1546P Impact of stratification factors on outcomes in limited-stage small cell lung cancer: Analysis of CALGB 30610 (Alliance)/RTOG 0538

Abstract

Phase 3 trials are typically designed to account for baseline patient characteristics and other factors that may influence treatment outcomes, but stratification factors are inconsistently employed across trials and the impact of commonly used variables is not well defined in the contemporary treatment of limited-stage small cell lung cancer. Better refinement of prognostic baseline characteristics may allow for reduced complexity of clinical trial design. Analysis includes 638 patients enrolled on CALGB 30610/RTOG 0538 randomized to either 45 Gy twice-daily radiotherapy or 70 Gy once-daily radiotherapy concurrent with 4 cycles of chemotherapy. Stratification factors for randomization included gender, ECOG performance status (PS), weight loss > 5% prior to study entry, timing of initiating radiotherapy (chemotherapy cycle 1 vs cycle 2), radiotherapy planning technique (3D conformal vs intensity modulated), and carboplatin or cisplatin-based chemotherapy. The only variable significantly associated with improved survival was female gender (HR 0.79, p = 0.021), while a trend toward improved survival was observed with the use of cisplatin vs. carboplatin (HR 0.81, P = 0.121). There was also a trend toward better survival for ECOG PS 1 vs. ECOG PS 2 (HR =0.78, p =0.15), but not for ECOG 0 vs. ECOG 1 (HR=1.05, p=0.59). Weight loss, timing of initiating radiotherapy, and radiotherapy technique did not appear to impact survival. A similar impact of each factor was observed regardless of treatment arm. These data may help simplify trial design and selection of stratification factors for future prospective trials. In particular, outcomes were virtually indistinguishable for ECOG PS = 0 and ECOG PS = 1 patients, suggesting these groups may be combined in future studies.