Sequential treatment with direct KRAS inhibitors: how to improve the management of KRAS G12C drugs

Lung cancer is responsible for 12% of cancer-related deaths worldwide (WCRF, https://www.wcrf.org). The most frequent diagnosed form of lung cancer is lung adenocarcinoma (LUAD). Mutant KRAS, particularly KRAS G12C, is the driver mutation in more than 25% of cases. KRAS driven cancers were thought to be directly undruggable for decades, but in 2021 two direct inhibitors, sotorasib (AMG510) and adagrasib (MTRX849), were FDA approved for the treatment of LUAD with KRAS G12C mutations. Even if they are effective, patients invariably develop resistance to sotorasib or adagrasib monotherapy, leaving limited possibility for further clinical treatment with targeted therapies. Our research aims to determine whether treating in a sequential mode patients who are resistant to sotorasib with adagrasib –and viceversa– can be successful in prolonging the response in vitro and in vivo. We generated Ras-less cells expressing KRAS G12C alone (LOH) or in presence of the wild-type endogenous KRas allele. We obtained adaptive-resistant cells to sotorasib (AMG-R) and adagrasib (MRTX-R) by treating them with an increasing concentration of drugs up to 5 uM for sotorasib and 2 uM for adagrasib and we sequenced them in order to identify potential selection of acquired alterations. In parallel, we also generated bona fide resistant Ras-less cells with mutations in cis with G12C (Y96D/R68S/H95R/H95Q). We investigated the sequential treatment efficacy in vitro and in vivo. Proliferation assays on adaptive-resistant cells revealed that, while MRTX-R cells are resistant to AMG, AMG-R cells are still sensitive to MRTX treatment. Afterwards, we studied the effects on MAPK pathway via WB analysis, which showed that both models are characterized by lower pERK levels upon sequential treatment with the reciprocal drug. Constitutively resistant models showed different drug-response profile dependent on the mutation: KRAS G12C/Y96D and G12C/R68S are resistant to both drugs and their combination, whereas KRAS G12C/H95R and G12C/H95Q are resistant to MRTX849 but still slightly sensitive to AMG510. In this situation, combination therapy appears to be more effective. Adaptive resistant in vivo models confirmed that AMG-R tumors are partially responsive to MRTX treatment whereas MRTX-R tumors are resistant to AMG treatment. MRTX-R models are resistant to sotorasib treatment whereas AMG-R cells are still partially sensitive to adagrasib. G12C/Y96D and G12C/R68S double mutant cells are resistant to both drugs in vitro. G12C/H95Q and G12C/H95R are resistant to adagrasib, but slightly sensitive to sotorasib in vitro. In vivo, MRTX-R tumors are resistant to sotorasib treatment whereas AMG-R tumors are still partially sensitive to adagrasib.