Merrimack, Novartis, Peptomyc, Roche, Sanofi, Symphogen and Taiho. E. Garralda: Advisory role: Roche, NeoMed Therapeutics, Ellypses Pharma. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

111P - Combination of baseline LDH, performance status and age to identify solid tumor patients with higher probability of response to anti-PD1 and PDL1 monoclonal antibodies

Presentation Number
111P
Lecture Time
12:30 - 12:30
Speakers
  • Maria Silvia Cona (Milan, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The advent of immune check point inhibitors (ICIs) has improved prognosis of various cancers. To better select responding patients (pts) and for a more accurate management of economic resources, biological and biochemical factors have been investigated. To date, no predictive biomarkers have been validated. The aim of this study is to identify manageable and routinely detectable parameters to use in clinical practice to select pts with higher change of response to ICIs.

Methods

271 consecutive metastatic solid tumor pts treated in our Institute from 2013 to 2017 with ICIs were evaluated for baseline LDH serum level, ECOG score, age, type of ICI, number of metastatic sites, histology and sex. A training and validation set were used to build and test models, respectively. Variables’ effects were assessed through odds ratio estimates (OR) and area under the receive operating characteristic curves (AUC), from univariate and multivariate logistic regression models. The validated estimates were used to develop an Excel algorithm to calculate probabilities of response.

Results

As best response, 55.4% of pts achieved disease control and 44.7% had progressive disease. On the training set, LDH, age and ECOG showed a significant OR (p:<.001, 0.009, 0.042, respectively) and were combined in a multivariate model with an AUC of 0.771 (95%CI: 0.701;0.842). These results were statistically validated on the validation set (AUC: 0.685, 95%CI: 0.569;0.801). By fitting the validated model on all pts, the 3 variables retained a significant OR and a satisfactory cross-validated AUC.

Conclusions

We confirm, as reported in literature, that baseline LDH serum levels are inversely associated with response probability. It’s reasonable to jointly consider age and ECOG, which give a significant contribution to model performance. The developed algorithm, once validated on an independent prospective series, might be a base to guide physicians in clinical practice to better plan ICI therapy tailored on pts characteristics.

Legal entity responsible for the study

Fondazione IRCCS Istituto Nazionale Tumori.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

112P - Effects of concomitant genetic alterations on cancer patient overall survival

Presentation Number
112P
Lecture Time
12:30 - 12:30
Speakers
  • Yu Wang (Ji'nan, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Genomic alterations have a profound impact on all aspects of cancer biology and therapy. Concomitant mutations involving multiple genes are intrinsically complex, and their relationship with biomarkers such as TMB and clinical outcomes such as drug response is only beginning to be explored. In this study, we systematically analyzed the impact of pairwise co-mutations of common genes on patient overall survival (OS).

Methods

We downloaded the publicly available TCGA datasets and used the LUAD lung adenocarcinoma subset for our pilot study. To limit noise, we restricted our attention to the top 48 most commonly mutated genes among LUAD patients. For each pair of genes g1 and g2 in this list, we compared the survival data of three groups of patients: those with g1 mutations only, those with g2 mutations only, and those with both g1 and g2 mutations. Kaplan-Meyer survival curves were plotted and p-values computed.

Results

We obtained a large number of double mutants (223 out of 1128 possible pairs, ∼20%) with significantly different OS from single mutants. There was a wide spectrum of “co-mutation potential”: on one hand, genes such as AHNAK2 and ANK2 readily co-mutated with many other genes, all leading to double mutants with distinguishing OS; on the other hand, genes such as DMD or DNAH9 co-mutated with few or no other genes that led to distinguishing double mutants. In terms of OS, double mutants could be put into three broad categories: those better than either single mutant (“synthetic rescue”), those worse (“synthetic lethal”), and those in between (“averaging”). Surprisingly, many double mutants exhibited synthetic-rescue behaviors. For example, ANK2- and LRP1B-mutant patients had very similar OS, but double-mutants exhibited significantly better OS than either single mutant (p < 0.001 in each case).

Conclusions

In our proof-of-concept study we systematically explored the impact of co-mutations on OS. A large number of double mutants exhibited “synthetic rescue” behaviors, and we pinpointed many distinguishing gene pairs for further investigation. It remains to be seen whether co-mutations of the same pair of genes always have the same effect across cancer types, and how they interact with other bio- and clinical markers.

Legal entity responsible for the study

OrigiMed.

Funding

Has not received any funding.

Disclosure

Y-A. Dong: Employee: OrigiMed. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

113P - Circulating Exosomal Integrin _v_5 Predicts Liver Metastasis and Prognosis in Human Colorectal Cancer

Presentation Number
113P
Lecture Time
12:30 - 12:30
Speakers
  • Dake Chu (Xi'an, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Ever since the “seed-and-soil” hypothesis, the mechanism of cancer metastatic organotropism is still an unsolved mystery. In colorectal cancer (CRC), there is still no robust metastasis predictive biomarkers for distant organ metastasis, which is the most common cause of deaths. In spite of the function of exosome in RNA and protein delivery, its clinical significance in CRC metastasis remains uncertain. Here, we evaluated the potential role of serum exosome integrin in CRC metastasis.

Methods

Tissue integrin αvβ5 was quantified by quantitative reverse-transcription PCR in 31 pairs of primary CRC and corresponding matched liver metastasis (LM), with non-LM as control. Serum exosomal integrin αvβ5 was accessed by ELISA in 126 CRC patients with LM and 166 CRC patients without, as well as when LM was diagnosed in these 166 patients in exploratory cohort. In prospective validation cohort, serum exosomal integrin αvβ5 was investigated in 135 initially diagnosed CRC patients without metastasis. CRC-associated metastasis mouse models were established to verify the role of serum exosomal integrin αvβ5.

Results

Integrin αvβ5 level in LM was significantly increased compared with that in non-LM, which was correlated with its expression in primary CRC. Serum exosomal integrin αvβ5 was significantly increased in CRC patients with LM than those without, in a TNM stage-dependent manner. Moreover, it was found that serum exosomal integrin αvβ5 in CRC patients was significantly upregulated when LM occurred and associated with unfavorable survival. In validation cohort, increased serum exosomal integrin αvβ5 indicated higher risk of LM and unfavorable prognosis. Serum exosomal integrin αvβ5 was significantly increased in mice with LM compared with controls.

Conclusions

Our clinical and animal model data indicate that increased levels of serum integrin αvβ5 associate with CRC LM and unfavorable survival. These results suggested that circulating integrin αvβ5 could be a promising non-invasive predictor for CRC LM and prognosis.

Legal entity responsible for the study

Xi'an Jiaotong University.

Funding

NSFC.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

114P - Circulating and tumor-associated caspase-4: a novel diagnostic and prognostic biomarker for Non-Small Cell Lung Cancer patients?

Presentation Number
114P
Lecture Time
12:30 - 12:30
Speakers
  • Rosalinda Sorrentino (Fisciano, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Late diagnosis limits therapeutic options and survival rate of non-small cell lung cancer (NSCLC) patients. Therefore, the identification of biomarkers represents an emerging medical need.

Methods

A highly sensitive and specific ELISA test was developed to identify/quantify a novel/selective diagnostic biomarker for NSCLC patients, caspase-4, which was detected into the plasma and tissues of NSCLC patients. This test was validated by using plasma from 125 NSCLC patients and 79 healthy (non-pathological) subjects. Caspase-4 quantification was also assessed in the lung tumor mass of 98 paired-matched NSCLC patients compared to 10 non-tumor lung tissues (i.e. tuberculosis).

Results

Circulating caspase-4 was detected in both healthy and NSCLC patients; however, at different range values: 2.603-3.372 ng/ml for NSCLC patients (95% CI) compared to 0.3994-0.6219 ng/ml for healthy subjects (95% CI). The sensitivity of the test ranged from 97.07% to 100%; the specificity was 88.1% with a positive predictive value of 92.54%, accuracy of 95.19% and AUC of 0.971. Tissue levels of caspase-4 in the tumor mass showed that 72 (72.7%) out of 99 patients were positive. More importantly, higher levels (cut-off value= 0.307 ng/ml) of caspase-4 in the tumor mass were associated to reduced overall survival (median 0.92 years) compared to NSCLC patients with lower levels (median 3.02 years).

Conclusions

We report for the first time caspase-4 as a novel diagnostic and prognostic biomarker, opening new therapeutic perspectives for NSCLC patients.

Legal entity responsible for the study

ImmunePharma srl.

Funding

Invitalia-Italian Ministry of Economy (MISE).

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

115P - Smokers and COPD patients have high circulating caspase-4 levels: is it an alarm?

Presentation Number
115P
Lecture Time
12:30 - 12:30
Speakers
  • Michela Terlizzi (Fisciano (SA), IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Lung cancer is the leading cancer-related death disease worldwide. This alerting data is mainly due to late diagnosis of lung cancer, especially non-small cell-lung cancer (NSCLC), limiting the therapeutic options. Therefore, the identification of non-invasive, selective, sensitive and specific biomarker/s represents the emerging medical need for the clinical practice to avoid late diagnosis and ameliorate the personalized therapy with an ensuing higher survival rate. It is well known that smoking and chronic obstructive pulmonary disease represent two high risk factors for NSCLC. Therefore, in this study we aimed to evaluate the levels of a novel diagnostic tool for NSCLC patients in order to understand whether caspase-4 could represent a predictive biomarker.

Methods

In order to evaluate the circulating caspase-4 in the blood, we developed an ELISA test.

Results

Smokers (≥15 cigarettes/day) had higher levels of circulating caspase-4 (95% CI, 1.331-1.94 ng/ml) than healthy subjects (95% CI, 0.395-0.619 ng/ml). Though, these levels were statistically lower than those observed in NSCLC patients. Moreover, there were no statistical differences between the levels of circulating caspase-4 in smokers younger or older than 60 years. Similarly, no gender differences were noted. Similarly, COPD patients, who are smokers and former smokers, had higher levels of circulating caspase-4 (95% CI, 1.703-2.995 ng/ml). According to χ2 test, the expected frequency of smokers, positive to the circulating caspase-4, who could develop NSCLC is robustly significant (calculated χ2=82.884 vs tabulated χ2=3.845, df = 1). Moreover, according to the independence test, smoker who were positive to the circulating caspase-4 are at high risk to develop NSCLC.

Conclusions

In conclusion, we report for the first time that the circulating caspase-4 could represent a predictive diagnostic tool to avoid the occurrence of NSCLC.

Legal entity responsible for the study

ImmunePharma s.r.l.

Funding

Italian Ministry of Economy (MISE) Invitalia.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

116P - Genomic characteristics of standardized uptake value of 18F-fluorodeoxy-glucose positron emission tomography in breast cancer

Presentation Number
116P
Lecture Time
12:30 - 12:30
Speakers
  • Seon-Kyu Kim (Daejeon, KR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Standardized uptake value (SUV), an indicator of the glucose uptake degree in 18F-fluorodeoxy-glucose positron emission tomography (FDG-PET), has been used as a prognostic factor in various malignant tumors. The aim of this study is to identify a molecular signature reflecting prognostic SUV characteristics in breast cancer (BRC).

Methods

We sought to identify a molecular signature associated with SUV by a gene expression profiling using a dataset obtained from 60 BRCs who underwent preoperative FDG-PET. The prognostic value of the signature was verified in three BRC cohorts including TCGA dataset (n = 1,616). Various statistical methods, including log-rank and Cox regression analyses, were applied to estimate an association between the signature and BRC prognosis. To compare somatic variants between two patient subgroups divided by the signature, we obtained predefined gene sets involved in oncogenic or metabolic pathways and estimated a difference of their mutation frequencies between subgroups in the TCGA cohort.

Results

By a gene expression profiling, we defined a signature, namely SUV signature, consisting of 723 genes significantly associated with SUV (Pearson correlation test, |r| > 0.35, p < 0.001). The patient subgroups classified by the signature [i.e., SUV-high-cluster and SUV-low-cluster] were significantly similar with patient classification by SUV [Fisher exact test, odds ratio 8.02, 95% confidence interval (CI) = 2.45-29.3, p < 0.001]. When estimating prognostic value of the SUV signature in three cohorts, the signature showed a strong prediction ability (log-rank tests, each p < 0.05) and an independent clinical utility (multivariate Cox regression model, hazard ratio = 1.51, 95% CI = 1.07-2.22, p = 0.01) in BRC prognosis. Gene network and mutation analyses revealed that a signaling defined by TP53-FOXM1 and its downstream effectors involved in glycolysis-gluconeogenesis might be important mediators in FDG-PET operation.

Conclusions

Our results uncover genomic and metabolomic characteristics of glucose uptake captured by FDG-PET, supporting an understanding of glucose metabolism as well as a poor prognosis in BRC patients with high SUV.

Legal entity responsible for the study

Korea Research Institute of Bioscience and Biotechnology Gananam Severance Hospital.

Funding

Korea Research Institute of Bioscience and Biotechnology.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

117P - Characterization of PD-L1, CD8, CD3, CD68 and PanCK in Tumor Microenvironment of Gl Tract Tumors with respect to Patients’ Mismatch Repair Status and Anti-PD-1 Treatment Outcome using 5Plex IHC and Whole Slide Image Analysis

Presentation Number
117P
Lecture Time
12:30 - 12:30
Speakers
  • Wenjun Zhang (Tucson, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. However, not all patients with mismatch repair deficiency respond to the PD-1 blockade treatment. To understand the different responses, we evaluated tumor micro-environment such as PD-L1 expression in relationship with tumor infiltrating immune cells.

Methods

FL multiplex IHC for PD-L1, CD8, CD3, CD68, and pan-cytokeratin (panCK) on BenchMark ULTRA instrument stained 54 pre-pembrolizumab treatment patient resection and biopsy specimens including pancreatic, colorectal and cholangiocarcinoma. 17 are mismatch-repair proficient (13 PD, 3 SD, and 1 CR) and 37 deficient (5 PD, 10 SD, 9 CR, and 13 PR). Whole slide images by Zeiss AXIO Z1 were analyzed with in-house dPath automated algorithm. PD-L1+/- T-cells (CD3+), cytotoxic T-cells (CD3+ and CD8+), T helper cells (CD3+ and CD8-), macrophages (CD68+), and tumor cells (panCK+) were identified. Fractions of PD-L1+ phenotypes, area density and spatial relationships of phenotypes in pathologist-annotated tumor/peritumor regions, the panCK+ epithelial tumor and panCK- stroma were computed.

Results

Random forest modelling, logistic regression and Relieff feature selection followed by quadrant discriminant analysis were used to assess the relationship of multiplex IHC readouts to anti-PD-1 treatment responses. Different groupings were used, e.g. PR + CR vs. SD + PD, and PR + CR + SD vs. PD, etc. Fraction of PD-L1+ macrophages and fraction of PD-L1+ T helper cells in tumor region, stroma, and epithelial tumor were identified most important features. Mpx IHC achieves 89% accuracy over 70% with mismatch repair status alone.

Conclusions

Multiplex IHC together with automated image analysis provides a tool to evaluate multiple biomarkers and their special relationships in the tumor micro-environment. In a cohort of 54 patient specimens, exploratory analysis of multiplex IHC data suggests that the knowledge of PD-L1 expression on various immune cell phenotypes aids in better predicting response to anti-PD-1 therapy compared to mismatch repair status alone.

Legal entity responsible for the study

Roche Diagnostics/Ventana Medical Systems, Inc.

Funding

Roche Diagnostics/Ventana Medical Systems, Inc.

Disclosure

W. Zhang, M. Khojasteh, A. Hubbard, X. Wang, J.L. Munoz-Rodriguez, D. Jiang, Z. Cai, J. Li, L. Pestic-Dragovich, L. Tang: Employee: Roche Diagnostics. J. Martin, S. Kamthamraju: Contractor employee: Roche Diagnostics. R. Anders: Financial relationships: Merck, BMS, 5 Prime Therapeutics, FLX Bio, Adaptive Biotech. L. Diaz: Founder: Personal Genome Diagnostics, PapGene, PagerBox.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

118P - Exhaustion of Platelet Kinetics and its Implication in Post-resection HCC Recurrence

Presentation Number
118P
Lecture Time
12:30 - 12:30
Speakers
  • Bibek Aryal (Kagoshima, JP)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Platelet activation and the release of various growth factors from platelet granules in response to tumor cell profoundly influence tumor biology. Platelet hyperreactivity in cancer patients is a frequently discussed phenomenon; however, cancer patients were also found to have platelet dysfunction. Considerably less is known about exhaustion of platelet kinetics in cancer.

Methods

In this study, we investigated kinetics of intra-platelet growth factors in hepatocellular carcinoma (HCC) recurrence. We prospectively recruited forty patients diagnosed with HCC who were undergoing liver resection. The patients were followed every three months with imagings after the resection. Platelet fractions were separately isolated before and after a month of radical resection of the tumor. Growth factors/cytokines were measured using enzyme-linked immunosorbent assay (ELISA) kits. Follow-up was standardized to two years. HCC recurrence was diagnosed on the basis of imaging.

Results

Fifteen patients developed post-resection HCC recurrence during 2-year follow-up. The concentrations of platelet-alpha granules secreted growth factors [vascular endothelial growth factor-A (VEGF-A), angiopoietin-1(Ang-1)] and dense granules secreted growth factors (serotonin) were significantly depleted in patients with early HCC recurrence. In addition, the post-resection platelet count was also significantly lower in patients with recurrence than those without recurrence. A combined receiver-operating characteristic (ROC) curve generated to determine the cut-off values for these growth factors yielded both specificity and sensitivity of greater than 80%, area under curve (AUC) > 0.8, and P < 0.001. Furthermore, in the binary logistic regression model, VEGF-A was able to independently predict early HCC recurrence (P < 0.05); accordingly, the disease-free interval was substantially worse in accordance with the exhausted intra-platelet growth factors.

Conclusions

We found a qualitative and quantitative platelet crisis, and its prognostic implication in patients with post-resection HCC recurrence. This study provides an avenue to identify the pathophysiological mechanism of the impaired platelet-kinetics and its relevance in cancer biology.

Legal entity responsible for the study

Bibek Aryal.

Funding

Japan Society for the Promotion of Science (JSPS).

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

119P - Progastrin, a new blood biomarker for the diagnostic and therapeutic monitoring, in gastro-intestinal cancers: A BIG-RENAPE project.

Presentation Number
119P
Lecture Time
12:30 - 12:30
Speakers
  • Benoit You (Pierre Bénite, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Progastrin is abnormally released in the blood of patients with different cancers (colorectal, gastric, ovarian, breast, uterus, melanoma…), as the progastrin gene is a direct target of the WNT/ß-catenin oncogenic pathway involved in tumorigenesis of many organs (ASCO 2017, Prieur et al CCR 2017). The value of progastrin as a diagnostic, and as a therapeutic monitoring marker, was assessed in patients treated for peritoneal carcinomatosis from gastro-intestinal (GI) cancers in the prospective BIG-RENAPE study (NCT02823860).

Methods

Patients were enrolled during management of peritoneal carcinomatosis (before or after neo-adjuvant chemotherapy, or surgery) and then regularly sampled for blood. Progastrin was measured using the ELISA DECODE LAB test®. The diagnostic value of progastrin concentrations at inclusion in 190 GI cancer patients (test set) was assessed against 80 samples from French donors (control set with non-cancer subjects). The longitudinal therapeutic monitoring value of progastrin test was also investigated.

Results

The Area Under the ROC curve of prograstrin for cancer diagnosis was 0.87, 95% CI [0.83-092]. Progastrin was significantly elevated at inclusion in all GI tumor subtypes (p < 0.0001; median 3.08, 95% CI [1.15-7.23], including colo-rectal & small bowel (n = 151; median 2.78) and oeso-gastric (n = 33; median 4.75) carcinomas). During monitoring, progastrin levels decreased after neoadjuvant treatment (median value 2.20, n = 23), decrease that became significant after surgery (p < 0.0001, median value 1.57, n = 84), with patients going back to normal value and others not. A trend for better PFS was observed in patients with progastrin decline after surgery. Progastrin baseline value did not correlate with renal function.

Conclusions

Progastrin assay is a simple and inexpensive blood test exhibiting high diagnostic accuracy in patients with GI carcinomas, along with promising therapeutic longitudinal changes across sequential managements. Assessment of progastrin value as a multi-tumor screening assay, and as a monitoring test, is on-going.

Legal entity responsible for the study

BIG-RENAPE.

Funding

ECS-Screening.

Disclosure

A. Prieur, P. Liaud, M. Flaceliere, D. Joubert: Employee: ECS-Screening. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

120P - A new biomarker of breast cancer stage and patient response to neoadjuvant chemotherapy: HLA-DR expression in cytotoxic and regulatory T cells

Presentation Number
120P
Lecture Time
12:30 - 12:30
Speakers
  • Diana P. Saraiva (Lisbon, PT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Neoadjuvant chemotherapy (NACT) is the treatment option for locally advanced breast cancer (BC). However, approximately half of the patients have no response. To promptly direct non-responders to personalized therapies, there is an urgency to find a clinical biomarker that could predict treatment response. Tumor infiltrating lymphocytes, namely CD8+ T cells (CTLs) and regulatory T cells (Tregs) are being appointed as biomarkers of response. Nonetheless, tumor cells can escape the immune system by releasing cytokines or expressing immune checkpoint inhibitors, dampening CTLs and increasing Tregs activation. CTLs and Tregs with HLA-DR, a T cell activation marker, by reflecting the tumor immune status, should be a more reliable biomarker of NACT success.

Methods

Fresh biopsies, surgical specimens and blood were collected from 150 BC patients. Immunophenotype was performed by flow cytometry, ELISA and qRT-PCR. Peripheral blood mononuclear cells (PBMCs) were isolated and cultured under canonical stimuli.

Results

67.3% of BC analysed were ER+, 15.5% HER2+ and 17.2% triple negative. Prior to treatment and independent of BC type, BC with no metastasis in the lymph nodes (53%) have HLA-DRhi CTLs (p = 0.003) and HLA-DRlo Tregs (p = 0.002), although the average percentage of lymphocytes and myelocytes are similar between more or less advanced disease. Biopsies from NACT responders also have HLA-DRhi CTLs (p = 0.0006) and HLA-DRlo Tregs (p = 0.0002). A ROC curve revealed a threshold of HLA-DR in CTLs bellow which patients will not respond to NACT. Moreover, HLA-DR+ CTLs express IFN-g, Granzyme B, Perforin, Eomes and TNF-a, essential for CTLs cytolytic activity. HLA-DR+ CTLs negatively correlate with pro-tumorigenesis molecules, such as TGF-b, PD-L1, IL-6, IL-1b and IL-8 (p < 0.005); while HLA-DR+ Tregs positively correlate with them. HLA-DR expression in tumor T cells correlates with its level in systemic T cells (CTLs: r = 0.58 p = 0.001; Tregs: r = 0.65 p = 0.0002). PBMCs stimulated in vitro from NACT responders reveal higher IFN-g and lower IL-10 (p = 0.04).

Conclusions

We propose HLA-DR levels in T cells as a biomarker of BC stage and response to NACT, with the advantage of being systemically evaluated.

Legal entity responsible for the study

CEDOC, Nova Medical School.

Funding

Fundação para a Ciência e Tecnologia (FCT), iNOVA4Health.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

121P - LungBEAM: A prospective multicenter trial to monitor EGFR mutations using BEAMing technology in Stage IV NSCLC patients

Presentation Number
121P
Lecture Time
12:30 - 12:30
Speakers
  • Pilar Garrido Lopez (Madrid, ES)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Liquid biopsy is a promising approach to improve the management of NSCLC patients, as it offers a minimally-invasive alternative to tumor tissue testing and enables timely monitoring of patients on-therapy. The goal of the present study was to evaluate the clinical value of longitudinal testing of EGFR mutation status in plasma of tissue EGFR mutation-positive NSCLC patients during first-line EGFR TKI therapy across 19 Spanish hospitals to: 1) determine the timing of T790M mutation emergence and 2) monitor EGFR mutation levels in plasma during first-line EGFR TKI therapy with respect to radiological progression.

Methods

Blood samples from 109 therapy-naïve advanced NSCLC patients were collected at baseline and monthly throughout EGFR TKI standard therapy. Results from OncoBEAM EGFR mutation were performed by Sysmex in Hamburg, and compared to those obtained by the EGFR tissue testing obtained at the referring hospital. The times at which T790M were first detected in blood were compared to the date of progression as determined by radiological imaging in standard clinical practice.

Results

At baseline, the initial positive percent agreement (PPA) for EGFR mutation status in 78 out of 109 patients enrolled in this study was 71.6%. From a total of 60 patients out of 89 who completed the study showing either clinical or radiological progression, 20 patients (33.3%) showed presence of the T790M mutation in plasma during follow-up. In 13 of these patients plasma T790M-positivity was detected an average of 14 weeks prior to radiological progression. Furthermore, the clearance of EGFR mutations in plasma at 8 weeks after initiation of EGFR TKI was a favorable indicator for PFS (37.3 weeks with clearance vs 25.5 weeks in patients without clearance). Patients showing EGFR mutation clearance at 8 weeks had an average baseline MAF of 3.6%, whereas patients with detectable mutations at 8 weeks showed an average baseline value of 13% MAF.

Conclusions

Overall, these results show high PPA of plasma and tissue EGFR mutation status at baseline. Early EGFR mutation clearance may be predictive of response to first-line EGFR TKI therapy. Plasma detection of T790M mutation anticipates clinical progression.

Clinical trial identification

SYS-ONC-2015-01.

Legal entity responsible for the study

Sysmex Inostics GmbH.

Funding

Sysmex Inostics GmbH.

Editorial Acknowledgement

Not applicable

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

122P - Immune checkpoints and liver resection after neoadjuvant chemotherapy including bevacizumab in patients with colorectal liver metastases

Presentation Number
122P
Lecture Time
12:30 - 12:30
Speakers
  • Stefan Stremitzer (Vienna, AT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Liver resection after neoadjuvant chemotherapy including bevacizumab offers the possibility of cure in patients with colorectal liver metastases. The clinical value of immune checkpoint expression as prognostic biomarker is unclear.

Methods

Expression analyses of IDO-1, PD-L1 and CTLA-4 were performed by immunohistochemistry in resected colorectal liver metastases in patients who underwent liver resection after neoadjuvant chemotherapy including bevacizumab (2005-2011). Association of expression of immune checkpoints in tumor cells and immune cells with response, RFS and OS was investigated.

