%0 Generic
%A Rübsam, Marc
%C Heidelberg
%D 2025
%F heidok:37453
%R 10.11588/heidok.00037453
%T Assessing defects in homologous recombination repair with a composite biomarker
%U https://archiv.ub.uni-heidelberg.de/volltextserver/37453/
%X Homologous recombination repair (HRR) deficiency (HRD) creates a therapeutic vulnerability in cancer, exploitable by PARP inhibitors (PARPi). However, current HRD biomarkers are often limited in patient identification due to their BRCA1/2-centric view or high input requirements. A widely applicable composite biomarker is crucial for improving selection for HRD-targeted therapies across diverse cancers.    This study analyzed HRD in 739 patients from two precision oncology cohorts, including many rare entities. Germline and somatic variants in 182 HRR-associated genes were characterized, including their zygosity, and consequential genomic patterns such as mutational signature 3 exposure and HRD-LOH/LST events. This HRD landscape analysis revealed complex HRD feature relationships, including HRR gene alterations beyond BRCA1/2 strongly associated with genomic patterns, and a strong influence of zygosity and cancer entity. No individual criterion was capable of fully capturing this complexity.    Consequentially, the broadly applicable TOP-ART score biomarker was developed, informed by the HRD landscape analysis, combining germline and somatic variants, genomic instability, and mutational signature 3 into a composite scoring schema.    The TOP-ART score effectively captures the observed complexities and discordant signals and predicts the HRD status based on a rigorously defined positivity threshold. It classified 38% of the cohort as TOP-ART positive, including patients across all 20 entity baskets.    Functional validation on 46 patient-derived cancer models showed TOP-ART positivity significantly correlated with higher PARPi sensitivity. Compared to two established biomarkers HRDetect/CHORD, the TOP-ART score showed superior drug sensitivity prediction and identified more HRD-positive cases, especially in non-breast cancers.    Exploratory analyses suggest insightful trajectories into methylation analysis, signature set expansion, and characterization of signal dynamics. The prospective clinical evaluation in the TOP-ART trial will elude the potential of the TOP-ART score to improve patient selection for HRD-targeted therapies.