24P - Mutational load by targeted next generation sequencing (NGS) panels as potential biomarker of response to checkpoint inhibitors

Date 20 November 2015
Event ESMO Symposium on Immuno-Oncology 2015
Session Welcome reception and general Poster viewing
Topics Cancer Immunology and Immunotherapy
Translational Research
Presenter Juan Martin-Liberal
Citation Annals of Oncology (2015) 26 (suppl_8): 5-14. 10.1093/annonc/mdv514
Authors J. Martin-Liberal, M. Ochoa de Olza, C. Hierro, C. Cruz, C. Rodriguez, I. Brana, A. Vivancos, R. Dienstmann, J. Rodon, J. Tabernero
  • Medical Oncology, Vall d`Hebron University Hospital Institut d'Oncologia, 08035 - Barcelona/ES

Abstract

Aim

Mutational load by whole-exome sequencing has been correlated with clinical benefit to checkpoint inhibitors in melanoma and non-small cell lung cancer. We aim to explore if mutational burden found with broadly used targeted NGS panels is a potential biomarker of response to checkpoint inhibitors.

Methods

Retrospective study of 14 patients (pts) treated from Jan/2013 to Jun/2015 with checkpoint inhibitors (10 atezolizumab, 3 ipilimumab, 1 nivolumab) in phase 1 and 2 trials at Vall d'Hebron Institute of Oncology (VHIO). Tumors were analyzed for mutations (mut) with Amplicon MiSeq (Amp) (8 pts; panel of 71 genes) and FoundationOne (FO) (6 pts; panel of 325 genes). Mutational findings were correlated with response using immune-related response criteria (irRC).

Results

Tumor types included 5 triple negative breast cancer (TNBC), 5 head & neck tumors, 2 gastric cancers and 2 glioblastoma multiforme. Median number of prior therapies was 3 (range 1-5). Median and mean mut count was 2 and 8 (range 0-62) respectively, and median mut rate (mut count/size panel) was 3% (range 0-20%). Eight pts had tumors PDL1 positive by IHC and 2 negative (4 missing). Best response was progressive disease (PD) in 8 pts, stable disease (SD) in 4 pts and partial response (PR) in 2 pts. No correlation between mut count or rate and response was found (p = 0.93 and 0.311, respectively). PDL1 expression did not correlate with mut count (p = 0.4992) or rate (p = 0.1371). One TNBC pt, PDL1 negative, with 3 mut (TP53, CDKN2A, frameshift GATA3) by FO had PR to atezolizumab (-78% in tumor burden). Another patient with PR to atezolizumab (-56% in tumor burden) had a squamous oropharyngeal carcinoma, PDL1 positive with 2 mut (CDKN2A, FGFR3) by Amp. One pt with mucoepidermoid parotid carcinoma with 8 mut (including variant of unknown significance BRCA2) and 1 pt with nasosinusal carcinoma with 62 mut (including MSH2, MSH6, MUTYH, PMS2), both by FO, had PD to ipilimumab.

Conclusions

In our small heterogeneous cohort, targeted NGS panels were not able to predict response to checkpoint inhibitors. The role of NGS panels as tools to predict clinical benefit to immunotherapy merits further research. Updated results with a growing cohort of pts will be presented.

Clinical trial identification

Disclosure

J. Rodon: Served in advisory board for Novartis, Lilly, Servier, Leti y Oncompass. J. Tabernero: Consultant/Advisory Board for Amgen, Boehringer Ingelheim, Celgene, Chugai, Imclone, Lilly, Merck, Merck Serono, Millennium, Novartis, Roche, Sanofi, Symphogen and Taiho. All other authors have declared no conflicts of interest.