Biomarker-driven indication selection in JTX-2011 ICONIC clinical trial.

Abstract

11602 Background: ICOS (Inducible T cell CO-Stimulator) is a co-stimulatory molecule expressed primarily on T lymphocytes. Clinical and preclinical data suggest that ICOS mediates anti-CTLA-4 driven anti-tumor responses. JTX-2011 is an ICOS agonist antibody in clinical development in advanced solid tumors (ICONIC trial). JTX-2011 is designed to generate an anti-tumor immune response via stimulation of T effector cells and preferential reduction of intra-tumoral T regulatory cells. Single agent preclinical efficacy correlates with the percentage of ICOS-expressing T cells within the tumor. We report indication selection and patient enrichment strategy for ICONIC using in silico and IHC analysis and assessment of potential predictive biomarkers for JTX-2011 using ex vivo tumor histoculture. Methods: Integrated analysis was performed from the TCGA for ICOS expression in histologic and molecularly defined tumors and immune cell signatures. ICOS expression was analyzed by IHC in a subset of indications based on in silico analysis. ICOS expression on intra-tumoral Tregs and PD-L1 were analyzed in a cohort of 126 head and neck squamous cell carcinomas (HNSCC). Ex vivo histoculture assays of human HNSCC was treated with JTX-2011 and assessed for IFNg gene signature induction. Results: ICOS mRNA expression was analyzed in ~10,000 solid tumors samples across ~30 indications. ICOS expression in key indications was confirmed using IHC. Based on frequency of high ICOS expression, non-small cell lung cancer, HNSCC, triple negative breast carcinoma, gastric cancer, and melanoma were selected as indications for ICONIC. Results were confirmed in clinical samples using multiplex immunofluorescence and ICOS IHC. A wide range of ICOS expression was observed suggesting that identification of an ICOS “high” group may enrich for patients likely to benefit from ICOS agonist therapy. In ex vivo histoculture assays of human HNSCC tumors treated with JTX-2011, ICOS IHC and ICOS RNA gene signatures correlated to response endpoints. Comparison of ICOS and PDL1 expression identified subsets of tumors in multiple indications with high ICOS but low PDL1 expression. Conclusions: These data support prioritization of specific tumor types the ICONIC trial.