Findings presented at the 2025 San Antonio Breast Cancer Symposium showed that in cohort 1, the overall response rate (ORR) was higher in patients with RAD51-low (n = 41) vs -high (n = 14) tumors, at 68.3% (95% CI, 53%-80.4%) vs 21.4% (95% CI, 4.7%-50.8%), resulting in a difference of 46.9% (odds ratio [OR], 7.90; 95% CI, 2.07-39.54; P = .002).
Best overall response in the RAD51-low group comprised a complete response (CR) rate of 7.3% (95% CI, 1.5%-19.9%), and partial response (PR) rate of 61.0% (95% CI, 45.7%-74.3%). Stable disease (SD) rate was achieved in 22.0% (95% CI, 10.6%-37.6%) of patients, and 9.8% (95% CI, 2.8%-23.1%) experienced progressive disease (PD). The clinical benefit rate (CBR) was 75.6% (95% CI, 60.7%-86.2%). Corresponding CR, PR, SD, PD, and CBR rates in the RAD51-high group were 0% (95% CI, 0%-23.2%), 21.4% (95% CI, 4.7%-50.8%), 50% (95% CI, 23%-77%), 28.6% (95% CI, 1.8%-42.9%) and 57.1% (95% CI, 28.9%-82.3%), respectively.
“This trial supports the value of a functional RAD51-based HRD assessment to refine patient selection for PARP inhibitors,” presenting author Isabel Pimentel, MD, a breast oncologist at the Vall d' Hebron University Hospital and clinical and translational investigator at the Vall d`Hebron Institute of Oncology in Barcelona, Spain, stated.”[With] PARP inhibitors hopefully advanced to the neoadjuvant setting, these findings could [help] increase the likelihood of response.”
What was the rationale for evaluating the predictive value of RAD51 for olaparib sensitivity in HER2-negative breast cancer?
RAD51 recombinase is a key mediator of DNA repair via homologous recombination repair (HRR).2 After DNA damage occurs, RAD51 accumulates at break sites and forms discrete nuclear foci. Accordingly, RAD51 foci formation may serve as a surrogate biomarker of HR proficiency. Although RAD51 gene expression is normally low and tightly controlled in healthy cells, BRCA-deficient tumors demonstrate markedly reduced or absent RAD51 foci formation and increased susceptibility to PARP inhibitors.
“PARP inhibitors improve survival outcomes in patients with early and advanced HER2-negative breast cancer harboring germline BRCA1/2 pathogenic mutations,” Pimentel explained in her presentation.1 “However, germline BRCA status fails to capture the dynamic changes in HRR functionality and HRD from epigenetic silencing or other HRR gene mutations.”
Accordingly, Pimental and colleagues developed and retrospectively validated a RAD51-based functional HRD assessment. Based on data from this retrospective analysis, they hypothesized that RAD51-low tumors would be predictive of PARP inhibitor sensitivity in patients withHER2-negative breast cancer with BRCA1/2-, PALB2, or RAD51 CRD PVs.
What is the design of RADIOLA?
RADIOLA was a multicenter, open-label, 2-cohort study that enrolled 65 patients with unresectable locally advanced/metastatic HER2-negative breast cancer who had received 2 or more prior lines of chemotherapy in the metastatic setting; had measurable disease; and had not previously experienced disease progression on platinum-based chemotherapy. Of note, if patients did have prior exposure to platinum-based chemotherapy, a biopsy was performed after treatment. Patients with HR-positive disease were required to have relapsed on at least 1 line of endocrine therapy plus a CDK4/6 inhibitor.
All patients underwent RAD51 testing at screening. HRD was defined as no more than 10% of RAD51-positive cells in the tumor; these tumors were categorized as RAD51 low.
The study comprised 2 cohorts. Cohort 1 consisted of 55 patients with somatic or germline mutations in BRCA1/2-, PALB2, or RAD51 CRD PVs, and were RAD51-evaluable. Cohort 2 included 10 patients with wild-type or unknown BRCA1/2-, PALB2, or RAD51 CRD status and RAD51-low tumors. Patients in cohort 1 received 300 mg of oral olaparib twice-daily until disease progression or unacceptable toxicity. Circulating tumor DNA (ctDNA) was evaluated at baseline, on treatment, and at the time of progression.
