Phase II randomized study of biomarker-directed treatment for non-small cell lung cancer (NSCLC): The BATTLE (Biomarker-Integrated Approaches of Targeted Therapy for Lung Cancer Elimination) clinical trial program.

8024 Background: Treatment for patients (pts) with recurrent NSCLC has limited efficacy despite the use of new targeted agents. Identifying biomarkers to predict tumor response will help personalize therapy for individuals. METHODS Eligible pts had prior chemotherapy, ECOG PS 0-2, and when enrolled, required 2 fresh core needle biopsy specimens to test 11 biomarkers related to 4 molecular pathways in NSCLC: EGFR, Kras, and Braf gene mutation (PCR-based sequencing), EGFR and Cyclin D1 copy number analyses (FISH), and 6 proteins via IHC (VEGF/R and RXR receptors/Cyclin D1). Based on eligibility and tumor biomarker analyses, pts were randomized into 1 of 4 treatments: erlotinib (E) 150 mg qd; sorafenib 400 mg bid, vandetanib 300 mg qd; E 150 mg + bexarotene 400 mg/m2 qd. The primary endpoint is 8-week progression-free status. The first 97 pts were equally randomized, with subsequent pts adaptively randomized under a Bayesian framework. RESULTS Since 11/2006, 227 pts have been enrolled and 171 randomized and treated (median age 62 yrs, 89 males, 137 Caucasians, 11 Asians, 107 former and 38 never smokers). Accrual has proceeded well (average 9 pts per month). 112 of the biopsied lesions were lung, with a pneumothorax rate of 12.1% (15 of 124 pts; grade 1-2 only; lung, mediastinal and pleural sites). 168 pts (74%) have complete biomarker profiles and 7 (3%) have partial. HISTOLOGY adenocarcinoma (75%), squamous (11%), large cell (13%). Mutation frequency was 19% EGFR, 15.4% Kras, and 2.9% Braf. 2 tumors contained both EGFR and Kras mutations, and 2 cases had both Kras and Braf mutations. EGFR polysomy by FISH was 36.4%, and gene amplification 21%. Cyclin D1 amplification was 16.3%. Treatment-related adverse events (grade 3 or more) were 6.5%. CONCLUSIONS We have demonstrated that biopsies are safe and feasible for biomarker-directed studies. The study continues enrollment to 200 evaluable pts at which point the results will be unblinded. BATTLE is one of the first studies in advanced lung cancer to prospectively utilize biomarker analysis of fresh biopsies to direct pt treatment, and is a step towards personalizing therapy in NSCLC. Supported by grant DoD W81XWH-6-1-0303. No significant financial relationships to disclose.