Results

One hundred forty-six patients were enrolled [88 (60.3%) male/58 (39.7%) female, median age 63.0 years (31.0-80.4)]. High expression of CTLA-4 in tumor cells was associated with shorter OS (median OS 48.2 months versus not reached, HR 2.04, P = 0.028). High expression of IDO-1 and PD-L1 in immune cells was associated with longer OS (not reached versus 47.1 months, HR 0.43, P = 0.016 and not reached versus 47.1 months, HR 0.41, P = 0.017). Results of IDO-1 remained significant in multivariable analysis (HR 0.29, P = 0.006). Low expression of CTLA-4 in tumor cells was associated with better histologic response (26 major, 19 partial, 18 none versus 14 major, 23 partial, 30 none, P = 0.032). No association of expression was found with RFS and radiologic response.

Conclusions

The clinical meaning of immune checkpoint expression and its association with response and survival were dependent on the expressing cell types. IDO-1 and CTLA-4 may be new prognostic and/or predictive biomarkers in patients with colorectal liver metastases. The role of immune checkpoint inhibitors in a multidisciplinary treatment approach remains to be elucidated.

Legal entity responsible for the study

Stefan Stremitzer.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

123P - Gene expression (GE)-based biomarkers associated with nivolumab response in a real-life cohort of patients with metastatic non-small cell lung cancer (mNSCLC)

Presentation Number
123P
Lecture Time
12:30 - 12:30
Speakers
  • Nina Radosevic-Robin (Clermont-Ferrand, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Robust and versatile biomarkers are urgently needed to identify cancer pts who are likely to benefit from immune checkpoint inhibitors (ICI). The Tumor Inflammation Signature (TIS) is an 18-gene signature measuring the suppressed adaptive immune response necessary for pts clinical benefit from ICI (Ayers, J Clin Invest 2017). This retrospective study evaluated the performance of TIS and additional GE signatures assessing pathways associated with immune evasion in mNSCLC pts, treated with nivolumab as per label (2nd+ line), in two French cancer centers.

Methods

RNA from primary FFPE tumor samples from 77 immunotherapy-naïve mNSCLC pts, treated with nivolumab monotherapy, was profiled with a β-version of the NanoString® IO 360 GE panel, which includes the TIS and other tumor and immune biology signatures. The statistical analysis treated associations of GE and clinical response.

Results

In the whole cohort analysis, samples from pts who experienced clinical benefit showed an “inflamed” phenotype. Specifically, TIS was significantly higher in the responder group compared to non-responders (p = 0.005, non-adjusted). A similar association was observed for myeloid and macrophage scores (p = 0.001 and p = 0.002, respectively) as well as for PDCD1 (PD1), CD274 (PDL1), and CTLA4 (p = 0.05, 0.001, 0.02, resp.). In a subtype analysis, squamous carcinomas showed a less inflamed phenotype than adenocarcinomas but had elevated proliferation, glycolysis and hypoxia scores, as well as increased ARG1 and NOS2. In contrast, in the independent subtype analysis, both TIS and additional immune signatures remained associated with clinical response.

Conclusions

TIS and other immune signatures predicted response to nivolumab single agent in a real-life cohort of pts, in both adeno- and squamous mNSCLC. Thus, assessing GE patterns can give insight into different mechanisms of immune evasion operating at the single patient level. Additional analyses of this cohort, evaluating the NanoString® IO 360 signatures and single gene expressions, with regard to various other clinical and molecular tumor features, will be presented.

Legal entity responsible for the study

NanoString Technologies, Inc.

Funding

NanoString Technologies, Inc.

Disclosure

S. Ong, S.E. Warren, P. Morel, A. Cesano: Employee and stockholder: NanoString. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

124P - Association of microRNA-21 (miR-21) with efficacy of cetuximab (cet) and bevacizumab (bev) in patients with metastatic colorectal cancer (mCRC) within the FIRE-3 study (AIO KRK-0306)

Presentation Number
124P
Lecture Time
12:30 - 12:30
Speakers
  • Lisa Miller-Phillips (Munich, DE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

FIRE-3 compared first-line therapy with FOLFIRI plus either cet or bev in KRAS exon 2 wildtype (wt) patients with mCRC. Identification of RAS mutations as a predictor for the efficacy of EGFR-antibody therapy raised the question of whether miR-21 as a potential regulator of the EGFR dependent pathway influences therapy outcome.

Methods

A reverse-transcription quantitative polymerase chain reaction (RT-qPCR) assay was used to identify quantitative miR-21 expression in formalin-fixed paraffin-embedded (FFPE) tumor samples of FIRE-3 patients. The median of miR-21 expression within the FIRE-3 population was determined and subsequently used to segment the population into low and high miR-21 expression groups. Overall response rate (ORR) between treatment groups was compared using Fisheŕs exact test. Median progression free (PFS) and overall survival (OS) were calculated using Kaplan-Meier estimation and compared through log-rank test.

Results

RAS wt population with low miR-21 expression (n = 146) showed a significantly higher ORR when cet instead of bev was added to FOLFIRI chemotherapy (80.0% vs. 57.9%; p = 0.005). High miR-21 expression in RAS wt population (n = 149) showed no significant difference in ORR between treatment groups (74.6% vs. 64.0%; p = 0.21). ORR of RAS mutated patients with high miR-21 expression (n = 48) showed no significant difference between cet or bev when added to FOLFIRI (38.1% vs. 59.3%; p = 0.24). ORR (50.0% vs. 48.3%; p > 0.99) also showed no significant difference in RAS mutated patients with low miR-21 expression (n = 59). The following table presents statistical results of PFS and OS by comparing FOLFIRI plus cet versus FOLFIRI plus bev depending on RAS and miR-21 status.

RAS wt
RAS mut
low miR-21 (n = 166)high miR-21 (n = 167)low miR-21 (n = 67)high miR-21 (n = 62)
PFS (months)10.6 vs. 10.3 p = 0.310.1 vs. 9.9 p = 0.58.5 vs. 12.2 p = 0.37.7 vs. 8.9 p = 0.044
OS (months)35.8 vs. 25.9 p = 0.00524.5 vs. 23.8 p = 0.420.2 vs. 26.0 p = 0.416.4 vs. 20.2 p = 0.2

Conclusions

Along with RAS status, miR-21 expression level may be a promising predictive biomarker for anti-EGFR-therapy.

Clinical trial identification

NCT00433927; Study Start Date: January 2007; First Posted: February 12, 2007.

Legal entity responsible for the study

University Hospital, LMU Munich.

Funding

Merck.

Disclosure

C. Vazart, K. Fontaine: Employee: Integragen. D.P. Modest: Honoraria and advisory boards: Amgen, Bayer, Merck, MSD, Roche, Servier, Taiho, BMS, Sirtex; Travel support/research funding: Amgen, Bayer, Merck, Roche, Servier. V. Heinemann: Advisory boards: Merck KgaA, Roche AG, Amgen, Sanofi, Lilly, Sirtex, Boehringer Ingelheim, Taiho, Servier; Honoraria for talks: Merck KgaA, Roche AG, Amgen, Sanofi, Sirtex, Servier, MSD; Travel expenses to meetings: Merck KgaA, Roche AG, Amgen, Sirtex, Servier, MSD, BMS; Funding of research activity: Merck KgaA, Pfizer, Amgen, Roche, Servier, Sirtex. S. Stintzing: Honoraria for talks and travel support: Amgen, Bayer, Roche, Merck KgaA, SANOFI, Lilly, Takeda. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

125P - Molecular heterogeneity assessment by NGS in non-small cell lung cancer (NSCLC) harboring EGFR mutations: results of the French Cooperative Thoracic Intergroup (IFCT) Biomarkers France study.

Presentation Number
125P
Lecture Time
12:30 - 12:30
Speakers
  • Hélène Blons (Paris, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

NSCLC is characterized by genome alterations that promote cancer cell growth. TO what extent co-mutations cooperate with the EGFR driver and explain the variable response to EGFR TKI is not well understood. Here, we screened additional co-mutations (ACMs) and evaluate their clinical impact in patients (pts) with advanced NSCLC.

Methods

We identified 204 pts with EGFR mutated tumors from the IFCT Biomarkers France study cohort with available tumor DNA who received first or second line first-generation EGFR-TKIs. Samples were assessed by NGS using the AmpliSeq Cancer Hotspot Panel v2. Among the 204 samples, 167 were fully contributive.

Results

EGFR mutations were: classical with del 19 (74; 44%) and L858R (59; 35%), complex (17; 10%) including T790M (9; 5%) or uncommon (17; 10%). EGFR was amplified in 27 (17%) samples. ACMs were identified in 120 (72%) samples with an average of mutations at 2.9 (2-8). Recurrent ACMs (more than 6 samples) were in TP53 (84; 50.3%), CTNNB1 (16; 10%), PI3KCA (15; 9%), RB (12; 7%), APC (10; 6%), PTEN (8; 5%) and ATM (7; 4.5%). EGFR complex mutations were more frequent in smokers (p = 0.01) whereas RB1 mutations were more frequent in non-smokers (p = 0.03). CTNNB1 mutations were mutually exclusive with TP53 (p = 0.01) or PI3KCA (p = 0.05) mutations. High EGFR variant allelic fraction was associated to EGFR amplification (p < 0.001) suggesting mutant allele amplification and to TP53 mutations (p = 0.003). Non-classical or complex EGFR mutations were linked to rapid (< 3 months) versus normal (3-20 months)/slow (> 20 months) progression (p = 0.07 and p = 0.05 respectively). In the non-T790M group, ATM and PTEN mutations were negative predictors of first-line TKI efficacy (mPFS 3.7 versus 8.9 months, HR 2.85, 95CI% 1.14-7.15 and mPFS 5.6 versus 9.0 months, HR 2.46, 95CI% 1.16-5.9, respectively).

Conclusions

EGFR mutated NSCLC have heterogeneous molecular profiles. This work suggests that PTEN and ATM mutations could limit EGFR inhibitor efficacy. However, large series of EGFR mutated NSCLC will be needed to validated links between clinical outcomes and specific EGFR altered pathways.

Legal entity responsible for the study

IFCT.

Funding

AstraZeneca: INCa (French NCI); IFCT (Alain Depierre Research Award 2015).

Disclosure

D. Debieuvre: Consulting: Roche; Honoraria: AstraZeneca, Chugaï, Lilly, Roche, Novartis, Pfizer, MSD, BMS; Grants: Roche, AstraZeneca, Lilly, BMS, Boehringer Ingelheim, Chiesi, Chugaï, Janssen, Pfizer, MSD, Novartis, GSK, Sandoz; Board: Roche, Boehringer Ingelheim, Pfizer, MSD, BMS, Novartis; Conferences: Roche, Boehringer Ingelheim, Novartis, Pierre Fabre, Pfizer, MundiPharma, BMS. C. Audigier Valette: Clinical trials (PI): AstraZeneca, Boehringer Ingelheim, BMS, Novartis, Roche; Consulting: AstraZeneca, Boehringer Ingelheim, BMS, Lilly, Novartis, MSD, Pfizer, Roche; Conferences: AstraZeneca, Boehringer Ingelheim, BMS, Lilly, Novartis, Pfizer, Roche. S. Moreau Fraboulet: Personal fees and non-financial support: Novartis; Non-financial support: Lilly. D. Moro-Sibilot: Personal fees: Roche, Eli Lilly, Pfizer, Novartis, AstraZeneca, BMS, MSD, Boehringer Ingelheim. A. Lemoine: AstraZeneca, Boehringer, Roche. F. Barlesi: Personal fees: AstraZeneca, Bristol-Myers Squibb, Boehringer Ingelheim, Clovis Oncology, Eli Lilly Oncology, F. Hoffmann–La Roche Ltd, Novartis, Merck, MSD, Pierre Fabre, Pfizer. J. Cadranel: Clinical research (PI): AbbVie, AstraZeneca, Bayer, Boehringer–Ingelheim, BMS, MSD, Novartis, Roche, Takeda; Consulting: AstraZeneca, Boehringer Ingelheim, BMS, Lilly, Novartis, MSD, Pfizer, Roche; Conferences: AstraZeneca, BMS, MSD, Pfizer. M. Beau-Faller: Board: BMS, Boehringer-Ingelheim, AstraZeneca; Non-financial support: AstraZeneca, Roche. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

126P - Simultaneous identification and profiling of tumor-specific T cells by mass cytometry

Presentation Number
126P
Lecture Time
12:30 - 12:30
Speakers
  • Michael Fehlings (Singapore, SG)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

immunoSCAPE leverages the high-dimensional immune profiling capabilities of mass cytometry combined with a unique technology for the identification of antigen-specific T-cells to support the development of immunotherapy strategies in cancer and immune-related diseases. In cancer, there is now strong evidence that immunotherapy-mediated tumor rejection can rely on tumor-specific (neoantigen-specific) CD8+ T-cells. Consequently, the discovery of neoantigens becomes valuable for personalized cancer immunotherapies. Although in silico pipelines exist, that are capable of predicting non-synonymous mutations potentially giving rise to tumor-specific neoantigens, it is not clear how accurate these methods are in identifying immunogenic and therapeutically relevant epitopes, since T-cell epitope usage can be influenced by many factors. Moreover, analysis of T-cells in cancer patients is challenging as it requires detecting rare antigen-specific T-cell populations in samples that are usually limited in volume and availability.

Methods

By applying cytometry by time of flight in conjunction with combinatorial peptide-MHC tetramer staining and high-performance dimensional analysis tools, we are able to map broadly MHC-class I epitope with a high sensitivity for rare antigen-specific T-cells and perform concurrently in-depth characterization of these cells.

Results

We will show here the application of this technology in the context of immunotherapy, through the example of a murine in vivo tumor model responsive to checkpoint blockade inhibitors, as well as through the analysis of different human cancer samples.

Conclusions

Together, by providing insights into the nature of neoantigen-specific T-cells, immunoSCAPE’s unique target discovery and high-dimensional immune profiling platform is a valuable tool for the development of novel diagnostic biomarkers and therapeutic strategies at different stages of drug development.

Legal entity responsible for the study

A*STAR / Singapore Immunology Network and immunoSCAPE.

Funding

immunoSCAPE.

Disclosure

E. Newell: Board director and shareholder: immunoSCAPE Pte. Ltd. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

127P - Over-expression of S100B protein as a serum marker of brain metastasis in non-small cell lung cancer and its prognostic value

Presentation Number
127P
Lecture Time
12:30 - 12:30
Speakers
  • Lijuan Chen (Zhengzhou, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Validated serum biomarkers for patients suffering from non-small cell lung cancer (NSCLC) brain metastasis are urgently needed for early diagnosis, treatment monitoring, and prognostic classification in daily clinical practice and trials. Serum S100B was reported to be a marker of leaky blood-brain barrier (BBB), which was often caused by brain tumors. This study aimed to investigate the role of S100B and S100B antibody in the early detection of NSCLC brain metastasis and the prognostic significance.

Methods

100 patients with NSCLC brain metastasis, 50 patients of stage IV NSCLC without brain metastasis, and 50 patients with cerebrovascular diseases were enrolled in this prospective study. S100B and S100B antibody were measured in serum samples of all patients before and after treatment by ELISA, and the correlations with brain metastasis were assessed by ANOVA. Kaplan-Meier survival analyses and COX regression were used to unveil the prognosis significance.

Results

The results showed that serum S100B correlated significantly with NSCLC brain metastasis (p < 0.001), but not S100B antibody (p > 0.05). When evaluated by the ROC curve, at the cutoff point 13.83 pg/ml, the sensitivity and specificity were 94% and 93%, respectively (AUC= 0.938, p < 0.001). The PFS and OS of NSCLC patients with brain metastasis were significantly shorter in the patients with high levels of serum S100B. In addition, S100B was an independent prognostic factor.

Conclusions

In conclusion, serum S100B was a sensitive and specific marker for early detection of brain metastasis in NSCLC and could be used as a surveillance tool for prognosis evaluation.

Legal entity responsible for the study

Affiliated Cancer Hospital of Zhengzhou University.

Funding

Has not received any funding.

Editorial Acknowledgement

We thank Bart's Cancer Institute, Queen Mary University of London for supporting our work.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

128P - Differential gene expression profiles in poor vs good responders to maintenance vinflunine in patients (p) with advanced urothelial carcinoma (aUC): Preliminary results of biomarker analyses from the MAJA trial (SOGUG 2011/02)

Presentation Number
128P
Lecture Time
12:30 - 12:30
Speakers
  • Jose Luis Ramirez (Badalona, ES)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Vinflunine is an antimicrotubule agent approved by the EMA for second-line treatment in p with aUC. However, no molecular biomarkers are currently available that can predict response to vinflunine in aUC. In the randomized phase II MAJA trial (NCT01529411) in p with aUC with disease control after a platinum-based regimen, maintenance vinflunine conferred a significant improvement in progression-free survival compared to best supportive care (Garcia-Donas et al. Lancet Oncol 2017). Pre-planned gene expression analyses aimed to identify biomarkers to predict response to maintenance vinflunine.

Methods

In the MAJA trial, 44 p received vinflunine. We have compared the gene expression profiles of eight poor responders to vinflunine (<4 cycles) and nine good responders (>12 cycles). RNA was isolated from FFPE tumor tissue collected during screening using the Covaris kit and gene expression levels were analyzed with Clariom S array (Thermo Fisher). Differential expression (DE), defined as p < 0.05 and |FC|>1.5, was determined with linear models for microarray data included in the limma and sva packages. Pre-ranked Gene Set Enrichment Analysis (GSEA) was used for the functional classification of the DE genes.

Results

Hierarchical clustering of genes showed a DE between good and poor responders. DE were found in 31 genes, 13 of them were unregulated in good responders and 18 were unregulated in poor responders. In good responders, GSEA revealed overexpression of 72 genes related to G2M-checkpoint and of 61 genes related to E2F transcription factor. In poor responders, 73 genes related to epithelial-mesenchymal transition and 39 related to IL6/JAK/STAT3 were downregulated. We are currently validating these genes using qPCR to determine a gene expression profile associated with response to maintenance vinflunine.

Conclusions

Our preliminary results suggest that microarray analysis could identify a gene expression signature to predict response to maintenance vinflunine, which will be useful in selecting treatment for p with aUC. Complete results of the analyses will be reported.

Clinical trial identification

NCT01529411; EudraCT: 2011-001271-39.

Legal entity responsible for the study

Spanish Oncology Genitourinary Group.

Funding

Spanish Oncology Genitourinary Group.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

129P - Pre-diagnostic measurements of high-sensitive C - reactive protein and risk of Prostate Cancer. The PROCA-life study

Presentation Number
129P
Lecture Time
12:30 - 12:30
Speakers
  • Einar Stikbakke (Tromso, NO)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Inflammation may promote prostate cancer development, which can be characterized by elevated circulating levels of inflammation markers, such as high-sensitivity C-reactive protein (hs-CRP). Whether pre-diagnostic measurements of hs-CRP are associated with prostate cancer remains unknown.

Methods

In the Prostate Cancer Study throughout life (PROCA-life), a total of 11 064 initially healthy men, who participated in the Tromsø Study between 1994 and 2008, were included. Pre-diagnostic hs-CRP was assessed and height and weight were measured at study entry. During a mean follow-up time of 14.2 years, a total of 459 men developed histological verified prostate cancer and detailed medical and histological records were obtained.

Results

At study entry, the cohort participants had a mean age of 60.5, mean level of hs-CRP of 2.18 mg/l and a mean body mass index (BMI) of 25.8 kg/m2. The 459 prostate cancer cases identified had a mean age at diagnosis of 72.0 years. Among normal weighted men (BMI<25kg/m2), we observed a positive linear relationship between pre-diagnostic hs-CRP levels and prostate cancer risk after adjustments, both when using single and repeated measurements of hs-CRP, with hazard ratio 1.09 (95% CI 1.03-1.14) and 1.08 (95% CI 1.01-1.16), respectively. This relationship was not present in the overweight (BMI 25-30 kg/m2) or obese (BMI>30kg/m2) group.

Conclusions

Our study supports the hypothesis that inflammation may play a role in prostate cancer development, but this association may vary by body composition.

Legal entity responsible for the study

University of Tromsø.

Funding

University of Tromsø.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

130P - Training and validation of a gene expression signature for microsatellite instability

Presentation Number
130P
Lecture Time
12:30 - 12:30
Speakers
  • Sarah E. Warren (Seattle, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Clinical response to cancer immunotherapy can be predicted from two biologically distinct variables: tumor foreignness, manifested in some tumors as deficient DNA mismatch repair (dMMR)/microsatellite instability (MSI), typically measured by qPCR or IHC, and anti-tumor immunity, typically measured by gene expression signatures or IHC. Clinical benefit should be more accurately predicted by the combination of the two variables than by either alone, but measuring both variables requires multiple clinical assays. Here, we investigate the ability of gene expression alone to provide a surrogate measure of both tumor foreignness and immunogenicity, empowering a single assay to measure both axes of predictive biology. In addition, we explore the relationships between tumor foreignness and immunogenicity in both dMMR and MMR-proficient (pMMR) tumors.

Methods

Using TCGA datasets from colon, esophageal, stomach and uterine cancers, we trained two algorithms predicting hypermutation: the first detecting loss of expression of mismatch repair genes (MLH1, MSH2, MSH6 and PMS2) and the second identifying expression patterns shared by hypermutated tumors across these four cancer types. A final algorithm synthesizes the two above algorithms into a single score. For independent validation, we evaluated 60 colorectal cancer (CRC) FFPE samples, 30 dMMR and 30 pMMR as previously identified by IHC, and a cohort of 10 MSI and 5 MSS endometrial and neuroendocrine tumors with the NanoString nCounter® platform.

Results

We show that our algorithms successfully predicted hypermutation phenotypes and MMR status in TCGA training data and in the independent CRC and endometrial datasets analyzed with NanoString; with an AUROC of 0.94. Additionally, we demonstrate that higher mutational burden is linked to a heightened tumor immune environment as shown by adaptive immune gene signatures.

Conclusions

Gene expression proves to be a powerful predictor of microsatellite instability and hypermutation in cancers where dMMR subtypes are known to exist. This discovery raises the possibility that a gene expression algorithm measuring both hypermutation and immune activity may be advantaged in predicting response to checkpoint inhibition in these cancer types.

Legal entity responsible for the study

NanoString Technologies.

Funding

NanoString Technologies.

Disclosure

S.E. Warren, S. Ong, N. Elliott, A. Cesano: Employee and stockholder: NanoString Technologies. P. Danaher: Stockholder: NanoString Technologies.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

131P - Development and analytical validation of a plasma-based tumor mutational burden (TMB) score from next-generation sequencing panels

Presentation Number
131P
Lecture Time
12:30 - 12:30
Speakers
  • Katie Quinn (Redwood City, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The advantages of plasma-based tumor mutational burden (TMB) include non-invasive, real-time assessment of mutational load, without the limitations of insufficient tissue. However, at low levels of tumor DNA shedding, TMB may be underestimated if a fraction of the genomic alterations is below assay limit of detection. Currently available blood TMB panels report a 1% tumor content limit of detection, which would result in about half of all clinical plasma samples (based on > 30,000 patients) with unevaluable TMB. Hence, clinically effective blood-based diagnostics must be highly sensitive and account for tumor shedding. Here, we present a comprehensive cfDNA-based TMB using a 500-gene (GuardantOMNI) and a 73-gene (Guardant360) panel.

Methods

We developed a statistical model to calculate TMB on plasma samples with low cell-free circulating tumor DNA (ctDNA) content. Theoretical panel performance was assessed in silico by subsetting mutations from whole exome sequencing (WES) to the Guardant panel space (2Mb for GuardantOMNI and 200Kb for Guardant360) from 9,104 TCGA samples and 30 lung cancer samples with published immunotherapy outcomes. Sensitivity was evaluated using 50 serially diluted cfDNA specimens. Analytical validation was performed against tissue-based WES TMB using matched plasma and tissue samples across multiple tumor types.

Results

High correlation was observed between TMB called on the Guardant panel and WES mutations from the TCGA dataset (r = 0.99 with GuardantOMNI; r = 0.92 with Guardant360). Subsetting WES from clinical outcome cohorts to each panel recapitulated the association with PFS on immunotherapy (HR = 0.41 with GuardantOMNI; HR = 0.27 with Guardant360). The sensitivity of detection was assessed down to 0.3% tumor content and 5ng cfDNA input. Lastly, we show high quantitative concordance between matched plasma and tissue WES samples for both GuardantOMNI and Guardant360.

Conclusions

We describe a plasma-based TMB score that correlates with tissue-derived TMB at tumor fractions down to 0.3%, enabling TMB calculation on > 70% of all clinical samples. Accurate reporting of TMB from a plasma sample has the potential to accelerate clinical trial enrollment and improve outcomes.

Legal entity responsible for the study

Guardant Health, Inc.

Funding

Guardant Health, Inc.

Disclosure

K. Quinn, E. Helman, T. Nance, C. Artieri, J. Yen, J. Zhao, S. Fairclough, M. Sikora, D. Chudova, R.B. Lanman, A. Talasaz: Employee and ownership (stock): Guardant Health, Inc.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

132P - Carbohydrate antigen 19-9 and apolipoprotein A2 isoform as early detection biomarkers for pancreatic cancer: a prospective evaluation by the EPIC study

Presentation Number
132P
Lecture Time
12:30 - 12:30
Speakers
  • Kazufumi Honda (Tokyo, JP)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

This study prospectively evaluated the performance of an apolipoprotein A2 isoform (ApoA2-ATQ/AT) in combination with carbohydrate antigen 19-9 (CA19-9) as early detection biomarkers for pancreatic cancer.

Methods

Using ELISA, we measured CA19-9 and ApoA2-ATQ/AT in plasma samples collected ≤60 months before diagnosis from 159 pancreatic cancer patients and 217 matched controls within the EPIC (European Prospective Investigation into Cancer and Nutrition) cohort. The diagnostic sensitivity, specificity, and C-statistics were calculated for risk scores by strata of the time before diagnosis.

