What question did this study set out to answer?

The aim is to assess the prevalence of a BRCAness phenotype in prostate cancer patients lacking homologous repair gene mutations.

March 4, 2026

Profiling BCRAness phenotype in public prostate cancer datasets without homologous repair gene mutations.

Key Points

The aim is to assess the prevalence of a BRCAness phenotype in prostate cancer patients lacking homologous repair gene mutations.
Analyzed RNA sequencing data from public datasets to identify oncogenic DNA repair mutations.
Used the RNAmut tool to detect expressed somatic variants from RNA-seq data.
Filtered variants with a variant allele frequency (VAF) of ≥0.4.
Evaluated the BRCAness phenotype using the expHRD algorithm to derive HRD scores.
Included datasets from treatment-naïve localized tumors and metastatic samples from mCRPC.
In TCGA-PRAD, 1.3% of patients without HRR mutations exhibited a BRCAness profile.
In GSE216490, 4.2% showed BRCAness, and in GSE54460, 2.0% were classified similarly.
20.1% of metastatic tumors without HRR mutations displayed high BRCAness scores, indicating a higher prevalence in metastatic disease.

Abstract

223 Background: Co-inhibition of PARP and androgen receptor activity may enhance antitumor efficacy irrespective of HRR alterations, as suggested by TALAPRO and PROPEL. This may result from pathway crosstalk or alternative mechanisms generating a BRCAness phenotype not captured by standard genomic panels. Identifying these mechanisms is critical to optimize patient selection. We aimed to determine the prevalence of transcriptomic BRCAness in prostate cancer patients without HRR mutations across publicly available datasets, including primary tumors and metastases. Methods: RNA sequencing data were analyzed to identify oncogenic DNA repair mutations using RNAmut, a tool that detects expressed somatic variants from RNA-seq data. Variants identified as oncogenic were subsequently filtered, retaining those with a variant allele frequency (VAF) ≥0. 4. The BRCAness phenotype was evaluated with expHRD, a transcriptome-based algorithm deriving a homologous recombination deficiency (HRD) score. Tumors with expHRD ≥42 were classified as HRD-high (BRCAness), consistent with the Myriad MyChoice CDx cutoff. Datasets included treatment-naïve localized tumors (TCGA-PRAD, n=497; GSE216490, n=95; GSE54460, n=98) and metastatic samples from castration-resistant prostate cancer (mCRPC; GSE147250, n=138). Results: In primary prostate cancer patients from the TCGA-PRAD dataset, a high BRCAness score was associated with faster disease progression, validating the tool's ability to identify this profile. Within the TCGA dataset, 1. 3% of patients without mutations in homologous recombination repair (HRR) genes exhibited a BRCAness profile. Similar findings were observed in the GSE216490 dataset, where 4. 2% of patients showed BRCAness, and in the GSE54460 dataset, where 2. 0% of patients were classified as such. A higher prevalence of the BRCAness phenotype was observed in metastatic compared with localized disease: among patients without HRR mutations, 20. 1% of metastatic tumors displayed high BRCAness scores. Conclusions: A high prevalence of BRCAness was observed in HRR-negative mCRPC, which may partly explain the clinical benefit of PARP inhibitor and ARPI combinations in all-comer populations. Validation of this strategy in pivotal trials is essential to refine patient selection and optimize therapeutic strategies.

Perguntar à IA

Bookmark