Survey data from 140 laboratories showed variability in homologous recombination deficiency (HRD) testing, with approaches ranging from targeted next-generation sequencing panels to whole-exome and whole-genome sequencing.
A standardized definition and consistent laboratory methods for detecting HRD have not been established, creating challenges in clinical implementation. To address these issues, the Association for Molecular Pathology, in collaboration with the Association of Cancer Care Centers and the College of American Pathologists, convened a multidisciplinary working group to develop consensus recommendations for clinical molecular laboratories.
The recommendations, published in The Journal of Molecular Diagnostics, were informed by a literature review, survey of laboratory practices, and expert consensus. A total of 12 statements were issued, addressing three categories: sample considerations, assay design, and reporting practices. For samples, the group advised documenting neoplastic cellularity, specimen type, and comparator requirements during validation.
In terms of assay design, recommendations included analysis of BRCA1 and BRCA2 with deletion and duplication assessment, and consideration of additional homologous recombination repair genes such as PALB2, RAD51C, and RAD51D. The group also outlined potential roles for promoter methylation analysis and emphasized defining genomic scar components, including loss of heterozygosity, large-scale transitions, and telomeric allelic imbalance. Laboratories using mutational signatures were advised to specify which signatures are evaluated and to validate assay performance accordingly.
For reporting, the consensus recommended that laboratories include key features of the assay, thresholds, and limitations in clinical reports, and note biallelic inactivation when available.
Reporting practices and adoption of promoter methylation analysis also varied. Barriers identified included lack of standardization, uncertainties around reimbursement, and limited expertise.