Background: Patients with small-cell lung cancer (SCLC) that progress after first-line therapy have a dismal prognosis. Second-line chemotherapy is effective for some patients, but there is little evidence for third-line chemotherapy. The purpose of this study is to evaluate the efficacy of third-line chemotherapy and clarify prognostic factors for patients receiving it. Methods: Between November 2001 and October 2011, 202 of 648 patients who were consecutively diagnosed with SCLC at the National Cancer Center Hospital East received third-line chemotherapy. We performed multivariate Cox regression analysis to identify the prognostic factors of overall survival after third-line chemotherapy (OS). Results: At the start of third-line chemotherapy, median age was 66 years (range 38-83), male/ female: 168/34, ECOG performance status (PS) 0/1/2/3: 22/122/49/9, stage at diagnosis LD/ED: 88/114, response to second-line chemotherapy CR/PR/SD/PD: 3/95/56/48, and median progression free survival after second-line chemotherapy (PFS2) : 4.4 months. 155 and 47 patients received single-agent and platinum-based regimen, respectively. Overall response rate was 17.8%, which was significantly associated with response (CR/PR) to second-line chemotherapy (23.5% vs. 12.5%, p=0.042) and PFS2 (25.6% vs. 7.4%, p=0.024). Median PFS and OS from third-line chemotherapy were 2.7 months and 5.1 months, respectively. Multivariate Cox analysis identified PS 0-1 (hazard ratio [HR], 0.38; 95% confidence interval [CI], 0.27-0.55; p < 0.001) and PFS2 ≥5 months (HR, 0.59; 95%CI, 0.42-0.81; p=0.001) were the independent prognostic factors indicating better OS. An optimal threshold of 5 months of PFS2 (≥5/<5 months: 82/120) was selected based on the concordance index adjusted by PS. Conclusions: PS 0-1 and PFS2 ≥5 months were associated with favorable prognosis among SCLC patients receiving third-line chemotherapy. These two factors would be worth considering as stratification factors when conducting future clinical trials.