Ontology highlight
ABSTRACT: Background
Homologous recombination deficiency (HRD) is closely associated with patient prognosis and treatment options in prostate cancer (PCa). However, there is a lack of quantitative indicators related to HRD to predict the prognosis of PCa accurately. Methods
We screened HRD-related genes based on the HRD scores and constructed an HRD cluster system to explore different clinicopathological, genomic, and immunogenomic patterns among the clusters. A risk signature, HRDscore, was established and evaluated by multivariate Cox regression analysis. We noticed that SLC26A4, a model gene, demonstrated unique potential to predict prognosis and HRD in PCa. Multi-omics analysis was conducted to explore its role in PCa, and the results were validated by qRT-PCR and immunohistochemistry. Results
Three HRD clusters were identified with significant differences in patient prognosis, clinicopathological characteristics, biological pathways, immune infiltration characteristics, and regulation of immunomodulators. Further analyses revealed that the constructed HRDscore system was an independent prognostic factor of PCa patients with good stability. Finally, we identified a single gene, SLC26A4, which significantly correlated with prognosis in three independent cohorts. Importantly, SLC26A4 was confirmed to distinguish PCa (AUC for mRNA 0.845; AUC for immunohistochemistry score 0.769) and HRD (AUC for mRNA 0.911; AUC for immunohistochemistry score 0.689) at both RNA and protein levels in our cohort. Conclusion
This study introduces HRDscore to quantify the HRD pattern of individual PCa patients. Meanwhile, SLC26A4 is a novel biomarker and can reasonably predict the prognosis and HRD in PCa. Supplementary Information
The online version contains supplementary material available at 10.1186/s12967-022-03513-5.
SUBMITTER: Luo C
PROVIDER: S-EPMC9281181 | biostudies-literature |
REPOSITORIES: biostudies-literature