Background: The use of platinum-based doublet chemotherapy combined with immune checkpoint inhibitors (ICIs) has demonstrated promising outcomes in the first-line treatment of extensive-stage small cell lung cancer (ES-SCLC). Relapsed SCLC has been classified into "sensitive" or "refractory" relapse types according to cutoff values (60 or 90 days) of duration from the last chemotherapy administration to disease progression. However, it is unclear whether these cutoff values can be applied to ICI combination therapy. Methods: We retrospectively analyzed a multicenter database of ES-SCLC patients who received second-line therapy after the platinum-etoposide plus ICI (atezolizumab or durvalumab). An optimal cutoff value for the platinum-free interval (PFI) was selected, which minimized the two-sided p-value and maximized hazard ratio (HR) regarding relapse type (sensitive or refractory according to a cutoff value) calculated from a multivariable Cox regression model for overall survival (OS) including performance status (PS) and sex as covariates. The internal validity of the selected cutoff value was assessed via two-fold cross-validation (CV) manner. Results: A total of 101 patients (61 deaths) from 10 hospitals were included in the study. The median follow-up period was 21.1 months. The optimal cutoff value was 75 days (p=0.0002), and when applying this cutoff value, median OS was 15.9 and 5.0 months of sensitive- (n=51) and refractory- (n=50) relapsed patients, and the HR calculated from a two-fold CV was 3.13 (95% confidence interval [CI], 1.66 to 5.90). Additionally, traditional cutoff values of 60 and 90 days also predicted prognosis better, but cutoff values of 110 days or longer did not (Table). Conclusions: Even in the era of combined immunotherapy in ES-SCLC patients, the threshold days for classifying as sensitive- or refractory- relapse did not exhibit a significant change compared to the pre-immunotherapy era. Although further validation studies with a larger sample size would be needed, relapse type classification using the selected cutoff value of 75 days is worth considering as a new prognostic factor for relapsed ES-SCLC patients.

*: p<0.05, **: p<0.01, ***: p<0.001, n.s: not significant.