Abstract 256: Identification of additional cancers likely to respond to anti-PD-1 therapy (pembrolizumab): Evaluation of PD-L1 expression in a large molecular tumor profiling gene expression database

Abstract Increased PD-L1 expression has been associated with clinical activity of anti-PD-1/PD-L1 therapies in both melanoma and non-small cell lung cancer (NSCLC). Our objective was to identify other cancers that show increased PD-L1 expression in order to target them for treatment with the PD-1 inhibitor pembrolizumab (MK-3475). A collaboration between Merck and the Moffitt Cancer Center was previously established to build a molecular profiling database of >16,000 primary and >3000 metastatic tumors representing 25 different cancers. All tumor samples were profiled on a standardized platform (Affymetrix-Merck Custom GeneChip). For each profiled sample, the PD-L1 cutpoint determined for positivity was defined as the Affymetrix pan-cancer 75th percentile of PD-L1 probes mean. This cutpoint was projected for each specific tumor type and percentages of tumors with PD-L1 above the cutpoint were determined. Cancer indications were rank ordered from highest to lowest percentage of PD-L1 positivity. The analysis identified NSCLC (42% PD-L1+) and melanoma (41% PD-L1+) among the top indications for which single-agent clinical activity of anti-PD1/PD-L1 therapies has been reported. At the bottom of the rankings were prostate cancer (14% PD-L1+) and pancreatic cancer (4% PD-L1+), indications for which limited clinical activity had been reported. This observation confirmed that ranking tumor types by PD-L1 expression across the profiling database could be used to identify other cancers that may respond to anti-PD-1 therapy. Of interest were the tumor types with high PD-L1 expression for which no clinical studies evaluating an anti-PD-1/PD-L1 agent had been initiated. Among the indications with high PD-L1 expression were head and neck (59% PD-L1+), urothelial (42% PD-L1+), and triple-negative (TN) breast (29% PD-L1+) cancer; these indications were chosen for evaluation with pembrolizumab in the KEYNOTE-012 study. By accessing cohorts of Asian patients with lung, liver, and gastric cancer and use of similar gene expression microarray profiling, we were able to extrapolate the PD-L1 rankings for these cancers; this resulted in the addition of gastric cancer to KEYNOTE-012. Recently, clinical results from KEYNOTE-012 have shown strong clinical activity for pembrolizumab in all 4 selected indications: head and neck, 20% ORR; urothelial, 21% ORR; TN breast, 18% ORR; and gastric, 31% ORR. The strategy of using a tumor profiling gene expression database for evaluation of PD-L1 expression enabled rapid expansion of pembrolizumab development into indications for which the likelihood of demonstrating clinical activity was high. In turn, this should help accelerate the approval of pembrolizumab for additional indications and, ultimately, provide help to patients suffering from cancer. Citation Format: Mark D. Ayers, Michael Nebozhyn, Razvan Cristescu, Terrill K. McClanahan, Heather A. Hirsch, Jonathan D. Cheng, Andrey Loboda. Identification of additional cancers likely to respond to anti-PD-1 therapy (pembrolizumab): Evaluation of PD-L1 expression in a large molecular tumor profiling gene expression database. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 256. doi:10.1158/1538-7445.AM2015-256

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.