Combining RASG12C(ON) inhibitor with SHP2 inhibition sensitises immune excluded lung tumours to immune checkpoint blockade: a strategy for turning cold tumours hot

ABSTRACT Mutant selective drugs targeting the inactive, GDP-bound form of KRASG12C have been approved for use in lung cancer, but responses are short-lived due to rapid development of resistance. In this study we use a novel covalent tri-complex inhibitor, RMC-4998, that targets RASG12C in its active, GTP-bound form to investigate treatment of KRAS mutant lung cancer in various immune competent mouse models. While this RASG12C(ON) inhibitor was more potent than the KRASG12C(OFF) inhibitor adagrasib, rapid pathway reactivation was still observed. This could be delayed using combined treatment with a SHP2 inhibitor, RMC-4550, which not only impacted RAS pathway signalling within the tumour cells but also remodelled the tumour microenvironment (TME) to be less immunosuppressive and promoted interferon responses. In an inflamed, “hot”, mouse model of lung cancer, RASG12C(ON) and SHP2 inhibitors in combination drive durable responses by suppressing tumour relapse and inducing development of immune memory, which can also be induced by combination of RASG12C(ON) and PD-1 inhibitors. In contrast, in an immune excluded, “cold”, mouse model of lung cancer, combined RASG12C(ON) and SHP2 inhibition does not cause durable responses, but does sensitise tumours to immune checkpoint blockade, enabling efficient tumour rejection, accompanied by significant TME reorganization, including depletion of immunosuppressive innate immune cells and recruitment and activation of T and NK cells. These preclinical results demonstrate the potential of the combination of RASG12C(ON) inhibitors with SHP2 inhibitors to sensitize anti-PD-1 refractory tumours to immune checkpoint blockade by stimulating anti-tumour immunity as well as by targeting KRAS-driven proliferation in tumour cells. Competing Interest Statement J.D. has acted as a consultant for AstraZeneca, Jubilant, Theras, Roche and Vividion and has funded research agreements with Bristol Myers Squibb, Revolution Medicines and AstraZeneca. S.C.T has acted as a consultant for Revolution Medicines. C.B., E.Q. and J.A.M.S. are employees of Revolution Medicines. The other authors declare that they have no competing interests. Footnotes In this revision, corrections have been made to the bibliography which contained errors in referencing. In addition, minor changes have been made to the methods and acknowledgement sections. The rest of the paper is unchanged.