P2.12-15 Prognostic and Predictive Covariates in Limited-Stage Small-Cell Lung Cancer: Analysis of the Phase 3 CONVERT Trial

Abstract

The majority of patients with limited-stage small cell lung cancer (LS-SCLC) progress after concurrent chemo-radiotherapy (cCTRT). Data from the CONVERT trial was analysed to investigate prognostic and predictive covariates in LS-SCLC patients. CONVERT is an international multi-centre phase III trial that randomly assigned fit patients to receive either twice-daily (45Gy in 30 fractions) or once-daily (66Gy in 33 fractions) radiotherapy starting on day 22 of chemotherapy cycle 1 (ClinicalTrials.gov NCT00433563). Chemotherapy consisted of 4 or 6 cycles (centres choice) of cisplatin and etoposide. Prophylactic cranial irradiation was offered, if indicated. The following covariates were investigated for prognostic and predictive significance (benefit from twice-daily radiotherapy and completion of cCTRT): clinical (age, performance score, TNM staging, smoking status, weight loss >10% and lung function), laboratory (alkaline phosphatase, sodium and lactate dehydrogenase) and dosimetric (gross tumour volume (GTV), heart and lung dose). Completion of chemotherapy was defined as delivery of all planned cycles while completion of radiotherapy was defined as delivery of all fractions. Results of the multivariate regression analysis of overall survival (OS) and progression-free survival (PFS) were reported after correcting for multiple comparisons. Of 547 patients recruited to CONVERT, 449 with complete covariate and outcome data were eligible for this analysis. GTV was the strongest prognostic covariate of OS (hazard ratio (HR) 1.37 (95% confidence interval (CI) 1.21-1.56); p<0.001). The addition of weight loss and performance score modestly improved the concordance probability (0.59 to 0.61) of this model. The HR for OS between high and low risk groups using this model was 2.72 (95% CI 1.94-3.81), median OS: 21 months (95% CI 19-26) vs 44 months (95% CI 36-not reached), respectively. For PFS, the HR between high and low risk groups was 2.55 (95% CI 1.86-3.5), median PFS: 13 months (95% CI 12-15) vs 26 months (95% CI 18-48), respectively. None of the tested covariates predicted patient benefit from twice-daily radiotherapy. Increase in patient age (continuous variable) predicted non-completion of planned chemotherapy (p=0.002). Due to the high completion of radiotherapy (86% in twice-daily and 80% in once-daily group), a multivariate analysis to predict radiotherapy completion was not performed. We report a clinical prognostic model in LS-SCLC, providing information that clinicians can relay to their patients to aid clinical decisions. The addition of biological covariates could help refine these prognostic models in the future.