Abstract A010: Cancer associated fibroblast dependency in a novel immunocompetent mouse model of bladder cancer

Abstract Introduction: The relationship between genotype and phenotype in targeted therapy has proven to be crucial for the successful use of most kinase inhibitors. In bladder cancer, the PI3K/AKT/mTOR pathway is widely activated by mutation but untargeted. Our study focuses on developing an immunocompetent, CRISPR-edited bladder cancer mouse model that mirrors common PI3K/AKT/mTOR pathway mutations. We established effective genotype-kinase inhibitor pairings using CRISPR-edited organoids and cancer associated fibroblasts (CAFs) to create a pro-tumorigenic environment in vivo. Methods and Results: Urothelial cells were isolated from the bladder of Rosa26 Lox-Stop-Lox-Cas9/GFP transgenic mice and grown as organoids in matrigel. These organoids were co-edited for Trp53 and Rb1 KO, and further editing of the Pten gene resulted in a nine-fold increase in sensitivity to the kinase inhibitor ipatasertib. Despite successful initial engraftment into mouse bladder, these edited cells regressed before tumor formation, and subcutaneous injection did not yield subcutaneous tumors. An alternate model featuring a 9p21 deletion with Trp53 KO also displayed similar regression post-engraftment into bladder or subcutaneous injections. However, co-injection with a previously established CAF line subcutaneously yielded palpable subcutaneous tumors that enlarged for 12 weeks. We then established a BBN bladder-tumor-derived CAF line, leading to the formation of tumors with squamous differentiation upon intramural co-injection with edited organoids. These models displayed immune-excluded architecture and a predominance of M2 over M1 macrophages. We also aimed to enhance the model's adaptability by introducing common PI3K/AKT/mTOR mutations, specifically TSC1/TSC2 KO, and Pik3ca hotspots, into our 9p21/Trp53 KO model. While TSC1/TSC2 KO lines exhibited elevated P-S6K signaling, they displayed resistance to Torin 1, Voxtalisib, and Everolimus—kinase inhibitors targeting mTORC1/2. In parallel, we employed CRISPR-mediated homologous recombination to successfully introduce a Pik3ca H1047R hotspot into the 9p21/Trp53 KO genotype, as evidenced by increased Phosphorylation of AKT compared to wild-type 9p21/Trp53 KO cell lines. This genotype resulted in approximately 80-fold selective increased sensitivity to pictilisib, a PI3K inhibitor compared to Pik3ca WT control. Conclusions: In this model, CAFs were required for successful engraftment of tumors into the bladder wall or subcutaneously. BUTU is a customizable model of bladder cancer and is suitable for direct targeting by kinase inhibitors while providing a platform for understanding the genotype-phenotype relationship in bladder cancer. We hope to use it as a tool to develop preclinical evidence on which to base precision medicine clinical trials. Citation Format: Philip H. Abbosh, Henkel Valentine. Cancer associated fibroblast dependency in a novel immunocompetent mouse model of bladder cancer abstract. In: Proceedings of the AACR Special Conference on Bladder Cancer: Transforming the Field; 2024 May 17-20; Charlotte, NC. Philadelphia (PA): AACR; Clin Cancer Res 2024;30 (10Suppl): Abstract nr A010.