Conclusions
- We report a large real-world dataset evaluating outcomes with check-point inhibitors in NSCLCs withKRAS and specific co-mts. Across the subgroups, KKL (KRAS mt/STK-11 mt/KEAP-1 mt) demonstrated universally poor outcomes in all KRASsubtypes; irrespective of PD-L1 expression.
- Among all KRAS co-mutant groups, K-only NSCLC tumors showed the best prognosis, followed by KL and KP groups while KKL showed the worst outcome.
- In the PD-L1 <1% group, KP group showed worse outcome than K-only however comparable outcomes to K-only in PD-L1 positive (TPS ≥1%).
- Interestingly, positive TPS score was not associated with significantly improved outcome in the molecular groups investigated and was in fact, associated with worse outcome in KKL. Pts with KKL co-mts have adverse post-IO outcomes in TPS ≥1% but favorable in TPS <1%.
- These observations emphasize that co-mutation patterns have a clear association with clinical outcomes inKRAS-mt NSCLC and must be used in predictive models for individualized therapy while the role of PD-L1 score may be limited in KRAS-mt NSCLC.