Abstract 1888: Identification and optimization of novel small molecule modulators of immune checkpoint resistance with a unified representation space for genomic and chemical perturbations

Abstract

Abstract Immune checkpoint inhibitors have revolutionized cancer treatment, producing a durable response consistent with immunologic memory in a subset of patients. However, the majority of patients demonstrate innate or acquired resistance that must be characterized and overcome to induce successful treatment. Advancements in human reverse translation and scaled in vivo CRISPR screening have uncovered novel molecular and genomic correlates of resistance, and promising druggable mechanisms - driven by highly complex interactions between tumor cells and the immune system. It is this core biology that must be disentangled to build the treatment paradigms of the future. Here we demonstrate a technique for massively parallel prioritization of new immuno-oncology hypotheses using industrial-scale experimentation and machine learning. Leveraging high-content imaging data from whole-genome CRISPR knockout and a library of >250,000 compounds, a deep learning model was trained to construct a batch-invariant low dimensional representation of each perturbation. Millions of perturbations in multiple cell types were embedded in a unified representation space that was leveraged to increase the rate of discovery, accelerate reverse translation, yield novel biological insights, and guide the advancement of lead molecular series through SAR. Here we highlight multiple discovery programs driven by inferred relationships between small molecules and gene knockout with translation from inference to in vivo efficacy. Specifically, we prioritize molecules with activity in STK11-deficient tumors and additional immune checkpoint sensitizers. Citation Format: Ashish Bhandari, Michael F. Cuccarese, Kevin Fales, Kiran Nadella, Rebecca Sarto Basso, Daria Beshnova, Hayley Donnella, Bahar Shamloo, Jacob Cooper, Imran Haque, Ron Alfa, Jacob Rinaldi. Identification and optimization of novel small molecule modulators of immune checkpoint resistance with a unified representation space for genomic and chemical perturbations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1888.