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

Area of Science:

• Oncology
• Genomics
• Bioinformatics

Background:

• Small cell lung cancer (SCLC) presents a poor prognosis with limited treatment advancements.
• The role of surgery and patient selection for SCLC remain debated.
• Transcriptomic alterations in SCLC prognosis are not well-defined.

Purpose of the Study:

• To identify transcriptomic alterations linked to prognosis in SCLC.
• To develop a gene expression-based risk signature for predicting overall survival (OS) in surgically treated SCLC patients.
• To assess the signature's performance against existing prognostic parameters.

Main Methods:

• Integrative transcriptome analysis of three SCLC gene expression datasets (GSE30219, GSE43346, GSE149507).
• Identification of differentially expressed genes (1734 up-regulated, 2907 down-regulated).
• Application of Cox-Mantel, Cox regression, and Lasso regression for risk signature development.
• Internal and external validation using EGAD00001001244 and GSE60052 cohorts.

Main Results:

• A novel gene expression-based risk signature was developed and validated.
• High-risk score patients exhibited significantly poorer overall survival.
• The risk signature demonstrated superior discriminatory performance compared to other parameters.
• Multivariate analysis indicated the signature's potential as an independent prognostic predictor.

Conclusions:

• The identified prognostic genes are involved in key pathways like transcription regulation, cell cycle, metabolism, and angiogenesis.
• The developed risk signature can predict overall survival in SCLC patients undergoing surgery.
• Further validation in larger cohorts is necessary to confirm clinical utility for decision-making in SCLC treatment.