Oncogenic MAPK signalling induces epithelial cell state change and therapeutic resistance in vivo II

The colorectal epithelium is a rapidly renewing tissue, with a remarkable capacity to regenerate following injury. In colorectal cancer (CRC), this regenerative capacity can be co-opted to drive epithelial plasticity. Activation of oncogenic MAPK signalling in CRC is common, with mutations frequently found in both KRAS (40-50%) and BRAF (10%), alongside other mechanisms of EGFR pathway upregulation. Nonetheless, inhibition of this pathway is often linked to rapid therapeutic resistance in patients. Given the recent development of KRAS inhibitors and the licensing of BRAF+EGFR inhibitor combinations we have investigated mechanisms of resistance to these agents in complex mouse models of CRC. Here we show that MAPK signalling supports the regenerative/revival stem cell phenotype in vivo, while its inhibition leads to rapid remodelling of both Kras and Braf mutant mouse models to a high Wnt activity canonical stem cell phenotype marked by expression of Lgr5. This results in acute therapeutic resistance in Kras driven, and delayed resistance in Braf driven models. Importantly, where therapeutic MAPK targeting is applied in the presence of a single regenerative epithelial population, such as early metastatic disease, or where plasticity is restrained, for example through Wnt pathway restriction, a marked impact on tumourigenesis is observed. Modelling of an BRAF and RNF43-mutant patient population, clinically super-responders to BRAF/EGFR inhibitor therapy, demonstrates excellent response to therapy due to restricted epithelial plasticity, highlighting the criticality of cellular plasticity in therapeutic response. Together, our data provide clear insight into the mechanisms underpinning resistance to targeting oncogenic MAPK signalling in CRC. Moreover, we demonstrate that strategies that aim to restrict epithelial plasticity or corral into or intervene when tumours are in a regenerative cell fate may improve efficacy of MAPK targeting agents in CRC. Overall design: Mice were induced with a single 70uL injection of 100uM 4-OH tamoxifen into the colonic submucosa. When an established tumour was identified on colonoscope (beginning weekly two weeks post injection), mice were placed on a treatment regime. Mice were treated once daily with AZD4625 (OG 100uL 100mg/kg) or vehicle (0.5% HPMC 0.1% Tween 80) for 4 days and sampled two hours post dose on the fourth day.