In:
Journal of Clinical Oncology, American Society of Clinical Oncology (ASCO), Vol. 35, No. 15_suppl ( 2017-05-20), p. 4025-4025
Abstract:
4025 Background: Immuno-oncology (IO) with anti-PD-1 and –PD-L1 antibodies (Abs) is active in EGC but only benefits a minority of Pts. Biomarkers are needed to identify responders. Methods: We reviewed our experience of Pts treated with anti-PD-1/PD-L1 Abs and correlated their outcomes with PD-L1 and mismatch repair protein (MMR) status by immunohistochemistry (IHC), as well as MSK-IMPACT (≥340-gene) NGS profile. MSIsensor from IMPACT assesses microsatellite instability phenotype, while ≥20 mutations (or 17 mutations/Mb) strongly correlates with MMR-deficiency (dMMR) by IHC (J Clin Oncol 2016;34:2141). Progression-free (PFS) and overall survival (OS) were analyzed from the start of IO. Results: 71 Pts were identified, with 3 Pts receiving 2 IO regimens. 66 had adenoCAs and 5 had squamous CAs. Median age 58, 77% male, 96% had received ≥2 prior chemo regimens. 39 (55%), 18 (25%) and 17 Pts (24%) respectively received anti-PD-1, anti-PD-L1 and anti-CTLA-4 plus anti-PD-1/PD-L1 Abs. 6 Pts (8%) had objective response (2 complete responses or CRs) and the median PFS and OS are 1.6 and 4.7 mos; 2-yr OS is 17%. PD-L1 IHC was performed in 16 Pts (23%; 7 +ve), MMR was tested in 20 Pts (28%; 4 dMMR) and IMPACT was obtained in 44 Pts (62%). All 4 dMMR tumors were also MSI by MSIsensor and had a median of 46 mutations (range, 29-63) or, equivalently, 33 mutations/Mb (range, 21-46); 2 of 2 dMMR tumors tested PD-L1 +ve. 3 of the 4 Pts with dMMR/MSI tumors had a response (including 1 CR) and the median OS of these 4 Pts is not reached with 23+ months of follow-up. Finally, a patient whose tumor is MMR-proficient, not MSI but has 15 mutations (including in POLD1), achieved an ongoing CR at 37+ mos. For the 44 Pts with IMPACT testing, there appeared to be improved OS for tumors with ≥10 vs. 〈 10 mutations/Mb (2-yr OS 80% vs. 12%, p=0.03). Conclusions: Pts with tumors that are MSI or have ≥10 mutations/Mb on MSK-IMPACT appear to derive significant benefit from IO. MSK-IMPACT can offer novel information, identify novel mutations (e.g. POLD1) and may be used to help select Pts for IO. We are seeking to define a mutation no. cut-off that can serve as a biomarker and updated data will be presented.
Type of Medium:
Online Resource
ISSN:
0732-183X
,
1527-7755
DOI:
10.1200/JCO.2017.35.15_suppl.4025
Language:
English
Publisher:
American Society of Clinical Oncology (ASCO)
Publication Date:
2017
detail.hit.zdb_id:
2005181-5
