Impact of KRAS mutation subtype and concurrent pathogenic mutations on non-small cell lung cancer outcomes

Concurrent genetic mutations are prevalent in KRAS-mutant non-small cell lung cancer (NSCLC) and may differentially influence patient outcomes. We sought to characterize the effects of KRAS mutation subtypes and concurrent pathogenic mutations on overall survival (OS) and PD-L1 expression, a predictive biomarker for anti-PD-1/PD-L1 immunotherapy.

We retrospectively identified patients with KRAS-mutant NSCLC at a single institution and abstracted clinical, molecular, and pathologic data from electronic health records. Cox regression and multinomial logistic regression were used to determine how KRAS mutation subtypes and concurrent pathogenic mutations are associated with OS and tumor PD-L1 expression, respectively.

A total 186 patients were included. Common KRAS mutation subtypes included G12C (35%) and G12D (17%). Concurrent pathogenic mutations were identified in TP53 (39%), STK11 (12%), KEAP1 (8%), and PIK3CA (4%). On multivariable analysis, KRAS G12D mutations were significantly associated with poor OS (hazard ratio [HR] 2.43, 95% confidence interval [CI] 1.15-5.16; P = 0.021), as were STK11 co-mutations (HR 2.95, 95% CI 1.27-6.88; P = 0.012). Compared to no (<1%) PD-L1 expression, KRAS G12C mutations were significantly associated with positive yet low (1-49%) PD-L1 expression (odds ratio [OR] 4.94, 95% CI 1.07-22.85; P = 0.041), and TP53 co-mutations with high (≥50%) PD-L1 expression (OR 6.36, 95% CI 1.84-22.02; P = 0.004).

KRAS G12D and STK11 mutations confer poor prognoses for patients with KRAS-mutant NSCLC. KRAS G12C and TP53 mutations correlate with a biomarker that predicts benefit from immunotherapy. Concurrent mutations may represent distinct subsets of KRAS-mutant NSCLC; further investigation is warranted to elucidate their role in guiding treatment.