Clinical Trial Design in Small Cell Lung Cancer: Surrogate End Points and Statistical Evolution

BackgroundSmall-cell lung cancer (SCLC) is a disease for which few recent therapeutic advances have been achieved. SCLC trial design and reporting may have an impact on the interpretation of studies. Furthermore, the use of surrogate end points in SCLC has not been explored.Material and MethodsThrough examining SCLC trials published in the Journal of Clinical Oncology (JCO) (8471 patients from 66 trials between 1983 and 2010), we examined how SCLC trial reporting and design has evolved, determining if the type I error, power, and sample size calculations were provided. We assessed primary end points for all trials and sought to discover surrogate end points for overall survival (OS).ResultsThere was increased reporting of statistical design in power (16.7% in 1986-1996 to 77.8% in 2006-2010; P = .001) and type I error (22.2% in 1986-1996 to 72.2% in 2006-2010; P = .005). Of trials published in 1986 to 1996, 72.2% failed to report a primary end point, whereas only 5.56% of trials conducted in 2006 to 2010 failed to do so (P = .004). Of phase II trials, primary end points were identified as response rate (RR) in 65%, OS in 25%, and progression-free survival (PFS) in 10%.ConclusionThere is a strong correlation between RR and both PFS (P = .013) and OS (P = .012) in extensive disease (ED). RR (P = .029) exhibits a negative trend over time, with a dramatic and significant decrease in RR across all studies starting in 2005. A strong correlation exists between PFS and OS for limited disease (LD) (P = .036) and ED (P = .058). We found no change in OS (P = .383) over time.