The study’s primary end point was ORR by RECIST 1.1 criteria in RAD51-low vs -high tumors in cohort 1. Secondary end points included ORR in cohort 2, progression-free survival, CBR, duration of response, and time in response in both cohorts; ctDNA drop after 4 weeks of treatment in both cohorts, clinical benefit with olaparib by HRR mutations in cohort 1; and safety.
What were the baseline characteristics of patients on this study?
Patients were enrolled onto the study between April 2022 to June 2024. At the data cutoff of February 3, 2025, 4 patients in cohort 1 remained on treatment. In cohort 2, all patients were RAD51-low, and 2 remained on treatment at the data cutoff.
In cohort 1, the median age was 50 years (range, 31-8) in the RAD51-low group and 52 years (range, 30-77) in the RAD51-high group. In cohort 2, the median age was 62 years (range, 45-78). Across all 3 groups, most patients were female (cohort 1, RAD51-low = 95.1%; cohort 1, RAD51-high = 100%; cohort 2 = 100%), had visceral metastasis (85.4%; 78.6%; 60%), and had received prior systemic treatment for metastatic disease (90.2%; 78.6%; 100%). The median time between sample collection and day 1 of cycle 1 was 1.6 months (range, 0.3-31.3), 1 month (range, 0.2-27.4) and 14.8 months (range, 2.7-35.5) for these respective groups.
The majority of patients in cohort 1 had hormone receptor–positive disease (RAD51-low, 63.4%; RAD51-high, 57.1%) rather than triple-negative breast cancer (36.6%; 42.9%); half of patients in cohort 2 had hormone receptor–positive disease.
In the RAD51-low group of cohort 1, the most common germline mutation was in BRCA2 (46.3%), followed by BRCA1 (34.5%), PALB2 (14.6%). No RAD51D germline mutations were observed, and somatic BRCA2 and PALB2 mutations occurred in 2.4% of patients each. In the RAD51-high group, the most common germline mutation was in BRCA1 (57.1%) followed by BRCA2 (21.4%), PALB2 (7.1%), and RAD51D (7.1%). No somatic mutations were observed.
What additional efficacy outcomes were presented during the meeting?
Additional efficacy data from cohort 1 showed that, at a median follow-up of 6.8 months (95% CI, 6.2-9.2), the median PFS was longer in the RAD51-low vs -high group, at 7.1 months vs 5.6 months, respectively (HR, 0.51; 95% CI, 0.26-0.98). The 6- and 12-month PFS rates in the RAD51-low group were 52.5% and 19.6%, respectively. Corresponding rates in the RAD51-high group were 35.7% and 0%.
In cohort 2, the ORR was 10% (95% CI, 0.3%-44.5%), and was entirely composed of PRs. SD was achieved by 40% (95% CI, 12.2%-73.8%) of patients and 50% (95% CI, 18.7%-81.3%) experienced PD. The CBR was 40% (95% CI, 12.2%-73.8%). Pimental noted that no patients in this cohort underwent a biopsy prior study entry, which could limit the interpretation of these results.She concluded by sharing that translational work in both tissue and ctDNA is ongoing.
Disclosures: Pimental reported serving as a consultant for AstraZeneca, and receiving honoraria and/or travel support from AstraZeneca, Gilead, Merck, Novartis, and Pfizer.
References
- Pimentel I, Pascual T, Lema L, et al. Predicting olaparib sensitivity in patients with metastatic HER2-negative breast cancer with BRCA1/2, PALB2, RAD51C/D mutations according to their homologous recombination status by the RAD51 test: Primary analysis of the RADIOLA phase II trial. Presented at: 2025 San Antonio Breast Conference Symposium; December 9-12, 2024; San Antonio, TX. RF5-05.
- Orhan E, Velazquez C, Tabet I, Sardet C, Theillet C. Regulation of RAD51 at the Transcriptional and Functional Levels: What Prospects for Cancer Therapy? Cancers (Basel). 2021;13(12):2930. Published 2021 Jun 11. doi:10.3390/cancers13122930
