Two parallel randomized phase II studies of selumetinib (S) and erlotinib (E) in advanced non-small cell lung cancer selected by KRAS mutations.

Abstract

8026 Background: KRAS mutations are present in 20% of NSCLC and are associated with primary resistance to erlotinib (E). KRAS mutations result in constitutive activation of the Ras/Raf/MEK/ERK pathway. Selumetinib (S) (AZD6244, ARRY-142886) is a selective and uncompetitive inhibitor of MEK kinase. Preclinical studies demonstrated increased activity in NSCLC cell lines using S+E regardless of KRAS status. Methods: Advanced NSCLC patients with progressive disease after platinum based chemotherapy +/- one other treatment were stratified by KRAS status, which was centrally assessed using macrodissection and pyrosequencing for all mutations involving codons 12, 13, and 61. Patients were randomized to receive either the standard of care E or a combination of S+E in KRAS wild type (wt) and S alone or S+E in patients with KRAS mutations. Single agent E and S were administered orally at 150 mg daily and 75 mg twice daily respectively. Combination dosing were S 150 mg every morning and E 100 mg every evening. The primary endpoint for the KRAS wt group was PFS and for the KRAS mutant group objective response rate. Results: From March 2010 to January 2013, 79 patients screened; 78 enrolled: KRAS mutant, 39; KRAS wt, 40; M/F (39:39); median age: 64 years (33-84); median WHO PS 1(0-2); 66 former and 13 never smokers; 67 adenocarcinoma, 9 squamous cell. Three patients died of complications prior to first evaluation (1 coronary disease, 1 pulmonary fibrosis, and 1 disease progression) of which none were related to S. Dose reductions occurred in 5% E, 40% S, and 56% E+S. Most grade 3/4 AEs occurred in combination therapy; diarrhea (23%), fatigue (23%) lymphopenia (13%), myositis (10%), dyspnea (10%), rash (7%). Discontinuation due to AEs was 8% all occurring in S+E cohorts. Conclusions: This study failed to show improvement of combination therapy over single agent in KRAS wt and mutant patients. Toxicity was increased in the combination arms. Interestingly, the KRAS mutant cohort had longer PFS than KRAS wt patients, although not statistically significant (p=0.11). Clinical trial information: NCT01229150. [Table: see text]