168P - Transcriptional profiling of metastatic hormone sensitive prostate cancer (mHSPC) and distinct features are associated with clinical outcome

Background

Although many treatment options have been adopted in patients with metastatic hormone sensitive prostate cancer (mHSPC), such as early docetaxel and AR-targeting therapy, there are currently no optimal biomarkers to guide treatment in these patients. Here, we sought to define transcriptomic landscape in mHSPC patients treated by either early docetaxel or abiraterone acetate by performing RNA sequencing.

Methods

Transcriptomic profiling of 52 mHSPC was performed by whole transcriptomic sequencing (WTS). For molecular classification, NMF (Non-negative Matrix Factorization) algorithm was used. For perform Gene Set Enrichment Analysis (GSEA), the Hallmark gene sets from the Molecular Signatures Database (MSigDB) were used. Gene sets in this study such as oncogenes, tumor suppressor genes (TSGs), chromosomal instability (CIN25, CIN70), human embryonic stem cell (hES) were used. Single sample Gene Set Enrichment Analysis (ssGSEA) was computed using the “GSVA” package. Gene ontology (GO) analysis was performed David website. Data processing and analysis were performed using R/Bioconductor libraries.

Results

We classified 52 mHSPC patients into two molecular subtypes by using NMF algorithm, particularly subtype1 (S1) and subtype2 (S2), respectively. Of note, S2 had significantly shorter survivals (r-PFS, PSA-PFS, FFS, and Time to CRPC) compared to S1. Additionally, S2 was primarily characterized as highly mitotic cell cycle and cytokinesis. S1 was characterized by metabolic process. Next, ssGSEA and GSEA using Hallmark gene sets showed that G2M checkpoint, E2F targets, and MYC targets were significantly enriched in the S2. Besides, gene sets including oncogenes, TSGs, CIN25/70, hES associated with an aggressive phenotype were activated S2. Up-regulated genes in S2 was associated with poor prognosis using independent two data sets, and was activated primary prostate tissues and metastatic prostate tissues.

Conclusions

In sum, this study demonstrated the utility of molecular subtyping based on transcriptomic analysis to guide prognostication and potential selection of patients in mHSPC.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

National Research Foundation of Korea.

Disclosure

All authors have declared no conflicts of interest.