Results

The C-statistics of CA19-9 and CA19-9+ApoA2-ATQ/AT for distinguishing pancreatic cancer patients from cancer-free individuals were 0.87 and 0.68, respectively, for samples taken ≤6 months before diagnosis, and 0.74 and 0.72, respectively, for samples taken <6 to 18 months before diagnosis. The joint diagnostic model using CA19-9+ApoA2-ATQ/AT showed significantly improved diagnostic discrimination in samples taken ≤18 months before diagnosis. Before diagnosis, the specificity of CA19-9+ApoA2-ATQ/AT was 98%, while the sensitivities of CA19-9+ApoA2-ATQ/AT were 57%, 36%, and 43%, respectively, and those of CA19-9 alone were 50%, 29%, and 36%, respectively. This joint model also showed significantly improved C-statistics for the diagnostic discrimination of samples taken >6 to 18 months (0.80 for CA19-9+ApoA2-ATQ/AT, 0.74 for CA19-9; p = 0.004) and ≤18 months (0.8 for CA19-9+ApoA2-ATQ/AT, 0.78 for CA19-9; p = 0.003) before diagnosis.

Conclusions

Compared to CA19-9 alone, CA19-9+ApoA2-ATQ/AT showed improved diagnostic discrimination for early detection ≤18 months before diagnosis. This plasma biomarker panel may provide a useful first measure for detecting pancreatic cancer prior to imaging. We have reported those results on behalf of the EPIC Europe.

Legal entity responsible for the study

National Cancer Center.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

133P - Sarcopenia and inflammation predicts survival in advanced stage cancer patients (pts) treated with immunotherapy (IO)

Presentation Number
133P
Lecture Time
12:30 - 12:30
Speakers
  • Mehmet A. Bilen (Atlanta, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Sarcopenia is associated with poor prognosis in cancer pts and inflammation has been recognized as a hallmark of cancer. In this study, we investigated the synergistic effect of sarcopenia and inflammation on survival in advanced stage cancer pts treated with IO.

Methods

We performed a retrospective analysis of 90 pts treated on IO-based phase 1 clinical trials at Winship Cancer Institute of Emory University from 2009-2017. CT images at mid-L3 were obtained at baseline. Skeletal muscle density was obtained using SliceOmatic v5.0 by TomoVision and converted to skeletal muscle index (SMI) by dividing by height2. We defined sarcopenia as SMI<39. Neutrophil-to-lymphocyte ratio (NLR) was obtained at baseline and used as a surrogate of inflammation. Pts were categorized based on recursive partitioning methods into three groups: (1) low NLR, (2) high NLR without sarcopenia, (3) high NLR with sarcopenia. Overall survival (OS) and progression-free survival (PFS) were measured from date of first dose of IO to date of death or clinical or radiographic progression, respectively. Multivariable analysis (MVA) was carried out using Cox proportional hazard model.

Results

The majority of pts (n = 53) were males and most (68.9%) had at least 2 prior lines of therapy. Melanoma (33%) and GI (22.2%) tumors were the most common histologies. Low NLR was associated with longer OS and PFS (Table). Sarcopenic pts with high NLR had shorter survival than pts with high NLR who were not sarcopenic (Table).

Univariable analysis (UVA) and MVA of inflammation and sarcopenia with survival

UVA
MVA
Median Survival in Months (95% CI)NHR (CI)p-valueNHR (CI)p-value
OS
Group 1: NLR <2.924.3 (10.3, 44.8)36----36----
Group 2: NLR ≥2.9, SMI ≥ 399.4 (5.5, NA)332.65 (1.22-5.72)0.013*332.08 (0.90-4.77)0.085
Group 3: NLR ≥ 2.9, SMI < 393.8 (2.8, 5.9)218.40 (3.47-20.31)<0.001*217.93 (3.19-19.73)<0.001*
PFS
Group 1: NLR <2.94 (2.5, 5.4)36----34----
Group 2: NLR ≥ 2.9, SMI ≥ 392.8 (1.6, 4.1)331.62 (0.95-2.79)0.078331.35 (0.77-2.39)0.298
Group 3: NLR ≥ 2.9, SMI < 391.6 (1.2, 1.8)214.16 (2.26-7.67)<0.001*214.37 (2.26-8.48)<0.001*

The multivariable model was built by controlling for gender, checkpoint indication, # of previous treatment lines, royal marsden hospital (RMH) risk group, age, ECOG PS, race, # of metastatic sites, and histology. *statistical significance at alpha < 0.05.

Conclusions

Sarcopenia may have a synergistic effect with inflammation on decreasing survival in pts treated with IO. Prospective validation of the impact of body composition parameters on survival and whether adipose tissue plays a role in the relationship may be warranted. Equal contribution: MAB, DJM, JMS.

Legal entity responsible for the study

Emory University IRB.

Funding

Research reported in this publication was supported in part by the Biostatistics and Bioinformatics Shared Resource of the Winship Cancer Institute of Emory University and NIH/NCI under award number P30CA138292. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Disclosure

M.A. Bilen: Consulting/advisory role: Exelixis, Sanofi; Research funding: Bayer, Bristol-Myers Squibb, Genentech/Roche, Incyte, Nektar, AstraZeneca, Tricon Pharmaceuticals, Peleton, Pfizer. B.C. Carthon: Consulting/advisory role: Astellas Medivation, Pfizer, Blue Earth Diagnostics; Travel accommodations: Bristol-Myers Squibb. W.L. Shaib: Research funding: ArQule and Lilly. R. Kudchadkar: Consulting/advisory role: Bristol-Myers Squibb, Novartis, Array BioPharma; Honorarium: Bristol-Myers Squibb; Research funding: Merck. B. El-Rayes: Consulting/advisory role: Merrimack, BTG, Bayer, Loxo, RTI Health Solutions; Speakers’ bureau: Lexicon, Bristol-Myers Squibb; Honorarium: Lexicon, RTI Health Solutions, and Bayer; Research funding: Taiho Pharmaceutical, Bristol-Myers Squibb, Boston Biomedical, Cleave Biosciences, Genentech, AVeo, Pfizer, Novartis, Hoosier Cancer Research Network, Five Prime Therapeutics, PPD Inc., Merck, ICON Clinical Research. S.S. Ramalingam: Consulting/advisory role: Amgen, Boehringer Ingelheim, Celgene, Genetech/Roche, Lilly/ImClone, Bristol-Myers Squibb, AstraZeneca, Abbvie, Merck, and Takeda; Travel, accommodations: EMD Serono, Pfizer, and AstraZeneca. T.K. Owonikoko: Consulting/advisory role: Novartis, Celgene, Lilly, Sandoz, Abbvie, Eisai, G1 Therapeutics, Takeda, Seattle Genetics, Bristol-Myers Squibb, and MedImmune; Intellectual property of the following: “Overcoming acquired resistance to chemotherapy treatments through suppression of STAT3” and “Selective chemotherapy treatments and diagnostic methods related thereto”; Research funding: Novartis, Astellas Pharma, Celgene, Bayer, Stem CentRx, Regeneron, AstraZeneca/MedImmune, Abbvie, G1 Therapeutics, Bristol-Myers Squibb. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

134P - Prediction of irAEs in Ipilimumab-treated melanoma patients based on serum autoantibodies

Presentation Number
134P
Lecture Time
12:30 - 12:30
Speakers
  • Jessica C. Hassel (Heidelberg, DE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Checkpoint inhibition is an effective treatment in patients with metastatic melanoma (MM). T cell activation can induce tumor rejection but also possibly severe autoimmune side effects (irAE). Autoantibody biomarkers from serum have potential to predict irAEs such as an ipilimumab-induced colitis.

Methods

We use a cancer immunotherapy array consisting of 850 human protein antigens from 4 classes: 1. Tumor-associated antigens (TAA), 2. cancer pathway proteins, 3. autoimmune antigens, 4. cytokines/interleukins. Protein antigens were covalently coupled to magnetic beads and serum AABs were analyzed by Luminex FlexMap 3D. First, we screened pre-immunotherapy sera from 142 patients with MM (Heidelberg Cohort: 82 Ipilimumab (Ipi) treated; 11 Ipi/Nivolumab (Nivo); 40 Pembrolizumab (Pembro), 119 healthy controls (HC)). In this cohort, 41.5% (n = 59) experienced irAEs of any grade and 7% (n = 10) had colitis of grade 3 or 4. In a second study, 200 MM patients from 5 European sites (53 Ipi/Nivo; 111 Pembro, 100 HC) were analyzed. 25.6% (n = 42) had grade 3 or 4 irAEs, 12.8% (n = 21) had diarrhea or colitis of any grade and 9.8% (n = 16) had grade 3 or 4 colitis.

Results

40 different AABs were significantly more prevalent in MM compared to HC including NY-ESO1, NY-ESO 2 and other TAAs, cytokines, and nuclear proteins. Significant correlations of AABs were seen in Ipi-treated patients who experienced irAEs, both in mono- but also in combination therapy, allowing to dichotomize MM in risk groups. Also different sets of AABs were seen in Pembro-treated patients with irAEs. The protein antigens represent a variety of biological processes: they are involved in melanoma progression including transcription factors or components of the E3 ubiquitin ligase complex, cytokeratins, and proteins involved in cell adhesion.

Conclusions

In MM, screening of AABs prior to start of Checkpoint inhibition holds potential to predict risk for irAEs such as colitis. As irAEs are especially frequent in Ipi-based treatment regimes, AABs presented here may serve as useful biomarkers for a risk-based treatment decision.

Legal entity responsible for the study

Jessica C. Hassel and Protagen AG.

Funding

Protagen AG.

Editorial Acknowledgement

None

Disclosure

J.C. Hassel: Consulting role: Merck, Amgen; Honoraria: Bristol-Myers Squibb, Merck, Novartis, Roche and Pfizer; Science projects support: BMS. J. Mangana: Temporary advisory relationship and receives travel support: MSD, Merck. C. Pföhler: Consulting role: Merck Serono, Novartis, Roche, Amgen, BMS; Honoraria: Merck Serono, Novartis, Roche, Amgen, BMS. B. Weide: Consulting role: Curevac, Philogen, BMS; Honoraria: MSD, BMS, Roche, Amgen, Philogen; Science projects support: BMS, Philogen. L. Hakim-Meibodi: Travel grants: BMS. F. Meier: Honoraria: Roche, BMS, GSK, Novartis, MSD; Travel support: Roche, BMS; Research funding: Wyeth/Pfizer, Merck-Serono, Novartis. H.-D. Zucht, P. Budde; M. Tuschen: Employee: Protagen AG. P. Schulz-Knappe: Board member and chair holder: Protagen AG. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

135P - Combined tumor-based BRCA/TP53 mutation testing in ovarian cancer

Presentation Number
135P
Lecture Time
12:30 - 12:30
Speakers
  • Borcoman Edith (Villejuif, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Somatic or germline BRCA mutations remain the best predictive biomarker for PARP inhibitor benefit and > 95% of high grade serous ovarian cancers (HGSOC) have a clonal somatic TP53 mutation. Combined tBRCA/TP53 testing might provide the advantages of i) rapid results, ii) identification of somatic BRCAm, and iii) indirect evidence of the ‘2nd hit’ event, ie loss of heterozygosity (LOH).

Methods

Between 1/1/2016 and 1/2/2018 182 pts with HGOC underwent tBRCA/TP53 testing with a capture NGS panel and were oriented to a germline BRCA (gBRCA) testing via a dedicated genetics consultation. The ratio of allelic fractions (AF) for BRCAm/TP53m was calculated to estimate the proportion of cells carrying the BRCAm and derive LOH.

Results

At the time of data cut-off, gBRCA results were available for 125/182, and still pending for 61 pts. 15/125 (12%) demonstrated a deleterious (DEL) gBRCA1m (N = 12) or gBRCA2m (N = 3). Tumor testing was performed on 182 with a median testing turn-around time of 16 days (range 7-539 days). Twenty-seven (15%) were non-contributive. Among 155 contributive tumor samples, 31 DEL tBRCAm (21%) were identified. All gBRCAm (15/15) were identified on tumor testing including one large re-arrangement. 16 additional DEL BRCA1m or BRCA2m were detected: 10 somatic BRCAm in pts with confirmed wild-type (WT) germline status, and 6 among pts with pending germline results. Median TP53m AF was 0.48 (range 0.012-0.92) confirming a huge variability in tumor cellularity among samples. Among gBRCAm cases, ratio AF BRCAm/TP53m was always>1 confirming germline origin and suggesting LOH. AF BRCAm/TP53m was lower among known sBRCAm tumors (median AF BRCAm/TP53m=1.1) but always>0.8 suggesting acquired BRCA mutation was clonal and associated with LOH. For 3 gBRCA WT samples with <10% tumor cellularity and very low DEL BRCAm AF (0.04, 0.04 and 0.05), TP53m AF were also <0.05, thus validating the somatic BRCAm.

Conclusions

Combined BRCA/TP53 tBRCA testing is fast, sensitive and identifies somatic BRCA mutations. In addition, information on TP53 AF is useful to validate % neoplastic cells, identify somatic BRCAm in low cellularity samples and provides indirect evidence for LOH as the ‘2nd hit’.

Legal entity responsible for the study

Alexandra Leary.

Funding

Has not received any funding.

Disclosure

A. Leary: Honoraria: AstraZeneca; Consulting: AstraZeneca, GamaMabs Pharma, Clovis Oncology, Gritstone Oncology; Research funding (institution): AstraZeneca, GamaMabs Pharma, Clovis Oncology, AstraZeneca, Pfizer, MSD; Accommodation: AstraZeneca. P. Pautier: Honoraria: AstraZeneca; Consulting: AstraZeneca, Roche, Tesaro; Research funding (institution): AstraZeneca, Roche, ImmunoGen; Accommodation: AstraZeneca, PharmaMar. L. Lacroix, C. Genestie: Honoraria: AstraZeneca. J. Michels: Research funding (institution): Roche, MSD. E. Rouleau: Honoraria: AstraZeneca; Consulting: AstraZeneca, Roche, BMS UK; Research funding (institution): AstraZeneca. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

136P - EML4-ALK fusion variants associate with gender and age in Chinese NSCLC patients

Presentation Number
136P
Lecture Time
12:30 - 12:30
Speakers
  • Wei Zhao (Kunming, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

EML4-ALK fusion is one of the major activating kinase mutations in NSCLC. It occurs with varying frequency in different populations and is known to associate with clinical features such as smoking history and age. Based on different breakpoints in EML4 and ALK genes, there exist several fusion subtypes (variants), which have been reported to respond differently to ALK inhibitors. In this study, we investigated the landscape of EML4-ALK in Chinese NSCLC patients and correlated fusion variants with clinical factors.

Methods

FFPE tumor and matched blood samples of over 1000 Chinese NSCLC patients were collected for NGS-based 450 cancer genes panel assay. Genomic alterations including single nucleotide variations (SNV), short and long insertions/deletions (Indel), copy number variations (CNV) and gene rearrangements in selected genes were assessed.

Results

A total of 73 EML4-ALK fusion samples were identified. In our cohort, younger patients were more likely to harbor ALK fusions, consistent with previous reports. In terms of EML4-ALK fusion variants, we observed three major subtypes: 30 E6:E20 (variant 3) cases, 20 E13:E20 (variant 1), and 11 E20:E20 (variant 2). Interestingly, E6-E20 was enriched in male patients (18 vs 12), while E13-E20 was enriched in female patients (13 vs 7); furthermore, for the E13-E20 variant all 13 female patients were younger than 60, while 4 out of 7 male patients were older than 60 (p = 0.007). To complete the picture, other subtypes of EML4-ALK fusion in our cohort included: 4 E18:E20, 4 E14:E20, 3 E2:E20, and 1 E21:E20.

Conclusions

EML4-ALK fusion is known to associate with clinical features including race, age, and smoking history. Our profiling of EML4-ALK alteration in Chinese NSCLC patients not only revealed the composition of fusion subtypes, but also connected variants with demographic factors such as gender and age. Our initial results need to be validated in larger cohorts and/or different populations. Given that different fusion variants respond differently to ALK inhibitors, our findings could have important implications.

Legal entity responsible for the study

OrigiMed.

Funding

Has not received any funding.

Disclosure

A. Wang, H. Chen, Y-A. Dong: Employee: OrigiMed. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

137P - Multi-Dimensional Immuno-Oncology Assays for Understanding the Immune System and Tumor Microenvironment

Presentation Number
137P
Lecture Time
12:30 - 12:30
Speakers
  • Fiona Hyland (San Francisco, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The promise of immuno-oncology (IO) is the potential for activating the immune system of an individual with cancer to destroy the tumor and metastases, and cause a complete and persistent response. A percentage of patients treated with checkpoint inhibitors or Car-T cell therapies exhibit complete response. However understanding all the determinants of response, and what combination of therapies can maximize efficacy and minimize adverse events, is still poorly understood. Further understanding the properties of the tumor micro-environment, the immune system, and driver mutations in the tumor provides the most comprehensive picture, enabling personalized oncology research.

Methods

We describe a suite of Oncomine assays that use a single chemistry and instrument with 20ng of input material each. The first measures patterns of gene expression of 395 genes that capture interferon and chemokine signaling, T and B cell activation, checkpoint pathway, antigen presentation, and tumor proliferation, measuring the expression of markers specific to different effector cell types. We demonstrate highly sensitive detection of low expressing genes including Interferon-Gamma. The second is the T-cell Repertoire sequencing assay. This assay uses total RNA from blood for long-amplicon TCRB chain sequencing, covering CDR1, 2 & 3. This assay provides an estimate of T cell diversity and other properties. The third is the Tumor Mutation Load assay. This 400-gene panel measures somatic mutations/Mb on FFPE samples, without requiring a matched normal. We demonstrated high reproducibility on FFPE, concordance with matched normal, correlation with exome mutation load, and accuracy on control cell lines.

Results

Together, these IO panels provide sensitive, accurate, complementary information to further elucidate biological factors underlying response, resistance, and adverse events. A single software for these diverse assays supports joint interpretation of this data.

Conclusions

These IO assays enable deep, broad, multidimensional characterization of biomarkers to explore predictors of response, optimal combination therapy, and avoidance of adverse events, accelerating research into immunotherapy for personalized oncology.

Legal entity responsible for the study

Thermo Fisher Scientific.

Funding

Thermo Fisher Scientific.

Disclosure

F. Hyland, T. Looney, R. Chaudhury, A. Kamat, A. Pankov: Employee of Thermo Fisher Scientific.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

138P - Monitoring the effect of PI3K inhibition on HER2 therapy resistant breast cancer using serial analysis of PIK3CA mutant tumour DNA in plasma

Presentation Number
138P
Lecture Time
12:30 - 12:30
Speakers
  • Niamh Keegan (Dublin, IE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

PIK3CA is mutated in up to 20% of HER2 positive breast cancers, contributes to HER-2 therapy resistance and may be predictive of response to PI3K inhibitor therapy. PIK3CA mutations in breast cancer occur primarily at hotspots E545K at exon 9 and H1047R at exon 20. Copanlisib (C) is a pan-class I PI3K inhibitor that shows particular activity against PI3Kα, the isoform encoded by the PIK3CA gene. The aim of this study was to assess PIK3CA mutation status in matched tumour and plasma samples pre copanlisib treatment and to monitor PIK3CA mutation concentration changes in plasma over the course of PI3K inhibition therapy.

Methods

For 12 patients with advanced HER2 positive, breast cancer treated on a clinical trial of copanlisib and trastuzumab, we prospectively examined serial plasma samples to quantify the PIK3CA hotspot mutations in circulating tumour DNA by droplet digital PCR (ddPCR). Samples were taken pre-treatment, then every two weeks on treatment and immediately after radiological disease progression. Archival formalin fixed paraffin embedded (FFPE) primary tumour tissue were examined using MassArray® to detect PIK3CA mutation.

Results

PIK3CA mutations were detected in 6/12 (50%) archival FFPE primary tumours ; either an exon 9 (n = 2) or exon 20 (n = 4), all of which were also oestrogen receptor positive and had at least one prior line of anti Her2 therapy in the advanced cancer setting. There were 106 plasma samples included in the mutation analysis. PIK3CA mutation (H1047R or E545K) >500copies/mL were detected in 66% (70/106) of the samples. Of the six tumour samples that had no PIK3CA mutation detected, three had >500copies/mL (range: 0-25,500copies/mL) of mutated PIK3CA detected in serial plasma samples. Variations in plasma DNA mutation levels over time were found in all 12 patients.

Conclusions

Our data demonstrate that PIK3CA mutation is detectable in the plasma of a large proportion of a cohort of patients with HER2 therapy resistant advanced breast cancer, is potentially a more meaningful representation of current mutation status than archival primary tumour tissue given discordance and levels of mutation fluctuate with PI3K inhibition combined with trastuzumab.

Legal entity responsible for the study

Cancer Trials Ireland.

Funding

Bayer Pharmaceuticals.

Disclosure

B. Hennessy: Research funding: Bayer Pharmaceuticals. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

139P - Analysis of circulating biomarkers in A randomized phase II trial of maintenance oral metronomic vinorelbine in advanced NSCLC following platinum-based chemotherapy: A correlative MA.NI.LA. trial study

Presentation Number
139P
Lecture Time
12:30 - 12:30
Speakers
  • Mattia Boeri (Milan, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The efficacy of maintenance chemotherapy (CT) in advanced NSCLC is still under investigation. The multicenter, open label, randomized, phase II MA.NI.LA trial aimed to assess the activity of metronomic oral vinorelbine versus close observation as maintenance treatment following platinum-based CT in patients with advanced NSCLC. Serum angiogenesis biomarkers and the plasma microRNA signature classifier (MSC) associated with tumor aggressiveness in NSCLC patients were here investigated as prognostic and predictive factors.

Methods

119 advanced NSCLC patients with stable disease after platinum-based CT received maintenance metronomic oral vinorelbine (ARM A) or close observation (ARM B). The primary endpoint of the trial was progression free survival (PFS). Plasma samples for biomarkers evaluation were collected at the baseline from all patients. Concentrations of VEGFA and THBS1 were determined using EIA kit (Chemicon International). To determine MSC risk level, custom made microfluidic cards (Thermo Fisher) were adopted to analyze 8 samples simultaneously by RT-qPCR. Cox regression model was used to evaluate the association of biomarkers to PFS and OS.

Results

In the whole cohort, median age was 68.8; M/F: 79/40; stage IIIb/IV:14/105; adeno/other:73/46; PS 0-1/2:113/7. In front of a median follow up of 23.9 months, median PFS was 4.3 and 2.8 months for ARM A and ARM B, respectively (HR = 0.73; 90%CI 0.53-0.999; p = 0.0493). Considering all 119 patients, high VEGFA and THBS1 plasma levels and positive MSC expression were associated with shorter both PFS and OS. HR for 1000 unit, VEGF: 1.56 (90%CI 0.99-2.46; p = 0.054); HR for 10000 unit, THBS1: 1.16 (90%CI 0.97-1.38; p = 0.114); HR for MSC pos vs neg: 1.61 (90%CI 1.05-2.47; p = 0.028). Similar results were obtained considering OS. No interaction between treatment effect and biomarker values were detected.

Conclusions

In advanced NSCLC following platinum-based induction CT, metronomic oral vinorelbine showed a modest, but significant improvement in PFS. In addition, VEGF and MSC could be considered as prognostic but not predictive factors

Clinical trial identification

NCT02176369.

Legal entity responsible for the study

Clips srl.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

140P - Oral cavity immune response in pancreatic ductal adenocarcinoma (PDAC)

Presentation Number
140P
Lecture Time
12:30 - 12:30
Speakers
  • Juan P. Marquez (Ciudad Obregon, MX)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

We studied the humoral immune response in patients with pancreatic ductal adenocarcinoma (PDAC) using mouthwashes from the oral cavity. We found lower immune response against recall antigens, fusobacterium nucleatum and tumor associated antigens before immunotherapy treatment. We also found better prognosis when the antibodies against tested antigens got higher. Finally, we found significant differences in the humoral immune response in the oral cavity against serum.

Methods

Antigens from fusobacterium nucleatum, recall and tumor associated antigens were tested by IgG + IgM + IgA ELISA in both oral cavity and serum of PDAC patients n = 50. We provided 5 ml of saline solution to all the patients and controls to rinse the oral cavity hardly for 5 minutes and deposit the content in 15 ml tube. Sera were obtained. All was done after the approval of the local IRB committee.

Results

All the evaluated antigens were lower in the oral cavity of PDAC patients compared with controls (p = 0.001). Serum responses for all antigens were significantly different than oral cavity (p = 0.03). At the end of the immunotherapy treatment consisting in a combination of immunomodulation, multi peptide antigen specific active immunotherapy and immunogenic chemotherapy with doxorubicin and oxaliplatin the levels of antibodies were increased significantly. For example, FAP (p = 0.0001), EGFR (p = 0.005), Fascin-1 (p = 0001), VCP (P = 001). Also, the immunotherapy combination significantly increased antibody immune response against fusobaterium lysate (p = 0001).

Conclusions

PDAC is a challenge disease especially in late stages. PDAC patients had low antibody titters against several antigens in comparison with controls and before treatment the oral cavity. Serum antibodies were lower than oral cavity even after treatment. Oral cavity antibodies may be a useful early biomarker in patients in high risk of PDAC. This data also suggest that antibodies get low when the cancer or premalignant lesions of PDAC are present.

Legal entity responsible for the study

Centro de Investigacion de Cancer en Sonora (CICS).

Funding

CICS USA.

Editorial Acknowledgement

N/A

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

141P - Tumor Mutational Burden (TMB) Standardization Initiative: Establishing a Consistent Methodology for TMB Measurement in Clinical Samples

Presentation Number
141P
Lecture Time
12:30 - 12:30
Speakers
  • Albrecht Stenzinger (Heidelberg, DE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Clinical studies have established TMB, a measurement of mutations in the tumor genome, as a predictive biomarker for clinical efficacy of immune checkpoint inhibitors (ICIs). There is a lack of standardization for TMB estimation and reporting, which is critical for ensuring consistency for clinical implementation. An international collaboration organized by Friends of Cancer Research (Friends) and Qualitätssicherungs-Initiative Pathologie GmbH (QuIP) is establishing recommendations for achieving consistency in TMB estimation and reporting.

Methods

Friends and QuIP are using complementary TMB harmonization approaches. Friends will conduct in silico analyses where TCGA data will be compared between TMB estimates derived from whole exome sequencing (WES) and commercial targeted gene panels, followed by the use of patient-derived tumor cell lines to establish a universal reference standard for the alignment of panel-derived estimates. QuIP will compare TMB estimates from selected tissue (NSCLC and other solid tumors) using a WES-derived reference standard with commercial next-generation sequencing panels and lab-developed tests at several German academic institutions. These data will inform consistency of TMB estimation, assay comparability, and TMB cutoff values for potential clinical use.

