PD-L1 genomic alterations (GA) in solid tumors and hematologic malignancies: A comprehensive genomic profiling (CGP) study.

Authors

null

Laurie M. Gay

Foundation Medicine, Inc., Cambridge, MA

Laurie M. Gay , David Fabrizio , Garrett Michael Frampton , Lee A. Albacker , Ethan Sokol , Julia Andrea Elvin , Jo-Anne Vergilio , James Suh , Shakti H. Ramkissoon , Eric Allan Severson , Sugganth Daniel , Siraj Mahamed Ali , Jon Chung , Alexa Betzig Schrock , Vincent A. Miller , Jeffrey J. Pu , Robert John Corona , Gennady Bratslavsky , Jeffrey S. Ross

Organizations

Foundation Medicine, Inc., Cambridge, MA, Foundation Medicine, Inc., Boston, MA, Foundation Medicine, Cambridge, MA, Foundation Medicine, Inc., Morrisville, NC, SUNY Upstate Medical University, Syracuse, NY

Research Funding

Pharmaceutical/Biotech Company

Background: Amplification (AMP) of PD-L1 (CD274/B7H1) has recently been linked to enhanced benefit from immune checkpoint inhibitor (ICPI) treatment. We used CGP to evaluate the PD-L1 GA landscape in cancer. Methods: Specimens from 140,411 cancers representing > 450 individual disease ontologies were sequenced using a hybrid capture-based, next-generation sequencing assay. PD-L1 protein expression was measured by immunohistochemistry (IHC) in a subset of cases using the Dako 22C3 anti-PD-L1 antibody. Results: 1383 (0.9%) tumors had PD-L1 GA (1414 GA): 879 (62%) AMP, 471 (33%) short variants (SV) and 60 (0.4%) truncating rearrangements (RE). PD-L2 (PDCD1LG2) and JAK2 were co-amplified in 94% and 82% of PD-L1 AMP samples; other genes commonly altered with PD-L1 AMP included TP53 (77%), CDKN2A/B (28%/20%), MYC (21%), and TERT (13%). For samples with RE or SV, PIK3CA (19%; 21%) GA were common and TERT GA were in 7% or 14% of samples. Most SV GA were missense (88%), compared with truncating (10%) or splice site (2%) GA. Recurrent somatic variants of unknown significance (VUS) were observed, including frameshift and missense GA that may regulate PD-L1 stability. Of 10 common tumors, breast (213) and lung (210) carcinomas (CA) represented the most PD-L1 AMP cases. Select cancers with notable PD-L1 AMP frequencies were anaplastic thyroid CA (4%), head and neck squamous CA (HNSCC) (2.5%), cervical SCC (2.6%), and breast (1.4%), lung (0.7%) and bladder (0.6%) CA. Of common tumor types, such as colorectal, pancreatic, ovarian, or prostatic CA, melanoma and glioblastoma, ≤0.3% had PD-L1 AMP. There was strong correlation between PD-L1 AMP and PD-L1 expression by IHC in NSCLC, with 89% AMP positive samples showing > 50% IHC staining and 11% with positive but < 50% staining. Major clinical responses to ICPI therapies for tumors with PD-L1 AMP will be presented. Conclusions:PD-L1 AMP or other GA occur rarely across many cancer types. Many somatic PD-L1 GA are uncharacterized VUS. PD-L1 AMP correlates with high membrane expression of PD-L1 measured by IHC and is linked to durable response to ICPI therapies. Further evaluation of PD-L1 AMP as a potential biomarker for immunotherapy selection appears warranted.

Disclaimer

This material on this page is ©2024 American Society of Clinical Oncology, all rights reserved. Licensing available upon request. For more information, please contact licensing@asco.org

Abstract Details

Meeting

2018 ASCO Annual Meeting

Session Type

Poster Session

Session Title

Tumor Biology

Track

Tumor Biology

Sub Track

Tumor-Based Biomarkers

Citation

J Clin Oncol 36, 2018 (suppl; abstr 12092)

DOI

10.1200/JCO.2018.36.15_suppl.12092

Abstract #

12092

Poster Bd #

205

Abstract Disclosures

Similar Abstracts

First Author: Stephanie Leigh Alden

Abstract

2022 ASCO Genitourinary Cancers Symposium

Impact of PD-L1 expression on conventional urothelial bladder carcinoma (UBC) genomic alteration (GA) profile.

First Author: Petros Grivas

First Author: Vincent Lacasse