Unveiling novel transcriptomic prognostic biomarkers for specific breast cancer subtypes and treatment regimens

Breast cancer (BRCA) is the most common cancer in women worldwide, yet current gene expression panels offer limited insight into treatment responses across different subtypes and therapies. This study aimed to identify reliable biomarkers for predicting treatment outcomes in specific BRCA subtypes and treatment regimens.

This study analyzed transcriptomic data from The Cancer Genome Atlas to identify differentially expressed genes (DEGs) in patient groups treated with different combinations of hormone therapy (H), chemotherapy (C), radiotherapy (R), and targeted therapy (T). Non-negative matrix factorization clustering was performed to stratify patients into clusters representing different BRCA subtypes. Functional enrichment analysis was performed, and survival assessments were conducted using the METABRIC dataset.

A total of 1,148 DEGs were identified across treatment regimens, with 75 common DEGs shared across multiple regimens. Among these, 12 candidate biomarkers were associated with luminal subtypes treated with H, including LRP1B, of which high expression predicted cancer recurrence. In triple-negative breast cancer (TNBC) treated with C, 76 candidate biomarkers were identified, including TTYH1 for recurrence and ANXA8L1 and MPZ for non-recurrence. Functional analyses identified intermediate filament organization and keratinization as pathways associated with specific candidate biomarkers of TNBC following C. Survival analysis using METABRIC strengthened the prognostic ability of LRP1B and TTYH1 to predict worse survival and ANXA8L1 and MPZ to predict prolonged survival, with four additional prognostic biomarkers.

This study identified gene expression prognostic biomarkers for luminal and TNBC subtypes, thereby supporting personalized therapies. Further experimental validation is required to confirm these findings for clinical application.

No.