Results

Preliminary data indicate several components influence TMB estimation: preanalytical factors (eg, input material quality/quantity), sequencing parameters (eg, enrichment technologies), library preparation, bioinformatics (eg, filtering of germline variants), FFPE-induced deamination artifacts, mutation types, and clonal vs subclonal events. Analyses of panel size and composition suggest that larger panels may yield more reliable TMB estimation and that the panel should include actionable targets, genes associated with mutagenesis (eg, microsatellite instability), and potential negative predictors of response (eg, mutated β2M, JAK1/2, PTEN).

Conclusions

The Friends and QuIP collaboration will establish recommendations for reliable and reproducible TMB measurement to ensure consistent identification of patients who are likely to respond to ICIs.

Legal entity responsible for the study

Bristol-Myers Squibb.

Funding

Bristol-Myers Squibb.

Editorial Acknowledgement

Medical writing assistance was provided by Amrita Dervan, MSc, of Spark Medica Inc. (US), funded by Bristol-Myers Squibb.

Disclosure

A. Stenzinger: Advisory board and honoraria for talks: AstraZeneca, Novartis, BMS, Roche, Illumina, Thermo Fisher. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

142P - DNA damaging agents and immunotherapy in NSCLC: Is there a STING in the tale?

Presentation Number
142P
Lecture Time
12:30 - 12:30
Speakers
  • Carminia Maria Della Corte (Napoli, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

In NSCLC patients, several clinical trials are testing the efficacy of DNA damage response inhibitors (DDRi) and chemotherapy in combination with anti-PDL1 drugs. DDRi activate antitumor immune responses in cancer through release of cytosolic DNA leading to STING activation, stimulation of neo-antigens and release of pro-inflammatory cytokines. Our group has previously demonstrated a strong correlation between EMT and immune activation, showing that tumors with high EMT score have the highest levels of targetable immune markers.

Methods

We analyzed mRNA and protein expression of immune and EMT genes in the lung adenocarcinoma (LUAD) and lung squamous carcinoma (LUSC) TCGA NSCLC and in a panel of NSCLC cells, correlating them with the presence of somatic mutations in DDR genes. Contemporary, we treated NSCLC cell lines in vitro with cisplatin and various DDRi combinations, including PARP/ATR/ATM/WEE-inhibitors, to determine the effect on DNA damage and immune markers expression (by western blot and RPPA analysis).

Results

In both TCGA cohorts, immune markers mRNA expression clustered together and were positively correlated with EMT genes. In the LUAD cohort, high expression of CD274 (PDL1) was associated with high levels of other immune suppressive markers (LAG3, IDO1, PDCD1LG2, HAVCR2, CTLA4, ICOS, CD4, and CD40) and chemokines (CXCL10, CCL5, CCL2). Notably, expression of STING pathway mediators (TBK1 and TMEM173) and mesenchymal markers (TWIST1/2, SNAI1, SMO, and TGFB1) were positively related with CD274. Moreover, we found that mutations in DDR related genes TP53, RB1, POLE, FANCM and BRCA1 were allied with higher levels of targetable immune suppressive markers (LAG3, IDO1, and CD274) and the mesenchymal marker, TWIST1, but lower levels of TMEM173. Finally, in vitro treatments with DDRi and cisplatin increased DNA damage, as demonstrated by increased p-H2AX, and proportionally upregulated PDL1 and STING in some cell lines.

Conclusions

Our findings provide rationale to combine DNA damaging agents with immunotherapy drugs targeting immune suppressive markers in NSCLC. From our data, expression of EMT genes and deleterious mutations in DDR genes represent the best candidates to select patients that can benefit from these combinations.

Legal entity responsible for the study

MD Anderson Cancer Center.

Funding

AstraZeneca.

Editorial Acknowledgement

NA

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

143P - Tumor infiltrating lymphocytes (TILs) and PDL1 expression as prescreening enrichment biomarkers of clinical benefit to immune checkpoint inhibitors (CI) in early clinical trials (ECT)

Presentation Number
143P
Lecture Time
12:30 - 12:30
Speakers
  • Juan Martin-Liberal (Barcelona, ES)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Patients (pts) enrolled in ECT with CI are mainly selected based on tumor type. Genomic markers are informative in few cases. We assessed microenvironment markers as prescreening tool to identify pts with higher chances of clinical benefit from CI.

Methods

Pts treated with anti PD1/PDL1 drugs in monotherapy or in combination with other CI in ECT at our centre were evaluated for TILs on hematoxylin-eosin stained sections and PDL1 expression in tumor (tumcells) and immune cells (immcells) by immunohistochemistry (SP263 antibody). Results were correlated with clinical outcomes.

Results

From June 16 to June 17, 64 pts were recruited. TILs and PDL1 expression were available for all and 39 pts, respectively. Tumor types were melanoma (16 pts), neuroendocrine (9), gyne (8), breast (5), H&N (4), others (22). In total, 38 pts received anti PD1/PDL1 in monotherapy, the rest received anti PD1/PDL1 based combinations (12 pts had prior CI treatment). Response rate (RR) was 22%; median PFS was 4 months (m) (CI95% 3.30-5.57). We found no differences in PDL1 expression in tumcells according to tumor type (Kruskal test p = 0.33) and a weak correlation between TILs and PDL1 in tumcells (Pearson 0.44; p = 0.004) or immcells (Pearson 0.57; p = 0.0001). Median TILs was 7% (range 1-90), with no difference according to tumor type (Kruskal test p = 0.45). Median TILs was higher in pts with response to CI (17.5% v 5%, Kruskal test p = 0.06). RR in pts with TILs ≥7% was 32% v 9% if TILs <7% (Fisher test p = 0.06). In a multivariable logistic model adjusting for tumor type and CI regimen, RR was significantly higher in pts with TILs ≥7% (odds ratio 8.2; p = 0.05). Median PFS in pts with TILs ≥7% was 5 m v 3.7 m if TILs <7% (HR = 0.57 in a multivariable Cox model, p = 0.06). In univariate models, there was a trend for higher RR if PDL1 ≥1% in tumcells (35% v 12%, fisher test p = 0.28) or if PDL1 ≥1% in immcells (35% v 16%, fisher test p = 0.27). PFS was not correlated with PDL1 expression.

Conclusions

Quantifying TILs is a simple prescreening strategy that may help select pts for CI therapy in ECT from otherwise unselected population. The value of adding PDL1 expression needs further investigation.

Legal entity responsible for the study

VHIO.

Funding

Has not received any funding.

Disclosure

J. Tabernero: Advisory boards: Bayer, Boehringer Ingelheim, Genentech/Roche, Lilly, MSD, Merck Serono, Merrimack, Novartis, Peptomyc, Roche, Sanofi, Symphogen, Taiho. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

144P - Cyclin D1 differential activation and its prognostic impact among advanced breast cancer patients treated with trastuzumab.

Presentation Number
144P
Lecture Time
12:30 - 12:30
Speakers
  • Giannis Mountzios (Athens, GR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The cyclin D1-mediated molecular pathway depicts significant cross-talk with the ER/PR and HER2 pathways in patients with advanced breast cancer and these correlations may have clinical implications. We sought to determine the level of activation of the critical components of the cyclin D1-mediated pathway and to evaluate their prognostic significance across the different molecular subtypes of advanced breast cancer.

Methods

The study population included 219 trastuzumab-treated women with advanced breast cancer who had been found to have HER2-positive disease by local testing. For all tumors, central testing for HER2 was performed and cyclin D1 (CCND1) gene amplification and mRNA and protein expression were assessed by FISH, qRT-PCR and IHC, respectively.

Results

Only 134 of the 219 patients (61.2%) were HER2-positive (HER2 gene amplification and/or 3+ HER2 protein expression). After a median follow-up of 136.0 months, 105 HER2-positive patients (78.4%) and 76 HER2-negative patients (89.4%) had died, while 80.0% of the former and 87.1% of the latter had disease progression. Median PFS was 14.0 months for HER2-positive and 8.9 months for HER2-negative patients, while median survival was 48.1 months and 35.0 months, respectively. Cyclin D1 mRNA expression was higher in patients with positive ER/PgR. Cyclin D1 (as assessed by FISH, qRT-PCR and IHC) did not reach significance in terms of PFS or survival either in the entire study population or in HER2-positive patients. In the HER2-negative subgroup, negative cyclin D1 protein expression was associated with higher risk of progression (HR = 1.66, 95% CI 1.01-2.72, Wald’s p = 0.045), while in de novo metastatic patients, the risk of progression was higher for patients with non-amplified CCND1 tumors (HR = 2.00 95% CI 1.03-3.90, p = 0.041).

Conclusions

Aberrant activation of the cyclin D1-mediated pathway appears to reduce the risk of progression in HER2-negative tumors, but not in HER2-positive ones. If our results are validated by larger prospective trials, further evaluation of the cyclin D1-mediated pathway might identify prognostic and therapeutic implications in patients with advanced breast cancer.

Legal entity responsible for the study

Hellenic Cooperative Oncology Group (HeCOG).

Funding

Roche, Hellenic Cooperative Oncoogy Group.

Editorial Acknowledgement

NA

Disclosure

G. Mountzios, C. Christodoulou, P. Papakostas: Honoraria: Roche; Advisory role: Roche. G. Lazaridis: Honoraria: Roche. A. Koutras, G. Fountzilas: Advisory role: Roche. E. Razis: Advisory role, travel, honoraria: Roche; Research funding: Roche/Genentech. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

145P - Expression of TK1 and CDK9 in plasma-derived exosomes is associated with clinical response to CDK4/6 inhibitors in breast cancer

Presentation Number
145P
Lecture Time
12:30 - 12:30
Speakers
  • Stefania Crucitta (Pisa, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) improve PFS in patients with hormone receptor-positive (HR+) advanced breast cancer (Finn et al., 2016). In order to better characterize the response to these agents and increase our knowledge on the pharmacogenetic profile of CDK4/6i, the aim of this study was to analyse the expression of targets relevant to the activity of CDK4/6i in plasma-derived exosomes.

Methods

Blood samples were collected from patients affected by HR+, HER2- advanced breast cancer receiving palbociclib/ribociclib in association with hormonal therapy. Three ml of plasma were taken at the beginning of treatment (baseline) and at the first clinical evaluation (after 3 months). Objective responses were defined following the RECIST criteria v.1.1. RNA from plasma-derived exosomes was extracted by the ExoRNeasy kit (Qiagen) and analysed for the expression of thymidine kinase 1 (TK1), CDK 4, 6 and 9 by digital droplet PCR (BioRad). Mann-Whitney test was applied.

Results

Thirty-four metastatic breast cancer patients were prospectively enrolled in this study. The comparison of mRNA levels of TK1, CDK4, 6 and 9 between baseline and the first clinical evaluation was available in 22 patients treated with letrozole/anastrozole + palbociclib and 22 patients given fulvestrant + palbociclib. 18 patients had newly diagnosed advanced breast cancer while 16 patients received ≥1 line of treatment. Objective responses were: 1 (2,9%) CR, 4 (11,8%) PR, 16 (47,1%) SD and 13 (38,2%) PD. The comparison of changes in the expression between TK1, CDK 4, 6 and 9 at baseline compared to first evaluation was statistically significant for TK1 (PR+SD vs. PD p = 0.009), CDK4 (PR+SD vs. PD p = 0.020), CDK6 (PR+SD vs. PD p = 0.047) and CDK9 (PR+SD vs. PD p = 0.008). The univariate analysis didn’t find any significant correlation between patients clinical variable and PFS (i.e. type of hormonal treatment, the line of treatment, performance and menopausal status, visceral metastasis, bone only metastasis, number of metastasis, previous hormonal or lines of chemotherapy received).

Conclusions

Exosomal expression of CDK4, CDK6 and in particular of TK1 and CDK9 may be useful to early identify patients who are likely to respond to CDK4/6i.

Legal entity responsible for the study

Romano Danesi.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

146P - HER2 amplification is associated with higher tumor mutation burden in breast cancer

Presentation Number
146P
Lecture Time
12:30 - 12:30
Speakers
  • Yongmei Yin (nanjing, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

ErbB family consists of four transmembrane proteins (ErbB1/EGFR/HER1, ErbB2/HER2, ErbB3/HER3 and ErbB4/HER4) and plays a prominent role in the process of cell growth. Previous studies suggested that EGFR gene driven non-small-cell-cancer exhibited a suppressive immunity with lower tumor mutation burden (TMB) level. HER2 gene amplification (HER2+) is a well-known poor prognosis predictor in breast cancer and anti-HER2 target therapy has significantly improved the clinical prognosis of HER2+ patients. However, the association between HER2 gene alteration and TMB in breast cancer is still unclear.

Methods

Whole-exome sequencing data and clinical data of 366 breast tumors from The Cancer Genome Altas (TCGA) and next generation sequencing (NGS) data of 335 breast tumors from clinical dataset were analyzed to explore the association between HER2 gene alteration and TMB. TMB was defined as total number of somatic non-synonymous mutations in coding region.

Results

20.5% (75/366) of breast tumors in TCGA cohort and 20.3% (68/335) in clinical cohort harbored HER2 amplification. HER2 amplification was significantly associated with higher TMB in both TCGA cohort (P = 0.010) and clinical cohort (P = 0.008). HER2 somatic alteration occurred in 3.7% (12/366) of breast tumors in TCGA cohort and 7.2% (24/335) in clinical cohort. HER2 somatic alteration was also associated with higher TMB level in TCGA cohort (P = 0.016), but no association was observed in clinical cohort (P = 0.339). In addition, hormonal receptor (HR) + HER2- breast tumors exhibited the lowest TMB level compared with HR+ HER2 + (P = 0.001), HR-HER2 + (P = 0.052) and triple negative breast cancer (P = 0.000). Patients with low TMB level also tended to have a better overall survival than patients with higher TMB level (median, 216.6 vs. 112.0 months; HR, 0.572; 95% CI 0.31-3.05; P = 0.067).

Conclusions

HER2 amplification is associated with higher TMB in breast cancer. These findings may assist the selection of breast cancer patients likely to benefit from immunotherapy.

Legal entity responsible for the study

Yongmei Yin and Shaoqiang Zhou.

Funding

National Natural Science Foundation of China.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

147P - Application of CRISPR/Cas9 system for identification of genes involved in the regulation of pancreatic cancer cells platinum sensitivity

Presentation Number
147P
Lecture Time
12:30 - 12:30
Speakers
  • Vera Skripova (Kazan, RU)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Pancreatic cancer (PC) is an aggressive disease with high lethality rate due to multiple resistance mechanisms. We used in vitro CRISPR/Cas9 genetic drop-out screening to identify genes involved in the regulation of PC cell line sensitivity to platinum chemotherapy drugs.

Methods

We used two sgRNA libraries: 1) enriched for genes regulating cell cycle and nuclear proteins genes (CC, 50 000 sgRNA targeting 4 716 genes); 2) genome-wide (GW, 90 000 sgRNA targeting 18 164 genes). We performed screens in MIA PaCa-2 cells expressing doxycycline-inducible Cas9. Cells were treated with established IC30 of oxaliplatin (1 uM) or cisplatin (3 uM) for 9 cell divisions (12 days). Genomic DNA was extracted and sgRNA-containing regions were amplified and barcoded by PCR for further analysis by NGS. Statistical analysis for sgRNA enrichment or depletion was performed using R package comparing cells treated with the drugs vs. vehicle in the presence of Cas9/doxycycline.

Results

We identified 755 genes which significantly changed in cisplatin or oxaliplatin-treated cells (FDR 5%, p < 0.05). Candidate genes (n = 130) were further selected if at least 2 sgRNA per gene showed more than 2-fold change vs. vehicle. Among the 130 genes, 16 were known platinum sensitivity regulators involved in the double stranded break DNA repair pathway; 11 genes were positive platinum sensitivity regulators as their inactivation reduced sensitivity; 119 genes were negative platinum sensitivity regulators as their inactivation increased sensitivity. Gene Ontology analysis of the 130 candidate genes allowed us to identify regulators of cell cycle (n = 46), DNA replication and repair (n = 43), cellular compromise (n = 15), cellular assembly and organization (n = 35) and cell morphology (n = 48). Analysis of protein-protein interaction network showed that the majority of the hits (n = 74) are directly involved into cell cycle regulation and DNA repair processes.

Conclusions

We identified 130 candidate genes potentially involved in the modulation of platinum resistance most of which are regulating the cell cycle and DNA repair which is in keeping with the known DNA damaging mechanisms of action of platinum chemotherapy.

Legal entity responsible for the study

Kazan Federal University.

Funding

The work is performed according to the Russian Government Program of Competitive Growth of Kazan Federal University.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

148P - Elevated 70kDa heat shock protein (hsp70) and autophagy levels in peripheral blood mononuclear cells (PBMCs) in women with a malignant breast mass

Presentation Number
148P
Lecture Time
12:30 - 12:30
Speakers
  • Aikaterini I. Athanasiou (New York, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

PBMCs respond to adverse physiological conditions to maximize survival and to alleviate the threat. Hsp70 is produced to maintain protein function and activates an immune response. Autophagy is induced to provide sufficient nutrients in response to accelerated gene activation. We tested the hypothesis that PBMCS respond to the presence of a malignant breast mass by increasing production of hp70 and manifesting a higher level of autophagy.

Methods

In this pilot study seventy women had their breast mass evaluated by mammogram and/or breast ultrasound. A core biopsy and surgery was performed as indicated. PBMCs were isolated from peripheral blood, lysed and intracellular levels of hsp70 and p62 (a measure of autophagy) were quantitated by ELISA. Extracellular hsp70 in plasma was also measured. Differences in lab measurements between women with a diagnosis of a benign or malignant breast mass were determined. All assays were performed by personnel blinded to the clinical data.

Results

A breast malignancy was diagnosed in 42 women while 28 had a benign lesion. Plasma hsp70 levels were higher in women with a malignant lesion (p = 0.03). PBMCs from 46 women were available for analysis. Mean hsp70 levels were higher in PBMCs from 38 women with a malignant lesion than in 8 women with a benign breast mass (p = 0.04). The PBMC p62 levels were higher in women with a benign breast lesion than in those with a malignant breast mass (p < 0.0001). Since p62 is inversely related to the level of autophagy this indicates that autophagy is higher in PBMCs from women with a malignant breast lesion. There was no difference in the concentration of hsp70 or p62 between women with different histological types or stage of breast cancer.

Conclusions

Detection of elevated levels of hsp70 and autophagy in PBMCs, and higher plasma hsp70 levels, may differentiate between women with a malignant or benign breast lesion. Further studies on a larger sample are needed to confirm if the extent of autophagy and hsp70 induction in PBMCs may be of value in the preoperative triage of women with a breast mass.

Legal entity responsible for the study

Weill Cornell Medicine.

Funding

Weill Cornell Medicine.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

149P - Only estrogen receptor “Positive” is not enough to predict the prognosis of breast cancer Running head: Revisiting estrogen positive tumors in 8th AJCC staging era

Presentation Number
149P
Lecture Time
12:30 - 12:30
Speakers
  • Jai Min Ryu (Seoul, KR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Beginning in 2018, biomarkers including estrogen receptor (ER) status were incorporated in the 8th AJCC staging system. ER expression levels were not considered in these changes. We hypothesized that the levels of ER expression could affect the prognosis of breast cancer.

Methods

A retrospective review was conducted to identify all female patients with invasive breast cancer between 2003 and 2012. ER negative (group I), weakly ER-positive (group II), and strongly ER-positive (group III) were defined as Allred total scores of 0-2, 3-5, and 6-8, respectively. We examined a multigene panel, designated the BCT score, which is a newly developed prognostic model for predicting the risk of a distant metastasis.

Results

Among the 4,949 patients enrolled in this study, 1,310 (26.5%), 361 (7.3%), and 3,277 (66.2%) were categorized as group I, II, and III, respectively. Median F/U duration was 57.8 months. Compared to group III, patients in group II were younger, had larger tumors, and were also more likely to have PR-negative tumors, HER-2 amplification, high Ki-67, and high nuclear grade. Between group II and III, there was a significant difference in OS (P = 0.0764, .909, and 0.010, respectively). After adjusting for additional factors that may affect OS, the HR for OS showed higher in group II than in group III. The baseline median BCT score indicated that lower ER expression was associated with significantly higher BCT score (P < 0.0001) and significantly more likely to have high risk group (P < 0.0001) relative to higher levels of ER expression group.

Conclusions

ER expression levels affect the prognosis of breast cancer. The risk for patients with weakly ER-positive breast cancer should not be underestimated.

Legal entity responsible for the study

Institutional Review Board (IRB) of Samsung Medical Center in Seoul, Korea (IRB number: 2017-06-130).

Funding

This research was supported by Samsung Medical Center grant (SMO1170021) and by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HI17C1142).

Editorial Acknowledgement

This research was supported by Samsung Medical Center grant (SMO1170021) and by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HI17C1142).

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

150P - Adjuvant radiation therapy leads to an up-regulation of programmed death ligand 1 (PD-L1) on circulating epithelial tumor cells (CETCs) which might contribute to radioresistance in primary breast cancer patients

Presentation Number
150P
Lecture Time
12:30 - 12:30
Speakers
  • Dorothea Schott (Bayreuth, DE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Radiation therapy (RT) is an integral part of the treatment of breast carcinoma but unfortunately many patients experience local recurrence. During the inflammatory response that accompanies radiation tumor cells may develop multiple resistance mechanisms for example the up-regulation of PD-L1 on tumor cells which leads to immune evasion. Since CETCs arise from the tumor it is conceivable that under evolutionary pressure they might share some of the immune escape mechanism inherent to tumor cells. In this study we demonstrate that RT leads to a transitory adaptive up-regulation of PD-L1 expression on CETCs.

Methods

CETCs and the expression of PD-L1 and Ki-67 were analyzed from 25 patients with primary non-metastatic breast cancer using the maintrac method. The fraction of PD-L1 and Ki-67 positive CETCs were assessed at baseline, 3 and 6 weeks after start of RT and 6 weeks after end of therapy. Additionally, copy number status of PD-L1 was determined using FISH.

Results

Fractionated-dose RT leads to a significant increase in PD-L1 expression on CETCs with the highest expression level midterm of irradiation as compared to baseline (49% vs. 74%, p < 0.01). 6 weeks after end of RT the number of PD-L1 positive CETCs returned to baseline value. The up-regulation of PD-L1 was dose dependent. Patients who received higher total dose had significantly more PD-L1 positive CETCs as compared to patients treated with lower total dose midterm of RT (64% vs. 43, p < 0.05). Before start of therapy there was a correlation between the fraction of PD-L1 and Ki-67 positive CETCs (r = 0.6, p < 0.01). PD-L1 copy number gains were significantly associated with PD-L1 expression (r = 0.6, P < 0.05).

Conclusions

RT leads to an up-regulation of PD-L1 expression on CETCs, which could be a possible mechanism of acquired radioresistance. Combining immunomodulatory agents with radiation might have the potential to overcome this resistance and could improve clinical outcome in breast cancer.

Legal entity responsible for the study

Katharina Pachmann.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

151P - Recurrent and Metastatic Carcinomas of the Lacrimal Gland: High Frequency of ERBB2 Driven Disease

Presentation Number
151P
Lecture Time
12:30 - 12:30
Speakers
  • Nicolas Girard (Paris, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Lacrimal gland carcinomas (LGC) are uncommon primary malignancies that have a propensity to recur locally but rarely undergo metastasis. We performed comprehensive genomic profiling (CGP) on a series of 12 LGC to uncover genomic alterations (GA) that could possibly be used to design novel routes to targeted and immunotherapies for these rare neoplasms.

Methods

From a series of 158,360 clinically advanced cancer cases, FFPE tissue samples from 12 cases of LGC underwent CGP to find base substitutions, short indels, copy number changes and gene fusions. Microsatellite instability (MSI) was determined on 114 loci andtumormutational burden (TMB) was determined on 1.1 Mbp of sequenced DNA and reported as mutations (mut) per megabase (Mb).

Results

The 12 LGC patients ranged in age from 34 to 76 years (median 61 years) and 67% of the patients were male. The LGC included 7 (58%) adenocarcinomas, 2 (17%) squamous cell carcinomas, 2 (17%) adenoid cystic carcinomas and 1 (8%) undifferentiated carcinomas. There were 1 (8%) grade 1, 8 (67%) grade 2 and 3 (25%) grade 3 tumors. 2 (17%) of LGC were stage III and 10 (83%) were stage IV at the time of sequencing. There were an average number of 4.25 GA per tumor. Three (25%) of the LGC featured ERBB2(HER2) gene amplification. One (33%) of the 3 ERBB2amplified LGC also featured TOP2A amplification. ERBB2copy numbers in the amplified LGC ranged from 5 to 35 copies. Additional potentially targetable GA included PTENand PIK3CAboth at 25%, NF1 at 17% and RET, BRCA2, FGFR3and NTRK3all at 8% of cases. No (0%) LGC were MSI-High and the median TMB was 4.3 mut/Mb (range 0 to 12 mut/Mb) with no (0%) of LGC having >20 mut/Mb.

Conclusions

The high frequency ofERBB2amplification in clinically advanced LGC is similar to that seen in the approved indications for breast and upper gastrointestinal carcinomas raising opportunities for anti-HER2 therapies for these patients. In addition, a smaller cohort of LGC patients have opportunities for other targeted therapies with TKIs directed at RET, FGFR3 and NTRK. Given the lack of MSI-high and high TMB in LGC, the opportunities for immunotherapies for these patients appears limited.

Legal entity responsible for the study

Foundation Medicine, Inc, Cambridge, MA, USA.

Funding

Foundation Medicine Inc, Cambridge, MA, USA.

Disclosure

J. Chung, S.Z. Millis, L.M. Gay, J.A. Elvin, J-A. Vergilio, S. Ramkissoon, E. Severson, S. Daniel, J.K. Killian, S.M. Ali, A.B. Schrock, V.A. Miller: Employment: Foundation Medicine, Inc. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

152P - Circulating tumour DNA experience in patients with Cancer of Unknown Primary

Presentation Number
152P
Lecture Time
12:30 - 12:30
Speakers
  • Helen Winter (London, GB)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Improving the outcome for patients diagnosed with Cancer of Unknown Primary (CUP) is an unmet clinical need where survival is usually less than 1 year. Molecular characterisation of the disease may have diagnostic and therapeutic implications. The circulating tumour DNA test - Guardant 360™ is designed to detect gene alterations with a range of clinical utility.

