Abstract A17: A versatile mouse-modeling platform for rapid in vivo melanoma studies

Melanoma accounts for about 1% of skin cancers but causes the majority of skin cancer-related deaths. Despite progress in treating melanoma with immunotherapy and targeted therapy, melanoma patients still suffer from relapse due to resistance to these treatments. A better understanding of the molecular mechanisms that contribute to melanomagenesis and resistance may provide novel avenues for the development of improved treatment strategies. To accelerate the speed, increase the throughput, and lower the cost of delineating such mechanisms in vivo, we developed a novel embryonic stem cell-genetically engineered mouse-modeling (ESC-GEMM) platform. We generated ESC lines harboring twelve combinations of four common driver mutations (BRAF-V600E, NRAS-Q61R, loss of PTEN, loss of CDKN2A) that can be efficiently targeted by recombination-mediated cassette exchange (RMCE) to introduce inducible alleles of interest. Importantly, all of our ESC lines are capable of producing high-contribution chimeras, which exhibit the same melanoma kinetics and overall survival as conventionally bred mice. Thus, our platform offers readily adaptable melanoma models whose melanoma phenotypes range from long latency/low penetrance to aggressive melanoma with complete penetrance. We employed our ESC-GEMM approach to compare the efficiency of gene depletion by conditional knockout, CRISPR/Cas9, or RNAi. We found that CRISPR/Cas9 depletion of Pten on a Braf-V600E background produces fewer melanomas but with nearly the same latency as conditional knockout using the Cre-lox system. Silencing of Pten with inducible shRNA resulted in significantly increased latency as well as in melanomas with a distinct macroscopic appearance and widespread lymph node metastases. While inducible shRNAs allowed us to restore Pten expression in established melanomas, depletion of Pten with inducible CRISPR enabled us to control the number of moles/melanomas by limiting the duration of Cas9 expression. We also used inducible CRISPR to sequentially activate Braf-V600E and delete Pten, which more closely mimics the sequence of events in human melanomas. Interestingly, while sequential introduction of these genetic changes had no effect on the overall survival, a subset of mice developed a significantly greater number of tumors. Finally, we established melanoma cell lines from untargeted chimeras, which can be targeted with Tet-inducible expression cassettes and transplanted into syngeneic hosts. Taken together, we have established a speedy mouse-modeling platform that will stimulate and accelerate in vivo melanoma research, provide a powerful resource to investigate the pathobiology of melanomagenesis, and allow the genetic and pharmacologic evaluation of novel treatments. Citation Format: Ilah Bok, Jordan Reff, Arianna Nenci, Jose G. Gonzalez, Florian A. Karreth. A versatile mouse-modeling platform for rapid in vivo melanoma studies [abstract]. In: Proceedings of the AACR Special Conference on Melanoma: From Biology to Target; 2019 Jan 15-18; Houston, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(19 Suppl):Abstract nr A17.