Presentation Abstract

Abstract Number: LB-1
Presentation Title: The BATTLE trial (Biomarker-integrated Approaches of Targeted Therapy for Lung Cancer Elimination): personalizing therapy for lung cancer
Presentation Time: Sunday, Apr 18, 2010, 11:25 AM -11:45 AM
Location: Exhibit Hall D, Washington Convention Center
Author Block: Edward S. Kim, Roy S. Herbst, J Jack Lee, George R. Blumenschein Jr., Anne Tsao, Christine M. Alden, Ximing Tang, Suyu Liu, David J. Stewart, John V. Heymach, Hai T. Tran, Marshall E. Hicks, Jeremy Erasmus Jr., Sanjay Gupta, Garth Powis, Scott M. Lippman, Ignacio I. Wistuba, Waun K. Hong. University of Texas M. D. Anderson Cancer Center, Houston, TX
Abstract Body: Background: Patients (pts) with chemorefractory non-small cell lung cancer (NSCLC) have few options for effective treatment. BATTLE is a hypothesis-driven prospective study that identifies biomarkers (BMs) to predict tumor response and thus may help select personalized therapy for lung cancer pts.
Methods: Pts enrolled in this phase II adaptively randomized study required fresh core needle biopsy specimens to test 11 BMs from 4 NSCLC molecular pathways: EGFR, KRAS, and BRAF mutation (M) (by PCR), EGFR and Cyclin D1 copy number (by FISH), and VEGF, VEGFR, 3 RXR receptors and Cyclin D1 (by IHC). Based on eligibility criteria and tumor BM analyses, pts were adaptively randomized into treatments: erlotinib (E) 150 mg qd; sorafenib (S) 400 mg bid; vandetanib (V) 300 mg qd; E 150 mg + bexarotene (B) 400 mg/m2 qd. Pts with prior E were randomized only between S and V. The primary endpoint was 8-week disease control (DC).
Results: From 11/06 to 10/09, 255 pts were randomized to E (59 pts), V (54 pts), EB (37 pts), S (105 pts). Demographics: Median age 62 yrs (26-84); male 54%; ECOG PS 0-1 86%, PS 2 14%; Caucasian 82%, Asian 5%, other 13%; never/former/current smokers 22%/69%/9%; adenocarcinoma 63% squamous cell 18%, NSCLC NOS 19%; prior E 45%, median prior therapies for metastatic NSCLC: 2 (range 1-9); prior brain metastases 33%. 244 pts were evaluable for 8 wk DC. All 11 BMs were assessable in 215 pts. Biopsy sites were lung 55%, liver/adrenal 19%, other 26%. Pneumothorax incidence was 11.5%, and 6.5% of pts had treatment-related grade 3-4 toxicity. EGFR status included M in 15%, FISH amplification (A) in 16% and high polysomy in 28%; Other BMs were KRAS M in 20%; VEGF/R2 staining in 83%; RXR alpha nuclear staining in 79%; Cyclin D1 staining in 53%. Overall DCR at 8 weeks was 46%, median overall survival (OS) was 9 months, 1 year survival was 39%, and progression-free survival (PFS) was 1.9 months. Better DC was seen with EGFR M for E (p=0.04); Cyclin D1 IHC positivity (IHC+) (p=0.011) and EGFR FISH A (p=0.006) for E + B; VEGFR2 IHC+ for V (p=0.05); and absence of EGFR M (p=0.012) or high polysomy (p=0.048) for S. Pts with both EGFR M and FISH A had 100% DC (n=6) with E and 0% DC (n=6) with S. Pts with KRAS M tended to respond better with S (8-wk DC 61%) compared to other three regimens (8-wk DC 32%) (P=0.11). Pts with mutant KRAS Cys amino acid (aa) substitution had worse OS (all pts) and PFS (S-treated pts) compared to pts with wild type or all other KRAS aa substitutions (p=0.015 and p=0.013).
Conclusions: BATTLE is the first completed biopsy-mandated study in pretreated NSCLC. Our pre-specified hypotheses regarding BMs and targeted treatment were confirmed. Identifying proper BMs for a favorable population is a step towards personalizing therapy. BATTLE establishes a new paradigm to investigate BMs and molecularly targeted treatments in lung cancer pts. DoD W81XWH-6-1-0303
Webcast Available: Yes