Methods

Twenty-five patients were referred to Sarah Cannon Research Institute for the Guardant 360™ test. Panel version 2.10 reports single nucleotide variants in 73 genes, gene copy number amplifications in 18 genes, fusions/rearrangements in 6 genes as well as indels in 23 genes. The panel covers all NCCN somatic mutations. Digital SequencingTM technology essentially eliminates false positives allowing sequencing of targeted regions at very low DNA concentrations. Variants of unknown significance (VUS) were also measured. All patients were discussed at our institutional Genomics Review Board.

Results

Twenty-five patients (14 female; 11 male) were recruited from 24 August 2017 to 17 April 2018. Median age was 67 years (range 27-76). Main sites of disease were: lymph nodes (8); pelvis (8); liver (6); bone (3) and adrenal glands (2). The median turnaround time (TAT) from sample collection to report was 10 days (range 6-15). Seventeen patients (68%) had potentially actionable mutations; 4 patients had no mutations detected: 1 post resection; 2 were responding to chemotherapy; 1 was sampled prior to commencing chemotherapy. Genetic alterations detected included: BRAF V600E; KRAS; FGFR; MYC; KIT; PIK3CA and HER2. Twelve patients had ≥ 3 somatic mutations (including variants of uncertain significance (VUS)); ≥ 6 mutations were found in six of these patients.

Conclusions

ctDNA is feasible with an acceptable TAT and the identification of significant potentially actionable targets. Targetable mutations were detected including BRAF, V600E, HER2 and FGFR. Two patients now have access to BRAF and MEK inhibitors. Twelve patients had ≥ 3 mutations that is emerging as potential biomarker of response to immunotherapy. The burden of VUS and presence of actionable targets supports more research on personalised medicine in patients with CUP.

Legal entity responsible for the study

Sarah Cannon Research Institute.

Funding

Has not received any funding.

Disclosure

I. Faull: Employee: Guardant Health. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

153P - Reproducibility of the mRECIST criteria for the assessment of HCC treated by anti- VEGFR therapy: Impact of readers’ expertise

Presentation Number
153P
Lecture Time
12:30 - 12:30
Speakers
  • Hubert Beaumont (Valbonne, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The imaging criteria mRECIST, which was initially introduced to assess the efficacy of HCC vascular bed destruction induced by TACE, is now often used to assess subtler vascular effects induced by anti-VEGFR therapy. However, variability of mRECIST assessment has been only partially investigated. In this study, we evaluated the inter-reader variability and its sources in the mRECIST assessment of treatment response for HCC treated by anti-VEGFR drugs.

Methods

A subset of 41 advanced HCC tumors in 24 patients treated by anti-VEGFR drugs were selected from a phase I/II nalysedre study (the original study). These data were retrospectively reviewed according to mRECIST criteria by 3 mRECIST non-expert radiologists each having different levels of experience. Each liver lesion measurement was hand drawn using an electronic caliper. Results from these 3 radiologists were then compared to those extracted by an mRECIST expert at the time of the original study. The precision of measurements among the 3 non-experts and between the expert and the 3 non-experts were nalysed by assessing bias and standard deviation (SD) using the Bland-Altman method. The agreement of readers’ responses was assessed using the Kappa coefficient statistic. The causes of discrepancies were nalysed.

Results

Among the 3 non-experts, SD of measurements ranged [24.9%; 36.3%] and the Kappa coefficients were moderate 0.41 [0.28; 0.55]. SD in measurements of expert versus non-experts ranged [33.2%; 41.1%] and Kappa coefficients were poor 0.20 [0.06; 0.35]. Pooling the four readers together, the rate of discrepancy at declaring either Progressive Disease (PD) or Partial Response (PR) per patient was identical at 41.7% (10/24). The main cause of discrepancy at declaring PD came from the complexity of HCC enhancement patterns and the poor definition of tumors boundaries. Discrepancies at detecting PR came from the reader variability at selecting only the viable part or the entire liver tumor.

Conclusions

When used by mRECIST non-experts to assess the subtle vascular effect induced on HCC by anti-VEGFR therapy, mRECIST appears to lack reproducibility. It is therefore important when using mRECIST to require specific training to reduce readers variability.

Legal entity responsible for the study

Median Technologies.

Funding

Has not received any funding.

Disclosure

H. Beaumont, N. Faye, C. Klifa: Employee: Median Technologies. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

154P - PD-L1 expression pattern in large cell neuroendocrine carcinoma of the lung.

Presentation Number
154P
Lecture Time
12:30 - 12:30
Speakers
  • Diane Damotte (Paris, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Large cell neuroendocrine carcinomas of the lung (LCNECs) are rare neoplasms with limited therapeutics options. Pathological diagnostic of LCNECs is morphologically based, may be difficult and need immunohistochemical (IHC) analysis. Immune checkpoint inhibitors targeting tumoral and immune cells interaction have changed the NSCLC treatment but few data are available on LCNECs immune environment and particularly the expression of PD-L1 on both tumors (TC) and immune infiltrating (IC) cells. The objective of the present study is to determine the expression and pattern of PD-L1 staining in a cohort of LCNECs patients.

Methods

Clinical files and tumors biopsies of patients (pts) with a LCNEC diagnosed between 01.01.2014 and 31.12.2016 were retrospectively collected (GFPC 03-2017). All histological samples were centrally reviewed by six pathologists, according to the latest WHO 2015 classification. LCNEC was confirmed and PD-L1 expression was determined both in TC and IC, using the anti-PD-L1 antibody 22C3 (kit and automat Dako). PD-L1 expression was scored on TC as the percentage of PD-L1 positive cells (0 to 100%). PD-L1 expression on IC was determined as follows: IC0: positive IC representing < 1% of the tumor surface; IC1 : positive IC representing ≥1% but <5 % of the tumor surface ; IC2 : positive IC representing ≥ 5% but <10% of the tumor surface ; and IC3 : positive IC representing >10% of the tumor surface.

Results

86pts were initially included in the study, 28 (32%) were excluded for non-LCNEC diagnosis. Among the 58 pts with LCNEC, five (8%) had a composite LCNEC with a NSCLC component. The mean age of the population was 65 years, mainly mens (86%) and former or current heavy smokers (93%). PD-L1 was positive on TC for only 12% of the samples, while 76 % of the samples shows IC PD-L1 positive, with respectively 18 (35%) IC3, 8(14%) IC2, and 13(25%) IC1.

Conclusions

LCNEC display a particular PDL1 expression pattern, different from NSCLC and from SCLC and may suggest a potential effectiveness of therapeutic anti PD-L1 antibodies, this hypothesis have to be addressed in clinical trial.

Legal entity responsible for the study

GFPC.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

155P - Inter-rater reliability of programmed death ligand 1 (PD-L1) scoring using the VENTANA PD-L1 (SP263) assay in Non-Small-Cell-Lung Cancer (NSCLC)

Presentation Number
155P
Lecture Time
12:30 - 12:30
Speakers
  • Gareth H. Williams (Cambridge, GB)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The VENTANA PD-L1 (SP263) assay has been developed as a companion diagnostic for anti-PD-L1 immune checkpoint inhibitors. Here we investigate assay inter-rater reliability, applied to PD-L1 scoring of tumour cells (TCs) and immune cell (IC) infiltrates in NSCLC.

Methods

Six expert European pulmonary pathologists independently scored 200 NSCLC samples stained using the VENTANA PD-L1 (SP263) assay. Archival, commercially-sourced formalin-fixed paraffin-embedded resections were selected to represent the dynamic range of PD-L1 expression. Each pathologist scored the proportion of TCs expressing PD-L1 (TM score), tumour-associated IC population as a percentage of total tumour area (PIC value), and percentage of ICs expressing PD-L1 (IC score). Scores were analysed using intra-class correlation coefficient (ICC) and patient classification using Fleiss’ Kappa.

Results

Interim results were available for 3 pathologists and 180 cases. TM scoring between pathologists showed strong pair-wise correlations between individuals (R2>0.90) with an ICC >0.95. Pair-wise and overall agreement was ≥85% for TC ≥ 1% and >93% for TC ≥ 20%, TC ≥ 25%, and TC ≥ 50%. Fleiss’ Kappa showed substantial agreement for TC ≥ 1% and excellent agreement for TC ≥ 20%, TC ≥ 25%, and TC ≥ 50%. There were systematic and substantial differences in PIC values and IC scores between pathologists with poor pair-wise correlations. ICC indicated poor reliability for both IC score (0.36) and PIC values (0.044). Fleiss’ Kappa showed poor agreement for IC ≥ 25% (0.183).

Conclusions

Assessment of TM score in NSCLC was highly reproducible using VENTANA PD-L1 (SP263) assay, building confidence in the accuracy of this assay in patient selection for anti-PD-L1 therapy. However, expert pathologists were unable to reproducibly assess IC score in NSCLC suggesting assessment methodology is unreliable for this tumour type and assay. This contrasts with urothelial cancer (UC) in which pathologist agreement for PIC values and IC scores was generated as part of UC VENTANA PD-L1 (SP263) IHC assay CE marking and FDA approval. This difference in pathology of the different tumour types requires further investigation.

Legal entity responsible for the study

AstraZeneca.

Funding

AstraZeneca.

Editorial Acknowledgement

Editorial assistance, which was in accordance with Good Publication Practice (GPP3) guidelines, was provided by Kaveri Sidhu, PhD, of Cactus Communications (Mumbai, India).

Disclosure

A.G. Nicholson: Consultant: Abbvie and Oncologica; Educational grant: Pfizer. E. Thunnissen: Corporate sponsored research: Oncologica; Honorarium: Roche Ventana, Histogenex; Advisory boards: Roche. P. Cane: Speaker fees: Pfizer, Roche; Fees for interview: AstraZeneca. K.M. Kerr: Advisory boards: Roche, AstraZeneca, BMS, MSD, Pfizer, Merck Serono, Roche Diagnostics. M. Scott, P.W. Scorer, C. Barker: Employment: AstraZeneca; Stockholder: AstraZeneca. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

156P - Relationship between ring-type dedicated breast PET and tumor-infiltrating lymphocytes in early breast cancer

Presentation Number
156P
Lecture Time
12:30 - 12:30
Speakers
  • Shinsuke Sasada (Hiroshima, JP)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

FDG uptake on PET is related to biological parameters and prognosis in breast cancer. The predominance of stromal tumor-infiltrating lymphocytes (TILs) in breast cancer is a biomarker for prognosis and pathological complete response after neoadjuvant chemotherapy. However, whether whole-body PET (WBPET) and dedicated breast PET (DbPET) can reflect the amount of TILs is unclear. This study investigated the relationship between TILs and maximum standardized uptake value (SUVmax) in WBPET and ring-type DbPET.

Methods

A total of 125 invasive breast cancers underwent WBPET and ring-type DbPET and resected specimens were pathologically assessed. The impact on SUVmax on the tumor biological parameters and TILs was retrospectively evaluated. SUVmax was classified as high and low relative to the median values (WBPET-SUVmax: 2.2 and DbPET-SUVmax: 6.0).

Results

SUVmax correlated with tumor size, nuclear grade, Ki-67 labeling index, and TILs in both WBPET and DbPET (all P < 0.001). The cut-off values of tumor size, Ki-67 labeling index, and TILs predicting high SUVmax were 20 mm, 20%, and 20%, respectively. In multivariate analysis, the predictive factors for high SUVmax were tumor size and Ki-67 labeling index for WBPET and tumor size and TILs for DbPET. A high SUVmax in DbPET was related to high numbers of TIL tumors after propensity score matching analysis; however, WBPET was not (P = 0.007 and P = 0.624, respectively).

Logistic regression analysis for predicting high SUVmax tumor on WBPET and DbPET

FactorsUnivariate analysis
Multivariate analysis
Odds ratioPOdds ratioP
<WBPET>
Histology_IC-NST6.710.0031.350.655
T2–36.18< 0.00113.0< 0.001
Nuclear grade 36.48< 0.0012.700.071
ER positive0.390.2850.320.404
HER2 positive1.620.4860.310.211
Ki-67 labeling index ≥ 20%3.030.0053.680.020
TILs ≥ 20%1.990.0842.290.144
<DbPET>
Histology_IC-NST3.180.0031.680.393
T2–34.38< 0.0014.81< 0.001
Nuclear grade 34.43< 0.0011.820.238
ER positive0.980.9290.700.758
HER2 positive0.970.8570.530.450
Ki-67 labeling index ≥ 20%3.540.0011.390.513
TILs ≥ 20%7.50< 0.0016.98< 0.001

Conclusions

Unlike WBPET, the SUVmax in ring-type DbPET can represent the immune microenvironment after adjusting for tumor biological factors. DbPET might be a biomarker of pathological response to neoadjuvant chemotherapy and prognosis.

Legal entity responsible for the study

Shinsuke Sasada.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

157P - Short-term responders of non-small cell lung cancer patients to EGFR tyrosine kinase inhibitors display high prevalence of TP53 mutations and primary resistance mechanisms

Presentation Number
157P
Lecture Time
12:30 - 12:30
Speakers
  • Yun Fan (Hangzhou, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Non-small cell lung cancer (NSCLC) with activating EGFR mutations in exon 19 and 21 usually responds to EGFR tyrosine kinase inhibitors (TKI), but sometimes the responses can only be maintained for a few months. The underlying mechanisms of such short responses have not been fully elucidated.

Methods

The genomic profiles of sixteen short-term responders (SR) that had progression free survival (PFS) of less than 6 months on the first-generation EGFR TKI were interrogated, in comparison to twelve long-term responders (LR) that had more than 24 months of PFS. All patients were diagnosed with advanced lung adenocarcinoma and harbored EGFR 19del or L855R mutation before treatment. Paired tumor samples collected before treatment and after relapse (or at the last follow-up) were subjected to next-generation sequencing of 416 cancer-relevant genes.

Results

SR patients were significantly younger than LR patients (p < 0.001). 88% of SR patients have TP53 variations compared to 13% in LR patients (p < 0.001), and 37.5% SR patients carry EGFR amplification, which is much higher than LR patients (8%). In addition, 12 SR patients (75%) were identified with other potential primary resistance mechanisms in pre-treatment samples, including PTEN loss, BIM deletion polymorphism, amplifications of EGFR, ERBB2, MET, HRAS and AKT2. Comparatively, only 3 LR patients (25%) were detected with EGFR or AKT1 amplification that could possibly exert resistance.

Conclusions

The diversified pre-existing resistance mechanisms in SR patients revealed the complexity of defining treatment strategies even for EGFR sensitive mutations.

Legal entity responsible for the study

Fan Yun.

Funding

Has not received any funding.

Disclosure

X. Tong, X. Wu, Y.W. Shao: Employee: Geneseeq Technology Inc. J. Yan: Employee: Nanjing Geneseeq Technology Inc. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

158P - Assessment of cfDNA in patients with metastatic colorectal cancer treated with cetuximab monotherapy

Presentation Number
158P
Lecture Time
12:30 - 12:30
Speakers
  • Lindsay Angus (Rotterdam, NL)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Third line systemic treatment for patients with RAS wild-type metastatic colorectal cancer (mCRC) includes anti-EGFR monoclonal antibodies such as cetuximab. Here, we examined cell-free DNA (cfDNA) to gain insight into mechanisms underlying primary and acquired resistance in patients with mCRC receiving cetuximab.

Methods

34 patients with RAS wild-type mCRC (KRAS and NRAS exon 2-4) received biweekly cetuximab monotherapy (500mg/m2). cfDNA was isolated from plasma obtained at baseline, after 2 weeks of treatment and at disease progression (PD). ± 20 ng DNA was used for targeted next generation sequencing using the Oncomine™ Colon cfDNA Assay (14 genes, 242 hotspots). Mutation analysis of tumor tissue was performed according standard of care, at least including KRAS and NRAS. Outcome was defined as clinical benefit (CB; PD > 8 weeks, n = 21) versus no CB (NCB; PD ≤ 8 weeks, n = 13).

Results

Baseline cfDNA concentration correlated with the sum of diameters on CT (p = 0.043) and metabolically active tumor volume on [18F]FDG PET (p < 0.001). In 6/13 (46%) of patients with NCB, mutations in KRAS (n = 3) and BRAF (n = 3) were detected in baseline cfDNA. Two KRAS mutations were detected cfDNA, but not in tissue. All BRAF mutations in cfDNA were present in tissue, one BRAF mutation in tissue was not detected in cfDNA. In one patient (5%) with CB a polyclonal KRAS mutation was detected in cfDNA, which was not found in tumor tissue. In 9 patients with CB, cfDNA concentrations were measured and decreased from a median of 45 ng/mL plasma (range 13 – 784 ng/mL) at baseline to 19 ng/mL after 2 weeks of treatment (range 9 – 42 ng/mL) (p = 0.008). In patients with CB an enrichment of mutations in genes associated with resistance (KRAS, NRAS and BRAF) was found in 12/17 (70%) at PD compared to baseline. Moreover, in 8/17 (47%) of these patients EGFR mutations in codons coding for the epitope binding site of cetuximab emerged and in 9/17 patients (53%) multiple mutations in the same gene occurred suggesting the presence of multiple subclones.

Conclusions

A subset of mCRC patients with NCB could be identified based on baseline cfDNA mutations in genes associated with cetuximab resistance. By using cfDNA we can optimize patient selection for cetuximab therapy and elucidate mechanisms of resistance.

Clinical trial identification

NCT02117466.

Legal entity responsible for the study

Erasmus University Medical Center.

Funding

Dutch Cancer Society.

Disclosure

L. Angus: Advisory board: Merck B.V. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

159P - Quantifying circulating cell-free DNA as clinical biomarker

Presentation Number
159P
Lecture Time
12:30 - 12:30
Speakers
  • MEDDEB Romain (Montpellier, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

This is the first comprehensive study on the effect of pre-analytical and demographic parameters that could be a source of variability in the quantification of nuclear and mitochondrial circulating DNA (NcirDNA and McirDNA, respectively).

Methods

We set an optimal calculation of the simultaneous quantification of circulating nuclear and mitochondrial genome copy number based on a clinically validated q-PCR method. We report data from a total of 217 subjects, 99 healthy individuals and 118 metastatic colorectal cancer (mCRC) patients. We also investigated the influence of blood storage and collection time on cirDNA concentration from healthy volunteers.

Results

Approximately 26,650 and 3,000-fold more mitochondrial than nuclear genome copies were found in healthy subjects and mCRC patients, respectively. Neither NcirDNA nor McirDNA plasma concentrations depended on age in the healthy and mCRC cohorts taken as a whole. Remarkably however NcirDNA levels were significantly higher in healthy men as compared to women (n = 99, P = 0.010). Men and women did not differ in McirDNA levels. NcirDNA levels increased slightly with age in healthy women, suggesting a potential influence of menopause. A highly significant statistical difference was found between mCRC patients and healthy individuals for NcirDNA (P < 0.0001) and McirDNA (P = 0.019). In healthy volunteers, there was a higher level of NcirDNA at 9:00 AM with no food intake.

Conclusions

Nuclear and mitochondrial cirDNA levels do not vary in the same way with regards to blood stability, collection time, and pathological status. Our observations, of pre-analytical, analytical and demographical factors, could serve to set standard operating procedures and to transpose cirDNA analysis into clinical practice in oncology. Guidelines on the preanalytical conditions will be also presented from data from this study and a complete review of the literature.

Legal entity responsible for the study

Alain R. Thierry.

Funding

SIRIC Montpellier Cancer Grant INCa_Inserm_DGOS_12553.

Editorial Acknowledgement

We are grateful to A. Bauer and B. Ottolini for their help. We would like to thank the clinical investigators from the Kplex2 study: J.L. Raoul, R. Guimbaud, D. Pezet, P. Artru, E. Assenat, C. Borg, M. Mathonnet, C. De La Fouchardière, O. Bouché, and C. Gavoille for collecting the blood samples. The authors would like to thank Kevin Billings and Streck for providing the Cell-Free DNA BCT CE tubes. This work was supported by the INSERM (Institut National de la Santé et de la Recherche Médicale), Lilly (France), and the SIRIC Montpellier Grant (INCa_Inserm_DGOS_12553), France.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

160P - Gene Expression Profile (GEP) and Survival Among Patients With Advanced Ovarian Cancer

Presentation Number
160P
Lecture Time
12:30 - 12:30
Speakers
  • Estrid Høgdall (Copenhagen, DK)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The ability of T-cell–inflamed GEP (Ayers et al. J Clin Inv. 2017) to predict clinical outcome in ovarian cancer is not fully understood. A retrospective observational study was conducted to evaluate the prognostic value of GEP and its association with programmed death ligand 1 (PD-L1) expression in patients with advanced epithelial ovarian, primary peritoneal, or fallopian tube cancer (OvCa).

Methods

Patients diagnosed as FIGO stages II-IV OvCa from 2004 to 2012 at Aarhus University Hospital and Rigshospitalet, Copenhagen, Denmark, were included. Patients were considered platinum sensitive if treatment-free interval [TFI] was ≥6 months. PD-L1 was assayed using the 22C3 antibody, and positivity was defined as ≥ 1 stained tumor or immune cells per 100 tumor cells. T-cell–inflamed GEP score was defined as low (< –0.318), intermediate (–0.318 to < –0.162), or high (≥ –0.162). The log-rank test and Cox proportional hazards model were used for survival analyses, adjusting for age, stage, histology, residual tumor, surgery type, performance status, platinum sensitivity, and PD-L1 expression status.

Results

Median age of the 376 patients was 63 years (range, 26-86); 9%, 70%, and 20% were FIGO stages II, III, and IV disease, respectively. Of these patients, 80% had type II histologic type, and 76% were platinum sensitive; 49% had a GEP score of low, 16% had intermediate, and 35% had high. Baseline characteristics between GEP groups were similar; PD-L1 and GEP scores were correlated (Spearman r, 0.71; Kendall tau r, 0.57). Median overall survival (OS) was 43 months (95% CI, 38-49) in all patients and was similar for patients with low GEP (41 months) and intermediate GEP (40 months), compared with patients with high GEP (52 months). There was no significant association between GEP status (intermediate/low vs high) and OS among all patients (adjusted hazard ratio, 1.00 [95% CI, 0.72-1.38]), by platinum sensitivity or by PD-L1 expression status.

Conclusions

GEP correlated with PD-L1 expression in patients with advanced OvCa, but OS was not significantly different between GEP categories.

Legal entity responsible for the study

Merck & Co., Inc.

Funding

Merck & Co., Inc.

Editorial Acknowledgement

Medical writing and/or editorial assistance was provided by Matthew Grzywacz, PhD, of the ApotheCom pembrolizumab team (Yardley, PA, USA). This assistance was funded by Merck & Co., Inc., Kenilworth, NJ, USA.

Disclosure

P-T. Vo, W. Zhou: Employee and stockholder: Merck. L. Huang: Employee: Merck. M. Marton: Employee and stockholder: Merck; Stockholder of Biogen, CVS Health, Gilead Sciences, Intuitive Surgical, Johnson and Johnson, Pfizer and United Health Group. S.M. Keefe: Employment: Merck. M. Busch-Sørensen: Employment and stockholder: Merck. T. Steiniche: Research funding: Merck; Travel expenses: Merck. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

161P - Predicting survival benefit of capecitabine plus cisplatin in patients with metastatic gastric cancer patients using quantitative proteomics.

Presentation Number
161P
Lecture Time
12:30 - 12:30
Speakers
  • Dongyao Yan (Rockville, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Capecitabine plus cisplatin (XP) is a standard treatment for metastatic gastric cancer (mGC). Capecitabine activation requires the enzymes uridine-cytidine kinase 2 (UCK2) and orotate phosphoribosyl transferase (OPRT). We previously used mass spectrometry to quantitate UCK2 in tumor samples from 5-FU-treated patients with stage II/III colorectal cancer; UCK2 protein expression > 319 amol/ug of tumor protein was associated with improved survival. Here, we assessed whether these biomarkers would predict survival among mGC patients treated with XP.

Methods

Archived tumor samples from patients with mGC were microdissected and solubilized for mass spectrometric quantitation of 16 protein biomarkers. Kaplan-Meier survival curves were compared using a log-rank test. Multivariate Cox models of survival included clinical covariates and protein biomarkers.

Results

mGC tumor samples from 116 XP-treated patients were analyzed (males: 64%; median age: 55 years). All samples expressed OPRT protein (range: 202 – 1719 amol/ug), and 114 of 116 expressed UCK2 (range: 119 – 933 amol/ug). Patients with UCK2 expression above the pre-defined cutoff of 319 amol/μg (n = 30) had longer time to progression (TTP) (HR: 0.60; p = 0.020) than patients below the cutoff. Results for overall survival (OS) were similar (HR: 0.59; p = 0.015). OPRT protein expression > 790 amol/ug (n = 24) was associated with longer TTP (HR: 0.58; p = 0.019) and longer OS (HR: 0.60; p = 0.029). In multivariate analysis, UCK2 and OPRT remained independent predictors of survival after adjustment for age, gender, ECOG performance status, metastatic sites, and other clinical covariates.

Conclusions

We validated a pre-defined UCK2 expression cutoff and discovered an OPRT cutoff in an XP-treated mGC patient cohort. Patients with tumor expression of UCK2 and OPRT proteins above quantified thresholds survived longer than patients with lower expression. Mass spectrometric quantitation of these common tumor proteins at diagnosis may improve patient selection for XP. Studies to validate these and other chemopredictive biomarkers are ongoing.

Legal entity responsible for the study

NantOmics.

Funding

NantOmics.

Disclosure

D. Yan, E. An, Y. Tian, F. Cecchi, T. Hembrough: Employment: NantOmics. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

162P - Detection of microsatellite instability (MSI) with a novel set of 7 Idylla biomarkers on colorectal cancer samples in a multi-center study

Presentation Number
162P
Lecture Time
12:30 - 12:30
Speakers
  • Bram De Craene (Mechelen, BE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Detection of microsatellite instability (MSI) is recommended for all patients with colorectal cancer (CRC). Current clinical reference methods are immunohistochemical (IHC) staining of mismatch repair (MMR) proteins and/or PCR analysis of frequently mutated repetitive regions of DNA. The prototype Idylla™ MSI Test has been developed using a new set of short homopolymers located in the ACVR2A, BTBD7, DIDO1, MRE11, RYR3, SEC31A & SULF2 genes. This marker set allows probe-based detection with great specificity in a simplified workflow compared to current methods.

