HRProfiler Detects Homologous Recombination Deficiency in Breast and Ovarian Cancers Using Whole-Genome and Whole-Exome Sequencing Data

Sep 23, 2026

Uploaded by Putri

HRD breast and ovarian cancers show characteristic mutational signatures and are sensitive to PARP inhibitors and platinum chemotherapy. Conventional HRD detection relies on screening canonical genes such as BRCA1 and BRCA2. This study trained the Homologous Recombination Proficiency Profiler (HRProfiler), a machine learning model using six mutational features, on sequencing data from 1,043 breast and 182 ovarian cancers. Performance was benchmarked on independent datasets totaling 417 breast and 115 ovarian cases, including retrospective clinical-trial data from PARP inhibitor–treated patients. Individual mutational signatures alone were insufficient for consistent HRD detection or response prediction. HRProfiler uniquely and consistently identified HRD in whole-exome–sequenced breast and ovarian cancers, supporting its use as a predictive and prognostic HRD biomarker to guide clinical decision-making.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Introduction

Homologous recombination as genomic stability mechanism

Clinical relevance of HRD and therapeutic sensitivity

Study rationale and approach: HRProfiler with sequencing data

HRProfiler Detects Homologous Recombination Deficiency in Breast and Ovarian Cancers Using Whole-Genome and Whole-Exome Sequencing Data Ammal Abbasi1,2,3,4,5, Christopher D. Steele1,2,3, Erik N. Bergstrom1,2,3, Azhar Khandekar1,2,3,4, Akanksha Farswan6, Rana R. McKay3, Nischalan Pillay6,7, and Ludmil B. Alexandrov1,2,3,4,5 | | | A | | B | S | | T | R | | A | | C | T | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | Breast and ovarian cancers harboring homologous recombination deficiency (HRD) are sensitive to PARP inhibitors and platinum chemotherapy. Conventionally, detecting HRD involves screening for defects in BRCA1, BRCA2, and other relevant genes. Recent analyses have shown that HRD cancers exhibit characteristic mutational signatures due to the activities of HRD-associated mutational processes. At least three machine learning tools exist for detecting HRD based on mutational patterns. In this study, using sequencing data from 1,043 breast and 182 ovarian cancers, we trained Homologous Recombination Proficiency Profiler (HRProfiler), a machine learning method for detecting HRD using six mutational features. The performance of HRProfiler was assessed against prior approaches using additional independent datasets of 417 breast and 115 ovarian cancers, including retrospective data from a clinical trial involving patients treated with PARP inhibitors. Individual HRD-associated mutational signatures alone did not consistently detect HRD or predict clinical response across datasets. Notably, while all tools performed comparably for whole-genome– sequenced cancers, HRProfiler was the only approach that consistently identified HRD in whole-exome–sequenced breast and ovarian cancers, offering clinically relevant insights. Retrospective
􀀶 | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | analyses provided strong evidence that HRProfiler could serve as a valuable tool for predicting HRD and clinical response in breast and ovarian cancers. This study provides the rationale for large-scale prospective clinical trials to validate the potential of HRProfiler as a routine predictive and/or prognostic HRD biomarker to guide clinical decision-making.
Significance: HRProfiler is a machine learning approach that reliably identifies homologous recombination deficiency in whole-exome–sequenced breast and ovarian cancers, outperforming other tools and providing clinically useful insights.
See related commentary by Lim and Ju, p. 2348

Created in BioRender. Abbasi, A. (2025) [https://BioRender.com/pvmbc6i](https://BioRender.com/pvmbc6i) | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This article is part of a special series: Driving Cancer Discoveries with Computational Research, Data Science, and Machine Learning/AI. Introduction Repair of DNA double-strand breaks by homologous recombination is an essential cellular mechanism for maintaining genomic stability and preventing tumorigenesis (1). Prior studies have 1Depar

FAQs

What clinical vulnerability in HRD breast and ovarian cancers supports PARP and platinum treatment?

Defective homologous recombination repair disables recovery from DNA double-strand breaks, making cells vulnerable and sensitive to PARP inhibitors and platinum chemotherapy.

How is HRProfiler designed to detect HRD?

HRProfiler is a machine learning method trained to detect homologous recombination deficiency using six mutational features derived from whole-genome and whole-exome sequencing data.

What was the main performance finding comparing HRProfiler with other tools and mutational signatures?

While tools and individual mutational signatures performed similarly for whole-genome–sequenced cancers, HRProfiler was the only approach that consistently identified HRD in whole-exome–sequenced breast and ovarian cancers and provided clinically relevant insights.

Sep 23, 2026

Uploaded by Putri