A 15-Gene-Based Risk Signature for Predicting Overall Survival in SCLC Patients Who Have Undergone Surgical Resection

SIMPLE SUMMARY: Small cell lung cancer (SCLC) is a high-grade neuroendocrine carcinoma with a poor prognosis, accounting for approximately 15% of lung cancer cases. Acquired resistance to standard chemotherapy is quite common in SCLC patients, and the survival benefit of surgery and the selection of surgical candidates are still controversial. This highlights the necessity of identifying predictive biomarkers that can serve to select patients who will benefit from various treatments and discovering novel molecular targets for SCLC treatment. In this study, for the first time, the association between tumoral transcriptional changes and prognosis was examined, and a novel multigene prognostic risk signature with strong discriminatory power was constructed and validated to predict the overall survival of SCLC patients who have undergone curative-intent surgical resection. The risk signature worked better than existing clinical and demographic parameters in predicting overall survival in patients with resected SCLC. Prognostic genes were predicted to have roles in pathways including regulation of transcription, cell cycle, cell metabolism, and angiogenesis. ABSTRACT: Small cell lung cancer (SCLC) is a malignancy with a poor prognosis whose treatment has not progressed for decades. The survival benefit of surgery and the selection of surgical candidates are still controversial in SCLC. This study is the first report to identify transcriptomic alterations associated with prognosis and propose a gene expression-based risk signature that can be used to predict overall survival (OS) in SCLC patients who have undergone potentially curative surgery. An integrative transcriptome analysis of three gene expression datasets (GSE30219, GSE43346, and GSE149507) revealed 1734 up-regulated and 2907 down-regulated genes. Cox-Mantel test, Cox regression, and Lasso regression analyses were used to identify genes to be included in the risk signature. EGAD00001001244 and GSE60052-cohorts were used for internal and external validation, respectively. Overall survival was significantly poorer in patients with high-risk scores compared to the low-risk group. The discriminatory performance of the risk signature was superior to other parameters. Multivariate analysis showed that the risk signature has the potential to be an independent predictor of prognosis. The prognostic genes were enriched in pathways including regulation of transcription, cell cycle, cell metabolism, and angiogenesis. Determining the roles of the identified prognostic genes in the pathogenesis of SCLC may contribute to the development of new treatment strategies. The risk signature needs to be validated in a larger cohort of patients to test its usefulness in clinical decision-making.