Methods

Repeat length with this set of biomarkers was determined on 333 formalin-fixed and paraffin-embedded (FFPE) CRC samples using Idylla™ MSI Test prototype cartridges, which allow a fully automated workflow including sample preparation, DNA amplification and automated repeat length calling. A neural network based algorithm was built on a large cohort of reference/patients samples (n > 3000) obtained from different clinical sites (n > 10) and different ethnic groups (n = 5). Three-hundred fourteen samples were characterized by means of the Promega MSI analysis system and 272 samples by means of MMR protein IHC staining. Approximately 30% of the samples included in the study were previously characterized to be MSI-H by either one of these methods.

Results

Concordance analysis revealed an overall agreement of 98.7% (96.7%-99.5% 95% CI) with Promega and 97.6% (94.8%-98.9% 95% CI) with IHC analysis. Analysis of consecutive sections of 182 samples with the three methodologies revealed a higher number of invalid results for Promega (3.8%) and IHC (13.2%) compared to the prototype Idylla™ MSI Test (2.2%).

Conclusions

This study verified the robustness of the prototype Idylla™ MSI Test including novel MSI biomarkers to discriminate MSI-H from MSS status on a large and diverse set of CRC samples. The study was conducted in multiple centers demonstrating the possibility of a rapid and fully automated analysis for MSI testing close to the point of need. The prototype Idylla™ MSI Test provided accurate and reliable results within 150 minutes from just one FFPE tumor section (no normal reference sample required).

Legal entity responsible for the study

Biocartis NV.

Funding

Biocartis NV.

Disclosure

B. De Craene, J. Van de Velde, E. Bellon, M. Gazin, E. Rondelez, L. Vandenbroeck, T. Vanhoey, N. Elsen, K. Decanniere, E. Sablon, G.G. Maertens: Employee: Biocartis NV. E. Watkin: Advisory board, speaker’s fees: MSD. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

163P - Accurate measurement of tumor mutation burden in liquid biopsy (bTMB) using a 500 gene panel

Presentation Number
163P
Lecture Time
12:30 - 12:30
Speakers
  • Tingting Jiang (San Diego, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Tumor mutation burden (TMB) has been shown as a new predictive biomarker for immune checkpoint inhibitor in various cancer types. It is typically measured by processing tumor tissue with substantial input which limits its clinical utility in patients with metastatic or unresectable disease. Meanwhile, there is increasing interest in circulating tumor DNA (ctDNA) that act as a noninvasive real-time biomarker for cancer patients. Therefore, here we develop a new next generation sequencing assay that can identify patients with sufficient tumor fraction in plasma and accurately measures the TMB from blood (bTMB).

Methods

Cell free DNA (cfDNA) was extracted from plasma across four original tissue types by different tumor stages. CfDNA Assay was performed with unique molecular identifier (UMI) , sequenced on the Illumina® platforms and analyzed using an internal pipeline for variants down to 0.4%. By integrating the fragment size distribution and the clonal mutation frequency, we were able to estimate the tumor fraction per plasma sample. A bTMB score was also derived using all the coding variants on a 1.3M panel across 500+ genes. The matched TMB score is derived by the same assay using FFPE tissue.

Results

Our assay has generated sufficient results from 1-4ml plasma for variant detection down to 0.4%. Across four tissue types by various cancer stages, our assay and pipeline yield a variant concordance of 70% between cfDNA and FFPE. Majority of mutations only found in plasma may be associated with clonal hematopoiesis in genes such as TET2, DMBT3A and etc. By combining fragment size distribution and driver mutation frequency, we were able to estimate tumor fraction in plasma. Tumor fraction in plasma is significantly associated with tumor stage that >50% of metastatic cancers and > 25% early stage lung cancers contain high tumor fraction. In patients with at least 1% tumor content, there is high correlation between bTMB measured by plasma and TMB measured by FFPE (R2=0.92).

Conclusions

We have developed a ctDNA assay to detect somatic variants and determine bTMB with high accuracy and precision with input as low as 10 ng of cfDNA. Our assay yield accurate measurement of TMB compared to tissue biopsy.

Legal entity responsible for the study

Illumina, Inc.

Funding

Has not received any funding.

Disclosure

T. Jiang, S. Zhang, A. Jager, S. Katz, J. Lococo, P. Le, B. Andrian, C. Zhao, D. Baker, T. Pawlowski, S. Bilke: Employee and shareholder: Illumina, Inc.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

164P - Circulating tumor DNA and RNA as an exploratory biomarker to evaluate GT0918 in a Phase 1/II clinical trial in mCRPC patients

Presentation Number
164P
Lecture Time
12:30 - 12:30
Speakers
  • Mengyao Tan (Hayward, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Metastatic castration resistant prostate cancer (mCRPC) is a complex disease with distinct molecular features in relation to genomic instability and selective treatment pressure. Circulating tumor DNA and RNA fragments (ctDNA & ctRNA) found in blood offers the potential of disease diagnosis, monitoring and resistance mechanism interrogation by detecting genomic alterations from tumor. We explored ctDNA & ctRNA-based biomarkers from patient blood to assess their associations with clinical response of GT0918, a potent AR antagonist, in a Phase 1/2 clincal study in mCRPC who progressed after abiraterone or enzalutamide and docetaxol, or cannot tolerate either or both therapies.

Methods

We performed a retrospective analysis of blood samples from mCRPC patients collected at baseline, on- and after study during the trial. A highly sensitive ctDNA- & ctRNA-based NGS assay was developed to detect mutation, copy number gain, fusion and splicing variants. Statistical analyses were performed in R.

Results

20 blood samples were collected at multiple time points from 8 patients. CtDNA-based variants are detected in all of patients. The most frequent mutations are TP53 (55.0%) and AR (30.0%). Combined mutation rates in PTEN-PI3K-AKT and DNA damage repair pathways (BRCA1/BRCA2/ATM) are both 35.0%. Importantly, AR hotspot mutations (W742C, T878A, and S889G) and amplifications are detected in 4 subjects. AR splicing variants (AR-V3, AR-V7) were found in 3 patients by ctRNA assay. Interestingly, one AR-V3+ patient became negative during the treatment accompanied by a decrease of other molecular biomarkers including prostate-specific SPOP mutation and cfDNA yield. In contrast, another patient who was AR-V3+ at C4D1, had constantly high AR amplification and increasing cfDNA yields over treatment. Last, as a hallmark of prostate cancer, TMPRSS2-ERG fusion was also detected in 2 patients.

Conclusions

This is a preliminary study to explore genomic alterations in mCRPC in response to GT0918 treatment. As a non-invasive assay, the ctDNA & ctRNA-based assay was highly sensitive and provided useful molecular insights for monitoring treatment effect and deciphering drug sensitivity & resistance mechanisms.

Clinical trial identification

NCT02826772.

Legal entity responsible for the study

Suzhou Kintor Pharmaceuticals, Inc.

Funding

Suzhou Kintor Pharmaceuticals, Inc.

Disclosure

M. Tan, S. Zhang, Z. Zhao, A. Wang, D. Cheung, P. Du, J. Yu: Sockholder: Predicine, Inc. S. Jia: CEO and stockholder: Predicine, Inc. C. Guo, Y. Tong, K. Zhou: Stockholder: Suzhou Kintor Pharmaceutials, Inc.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

165P - The frequency of primary cilia, CD8+ tumor infiltrating lymphocytes and PD-1 expression in renal cell carcinoma of clear-cell type

Presentation Number
165P
Lecture Time
12:30 - 12:30
Speakers
  • Josef Dvorak (Prague, CZ)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Primary cilium (PC) is considered to represent a functional homologue of the immune synapse due to morphological and functional similarities in architecture. Both microtubule structures, i.e. primary cilia of cancer-associated fibroblasts and immune synapses between cytotoxic CD8+ tumor infiltrating lymphocytes (TILs) and antigen-presenting or cancer cells, are regularly found in varying amounts in the microenvironment of solid tumors. These could, in fact, represent two sides of the same coin. However, so far both parameters have not been evaluated simultaneously within the same group of patients.

Methods

The presence of PC in cells, programmed cell death protein-1 receptor (PD-1) expression and the frequency of intraepithelial CD8+ TILs was retrospectively evaluated in tumor tissue blocks of the resected specimens of the kidney in 104 patients with renal cell carcinoma of clear-cell type, 71 males and 33 females, with a median age of 64 years (range 38-82 years). Twenty-eight patients had stage I, 15 stage II, 31 stage III and 30 patients had stage IV tumor. Grade was as follows: grade 1 in 27 patients, grade 2 in 15 patients, grade 3 in 31 patient and grade IV in 30 patients.

Results

The median frequency of PC was 0.0028% (0-0,0465%). The frequency of intraepithelial CD8+ TILs was negative in 1 patient, <25% in 63, 26-50% in 29 and 26-50% in 11, respectively. The expression of PD-1 was <5% in 52 patients, 5-25% in 34 patients, 26-50% in 13 patients, 51-75% in 4 patients and 75% in 1 patient. During the follow-up, recurrence occurred in 42 patients and 43 patients died. Median PFS was 44% (95% CI: 34-67%) and median OS was 98% (95% CI: 84-117%).

Conclusions

The present study provides the first data on the potential association frequency of PC, PD-1 and CD8+ TILs in patients with renal cancer.

Legal entity responsible for the study

MH CZ-DRO (TH, 0064190).

Funding

MH CZ-DRO (TH, 0064190) and Ministry of Defense of the Czech Republic - long-term organization development plan Medical Aspects of Weapons of Mass Destruction of the Faculty of Military Health Sciences, University of Defense.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

166P - Association of MMR protein expression and MMR gene mutations in Chinese colorectal cancer patients

Presentation Number
166P
Lecture Time
12:30 - 12:30
Speakers
  • Shuang Wang (Guangzhou, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

DNA mismatch repair (MMR) deficiency is a genetic abnormality that has important clinical implications related to therapeutic option, familial cancer risk assessment, and checkpoint inhibitor response. It occurs in approximately 15% of colorectal cancers (CRC). Associations between MMR protein expression, microsatellite instability (MSI), and gene mutations remain under investigation.

Methods

Thirty-one FFPE samples from primary CRC patients (pts) were collected for immunohistochemistry (IHC) assay of MMR proteins, PCR-based MSI assay if available and NGS-based panel assay. Genomic alterations including single base substitutions, short and long insertions/deletions, copy number variations, and gene rearrangement were assessed. MSI status were predicted based from NGS data.

Results

Out of 31 sample, 12 CRC were identified as MMR deficiency (dMMR) by IHC including 5 males and 7 females (median age: 66 years), and 19 samples as MMR proficiency (pMMR) including 13 males and 6 females (Median age: 53 years). Eight of the 12 dMMR samples (67%) harbored at least one MMR gene mutations predicted as loss of functions,[u1] including nonsense mutations or truncations in MSH2, MSH6 or MLH1. No MMR gene mutation was detected in any of the 19 pMMR samples (p value<0.001). In addition, 2 BRAF and 6 KRAS hotspot mutations were detected in dMMR samples, and 8 KRAS, 1 BRAF and 1 NRAS mutations in pMMR samples. NGS panel based MSI algorithm successfully predicted the MSI status of all the 31 samples with 100% concordance with the MMR results. Neither copy number variation nor rearrangement was detected. Five pMMR samples were identified as microsatellite stability (MSS) by PCR, 6 dMMR were high level of microsatellite instability (MSI-H), and the rest were failed due to DNA contents.

nMMR gene mutations (n, %)KRAS mutations (n, %)NRAS mutations (n, %)BRAF mutations (n, %)NGS MSI-H status (n, %)
dMMR (IHC)128 (66.7%)6 (50%)0 (0%)2 (16.7%)12 (100%)
pMMR (IHC)190 (0%)8 (42.1%)1 (5.2%)1 (5.2%)0 (0%)

Conclusions

We observed a significant association between MMR deficiency and MMR gene mutations from deep DNA sequencing. The results suggested that CRC pts with MMR gene mutations could be more likely to have dMMR status.

Legal entity responsible for the study

OrigiMed.

Funding

Has not received any funding.

Disclosure

D. Chen, X. Dong, J. Hu, G. Chirn, W. Shi, M. Yao: Employee: OrigiMed. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

167P - Somatic and germline mutations of Chinese gastric cancer patients

Presentation Number
167P
Lecture Time
12:30 - 12:30
Speakers
  • Yusheng Wang (Taiyuan, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Gastric cancer is a high incidence malignancy in China, but the disease is often diagnosed at an advanced stage with limited therapeutic options and poor prognosis. Unfortunately, limited progress of target and immune therapy has achieved in gastric cancer due to tumor heterogeneity and the lack of effective biomarkers in the clinical. Thus the precise understanding of gastric cancer genomic profiling is urgent for exploring clinical strategy of this malignancy.

Methods

Formalin Fixed Paraffin Embedded (FFPE) samples of 110 Chinese gastric cancer patients were collected for next-generation sequencing (NGS)-based 450 genes panel assay. Genomic alterations including single base substitution, short and long insertions/deletions, copy number variations, gene fusions and rearrangement were assessed. Microsatellite instability (MSI) status and tumor mutational burden (TMB) were also acquired by NGS algorithm.

Results

There were 71 males (64.6%) and 39 females (35.4%) diagnosed as gastric cancer with a median age of 62 years old. The most frequent genomic alterations in Chinese gastric cancer patients were revealed as TP53 (66.4%), ARID1A (18.2%), LRP1B (17.3%), CDH1 (15.5%), CCNE1 (14.5%), FAT4 (13.6%), ERBB2 (11.8%), SMAD4 (10.9%), KMT2D (10.9%), RNF43 (10.9%), TGFBR2 (10.0%), KRAS (10.0%), APC (10.0%). Some actionable aberrantly activated mutations were identified in genes such as FGFR1/2/3 (10.0%), BRCA1/2 (6.4%) and PI3K/mTOR pathway including FBXW7, MTOR, PIK3CA, PTEN, STK11, TSC1 and TSC2 (20.0%). In total, 39.1% of patients carried one or more actionable genomic alterations, which indicated the potential clinical benefits of targeted therapies for them. The percentage of patients with germline mutations was 9.1% in current cohort. These germline mutation genes were mainly in homologous recombination (HR) pathways such as BRCA1/2, ATM and BARD1.

Conclusions

In this study, 39.1% of Chinese gastric cancer patients carried actionable genomic alterations which could potentially guide and influence personalized treatments. About 9.1% patients harbored germline mutations mainly in HR pathway. All these genomic identifications in Chinese gastric cancer cohort may provide useful information on drug discovery and biomarker exploration.

Legal entity responsible for the study

OrigiMed.

Funding

Has not received any funding.

Disclosure

Y. Zheng, Q. Cui, A. Wang, H. Chen, W. Shi, K. Wang, M. Yao: Employee: OrigiMed. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

168P - Validation of the MammaTyper® pathological complete response (pCR)-score as a predictor for response after neoadjuvant chemotherapy (NACT) in patients with early breast cancer (BC)

Presentation Number
168P
Lecture Time
12:30 - 12:30
Speakers
  • Peter A. Fasching (Erlangen, DE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Prediction of the response to NACT in early BC patients (pts) can lead to improved treatment decisions. The MammaTyper® test can be used to predict the probability of pCR after NACT by integrating accurate and reproducible assessment of ERBB2, ESR1, PGR and MKI67 mRNA into a standardized prediction model (Varga et al. Breast Cancer Research 2017).

Methods

Total RNA was extracted from formalin-fixed paraffin-embedded (FFPE) tumor samples of pts with cT1-2 BC enrolled in in the single arm phase II TECHNO trial (Untch et al. JCO 2011) and the randomised phase III PREPARE trial (Untch et al. Ann Oncol. 2011). MammaTyper®, a molecular in vitro diagnostic RT-qPCR test, was used to assess the expression of ERBB2 (HER2), ESR1 (ER), PGR (PR) and MKI67 (Ki67) genes from which a predefined continuous score was calculated. The study aimed to validate the MammaTyper® pCR-score for predicting pCR (ypT0 ypN0) after NACT in BC. Pts were classifed into a low or high score group according to a predefined cutoff: score ≤41 predicts a low probability of pCR; score ≥42 predicts a high pCR rate.

Results

A total of 324 pts with available FFPE samples and MammaTyper® measurements were analyzed. The MammaTyper® score was significantly associated with an increased pCR rate (AUC=0.805, p < 0.001). Similarly, pts with high MammaTyper® score (N = 159) had more frequently pCR compared to pts with low score (N = 165) (30.2% vs 3.0%, respectively; OR = 13.84 [95%CI 5.34-35.86], p < 0.001). In addition, the MammaTyper® pCR-score remained significantly predicitve when adjusted for age, nodal status, tumor grade and treatment (OR = 10.90 [95%CI 3.38-35.16], p < 0.001). Within the non-pCR subgroup, pts with low score had a significantly longer disease-free (DFS) and overall (OS) survival compared to pts with high score (DFS HR = 1.59, 95%-CI 1.04-2.44, p = 0.032; OS HR = 2.85, 95%-CI 1.62-5.00, p < 0.001).

Conclusions

The MammaTyper® pCR-score predicts pCR after NACT and seems to improve prognosis additionally to clinical predictors in pts with early BC. Its utility with regard to conventional ER, PR, Ki67 and HER2 has to be analyzed in future studies.

Legal entity responsible for the study

GBG Forschungs GmbH and BioNTech Diagnostics GmbH, Mainz.

Funding

BioNTech Diagnostics GmbH, Mainz, Germany.

Disclosure

P.A. Fasching: Personal fees: Celgene, during the conduct of the study; Grants and personal fees: Novartis; Personal fees: Pfizer, Roche, Teva, outside the submitted work. M. Laible: Personal fees: BioNTech Diagnostics GmbH, outside the submitted work; Employee: BioNTech Diagnostics GmbH); Patent WO 2015/024942 pending, new patent application pending. K.E. Weber: Grants and non-financial support: BioNTech Diagnostics GmbH, Mainz, Germany, during the conduct of the study; Patent EndoPredict issued. C. Denkert: Personal fees: Teva, Novartis, Pfizer, Roche, Amgen, MSD Oncology; Other: Sividon Diagnostics, outside the submitted work. K. Schlombs: Personal fees: BioNTech Diagnostics GmbH, outside the submitted work; Patent WO 2015/024942 pending; New patent application pending. F. Marme: Personal fees; Roche, AstraZeneca, Pfizer, Tesaro, Novartis, Amgen, PharmaMar, GenomicHealth, CureVac, Eisai, outside the submitted work. S. Loibl: Grants: AbbVie, Amgen, AstraZeneca, Celgene, Novartis, Pfizer, Roche, Teva, Vifor during the conduct of the study and outside the submitted work. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

169P - Selective induction of PD-L1 expression in plasma-derived exosomes by gemcitabine-nab-paclitaxel vs. FOLFIRINOX in pancreas cancer

Presentation Number
169P
Lecture Time
12:30 - 12:30
Speakers
  • Eleonora Rofi (Pisa, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Pancreatic ductal adenocarcinoma (PDAC) is considered a poorly immunogenic tumor and treatment with immune checkpoint inhibitors lacks efficacy in this disease. Recently, the use of immune checkpoint inhibitors with radiation therapy in PDAC has demonstrated synergistic activity. The aim of this study was to evaluate the effect of FOLFIRINOX and GEMnPAC on PD-L1 expression in plasma-derived exosomes of PDAC patients.

Methods

Four ml of plasma were obtained at baseline (before initiation of chemotherapy) and at the time of first radiological evaluation (3 months) from patients undergoing first-line FOLFIRINOX or GEMnPAC chemotherapy. Exosomes and RNA extraction from plasma were performed using the exoRNeasy kit (Qiagen®, Valencia, CA, USA); PD-L1 expression was evaluated by digital droplet PCR (Bio-Rad®, Hercules, CA, USA).

Results

A total of 22 pancreatic cancer patients were enrolled in this study; 15 (68.2%) were treated with GEMnPAC and 7 (31.9%) with FOLFIRINOX. In the GEMnPAC group one patient had a partial response (RP), 11 patients had stabilization of disease (SD) and 3 progressed (PD). In the FOLFIRINOX group there were 1 RP, 5 SD and 1 PD. Eleven patients treated with GEMnPAC had a significant increase of PD-L1 expression at 3 months vs. baseline. Indeed, the mean PD-L1 copies/ml was 90 at baseline and 170 at 3 months (p = 0.02). On the contrary, in the FOLFIRINOX group, PD-L1 levels were increased in 3 patients and remained stable/decreased in 4 subjects; the mean baseline copies/ml were 70 vs. 80 at 3 months (p = 0.4). The selective induction of PD-L1 expression was independent from tumor response.

Conclusions

These pilot data suggest that, due to its ability to increase PD-L1 expression, GEMnPAC regimen may be used as induction-treatment for immunotherapy in pancreatic cancer, either sequentially or concomitantly.

Legal entity responsible for the study

Romano Danesi.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

170P - Ecological diversity indices as measurements of tumor heterogeneity correlates with clinical outcomes in late stage small cell lung cancer (SCLC)

Presentation Number
170P
Lecture Time
12:30 - 12:30
Speakers
  • Stephanie Yaung (Pleasanton, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Tumor heterogeneity is difficult to characterize with tissue biopsies in late stage cancers, in which primary and metastatic tumors may have diverged in their mutational profiles. Assessing the diversity of mutations detected by sequencing circulating tumor DNA (ctDNA) from liquid biopsies can quantify the genetic complexity of tumors shedding DNA into the blood.

Methods

In a prospective, observational study, we obtained pre-treatment plasma samples from 56 subjects with Stage IV small cell lung cancer (SCLC) treated with first-line chemo or chemoradiation therapies. Plasma samples were analyzed with the AVENIO ctDNA Surveillance Kit, a targeted next-generation sequencing panel of 198 kilobases. We applied the Shannon and Simpson diversity indices by considering each somatic variant as a species and the number of detected duplex molecules with that mutation as the abundance of that species. Samples were ranked as low tumor heterogeneity if their plasma variant diversity score was below the first tertile of the cohort.

Results

Stage IV SCLC subjects with low tumor heterogeneity evaluated by the Shannon diversity had shorter overall survival (hazard ratio = 1.8; 95% CI 1-3.3; log-rank p = 0.034; median survival difference = 4.5 months). Furthermore, subjects with low tumor heterogeneity evaluated by the Gini-Simpson or inverse Simpson diversity index had shorter overall survival (hazard ratio = 1.8; 95% CI 1-3.3; log-rank p = 0.033; median survival difference = 4.5 months).

Conclusions

The molecular barcoding scheme in the AVENIO kit allows for each strand of the original double-stranded ctDNA molecule to be tracked. From this reconstructed profile of circulating duplex molecules harboring tumor variants, we derived a tumor heterogeneity measure based on the Shannon and Simpson diversity indices commonly used in ecology. We found that late stage SCLC subjects with low tumor heterogeneity had shorter overall survival, suggesting that highly heterogeneous SCLC tumors may respond better to chemotherapy or radiation. Studies to further validate these findings are ongoing.

Legal entity responsible for the study

Roche Sequencing Solutions, Inc.

Funding

Roche Sequencing Solutions, Inc.

Disclosure

S. Yaung, L. Xi, C. Woestmann, S. McNamara, B. Hinzmann, S. Froehler, C. Ju, A. Balasubramanyam, H-P. Adams, B. Wehnl, X.M. Ma: Employment: Roche. N. Tikoo: Employment: Roche; Stock ownership: BeiGene, Celgene, Denali Therapeutics, Exelixis, Gilead Sciences. F. Lasitschka: Consulting role: Roche; Research funding: Boehringer Ingelheim International GmbH. M. Meister, M. Schneider: Research funding: Roche. F.J.F. Herth: Stock ownership: Roche, Lilly, Novartis; Consulting role: Novartis. T. Muley: Royalty sharing agreement: Roche; Research funding: Roche. J. Palma: Employment: Roche; Stock ownership: Exact Sciences, Clovis Oncology, Johnson & Johnson, Roche. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

171P - MET exon 14 splicing mutation and its correlation with clinocopathological features in subjects with non-small cell lung cancer

Presentation Number
171P
Lecture Time
12:30 - 12:30
Speakers
  • Darko Katalinic (Zagreb, HR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Driver mutations are genomic alterations important for tumor initiation and growth. They are found in genes that control cellular proliferation and long-term survival. Mesenchymal-to-epithelial transition (MET) exon 14 splicing mutation occurs in about 3% of cases of non-small cell lung cancer (NSCLC). It has been recognized as an important biomarker to predict response to MET tyrosine-kinase inhibitor therapy. The aim of this study was to investigate possible connection among the MET exon 14 mutations and genomic as well as clinicopathological features in patients with NSCLC.

Methods

The study was performed among 270 patients (58% males and 42% females; mean age of 57.14 ± 16.48 years) with histologically confirmed diagnosis of NSCLC. The distribution of MET exon 14 splicing mutation was detected using the quantitative real-time polymerase chain reaction restriction fragment length polymorphism assay. RNA was extracted from formalin-fixed paraffin-embedded samples. The study was conducted according to the Declaration of Helsinki, the protocol was reviewed and approved by the institutional Ethics committee and all patients provided written informed consent.

Results

MET exon 14 splicing mutation was detected in 9 patients (3.4%). It was found in 7 adenocarcinomas (18.9%) and in 2 squamous cell carcinomas (5.4%). Most adenocarcinomas occurred in females and non-smokers. Squamous cell carcinoma predominantly occurred in male smoking patients. All subjects with MET exon 14 splicing mutation had earlier pathology stage of disease (IA, IB, IIA, IIB) (31%) and older age (>75 years) (43%). Overall survival (OS) of these patients were statistically longer than in patients with KRAS and EGFR mutations (2.2 vs. 1.3 months and 2.4 vs. 1.8 months).

Conclusions

We found that MET exon 14 splicing mutation occurs at a frequency of 3.4%, in older age, and mostly in early stage of NSCLC. OS of patients harboring MET exon 14 splicing mutation has lasted longer than in patients harboring KRAS and EGFR mutations. Patients with MET exon 14 splicing mutation may respond well to MET tyrosine-kinase inhibitor therapy. Further studies are needed to confirm our results.

