Homologous Recombination Deficiency: BRCA, ATM and HRD Testing

HRD Describes a Repair State, Not One Gene

Homologous recombination is most available in S and G2, when a sister chromatid can provide a template. BRCA1 helps coordinate end resection and the recruitment of PALB2–BRCA2; BRCA2 helps load RAD51 onto single-stranded DNA; RAD51 then performs the homology search and strand-invasion reaction. Replication-fork protection adds another layer that is not captured by a simple linear diagram.

Loss of both functional copies of a central HR gene can impair this system, but the phrase 'HRD positive' can also come from an assay that recognises patterns left behind by earlier repair dysfunction. A scar can persist even if a resistant tumour later restores HR through a reversion or another compensating event. Gene status, genomic history and present function are therefore related but distinct measurements.

BRCA1, BRCA2 and ATM Occupy Different Positions

BRCA1 and BRCA2 are core HR mediators with different jobs. BRCA1 participates early in pathway choice and end processing and helps connect damage sites to PALB2. BRCA2 acts later to regulate RAD51 filament formation. A pathogenic alteration may still need evidence of second-allele loss, promoter silencing or another event before a tumour can be assumed to lack the relevant function.

ATM is primarily a damage-response kinase that is recruited at double-strand breaks and coordinates checkpoint and repair signalling. ATM loss can affect the wider DNA-damage response, but it is not mechanistically identical to BRCA1 or BRCA2 loss. Evidence from a BRCA-defined population should not be transferred automatically to every ATM finding.

Germline and Tumour Tests Answer Different Questions

A germline test looks for an inherited change present throughout the body, usually using blood or saliva. A confirmed pathogenic or likely pathogenic BRCA1/2 result can inform an inherited-risk assessment and a discussion of testing relatives. It does not mean that cancer is inevitable or currently present. A VUS should not be treated as a pathogenic result.

Tumour sequencing asks which alterations are present in a cancer specimen. A harmful BRCA change in a tumour may be somatic or germline, so NCI advises discussing whether germline testing is appropriate. Tumour purity, variant allele fraction, copy-number state and assay design affect whether biallelic loss can be inferred; the gene name alone is incomplete.

What a Genomic-Scar HRD Test Measures

Genomic-scar assays combine one or more patterns such as loss of heterozygosity, telomeric allelic imbalance, large-scale state transitions, mutational signatures or rearrangement features. These are consequences accumulated across tumour evolution. Different platforms use different features, algorithms, specimen requirements and cut-offs, so scores from separate assays should not be treated as interchangeable numbers.

A scar result can broaden the search beyond BRCA1/2, but it may not identify the causal event and it does not show directly whether a living tumour cell can form RAD51 foci today. Conversely, a single HR-related gene alteration does not guarantee a high scar score. The appropriate conclusion is limited to what the validated assay was designed and studied to report.

Treatment Evidence Must Stay Within the Tested Setting

PARP inhibition can exploit selected homologous-recombination defects through catalytic inhibition, PARP trapping and replication-associated damage. However, the strength of evidence differs by cancer type, stage, line of therapy, maintenance or treatment setting, germline versus somatic status, assay and drug. A broad pathway rationale is not a universal prescribing rule.

Resistance can emerge through BRCA reversion, restoration of end resection or replication-fork protection, drug-efflux changes and other mechanisms. That is another reason a historical HRD scar cannot guarantee current sensitivity. Clinical interpretation should use the exact report alongside a current label, guideline and the patient's oncology context.

A Practical Reading Order for an HRD Report

First identify the specimen and test scope: germline, tumour sequencing, genomic-scar assay or a combination. Next preserve the exact variant classification and zygosity or copy-number wording. Then record the assay name, score, cut-off, tumour type and specimen-quality notes rather than copying only 'positive' or 'negative'.

Finally, separate the questions. Inherited risk belongs with genetics-trained care; tumour treatment evidence belongs with the treating oncology team; uncertain variants belong with the testing laboratory's reclassification process. This sequence prevents one result from being stretched beyond the question it was designed to answer.

Key Takeaways

• ·HRD is a functional concept; a BRCA variant, ATM alteration and genomic-scar score are not synonyms.
• ·Germline and tumour testing answer different questions, and a tumour BRCA result may warrant—but does not replace—germline evaluation.
• ·Genomic scars describe tumour history and may not prove the causal gene or current HR function.
• ·ATM findings should not be assumed to behave like BRCA1/2 loss.
• ·Treatment evidence and assay cut-offs are disease-, platform- and regimen-specific.