KRAS mutations and outcomes for patients with stage IV NSCLC treated with frontline platinum/pemetrexed based chemotherapy.

Abstract

e18139 Background: KRAS mutations are the most common driver mutation indentified in NSCLC, occurring in 20 - 30% of adenocarcinomas. While several studies suggest KRAS predicts for lack of response to TKI therapy, few data exist regarding its association with outcomes for patients treated with cytotoxic chemotherapy. This study explores the association between KRAS mutations and outcomes (RR, PFS) in a cohort of patients treated with frontline platinum/pemetrexed (PPm) based therapy. Methods: In this retrospective chart review, we evaluated RR and PFS for 16 KRAS + EGFR – pts treated with carboplatin (AUC 5-6) or Cisplatin 75mg/m2 and (Pm)pemetrexed (500 mg/m2) +/- (B)bevacizumab 15mg/kg. For comparators, we identified 19 KRAS - EGFR - patients treated with the same regimen. Maintenance therapy with Pm or Pm+B was given at the discretion of the treating physician. KRAS and EGFR mutational status were assessed by RT-PCR on tumor tissue collected at first diagnosis. RR was assessed using RECIST criteria. Kaplan-Meier estimates for PFS were evaluating using log rank test. Fisher exact test was used to assess the association between KRAS mutation status and response rate. Results: The groups were similar in age (KRAS + mean 61 vs. 60; p=0.87), gender (62% vs. 57% F; p= 0.9), ECOG 2 (0 vs. 10%,p=0.47), smoking hx (93% vs. 94% current/former smokers, p=0.7), brain mets (0% vs. 18% p=0.22), mean number induction cycles (4 in each, p=0.6), cisplatin and bevacizumab use (12% vs 10%, p > 0.1;10% vs. 40%, p=0.10). Pm maintenance was used in 31% KRAS+ (5/16) and 26% KRAS-(5/19) (p=0.79). P+B maintenance was used in 12% (2/16) and 5% (1/19) (p=0.70). RR was 56% in the KRAS + (9/16) vs. 36% KRAS- (7/19) respectively (p=0.3). There was a statistically significant improvement in PFS in the KRAS + group (10.3 mos vs. 5.7 mos, p =0.03). Conclusions: In this small retrospective review, KRAS mutations appeared to be associated with a non-significant improvement in RR and significant improvement in PFS for patients treated with frontline PPm based therapy. Future prospective studies should investigate and validate the predictive value of KRAS for this cytotoxic regimen. [Table: see text]