Legal entity responsible for the study

Cancer Genomic Group.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

172P - Cancer Stem Cell Markers in Pancreatic Ductal Adenocarcinoma

Presentation Number
172P
Lecture Time
12:30 - 12:30
Speakers
  • Fuat Aksoy (Bursa, TR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Pancreatic ductal adenocarcinoma (PDAC) is one of the leading causes of cancer deaths in both men and women world-wide. Despite development of scientific, PC remains lethal disease. Recently, cancer stem cells (CSC) of pancreatic cancer that often are resistant to treatment have been identified. Present study aims to investigate the existence and prognostic value of CSC markers in PDAC patients.

Methods

52 hematoxylin and eosin-stained slides cut from formalin-fixed, paraffin-embedded (FFPE) PDAC tissues were evaluated by a pathologist, and the areas of the slide representing tumor and normal were identified. The samples were analyzed for the presence and differential expression of LGR5, CD44 and CD133 using RT2 Profiler PCR Array Data Analysis (http://www.sabiosciences.com/pcr/arrayanalysis.php) to compare the PCR array analysis results and the characteristics of the tumors and cases.

Results

Of the 52 patients, 29 were men and 23 were women, with an average age of 63 years (range, 26-91 years). All patients underwent pancreaticoduodenectomy (Whipple prosedure). The median size of tumors was 2.3 cm (range, 0.5-6.0 cm). Lymphatic, vascular, and perineural invasions were observed in the tumors of 23 (44.2%), 9 (17.3%), and 5 (9.6%) patients, with 2 patients showing concurrent lymphatic, vascular, and perineural invasions. Tumors were classified as stage IA (n = 4), stage IB (n = 11), stage IIA (n = 7), stage IIB (n = 10), and stage III (n = 22). All surgically resected specimens showed negative (R0) resection margin status. The CD44 was not significantly expressed in eCC tumors compared to normal tissue. LGR5 and CD133 expression level was significantly higher in tumors than in corresponding normal tissues (4.5 fold, P = 0.034; 3.7 fold P = 0.045, respectively). Increased CD133 expression was associated with ampulla vateri tumor localization (P < 0.001). Over expression of LGR5 and CD133 were associated with short overall survival.

Conclusions

Here, we report that CD133 and LGR5 may acts as a functional CSC in the aggressivite of PDAC. Our results suggest that these molecules may serve as a candidate prognostic biomarker and target for new therapies in PDAC.

Legal entity responsible for the study

Ekrem Kaya.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

173P - Tumor mutation burden assessment on FFPE samples using a targeted next-generation sequencing assay

Presentation Number
173P
Lecture Time
12:30 - 12:30
Speakers
  • Ruchi Chaudhary (South San Francisco, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Tumor mutation burden (TMB) predicts durable benefit from immune checkpoint blockade in several cancer types. We demonstrate the ability of a targeted panel with fast turn-around time and low input needs to estimate TMB from research samples.

Methods

We developed a single sample analysis workflow to estimate mutation burden (TMB; nonsynonymous mutations/Mb) from FFPE and fresh frozen tumor research samples. The assay utilizes a PCR-based targeted panel that covers 409 genes and 1.7 Mb of genomic regions. The workflow requires only 10 ng of input DNA and enables a 2.5-day turn-around time from sample to the final report. Sequencing is performed on high throughput semiconductor sequencing platform to achieve sufficient depth (∼500x coverage) and accuracy. The workflow is tumor sample only, with no matched normal sample required; germ-line variants, along with background noise, are removed through filters based on population databases. The assay is research use only, not for diagnostic procedures.

Results

A comparison with whole exome sequencing (WES) on 12 FFPE tumors, where WES was performed on tumors and their matched normal using Agilent’s exome enrichment kit (∼150x coverage for tumor; ∼100x coverage for normal) on illumina platform and our assay ran on tumors only, showed high correlation (r2=0.83) between TMB estimates by our assay with that from WES. To assess reproducibility, we compared raw somatic mutations/Mb in library replicates for a cohort of 21 FFPE research samples (19 CRC, 2 Melanoma) and observed high correlation (r2=0.97). Our pipeline identifies mutation signatures consistent with specific mechanisms such as UV and tobacco damage, as well as substitutions from FFPE processing.

Conclusions

A simple workflow has been developed on the Ion Torrent sequencing platform with an AmpliSeq panel to estimate TMB from FFPE and fresh frozen tumor research samples. This solution will advance research in immuno-oncology.

Legal entity responsible for the study

Thermo Fisher Scientific.

Funding

Thermo Fisher Scientific.

Disclosure

R. Chaudhary, D. Cyanam, V. Mittal, W. Tom, J. Au-Young, C. Allen, S. Sadis, F. Hyland: Employment: Thermo Fisher Scientific.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

174P - Microsatellite instability-high (MSI-H) colorectal cancers with elevated microsatellite alterations at selected tetranucleotide repeats (EMAST) signature represent a target population for immune checkpoint and DNA damaging therapies

Presentation Number
174P
Lecture Time
12:30 - 12:30
Speakers
  • Kien Thiam Tan (Taipei, TW)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Elevated microsatellite alterations at selected tetranucleotide repeats (EMAST) is a different type of genomic instability in colon cancer (CRC) in contrast to mono-, and dinucleotide based instability microsatellite instability (MSI). In this study, we performed comprehensive genomic profiling (CGP) of CRC patients with different EMAST and MSI status to understand their genomic structure, which may help match them with relevant therapies.

Methods

99 formalin-fixed, paraffin-embedded (FFPE) CRC tissues consisting of four subtypes based on their EMAST and MSI status, namely (1) EMAST+ and MSI-high (MSI-H), (2) EMAST+ and microsatellite‐stable (MSS), (3) EMAST- and MSI-H, and (4) EMATS- and MSS, were subjected to next-generation sequencing (NGS) with a 440-gene panel to identify mutations and copy number variants (CNVs). Tumor mutational burden (TMB) was determined using mutations detected on exonic regions sequenced while CNV index was calculated to infer genome instability.

Results

In line with previous studies, the prevalence of TP53 (17.6%; n = 3) and APC (23.5%; n = 4) mutations was much lower whereas BRAF V600 mutation (41.2%; n = 7) was much higher in the subtype (1) CRCs which had both MSI-H and EMAST signatures. Interestingly, these dual positive tumors had a significant higher TMB and lower CNV index than other subtypes (TMB: (1) vs (2), (3) and (4); 54 vs 19, 25; and 16; p < 0.0001, CNV index: (1) vs (2), (3), and (4); 3.9 vs 13.7, 9.9, and 17.8; p < 0.0114, 0.006, and 0.0003), suggesting there are more likely to benefit from immune checkpoint inhibitors. Notably, ATM and ARID1A genes mutated in a mutually exclusive way in up to 13/17 (76.5%) of tumors with MSI-H and EMAST signatures, which may predict treatment benefit from the PARP inhibitors.

Conclusions

MSI-H and EMAST+ CRCs show distinctive genomic features that give them the potential opportunity for checkpoint inhibitors in combination with PARP inhibitors.

Legal entity responsible for the study

ACT Genomics.

Funding

Has not received any funding.

Disclosure

K.T. Tan, S-J. Chen: Shareholder: ACT Genomics. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

175P - Levels of circulating fibroblast growth factor (FGF) 23 and prognosis of cancer patients with bone metastasis

Presentation Number
175P
Lecture Time
12:30 - 12:30
Speakers
  • André B. Mansinho (Lisbon, PT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The FGF signaling network plays a key role in tumorigenesis and is recognized as a potential therapeutic target. FGF23 is predominately expressed in bone osteocytes and can act as an autocrine, paracrine and/or endocrine growth factor in cancer. In this study, we aimed to assess the role of circulating FGF23 levels in the prognosis of cancer patients with bone metastases.

Methods

This study included a cohort of 112 patients with cancer (63% breast;16% prostate) and metastatic bone disease treated with bone targeting agents (BTA), in which serum baseline FGF23 was quantified by ELISA and further dichotomized in two groups (FGF23high and FGF23low). Cut-off was defined by mean + one standard deviation. The association of FGF23 with overall survival (OS) and with time to skeletal related events (TTSRE) was investigated. Time to event outcomes was calculated using the Kaplan-Meier method and tested using univariate/multivariate Cox regression models controlling for established prognostic factors across patients with solid tumors and bone metastases: extra-bone involvement, urinary N-terminal telopeptide (uNTX), presence of bone fractures, and calcemia.

Results

Mean FGF23 was 38.16 ± 26.15 pg/mL (interquartile range [IQR] 19.77-50.72). 16.8% of patients were classified as FGF23high (n = 19). Baseline characteristics were balanced between groups, except for the median uNTX level, which was higher in the FGF23high group (824.30 vs 118.02 nmol BCE/mmol creatinine, p = 0.040). Median time from beginning of BTA was similar between groups (1.28 vs 1.10 months, p = 0.161). After a median follow-up of 26.0 months (IQR 13.0-47.0), median OS was 34.4 months in the FGF23low group and 12.2 months in the FGF23high group (multivariate HR 0.18, 95% CI 0.07 – 0.44, p = 0.001; univariate p = 0.001). As a continuous variable, FGF23 at baseline kept its prognostic association (p = 0.001). Patients with FGF23low status at baseline had a longer TTSRE (13.0 vs 2.0 months, p = 0.04).

Conclusions

In this exploratory cohort, patients in the FGF23low group had a longer OS and TTSRE. Further studies are warranted to define its role as a prognostic biomarker and as a potential predictor of response to drugs targeting the FGF axis.

Legal entity responsible for the study

Instituto de Medicina Molecular.

Funding

Has not received any funding.

Editorial Acknowledgement

Joana Cavaco Silva - Medical Writer (Medical Oncology Department - Hospital de Santa Maria - Centro Hospitalar Lisboa Norte).

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

176P - Clinical significance of RCAS1 and CD3 expression in non-small cell lung cancers in immunotherapy era

Presentation Number
176P
Lecture Time
12:30 - 12:30
Speakers
  • Nikolaos G. Tsoukalas (Kolonaki, GR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Lung cancer is the first cause of cancer related deaths. RCAS1 (Receptor-binding Cancer Antigen expressed on SiSo cells) is a protein that is expressed in different types of cancer and seems to be involved in the process of the tumour cells’ escape from the immune system surveillance (immunoescape). CD3 (cluster of differentiation CD3), is an antigen that is part of the T cell receptor (TCR) complex on a mature T lymphocyte. Tumor infiltrating lymphocytes (TILs) have been correlated with patients’ survival in several neoplasms.

Methods

The aim of this study was to evaluate the clinical importance of RCAS1 and CD3 expression in non-small cell lung cancer (NSCLC). Tissue microarrays of tumor specimens from 112 patients with newly diagnosed NSCLC were constructed. The sections were stained with monoclonal antibodies against RCAS1, Ki-67 and CD3 using immunohistochemistry and they were studied through classical pathological evaluation and computerized image analysis. Correlations among RCAS1, Ki-67 and CD3 expression, clinicopathological variables and survival were analyzed. In all cases p-value ≤ 0.05 was considered significant.

Results

112 patients were included in this study with mean age 63.6 years old and 83% were males. RCAS1 expression was higher in grade III tumors comparing with grade I (p = 0.004) and grade II (p = 0.005) regardless of the histological type and in adenocarcinomas with lymphovascular invasion (p = 0.014). A positive correlation between RCAS1 and Ki-67 levels was observed (p = 0.002). There was an inverse correlation of overall survival with RCAS1 and Ki-67 levels and patients with higher expression of RCAS1 or Ki-67 had a significantly shorter survival. Also, an inverse correlation between RCAS1 expression and the percentage of CD3(+) TILs was found. Finally, a positive correlation between the percentage of CD3(+) TILs and the patients’ overall survival (p = 0.094) was observed.

Conclusions

CD3 expression was negative correlated with RCAS1 and positive with overall survival in patients with NSCLC. RCAS1 could be a useful biomarker indicating tumor aggressiveness and immunoescape of cancer cells. Further studies needed to elucidate the possible role of RCAS1 as a biomarker in immunooncology era.

Legal entity responsible for the study

Medical School, National and Kapodistrian University of Athens.

Funding

Has not received any funding.

Editorial Acknowledgement

Hellenic Society of Medical Oncology (HeSMO).

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

177P - Comparative study of EGFR mutations detected in malignant pleural effusion, plasma and tumor tissue in patients with adenocarcinoma of the lung

Presentation Number
177P
Lecture Time
12:30 - 12:30
Speakers
  • Shuhang Wang (Beijing, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

The utility of malignant pleural effusion (MPE) as a source of tissue for determining EGFRmutations to guide EGFR TKI therapy in advanced adenocarcinoma of the lung (LUAD) remains unclear. This study compared MPE, plasma and tumor as sources of tissue for EFGR mutational analysis in LUAD patients.

Methods

MPE samples were collected from 295 LUAD patients. Matched tissue and plasma samples were available for 92 patients, and 248 patients had plasma samples. EGFRexon 19-deletion and exon 21-L858R mutation were detected. The concordance of EGFR mutation status in MPE, tissue, and plasma were evaluated, and the predictive value of EGFRmutations in MPE with respect to efficacy of EGFR-TKI was investigated.

Results

The EGFRmutation rate in MPE samples was 39.3% (116/295). The concordance between MPEs and tissues was 87.1% (Kappa=0.712); the sensitivity and specificity of EGFRmutation in MPEs according to tissues was 71.4% and 96.5%,, respectively. 219 patients received EGFR-TKI, and the objective response rate of patients with EGFRmutations was similar for patients with EGFRmutation either in MPE, tissues or plasma (57.6 % vs 56.0 % vs 47.4%, p = 0.51). Similar results were found in progression free survival (8.9 months vs 9.0 months vs 7.7 months, p = 0.077 and overall survival (29.8 months vs 25.9 months vs 25.3 months, p = 0.33).

Conclusions

MPE is a reliable surrogate for tumor tissue for identifying EGFRmutations. MPE could offer reference of EGFRmutation to EGFR-TKIs treatment decision for advanced LUAD patients even when tissue and plasma were available.

Legal entity responsible for the study

Peking University Cancer Hospital.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

178P - XmaI-RRBS DNA methylation screening resolves breast cancer epigenetic heterogeneity

Presentation Number
178P
Lecture Time
12:30 - 12:30
Speakers
  • Aleksandr S. Tanas (Moscow, RU)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Breast cancer (BC) heterogeneity calls for molecular subtyping that would assist in personalized treatment. An advantage of DNA methylation markers is that their detection in tumors is not compromised by the presence of normal tissues. With a perspective to develop methylation-based BC diagnostic markers, we have performed a genome-wide DNA methylation profiling of a collection of breast tissues and cell lines.

Methods

XmaI-RRBS method was used to profile DNA methylation of 170 BC samples obtained before chemotherapy, six BC cell lines, and 10 normal breast autopsy specimens. Unsupervised hierarchical cluster analysis was used to discern intrinsic DNA methylation BC subtypes; clustering uncertainty was assessed with pvclust R package using bootstrap permutation approach.

Results

We have identified 10 epigenetic BC subtypes different in the DNA methylation profiles. Of these, BC cell lines constitute a separate extremely high methylated subtype clustering far from any tissues assessed. In turn, BC tissues are classified into two major epigenetic subtypes, high- and low-methylated at the promoter regions of genes. We identified 114 genes that distinguish between high- and low-methylated BC subtypes. Noteworthy are the genes of adenylate cyclases ADCY4, ADCY8 and adenylate cyclase stimulants ADORA2B, ADCYAP1; proteins of cell adhesion and extracellular matrix (CDH4, NRXN2, MXRA5, COMP, integrins A8 & A11, ADAM19; potassium channels (KCNH8, KCNJ2, KCNG1, KCNK10, KCNK17, ATP1A3). More than a third among differentially methylated are homeobox genes (VAX2, TLX3, GSX1, IRX1, FOXC2, FOXE3, NKX6-2, VSX1, SOX21, POU4F1) and genes encoding proteins involved in early development and morphogenesis (ZIC1, SPOCK2, DPYSL3, ATOH1, ITGA8). Expectedly, there is no statistically significant difference in methylation of the classical tumor suppressor genes between epigenetic subtypes of BC, as their abnormal methylation is ubiquitous in cancers and thus non-discriminative between tumor types.

Conclusions

Intrinsically epigenetically heterogeneous BC may be classified into a reasonable number of DNA methylation subtypes, promising the discovery of new diagnostic and prognostic markers, as well as of new therapeutic targets.

Legal entity responsible for the study

Research Centre for Medical Genetics.

Funding

Russian Science Foundation (project No.18-15-00430).

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

179P - Prognostic value of change in neutrophil-lymphocyte ratio during treatment with first-line anti-PD1 therapy in metastatic melanoma

Presentation Number
179P
Lecture Time
12:30 - 12:30
Speakers
  • Jennifer A. Soon (Melbourne, AU)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

With the advent of immunotherapy, the overall survival (OS) of patients (pts) with advanced melanoma has seen significant improvement. Multiple studies have demonstrated a negative correlation between elevated baseline serum neutrophil-lymphocyte ratio (NLR) and OS in melanoma and other solid tumours. This retrospective analysis aimed to identify a relationship between change in NLR during treatment and OS.

Methods

83 consecutive pts with metastatic melanoma who received first-line anti-PD1 immunotherapy (mono or combination therapy) were identified at a single institution between May 2015 and August 2017. NLR was measured at baseline and following 4-6 weeks (4-6W) of therapy, with the result at each time point correlated with OS. An elevated NLR was defined as > 4.

Results

Median follow-up was 17 months. Median OS of pts with baseline NLR <4 was not reached (NR) compared with 17.1 months with NLR >4 (HR 0.29, 95% CI 0.13-0.65, p = 0.003). Pts whose NLR started and remained high after 4-6W of treatment performed significantly worse than pts whose NLR fell to < 4 at 4-6W (median OS 6.5 months vs NR, HR 0.18, p = 0.028). Survival in the latter group was comparable to those with a baseline NLR <4. NLR was more prognostic at 4-6W (HR 0.17, 95% CI 0.07-0.41, p = 0.000091) than at baseline (HR 0.29, 95% CI 0.18-0.65, p = 0.003). On Cox regression multivariate analysis including age, sex, M stage, lactate dehydrogenase level, presence of brain and/or liver metastases and NLR at the two time points, NLR >4 at 4-6W was the strongest prognostic factor (HR 0.14, 95% CI 0.06-0.37, p = 0.00005).

Conclusions

NLR is a simple and inexpensive prognostic biomarker in metastatic melanoma. NLR >4 at baseline is associated with a significantly poorer OS. In this cohort, NLR at 4-6W was the strongest predictor of outcome. Persistent elevation of NLR >4 at 4-6W after initiation of treatment was associated with a significantly poorer prognosis than those with a change in NLR to < 4 at this time point. Further analysis of a larger cohort may strengthen this association and potentially allow early identification of poor-risk pts and an opportunity to escalate treatment.

Legal entity responsible for the study

Andrew Haydon.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

180P - A serum microRNA expression signature for radiosensitivity of non-small cell lung cancer

Presentation Number
180P
Lecture Time
12:30 - 12:30
Speakers
  • Liyuan Fan (Ji'nan, CN)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Chemoradiotherapy represents the main treatment for non-small cell lung cancer (NSCLC), especially for the advanced lung cancer. However, the curative effect varies significantly. Many microRNAs are verified to be associated with it and microRNA signature may be a good biomarker to predict the radiosensitivity.

Methods

Genome-wide microRNA profiling was analyzed by microarray and validated by qRT-PCR in radio-resistant cell lines and their parent cell lines (A549 and PC9, the corresponding cell lines named A549-R and PC9-R). Then we used colony formation by transfecting miRNA-mimics into A549 and PC9 for functional verification. Finally, a potential microRNA signature was established by an independent set of non-small cell lung cancer (NSCLC) serum samples and validated by available corresponding formalin-fixed paraffin-embedded tissue (FFPE) samples.

Results

73 up-regulated and 24 down-regulated miRNAs were found by microarray and 11 up-regulated, 3 down-regulated and 3 non-different miRNAs were rechecked by qRT-PCR. A miRNA signature, including miR-1290, miR-2861, miR-25-5p and miR-92a-1-5p was selected for further exploration. Overexpression of miR-1290 and miR-2861 increased the radio resistance of A549 and PC9 while overexpression of miR-25-5p and miR-92a-1-5p reversed the radio resistance of A549-R and PC9-R. The four-miRNAs signature could predict the chemotherapeutic response with high accuracy, 83.4% and 79.5% in both the test (serum samples) and validation (FFPE samples) cohorts respectively.

Conclusions

It is the first report of a miRNA signature for cell lines, serum and tissues. Serum and tissue miRNAs represent novel biomarkers to predict radiotherapy response clinically and may represent potential molecular targets to sensitize resistant cancers.

Legal entity responsible for the study

Shandong Cancer Hospital affiliated to Shandong University.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

181P - Anti-tNASP antibodies as a diagnostic marker for malignant tumors

Presentation Number
181P
Lecture Time
12:30 - 12:30
Speakers
  • Oleg Alekseev (Lillington, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Nuclear Autoantigenic Sperm Protein (NASP), a facilitator of chromatin assembly, is expressed as two splice variants: tNASP, specific for testis and cancer cells, and sNASP, expressed in all somatic cells. Exposure of tNASP to the immune system induces a robust humoral immune response. We suggested that patients with malignancies have a higher level of serum anti-tNASP antibodies than those without malignancies. We hypothesized that detection of anti-tNASP antibodies in serum can be used as a cancer screening test.

Methods

Sera from cancer patients and healthy individuals (negative control) were tested for the presence of antibodies against tNASP using enzyme-linked immunosorbent assay (ELISA) with a recombinant tNASP fragment as bait. A total of 139 serum samples from patients with a known malignancy were tested. These included bladder (11), brain (12), breast (12), endometrial (10), gastrointestinal (10), lung (10), ovarian (10), prostatic (12), skin (10), soft tissue (12), thyroid (10), or urinary (10) malignancy, as well as sera from 10 control patients with no known cancers (negative control).

Results

The majority of samples (56.5%) demonstrated elevated levels of anti-tNASP antibodies compared to negative control: glioblastoma, astrocytoma, colorectal adenocarcinoma, pulmonary adenocarcinoma, large-cell and squamous cell lung carcinoma, ovarian serous carcinoma, ovarian adenocarcinoma, bladder urothelial and squamous cell carcinoma, adenocarcinoma of the urinary bladder, prostate adenocarcinoma, and endometrial adenocarcinoma. Samples from patients with melanoma, renal, thyroid, breast carcinomas, and different types of sarcomas demonstrated similar levels of anti-tNASP antibodies as control samples.

Conclusions

Serum anti-tNASP antibody levels are markedly elevated in the majority of cancer patients tested as compared to healthy controls. These data demonstrate that detection of anti-tNASP antibodies is a viable diagnostic approach and has the potential to be used as an early noninvasive screening method for detection of multiple cancers.

Legal entity responsible for the study

Oleg Alekseev.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

182TiP - EXPRESS study - A Multicenter, Prospective Trial In Progress Exploring The Association Between Low Level Of Genomic Alteration And Exceptional And Unexpected Response To Targeted Therapies In Patients With Solid Tumors

Presentation Number
182TiP
Lecture Time
12:30 - 12:30
Speakers
  • Antoine Italiano (Bordeaux, FR)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Most anticancer drugs are approved by making marginal improvements in terms of tumor response or survival in non-selected populations, and are highly heterogeneous at the molecular level. Studying patients who present an exceptional and unexpected response (ER) to these drugs could enable the rapid identification of novel treatment response biomarkers, accelerate drug development and, more broadly, lead to a better understanding of the biology of cancer cells. Several studies are currently recruiting to build cohorts of patients, in order to subsequently analyze their tumors and reveal in detail the molecular anomalies associated with exceptional response.

Trial design

This is an exploratory, multicenter, multicohort, prospective trial conducted in 264 adult patients, with advanced breast, lung, colorectal, ovarian, kidney cancers and melanoma, having presented an ER to an approved antineoplastic targeted therapy. ER is defined using the definition chosen by the NCI which combines the three criteria: -complete or partial response -lasting > 6 months -and not expected in > 10% of the patients in this drug – organ situation. The primary objective is to assess whether ER can be associated with a low level of genomic instability in the tumor. Low genomic instability is defined by the presence of less than the 5th quantile of genomic alterations (mutations, amplifications, deletions) to be expected in the given tumor type as per TCGA database. For each tumor type, the null hypothesis H0: π = 0.05 will be tested, against the one-sided alternative hypothesis π > 0.05. For each of the 6 cohorts, a sample size of 44 patients is necessary to achieve 80% power at π = 15 with a one-sided level 5% test. Patients presenting an ER will be identified retrospectively, in a nationwide manner, then monthly reviewed and validated for inclusion by a panel of pathology experts. As of May 2018, 56 patients have been included. The identification of molecular traits associated with ER might serve the development of predictive classifiers for precision medicine. This study also represents a unique opportunity to better understand cancer biology.

Clinical trial identification

NCT02701907.

Legal entity responsible for the study

UNICANCER R&D.

Funding

Fondation ARC.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

183TiP - Neoadjuvant Biomarker Research Study of Palbociclib Combined With Endocrine Therapy in Estrogen Receptor Positive/HER2 Negative Breast Cancer – the phase II NeoRHEA trial

Presentation Number
183TiP
Lecture Time
12:30 - 12:30
Speakers
  • Michail Ignatiadis (Brussels, BE)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Palbociclib (P) is a CDK4/6 inhibitor used in combination with endocrine therapy (ET) in metastatic estrogen receptor (ER)+/HER2- breast cancer (BC). The role of P in early BC treatment is currently being tested in phase III trials. Biomarkers that help us predict primary resistance to P may lead to better patient selection and thus avoid toxicity and reduce costs. In vitro studies suggest that CDK4 T172 phosphorylation is associated with sensitivity to P and an 11-gene expression signature has been developed that can predict the CDK4 modification profile in breast tumors. In order to identify biomarkers of resistance to P/ET and validate the 11-gene signature, we have launched the NeoRHEA study.

