Smooth Muscle Disruptive Feature Defines Molecular Subtyping, Risk Stratification, and Metastatic Potential in Prostate Cancer.

BACKGROUND

Smooth muscle (SM) invasion represents a critical feature of prostate cancer (PCa) progression and metastasis. This study aimed to develop a SM-based molecular score for progression and metastasis forecasting for PCa.

PATIENTS AND

METHODS

A total of 952 patients with PCa with transcriptomic sequencing data and corresponding clinical information were included in this study to develop and validate a SM-related signature (SM score). SM-related molecular subtypes were identified on the basis of 35 smooth muscle-related gene sets using unsupervised clustering. Weighted gene coexpression network analysis (WGCNA) was employed to identify key SM-related genes. The SM score was subsequently developed using machine learning across, The Cancer Genome Atlas Prostate Adenocarcinoma (TCGA-PRAD) training cohort and four independent external validation cohorts.

RESULTS

Utilizing 35 gene sets related to SM function, we identified pronounced SM disruption in PCa compared with normal prostate tissues. Unsupervised consensus clustering stratified patients into two distinct molecular subtypes, which exhibited significant differences in prognosis and immune infiltration profiles. We subsequently developed a robust prognostic signature termed SM score, and multiomics analyses confirmed a strong association between the SM score and SM disruption. The SM score consistently demonstrated superior prognostic performance across all cohorts and showed potential as a biomarker for responsiveness to immunotherapy. Notably, SM score was significantly elevated in metastatic lesions compared with primary tumors and normal tissues and accurately predicted metastatic potential.

CONCLUSIONS

SM disruption is associated with tumor progression in PCa. The SM score captures SM disruption and predicts PCa progression and metastasis potential, aiding personalized immunotherapy decisions.