Abstract 384: Differential response to a dual PI3K/mTOR inhibitor in PIK3CA mutant urothelial cancer patient derived xenografts

Background: PI3K pathway has been reported to be deregulated in up to 30% of bladder cancer patients. We have recently established two patient derived xenografts (PDX) models in our lab that carry unique helical domain mutations in the PIK3CA gene that is known to be deregulated in bladder cancer. This sets our models as unique tools for understanding the molecular background that would dictate an optimal response to targeting the PI3K pathway in vivo as well as the potential mechanisms of resistance. Methods: RNA-seq was carried out to understand the molecular make up of our PDX models (RP-B-01 and RP-B-02) and to identify clinically relevant drug targets. In-vivo drug testing was done using a dual PI3K-mTOR inhibitor to test the response to this class of drugs. Cell lines derived from the PDX models were used for in-vitro studies to validate our in-vivo data and to calculate the IC-50 for this class of agents in our cell lines. Western blot was used to assess the molecular effects of these agents using both in-vivo specimens as well as cell lines that were treated in vitro. Results: Our RNA-seq data have shown that our models (RP-B-01 and RP-B-02) are molecularly distinct where RP-B-01 has a basal-like phenotype while RP-B-02 is has a luminal like phenotype. Additionally, these models carry two unique helical domain mutations in the PIK3CA gene (E542K and E545K mutations respectively). Despite the molecular similarity between these two mutations, the two models responded differently to a dual PI3K-mTOR inhibiter in vivo. The RP-B-02 tumor model significantly responded to the drug compared to vehicle treated mice (P Value = 0.03) while the RP-B-01 model was resistant. Cell lines derived from these models were able to recapitulate the same phenotype in-vitro and IC-50 for each cell line was significantly different (IC50 for RP-B-01 cells = 204.1 nM; IC50 for RP-B-02 cells = 73.21 nM). Western blot analysis has shown that despite the ability of the dual PI3K-mTOR inhibitor to hit the target in RP-B-02 PDX, the treatment was not able to have the same biological effect in the other model (RP-B-01). Interestingly, supplementing media with insulin and nutrients switched the response phenotype and rendered cells derived from the RP-B-02 model more resistant to the PI3K-mTOR inhibitor. Conclusion: Response to PI3K pathway inhibitors in bladder cancer is not only dictated by the presence of targetable mutations but more importantly by the molecular make up of individual tumors which should be taken into account when using these agents in the clinic. Citation Format: May F. Elbanna, Eric Ciamporcero, Remi Adelaiye, Ashley Orillion, Sreenivasulu Chintala, Roberto Pili. Differential response to a dual PI3K/mTOR inhibitor in PIK3CA mutant urothelial cancer patient derived xenografts. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 384.