Trial design

Single arm, phase II trial, enrolling patients with ER+/HER2- breast tumors ≥ 15mm, N0-N1. Subjects will receive 4 months of neoadjuvant P/ET (tamoxifen± goserelin or letrozole). Subjects’ response to therapy will be evaluated before and after treatment by ultrasound, using WHO criteria. Biopsy samples will be collected at baseline and surgery. Primary objective is to identify biomarkers of resistance to P/ET (defined as stable or progressive disease by ultrasound) using RNA-sequencing of baseline samples. A key secondary objective is to validate the 11-gene signature as a biomarker of resistance to P/ET. Other secondary objectives include: to evaluate the safety of P/ET; to identify biomarkers of resistance to P/ET, defined as residual cancer burden of 3 or high tumor proliferation by the Genomic Grade Index; to understand mechanisms of resistance to P/ET by comparing tumors transcriptome at baseline and at surgery; to assess the role of plasma ctDNA in monitoring response to P/ET. Assuming a resistance rate of 20%-25% and a 10% dropout, 100 subjects are needed to develop a binary predictor. Any biomarker identified will be further validated in other studies e.g. NeoPAL (preliminary agreement in place). Accrual started in July 2017 and 38 patients have been enrolled thus far. The study is expected to be completed in 2019. Trial number is NCT03065621. Study sponsor is Institut Jules Bordet with a research grant from Pfizer.

Clinical trial identification

NCT03065621.

Legal entity responsible for the study

Institut Jules Bordet.

Funding

Pfizer.

Disclosure

M. Ignatiadis: Consulting or advisory role: Celgene, Roche, Pfizer, Seattle Genetics; Research funding: Roche; Patents: Université Libre de Bruxelles. S. Drisis: Travel, accomodation: Philips. M. Piccart: Honoraria: Pfizer. P. Vuylsteke: Honoraria: Roche, AstraZeneca, Pfizer; Travel, accommodation: Roche, Téva. C. Sotiriou: Consulting or advisory role: Pfizer; Research funding: Pfizer. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

Gynaecological cancers

Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

944P - QUADRA: A phase 2, open-label, single-arm study to evaluate niraparib in patients (pts) with relapsed ovarian cancer (ROC) in 4th or later line of therapy: results from the tBRCAmut subset

Presentation Number
944P
Lecture Time
12:30 - 12:30
Speakers
  • Kathleen N. Moore (Oklahoma City, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Therapeutic options in late line ROC offer limited efficacy, especially for pts who are considered platinum (plat) resistant (res) or refractory (ref). Pts whose cancers harbor BRCA mutations (BRCAmut) have been included in poly (ADP-ribose) polymerase inhibitor (PARPi) trials and derived modest benefit from treatment (ORR ≈25% for plat-res and 0-14% for plat-ref pts). QUADRA evaluated niraparib monotherapy in ROC pts regardless of their plat and biomarker status.

Methods

Eligible pts received treatment with single agent niraparib in 4th or later line of therapy. Pts were evaluated for BRCAmut and HRD status (MyChoice HRD Test). Pts received niraparib 300 mg once daily until progression; treatment emergent adverse events (AEs) were managed with dose reduction to 200 or 100 mg. The primary endpoint was ORR per RECIST v1.1.

Results

463 pts were treated. Median age was 65 years (range: 29-91). 162 pts were plat ref (defined as progression within 28 days of the last dose of plat); 152 plat res (defined as less than 6 months between last dose of plat and subsequent progression); 118 plat sensitive; 31 unknown. Results in HRD+ pts have been presented at a prior congress. We focus here on the tBRCAmut (both germline and somatic) PARPi-naïve subgroup. ORR for 4th line or later, PARPi-naïve tBRCAmut pts (n = 55) was 31% (95% CI: 19-45), including 18 plat-sensitive pts (ORR 39%), 21 plat-res pts (ORR 33%), and 16 plat-ref pts (ORR 19%). The combined ORR in the plat-res and -ref pts (n = 37) was 27%. The median DOR among all tBRCAmut pts was 9.2 months, with an estimated 43% of responding pts maintaining their response at 24 months. In the entire study cohort, 197 pts (42.5%) experienced a serious AE (SAE) and 91 pts (19.7%) a treatment-related SAE. The most frequent treatment-emergent SAEs were gastrointestinal disorders (19.9%), thrombocytopenia (8.4%), small intestinal obstruction (6.6%), and vomiting (5.1%).

Conclusions

Niraparib demonstrated meaningful and durable responses among the difficult-to-treat patient population, including platinum resistant and refractory tBRCAmut patients.

Clinical trial identification

NCT02354586.

Legal entity responsible for the study

Tesaro, Inc.

Funding

Tesaro, Inc.

Editorial Acknowledgement

Writing and editorial support, funded by Tesaro, Inc. (Waltham, MA, USA) and coordinated by Ted Paunescu, PhD of TESARO, Inc., was provided by Nicole Renner, PhD and Dena McWain of Ashfield Healthcare Communications (Middletown, CT, USA).

Disclosure

K.N. Moore: Honorarium and served on advisory boards: Tesaro, Genentech Roche, Clovis, Astra Zeneca (for agents not involved in the SOLO-a Study), Immunogen, VBL Therapeutics, Janssen. M. Geller: Advisory Role, Speakers' bureau & Research funding: Tesaro Inc. D.S. Miller: Advisory role: Eisai, ImmunoGen, Tesaro, Clovis Oncology, Genentech, AstraZeneca, Guardant Health, Alexion. Speakers' bureau: Genentech, Clovis; Research funding: Tracon, AstraZeneca, Tesaro, Janssen, Aeterna Zentaris, Genentech, Pfizer, Aprea AB, ImmunoGen, Takeda, Xenetic Biosciences. N.G. Cloven: Employment: Texas Oncology. G.F. Fleming: Research funding: Corcept Therapeutics. P. DiSilvestro: Advisory role: Tesaro, AstraZeneca; Research funding: AstraZeneca, Tesaro, Abbvie, Genentech, Roche, Janssen. A. Oza: Steering Committees for PARPi trials: Tesaro, Clovis, AstraZeneca; Honoraria: Intas Pharma. M. Cristea: Research funding: Trovagene. J.S. Berek: Advisory role: Atara Biotherapeutics; Research funding: Tesaro. J.K. Chan: Advisory role: Roche/Genentech, AstraZeneca, Janssen Oncology, Clovis, Mateon, Biodesix, Tesaro; Spearkers' bureau and Honoraria: Roche/Genentech, AstraZeneca, Clovis, Tesaro. Y. Li, K. Luptakova, R. Clark: Employment and stock and other ownership interests: Tesaro Inc. U.A. Matulonis: Advisory role: Merck KGaA, AstraZeneca, Immunogen, Tesaro, Genentech. B.J. Monk: Advisory role: GSK, Merck, Tesaro, Roche/Genentech, AstraZeneca,Gradalis, Advaxis, Verastem, Cerulean Pharma, Amgen, Vermillion, Immunogen, Bayer, NuCana BioMed, Insys Therapeutics, Clovis, Oxigene, Pfizer, Mateon, Precision, Perthera, Biodesix, Abbvie, Myriad, Incyte; Spearkers' bureau: Roche/Genentech, AstraZeneca, Janssen, Clovis, Tesaro; Honoraria: GSK, Merck, Tesaro, Roche/Genentech, AstraZeneca, Gradalis, Advaxis, Veraste, Cerulean, Amgen Vermillion, Bayer, NuCana BioMed, Insys Therapeutics, Clovis, Oxigene, Pfizer Mateon, Precision, Perthera, Biodesix, Abbvie, Myriad, Incyte, Janssen. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

945P - A phase 1/2 study of chemo-immunotherapy with durvalumab (durva) and pegylated liposomal doxorubicin (PLD) in platinum-resistant recurrent ovarian cancer (PROC)

Presentation Number
945P
Lecture Time
12:30 - 12:30
Speakers
  • Roisin E. O'Cearbhaill (New York, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Programmed cell death ligand 1 (PD-L1) expression and preliminary evidence of antitumor activity with anti-PD-1 therapy have been reported in ovarian cancer. PLD, a pegylated, liposomal form of doxorubicin, is a standard option for this population; durva is an anti-PD-L1 antibody. The primary objectives of this study are to determine the safety of the combination and to evaluate clinical efficacy by progression-free survival rate at 6 months (PFS6) using RECIST 1.1.

Methods

This is a phase 1/2, multicenter, open-label study (NCT02431559) of durva in patients (pts) with PROC, scheduled to receive PLD. The study includes a dose escalation (phase 1: 3 + 3 design; DLT evaluation over one 28-day cycle; n = 6-18) and a dose expansion (phase 2: n = 41). PLD has been reported to have a 25% PFS6. A sample size of 41 evaluable pts yields 80% power to test the null hypothesis of a PFS6 rate of ≤ 25% against the alternative hypothesis of a PFS rate of ≥ 45% at an alpha level of 0.05 (one-sided). Blood and tumor samples were also collected for assessment of correlative immunologic responses.

Results

First pt dosed: 09Aug2016. As of 05Mar2018, 40 female pts (median age: 65 [32-83] years) were enrolled in phase 2 of the study; each received at least 1 dose of study therapy (PLD 40 mg/m2 + durva 1500 mg Q4W) and are included in the safety analyses. Most frequent (in ≥ 25% pts) treatment-emergent adverse events (AEs, all causality) were palmar-plantar erythrodysesthesia syndrome (PPES)/rash, stomatitis, fatigue, abdominal pain, nausea, pyrexia, and vomiting. Grade 3 treatment-related AEs in ≥ 2 pts included PPES/rash, stomatitis, lymphocyte count decreased, lipase increased, and anemia. As of the cutoff date, 33 pts reached the timepoint for PFS6 assessment. Twelve pts were progression-free at 6 months; PFS6 = 30% (12/40 pts). The remaining data will mature by July 2018, and further improvement in PFS6 may occur. Updated PFS6 and preliminary correlative results will be presented at the meeting.

Conclusions

The combination of durva and PLD in women with PROC appears to have a tolerable safety profile and promising efficacy. PFS6 and translational endpoints are pending additional data.

Clinical trial identification

NCT02431559. May 1st 2015.

Legal entity responsible for the study

Ludwig Institute for Cancer Research.

Funding

Ludwig Institute for Cancer Research, Cancer Research Institute with funding also from VentiRx, and Medimmune. This study was funded in part through the NIH/NCI Support Grant P30 CA008748.

Disclosure

A. Wolfer: Advisory board: AstraZeneca. J.K. Bryan: CMO Novella Clinical, a contract research organization (January 2017-present); Previously CMO VentiRx Pharmaceuticals (2013-2016), with neither equity nor patent ownership on the technology at VentiRx. B.J. Monk: Consultant and speaker: AstraZeneca; Received honoraria for services. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

946P - Brain metastases in primary ovarian cancer: real-world data

Presentation Number
946P
Lecture Time
12:30 - 12:30
Speakers
  • Elena Ratner (New Haven, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Brain metastasis (BM) is infrequently reported in patients (pts) with ovarian cancer (OC), with past studies reporting a rate of approximately 1%.1 This study estimated real-world incidence of BM in OC and assessed whether BRCA mutation (BRCAmut) increased risk of BM in OC pts.

Methods

This retrospective study included 4515 pts diagnosed with OC between Jan 1, 2011 and Jan 31, 2018 from the Flatiron Health database. This is a longitudinal, demographically and geographically diverse database derived from electronic health record data from over 265 cancer clinics and over 2 million active US cancer pts. A time-to-event analysis was conducted to assess whether pts with a known BRCAmut were more likely to develop BM compared with BRCA wild type (BRCAwt) pts.

Results

Of 4515 OC pts, 473 were BRCAmut, 1679 were BRCAwt, and 2363 had unknown BRCA status. A total of 46 pts (1%) had a diagnosis of BM subsequent to OC diagnosis. Of those with BRCAmut, 3% (14/473) developed BM; the BM rate was 0.6% (10/1679) for BRCAwt. The K-M estimate for the proportion of pts with BM within 5 years of diagnosis was 5.7% in the BRCAmut population compared with 1.4% in BRCAwt. BRCAmut pts had a significantly higher risk of developing BM compared with BRCAwt (HR 4.44 [95% CI 1.97, 10.00], P < 0.0001). A total of 281 OC pts also had a breast cancer (BC) diagnosis (186 developed BC prior to OC, 95 developed BC after OC diagnosis). After excluding these pts from the analysis, the HR for developing BM among BRCAmut pts vs. BRCAwt pts was 3.84 (95% CI 1.60, 9.22), P = 0.001. Multivariate models adjusting for other pt characteristics yielded similar HRs. Among pts who developed BM, median time from OC diagnosis to BM diagnosis was 27 months in BRCAmut pts and 35 months in BRCAwt. Median survival after diagnosis of BM was 7.16 months (95% CI 3.48, 16.49). Overall survival after BM did not differ significantly by BRCA status.

Conclusions

OC pts with BRCAmut seem to have a higher risk of developing BM. Further research is needed to confirm these findings and better understand potential mechanisms and implications for management, given the poor prognosis in pts who develop BM. 1Pectasides et al. The Oncologist 2006;11:252–260.

Legal entity responsible for the study

Tesaro, Inc.

Funding

Tesaro, Inc.

Editorial Acknowledgement

Editorial support, funded by Tesaro, Inc. (Waltham, MA, USA) and coordinated by Michael Stillman PhD of Tesaro, Inc., was provided by Dena McWain of Ashfield Healthcare Communications (Middletown, CT, USA).

Disclosure

E. Ratner: Advisory board: Tesaro Inc. M. Bala, M. Louie-Gao, S. Hazard: Employment and stock and other ownership interests in Tesaro Inc. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

947P - Subgroup analysis of rucaparib in platinum-sensitive recurrent ovarian carcinoma: effect of prior chemotherapy regimens in ARIEL3

Presentation Number
947P
Lecture Time
12:30 - 12:30
Speakers
  • Domenica Lorusso (Milan, IT)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

In the randomised, placebo-controlled, phase 3 study ARIEL3, patients were randomised 2:1 to oral rucaparib (600 mg BID) or placebo as maintenance treatment following response to platinum-based chemotherapy. Rucaparib significantly improved progression-free survival (PFS) vs placebo in all patient populations regardless of biomarker status (Coleman et al. Lancet. 2017;390:1949-61). This post hoc exploratory analysis investigated the effect of the number of prior chemotherapy regimens on the primary and secondary endpoints of investigator-assessed and blinded independent central review (BICR)-assessed PFS in ARIEL3.

Methods

In ARIEL3, all patients received ≥2 prior platinum-based regimens in accordance with the protocol. PFS was explored in subgroups of patients who received 2 or ≥ 3 prior chemotherapy regimens. These subgroup analyses were presented for the following predefined cohorts: BRCA mutant; BRCA mutant or BRCA wild type/high loss of heterozygosity (LOH); and intent-to-treat (ITT) population (ie, all randomised patients).

Results

The visit cutoff dates for efficacy and safety were 15 April 2017 and 15 August 2017, respectively. In each predefined cohort, rucaparib significantly improved PFS compared to placebo irrespective of the number of prior chemotherapy regimens (ie, 2 or ≥ 3) (Table). Rucaparib’s safety profile was consistent between patients who received 2 or ≥ 3 prior chemotherapy regimens as assessed by the rate of all grade (100% and 100%) and grade ≥3 (59% and 59%) treatment-emergent adverse events (TEAEs) and dose modifications (ie, treatment interruptions and/or dose reductions due to a TEAE) (70% and 74%) in each respective subgroup.

CohortRucaparib, nPlacebo, nPFS (investigator review)
PFS (BICR)
HR* (95% CI)Median PFS, mo; P valueHR* (95% CI)Median PFS, mo; P value
Rucaparib vs placeboRucaparib vs placebo
Patients with 2 prior chemotherapy regimens
BRCA mutant73400.24 (0.14–0.40)21.9 vs 5.4; P<0.00010.24 (0.13–0.45)26.8 vs 5.5; P<0.0001
BRCA mutant or BRCA wild type/ high LOH136750.34 (0.23–0.49)14.1 vs 5.5; P<0.00010.33 (0.21–0.52)26.8 vs 5.5; P < 0.0001
ITT2311240.42 (0.32–0.55)10.4 vs 5.4; P<0.00010.37 (0.27–0.50)17.1 vs 5.4; P<0.0001
Patients with ≥3 prior chemotherapy regimens
BRCA mutant57260.21 (0.11–0.40)13.7 vs 5.4; P<0.00010.17 (0.08–0.35)18.0 vs 5.4; P<0.0001
BRCA mutant or BRCA wild type/high LOH100430.27 (0.16–0.44)11.1 vs 5.4; P<0.00010.30 (0.18–0.52)13.6 vs 5.4; P<0.0001
ITT144650.28 (0.19–0.41)11.1 vs 5.3; P<0.00010.36 (0.24–0.53)13.3 vs 5.3; P<0.0001

Cox proportional hazards model; P values for treatment-by-prior chemotherapy regimen subgroup interaction were nonsignificant for all analyses. Stratified log-rank P value. CI, confidence interval; HR, hazard ratio.

Conclusions

Maintenance treatment with rucaparib improved PFS vs placebo in all 3 predefined cohorts regardless of the number of prior chemotherapy regimens received.

Clinical trial identification

NCT01968213.

Legal entity responsible for the study

Clovis Oncology, Inc.

Funding

Clovis Oncology, Inc.

Editorial Acknowledgement

Writing and editorial support, funded by Clovis Oncology, Inc. (Boulder, CO, USA) was provided by Nathan Yardley, PhD, and Shannon Davis of Ashfield Healthcare Communications (Middletown, CT, USA).

Disclosure

D. Lorusso: Consulting or advisory role: AstraZeneca, Clovis Oncology, Roche, Tesaro, PharmaMar; Support for travel or accommodation: Roche, PharmaMar. R.L. Coleman: Grants: AstraZeneca, Roche/Genentech, Janssen, OncoMed, Millennium, Merck, Clovis Oncology, Esperance, AbbVie; Advisor: AstraZeneca, Roche/Genentech, Janssen, OncoMed, Millennium, Merck, Clovis Oncology, Esperance, Tesaro, GamaMabs, Pfizer, Genmab, Gradalis, Bayer, AbbVie. A.M. Oza: Steering committees: AstraZeneca, Clovis Oncology, Tesaro. C. Aghajanian: Steering committee: Mateon Therapeutics, Clovis Oncology; Advisory boards: Clovis Oncology, Cerulean Pharma, Bayer, VentiRx, Tesaro; Honorarium: Mateon Therapeutics, Clovis Oncology, Cerulean Pharma, Bayer, VentiRx, Tesaro. A. Oaknin: Advisory boards: Roche, AstraZeneca, PharmaMar, Clovis Oncology, Tesaro; Travel or accommodation: Roche, AstraZeneca, PharmaMar. A. Dean: Consulting or advisory role: Specialised Therapeutics Australia, Shire Pharmaceuticals, Precision Oncology Australia. N. Colombo: Consulting or advisory role: Roche, AstraZeneca, Tesaro, PharmaMar, Clovis Oncology, Pfizer, Advaxis. J.I. Weberpals: Research support: AbbVie and AstraZeneca; Advisory boards: AstraZeneca. A.R. Clamp: Advisory boards: AstraZeneca, Roche; Research funding: AstraZeneca; Travel and accommodation for congress attendance: AstraZeneca. G. Scambia: Consulting or advisory role: Roche, AstraZeneca, PharmaMar, Clovis Oncology, Tesaro. A. Leary: Advisory board: Clovis Oncology, Pfizer, PharmaMar; Institutional research grant support: GamaMabs, Merus; Boarding and travel expenses for congress activities: AstraZeneca. R.W. Holloway: Speakers bureau: AstraZeneca, Clovis Oncology, Tesaro; Advisory board: Clovis Oncology. P.C. Fong: Advisory boards: Clovis Oncology, AstraZeneca; Honoraria: AstraZeneca. J.C. Goh: Honoraria: Bristol-Myers Squibb, AstraZeneca; Consulting or advisory role: Bristol-Myers, Squibb, AstraZeneca, Janssen; Speakers bureau: AstraZeneca, Novartis; Support for travel or accommodation: Roche, Bristol-Myers Squibb, Astellas. D.M. O’Malley: Advisory board: Clovis Oncology, AstraZeneca, Janssen, Gynecologic Oncology Group, Myriad, Tesaro; Steering committees: Clovis Oncology, Amgen, Immunogen; Consultant to AbbVie, AstraZeneca, Tesaro, Health Analytics, Ambry; Institution received research support from Agenus, Ajinomoto, Array BioPharma, AstraZeneca, Bristol-Myers Squibb, Clovis Oncology, Ergomed Clinical Research, Exelixis, Genentech, GlaxoSmithKline, Gynecologic Oncology Group, ImmunoGen, INC Research, inVentiv Health Clinical, Janssen Research and Development, Ludwig Institute for Cancer Research, Novartis Pharmaceuticals, PRA International, Regeneron Pharmaceuticals, Serono, Stemcentrx, Tesaro, Tracon Pharmaceuticals. S. Banerjee: Institutional research support: AstraZeneca, Janssen-Cilag; Consulting or advisory role: AstraZeneca, Roche, GamaMabs Pharma, PharmaMar, AstraZeneca/MedImmune, Tesaro, Clovis Oncology; Support for travel or accommodation: AstraZeneca, Clovis Oncology, Tesaro. S. Goble, T. Cameron: Employee of Clovis Oncology and may own stock or have stock options in that company. J.A. Ledermann: Advisory role: Clovis Oncology, AstraZeneca, Pfizer; Speakers bureau for and research grants from: AstraZeneca, Merck/MSD. All other authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

948P - Intraperitoneal Chemotherapy (IP CT) after cytoreductive surgery benefits survival in epithelial ovarian cancer (EOC): Results of a pooled meta-analysis including 9 randomized clinical trials (RCT)

Presentation Number
948P
Lecture Time
12:30 - 12:30
Speakers
  • Mercedes Jimenez-Heredia (Valencia, ES)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

IP CT after upfront cytoreductive surgery has been the standard of care in EOC. Nevertheless, results from the recent GOG252 trial showed no impact of this strategy vesus IV CT in the concurrence of bevacizumab therapy. The aim of this study was to perform a meta-analysis of the survival benefit of IP vs IV CT after cytoreductive surgery including recent evidence.

Methods

A literature search in order to identify most relevant RCT comparing IV vs IP CT after cytoreductive surgery in EOC was performed. Non comparative studies or trials with no hazard or risk ratios were excluded. Statistical analyses were done using SPSS STATISTICS V.21. Fixed-effect meta-analysis for combining data was made. Degree of heterogeneity using the I2statistic was calculated. Endpoints meta-analysed were disease-free survival (DFS) and overall survival (OS).

Results

Initially 92 papers were identified. After exclusion criteria only nine RCT including pooled data from 3668 pts (2068 treated with IP CT and 1620 treated with IV CT) were included. In 8 RCTs IPCT was given after upfront surgery and in 1 (OV21/PETROC) after neoadjuvant CT. In all RCTs 6 CT cycles were foreseen in the IP arms. Seven RCTs including 3006 pts had DFS as endpoint. Among these, GOG252 cohorts A and B were analysed separately. IP CT showed a survival benefit over IV CT with HR = 0.86; (95% confidence interval (CI): 0.80 to 0.93). Heterogeneity was moderate (I2=37.3) Eight RCTs including 2289 pts had OS as endpoint. IP CT showed a significant impact on OS with HR = 0.81; 95% CI: 0.73 to 0.90). Heterogeneity was low (I2<0.01). Benefit of IP CT remained unchanged in the sensitivity analysis performed for both DFS and OS by eliminating the latest RCT of the meta-analysis.

Conclusions

IP CT benefits DFS and OS in this meta-analysis of pooled data from 9 RCTs and 3668 pts including latest negative results.

Legal entity responsible for the study

Hospital Clinico Universitario de Valencia. INCLIVA.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Collapse
Poster display session: Biomarkers, Gynaecological cancers, Haematological malignancies, Immunotherapy of cancer, New diagnostic tools, NSCLC - early stage, locally advanced & metastatic, SCLC, Thoracic malignancies, Translational research Poster Display session

949P - Mirvetuximab soravtansine, a folate receptor alpha (FR_)-targeting antibody-drug conjugate (ADC), with pembrolizumab in platinum-resistant ovarian cancer (PROC): Initial results of an expansion cohort from FORWARD II, a Phase Ib study

Presentation Number
949P
Lecture Time
12:30 - 12:30
Speakers
  • Ursula A. Matulonis (Boston, US)
Location
Hall A3 - Poster Area Networking Hub, ICM München, Munich, Germany
Date
20.10.2018
Time
12:30 - 13:30

Abstract

Background

Mirvetuximab soravtansine is an ADC comprised of a FRα-binding antibody linked to the tubulin-disrupting maytansinoid DM4. This agent activates monocytes and upregulates immunogenic cell death markers on ovarian tumor cells, providing a rationale for combining with immune checkpoint blockade. Mirvetuximab soravtansine is being evaluated in combination with pembrolizumab in patients with PROC.

Methods

Eligibility criteria include FRα positivity by IHC (≥ 25% of cells with ≥ 2+ staining intensity), 2-4 prior systemic treatments, and no prior immunotherapy. Mirvetuximab soravtansine (6 mg/kg; adjusted ideal body weight) and pembrolizumab (200 mg) were administered intravenously on day 1 of a 21-day cycle. Responses were assessed with RECIST 1.1 and adverse events (AEs) by CTCAE v4.03.

Results

During dose-escalation (n = 14 patients), the mirvetuximab soravtansine-pembrolizumab combination demonstrated favorable tolerability, with primarily ≤ grade 2 AEs observed. Overall, the AE profile was manageable and consistent with the known profiles of each agent. In addition, promising evidence of durable antitumor activity was observed, including a confirmed objective response rate of 43%, median duration of response of 6.9 months, and median progression-free survival of 5.2 months. These findings supported enrollment of an expansion cohort to further evaluate this combination in the setting of PROC. At time of analysis, 37/46 patients were enrolled, who received a median of 3 prior regimens. Median age was 62y (range 40-77), and 92% had tumors with high grade serous histology. Initial safety, antitumor activity, and exploratory biomarker data will be presented, including for the subset of patients with medium/high FRα expression, which showed the highest degree of clinical activity during escalation.

Conclusions

Preliminary data have demonstrated a manageable safety profile and encouraging signals of clinical activity for the mirvetuximab soravtansine-pembrolizumab combination in recurrent PROC. The results of this expansion cohort will guide further development of this novel combination.

Clinical trial identification

NCT02606305.

Legal entity responsible for the study

ImmunoGen, Inc.

Funding

ImmunoGen, Inc.