Background

The RAS pathway is frequently activated in cancers, leading to poor prognosis and resistance to therapy. We ask whether oncogenic RAS shapes the immune landscape of lung adenocarcinoma to find clinical targets for RAS-driven tumours.

Method

Following a reverse translation approach, we classified 517 human lung adenocarcinoma (LUAD) samples from TCGA according to their RAS activity using a RAS transcriptional signature (RAS84) we developed. We characterised the groups and we model them in preclinical studies to define new combinations of treatment for these patients.

Results

RAS84 classified LUAD in five RAS Activity Groups (RAG-0 to RAG-4). Each RAG was associated with specific genomic alterations and increased level of oncogenic RAS activity from RAG-0 to RAG-4. Despite having the worse prognostic, RAG-3 and RAG-4 were highly infiltrated with leucocytes (CIBERSORT and Danaher method) and strongly expressed IFN-stimulated genes (ISGs), necessary to achieve an anti-tumoural immune response. RAG-3 and RAG-4 also expressed high levels of several immune checkpoint genes (e.g. CD274 (PD-L1), PDCDL2 (PD-L2), PVR, TIGIT) and deregulated genes related to adenosine metabolism or function (e.g. NT5E (CD73), ADORA2B, SLC29A4 (ENT4)). The latter suggests an increase in interstitial adenosine, known to dampen the immune system. Using MEK or RASG12C inhibitors, we validated the RAS-MAPK-dependent regulation of adenosine genes in KRAS-mutant human lung cell lines and several Kras-driven mouse models of lung adenocarcinoma and the direct regulation of interstitial adenosine by RAS-MAPK. Finally, we established the KPAR1.3 (Kras^G12D/wt; Trp53^fl/fl; APOBEC3B) orthotopic model of RAG-4 and showed that a combination of anti-PD-1, anti-CD73 and anti-TIGIT improved survival of KPAR1.3 tumour-bearing mice.

Conclusion

This data suggest oncogenic RAS activity induces immune evasion by increasing interstitial adenosine. 

Impact statement

Patients with RAS-driven lung cancers, stratified using our method, could benefit from targeting adenosine metabolism (anti-CD73) or activity (A2AR inhibitors) in combination with immune checkpoint blockade.