Challenges with increased utilization of clinical tumor testing in ovarian cancer

Objectives: To describe challenges with applying tumor genomic testing among women with ovarian cancer to clinical practice in a large division with >90% germline genetic testing compliance. Methods: Institutional review board (IRB)-approved, retrospective review of tumor genomic results obtained within a single Gynecologic Oncology division. Clinicopathologic data were abstracted from medical records and tumor genomic results were evaluated for congruency and discrepancies between different labs. Results: A total of 261 patients with epithelial ovarian/fallopian tube/primary peritoneal cancer had tumor testing between 2017 and 2020. A total of 4 commercial labs with CLIA-approved testing were primarily used over the study period (Figure 1). A total of 15 patients had tumor tests at more than one clinical laboratory, with 4/15 (27%) found to have Loss of Heterozygosity (LOH) on the second test (LOH testing was not performed on the first test). An additional 4 patients (27%) had discrepant somatic BRCA (sBRCA) results. A single patient had a somatic large rearrangement in BRCA2 not detected through her first sBRCA test. For 3 patients, a sBRCA alteration was reported from a testing laboratory that subsequently changed testing/reporting practices that negated the original reports. Secondary sBRCA testing at a different laboratory revealed no sBRCA alterations in all three patients.26 (10%) had germline BRCA (gBRCA) PVs [15 BRCA1, 11 BRCA2]. All 26 gBRCA carriers had somatic testing. 5 of those somatic tests included LOH tests and 5/5 demonstrated LOH-high scores. 10/26 (38%) tumor tests were ordered at the time of initial surgery and before the first line of chemotherapy was completed. There were 48 tumors with sBRCA PVs and no gBRCA PV. LOH testing with a score was provided for 6 patients, 5/6 (83%) had LOH-high scores. Excluding the sBRCA results from the lab that later negated reports, there were 18 tumors with 20 sBRCA variants confirmed. One patient's sBRCA variant was also found through germline testing and was classified as a variant of uncertain significance by the germline testing lab. Download : Download high-res image (52KB) Download : Download full-size image Conclusions: Classification of somatic variants differs from germline variants. Whether all somatic BRCA variants indicate homologous recombination deficiency and sensitivity to PARP inhibitors is not known. Discrepancies between labs and the availability of novel test components (i.e. LOH testing) highlight that repeating tumor testing can be valuable and may allow additional treatment options. Tumor testing is rapidly changing; the test ordered and timing of testing need to be carefully considered for each patient. In retrospect, the addition of tumor testing in the upfront setting did not alter management in patients who were found to carry gBRCA PVs. For gBRCA PV carriers, it may be higher yield to reserve tumor testing for the recurrent disease setting.