Current HRD assays in ovarian cancer: differences, pitfalls, limitations, and novel approaches.

Ovarian carcinoma (OC) still represents an insidious and fatal malignancy, and few significant results have been obtained in the last two decades to improve patient survival. Novel targeted therapies such as poly (ADP-ribose) polymerase inhibitors (PARPi) have been successfully introduced in the clinical management of OC, but not all patients will benefit, and drug resistance almost inevitably occurs. The identification of patients who are likely to respond to PARPi-based therapies relies on homologous recombination deficiency (HRD) tests, as this condition is associated with response to these treatments. This review summarizes the genomic and functional HRD assays currently used in clinical practice and those under evaluation, the clinical implications of HRD testing in OC, and their current pitfalls and limitations. Special emphasis will be placed on the functional HRD assays under development and the use of machine learning and artificial intelligence technologies as novel strategies to overcome the current limitations of HRD tests for a better-personalized treatment to improve patient outcomes.

BRCA1 foci test as a predictive biomarker of olaparib response in ovarian cancer patient-derived xenograft models.

Standard therapy for high-grade ovarian carcinoma includes surgery followed by platinum-based chemotherapy and poly-ADP ribose polymerase inhibitors (PARPis). Deficiency in homologous recombination repair (HRD) characterizes almost half of high-grade ovarian carcinomas and is due to genetic and epigenetic alterations in genes involved in HR repair, mainly BRCA1/BRCA2, and predicts response to PARPi. The academic and commercial tests set up to define the HRD status of the tumor rely on DNA sequencing analysis, while functional tests such as the RAD51 foci assay are currently under study, but have not been validated yet and are available for patients. In a well-characterized ovarian carcinoma patient-derived xenograft platform whose response to cisplatin and olaparib, a PARPi, is known, we assessed the association between the BRCA1 foci score, determined in formalin-fixed paraffin-embedded tumor slices with an immunofluorescence technique, and other HRD biomarkers and explored the potential of the BRCA1 foci test to predict tumors' response to cisplatin and olaparib. The BRCA1 foci score was associated with both tumors' HRD status and RAD51 foci score. A low BRCA1 foci score predicted response to olaparib and cisplatin, while a high score was associated with resistance to therapy. As we recently published that a low RAD51 foci score predicted olaparib sensitivity in our xenobank, we combined the two scores and showed that the predictive value was better than with the single tests. This study reports for the first time the capacity of the BRCA1 foci test to identify HRD ovarian carcinomas and possibly predict response to olaparib.

Onvansertib treatment overcomes olaparib resistance in high-grade ovarian carcinomas.

Occurrence of resistance to olaparib, a poly(ADP-ribose) polymerase (PARP) inhibitor (PARPi) approved in ovarian carcinoma, has already been shown in clinical settings. Identifying combination treatments to sensitize tumor cells and/or overcome resistance to olaparib is critical. Polo-like kinase 1 (PLK1), a master regulator of mitosis, is also involved in the DNA damage response promoting homologous recombination (HR)-mediated DNA repair and in the recovery from the G2/M checkpoint. We hypothesized that PLK1 inhibition could sensitize tumor cells to PARP inhibition. Onvansertib, a highly selective PLK1 inhibitor, and olaparib were tested in vitro and in vivo in BRCA1 mutated and wild-type (wt) ovarian cancer models, including patient-derived xenografts (PDXs) resistant to olaparib. The combination of onvansertib and olaparib was additive or synergic in different ovarian cancer cell lines, causing a G2/M block of the cell cycle, DNA damage, and apoptosis, much more pronounced in cells treated with the two drugs as compared to controls and single agents treated cells. The combined treatment was well tolerated in vivo and resulted in tumor growth inhibition and a statistically increased survival in olaparib-resistant-BRCA1 mutated models. The combination was also active, although to a lesser extent, in BRCA1 wt PDXs. Pharmacodynamic analyses showed an increase in mitotic, apoptotic, and DNA damage markers in tumor samples derived from mice treated with the combination versus vehicle. We could demonstrate that in vitro onvansertib inhibited both HR and non-homologous end-joining repair pathways and in vivo induced a decrease in the number of RAD51 foci-positive tumor cells, supporting its ability to induce HR deficiency and favoring the activity of olaparib. Considering that the combination was well tolerated, these data support and foster the clinical evaluation of onvansertib with PARPis in ovarian cancer, particularly in the PARPis-resistant setting.