Factors associated with an inconclusive result from commercial homologous recombination deficiency testing in ovarian cancer

Abstract Introduction Homologous recombination deficiency (HRD) testing is used to determine the appropriateness of poly ADP‐ribose polymerase inhibitors for patients with epithelial ovarian cancer and no germline/somatic BRCA1/2 alterations. Myriad MyChoice CDx reports a genomic instability score (GIS) to quantify the level of HRD, with a positive score defined as ≥42. The authors sought to define factors associated with obtaining an inconclusive HRD test result. Methods GIS was retrieved for patients at their institution with epithelial ovarian cancer without germline/somatic BRCA1/2 deleterious alterations who underwent HRD testing from April 2020–August 2023. Clinical data were abstracted from the medical record. Results Of 477 HRD test results identified, 57 (12%) were inconclusive. High‐grade serous ovarian cancers had higher GIS than other histologic types (median 29 vs. 21, p < .001). Most HRD cases were of high‐grade serous histology; no cases with clear cell or endometrioid histology were HRD‐positive. On univariate analysis, interval versus primary cytoreductive surgery, other specimen sources versus surgical specimens, and chemotherapy exposure were risk factors for inconclusive HRD testing. On multivariable analysis, chemotherapy exposure, and tissue source were associated with an inconclusive test result, with surgical specimens more likely to yield a conclusive result than other sources (biopsy, cytology, other). Age, stage, self‐reported race, and histology were not associated with an inconclusive result. Conclusions Surgical tissue was more likely to yield a conclusive HRD test result versus other sources of epithelial ovarian cancer tissue acquisition. When feasible, laparoscopic biopsy before initiation of neoadjuvant chemotherapy may increase the likelihood of obtaining interpretable HRD test results.