Impact of STK11 and KRAS co-mutations on outcomes with immunotherapy in non-small cell lung cancer.

e15135 Background: While the use of monoclonal antibodies targeting the PD-1 axis in metastatic non-small cell lung cancer (NSCLC) continues to expand since initial FDA approval in 2015, factors predictive of response still remain to be determined. Mutation status may provide insight as to which subgroups exhibit resistance to checkpoint inhibitor therapy. Methods: We conducted a single center retrospective analysis of patients with metastatic NSCLC treated at the University of Miami / Sylvester Comprehensive Cancer Center who underwent next-generation sequencing (NGS) and identified patients that harbor either STK11 mutation alone (S) or co-mutations with STK11 and KRAS (S/K). Genomic results were obtained from Guardant360 and Foundation One testing in blood and tissue, respectively. Results: We identified 37 S patients and 36 S/K patients and determined no significant differences in progression-free survival (PFS). However, overall survival (OS) was significantly increased in patients with S/K co-mutation (20.3 ± 4.1 months) vs. patients with S alone (11.9 ± 1.9 months, p = 0.028). Furthermore, S/K patients who received immunotherapy had longer OS (20.7 ± 5.6 months) vs. S patients (13.6 ± 3.4 months, p = 0.049). We further investigated any population-specific factors that may contribute to the enhanced survival in the S/K cohort that had received immunotherapy, as previous studies have reported that STK11/KRAS co-mutations may confer a resistance to PD-1 axis-directed therapy. We found that S/K patients were older at diagnosis and were more likely to have received nivolumab (as compared to newer therapies pembrolizumab and atezolizumab). In addition, S/K patients were more likely to have longer smoking histories (81% smoked more than 30 pack-years at diagnosis) vs. S patients (53%, p = 0.01) and also had a higher number of additional targetable mutations found on NGS (4.7 ± 0.4 vs. 2.9 ± 0.3, p = 0.001). The most commonly identified mutations were TP53 (30%), KEAP1 (19%), CDKN2A/B (18%), SMARCA4/BRG1 (16%), and ARID1A (11%). Conclusions: Our study reveals an interesting analysis of potential predictors of resistance to immunotherapy with the utilization of precision medicine in combination with patient characteristics to identify the most appropriate treatment regimens for patients with NSCLC. Further studies will explore whether patients receiving immunotherapy as first line could overcome any inherent resistance to PD-1 axis-directed therapy from non-modifiable factors at diagnosis.

Duke Scholars

Author Fahmin Basher Medicine, Hematologic Malignancies and Cellular Therapy