Abstract 2856: Preclinical therapeutics in genetically engineered mouse models of inducible NRAS and BRAF mutant melanoma

Abstract

Abstract Melanoma is virtually incurable once it has spread to distant sites. Despite recent and exciting new therapeutics that extend lifespan in late-stage melanoma patients, much work remains in this direction. Ongoing clinical trials can be supplemented with preclinical therapeutics using mouse models, certain features of which are desirable, including: rapid cohort generation, imaging, and prescreening doses and drug combinations. Another feature of certain models is the reversibility of the driving oncogene, which provides a positive control for tumor regression. We have now generated the first such inducible mouse models of melanoma which use the most prominent signature mutation events, employing transgenic human NRASQ61K and BRAFV600E on an Ink/Arf-null background. In both models, respectively termed iNras (inducible NRAS) and iBIP (inducible BRAF Ink/Arf Pten), the expression of the transgenes are restricted to melanocytes; in addition, BRAF activation co-occurs with Pten deletion via Cre activation. Both models are completely homozygous, affording faster cohort generation. Melanomas form rapidly in both models and retain dependence on transgene expression: upon doxycycline withdrawal, tumors virtually disappear by two weeks. In the iNras model, the Mek inhibitor AZD6244 achieved cytostatic, but not cytotoxic effects on tumor growth even at the maximal tolerated dose. Newer Mek inhibitors are being tested. In the iBIP model, application of tamoxifen physically restricts the tumors to predefined anatomical locations. Restricting tumor formation to the ear provides a clean visual system for examining minimal residual disease. Histological examination of de novo tumor regions kept off dox for over two months revealed that the residual tumor cells were rare among an overwhelming stromal reaction. By contrast, the tumor regression induced by the potent Braf inhibitor PLX4720, though similarly sustainable over two months of treatment at high doses, retained a bulk of anaplastic tumor cells. Interestingly, global RNA and protein analyses of tumors regressing off dox or on PLX4720 revealed virtually identical profiles, suggesting that PLX4720 mimics genetic BRAF extinction but does not achieve complete remission. In xenografts in nude mice, tumors often escaped PLX4720 treatment. RT-PCR analysis revealed that these tumors progressively expressed higher levels of BRAF. This may provide a rapid model for PLX4032 resistance in patients, where reactivation of the ERK pathway is likely to come from a selection pressure for cells which directly circumvent BRAF inhibition. Preclinical combination therapies may prevent such escape. Overall, our two models represent platforms for preclinical drug testing that provide not only rapid cohort generation and a natural setting of tumor formation, but also a positive control for tumor regression which can be contrasted with clinical drug treatment. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 2856. doi:10.1158/1538-7445.AM2011-2856