Computational framework for prioritizing candidate compounds overcoming the resistance of pancancer immunotherapy.

Abstract

Combination therapy has emerged as an effective approach to overcome resistance to immunotherapy. However, only a small number of drugs have been identified with synergistic effects with immunotherapy. Here, we develop a computational framework (IGeS-BS) to recommend compounds that potentially overcome resistance to immunotherapy. A meta-analysis of approximately 1,000 transcriptomes from immunotherapy patients revealed 33 tumor microenvironment (TME) signatures that can robustly and accurately estimate immunotherapy responses. An immuno-boosting landscape for more than 10,000 compounds and 13 cancer types was subsequently generated on The Cancer Genome Atlas (TCGA) and The Library of Integrated Network-Based Cellular Signatures (LINCS) datasets. Furthermore, the immuno-boosting effects of several high-scoring compounds were evaluated by in vitro and in vivo experiments in hepatocellular carcinoma and other cancer types. The results showed that the two best compounds (SB-366791 and CGP-60474) significantly alleviate the resistance of hepatocellular carcinoma to anti-PD1 therapy by activating immune cells. Collectively, our research provides an efficient framework for discovering compounds that enhance immunotherapy responses.