BRCA Mutations Associated with Worse Survival in HRR-deficient mCRPC

David Olmos, MD, PhD

Patients with homologous recombination repair (HRR)–deficient mutations and metastatic castration-resistant prostate cancer (mCRPC) who also harbored BRCA mutations experienced poorer survival outcomes vs patients without BRCA mutations and those with non-BRCA HRR mutations, according to an analysis from the CAPTURE trial presented at the 2023 ASCO Annual Meeting.

Findings showed that patients with BRCA mutations had significantly worse radiographic progression-free survival (rPFS), time to second objective disease progression (PFS2), and overall survival (OS) than the full non-BRCA mutation population and worse PFS2 and OS than those who had other HRR mutations but were BRCA wild-type.

“These results further support the importance of screening for germline and somatic BRCA1/2 alterations to deliver more precise care for our patients, especially in light of phase 3 trials that suggest poor outcomes for BRCA patients, and maybe other HRR [deficiencies], can be averted by adding a PARP inhibitor,” David Olmos, MD, PhD, of Hospital Universitario 12 de Octubre in Madrid, Spain, stated in his oral presentation.

Cohort 1 of the CAPTURE trial, part of the PROCURE biomarker study platform, combined the populations with mCRPC enrolled in 4 separate trials: PROREPAIR-B (NCT03075735), PROSENZA (NCT02922218), PROSTAC (NCT02362620), and PROSABI (NCT02787837). They were evaluated in a custom next-generation sequencing panel with paired somatic/germline DNA analysis for HRR mutations in BRCA1, BRCA2, BRIP1, CDK12, CHEK2, FANCA, HDAC2, PALB2, RAD51B, and RAD54L. Patients were considered to have HRR deficient status if they had mutations in at least 1 allele of at least 1 gene.

The patients either received novel hormonal therapy (60.4%) or a taxane (39.6%) as first-line mCRPC treatment, including docetaxel (Taxotere), cabazitaxel (Jevtana), abiraterone acetate (Zytiga), and enzalutamide (Xtandi). Over 80% went on to receive second-line treatment after progression.

A total of 729 patients on the trials volunteered to undergo genomic analysis, and were hierarchically classified as BRCA1/2, HRR non-BRCA, and non-HRR. They were also classified by germline, somatic, or none, and biallelic, monoallelic, or none. Of the patients, 223 (30.6%) had HRR mutations, 96 (13.2%) of whom had BRCA mutations, and 127 (17.4%) of whom had HRR non-BRCA mutations. Twenty-five (3.4%) had germline BRCA mutations and 71 (9.8%) had somatic BRCA mutations. Of the HRR non-BRCA mutations, the most common were ATM in 64 patients (8.8%) and FANCA in 38 patients (5.2%).

Investigators did not see an association between the choice of first-line treatment and the BRCA/HRR status, but Olmos said they noted a difference in second-line treatment, with more than 90% receiving additional therapy compared with less than 80% for other subgroups.

Outcome analyses comparing BRCA1/2 patients with other groups used a propensity score weighted Kaplan-Meier curves and adjusted hazard ratios (HR) and p-values by inverse probability weighted Cox models. The BRCA1/2 subgroup had a median rPFS of 7.1 months (95% CI, 6.4-8.6) vs 10.3 months (95% CI, 9.6-11.5) for all non–BRCA-mutated patients (HR, 1.70; 95% CI, 1.32-2.19; P < .0001), suggesting outcomes were worse for BRCA patients. Data for rPFS were missing for 2 patients in this analysis. The median PFS2 for BRCA1/2 group was 12.6 months (95% CI, 11.3-13.9) vs 15.9 months (95% CI, 15.1-16.8) for the non-BRCA group (HR, 1.81; 95% CI, 1.44-2.27; P < .0001). The median OS for the BRCA1/2 group was 19.4 months (95% CI, 16.8-21.1) vs 27.9 months (95% CI, 25.8-29.6) for the non-BRCA group (HR, 1.95; 95% CI, 1.55-2.45; P < .0001).

When compared with the HRR non-BRCA group, median rPFS for BRCA1/2 was 7.1 months (95% CI, 6.2-8.5) vs 9.0 months (95% CI, 7.1-10.8) for HRR non-BRCA patients (HR, 1.34; 95% CI, 0.98-1.81; P = 0.0641), not reaching statistical significance. The median PFS2 for the BRCA1/2 group was 12.3 months (95% CI, 11.1-13.5) vs 13.6 (95% CI, 11.9-15.9) for the HRR non-BRCA group (HR, 1.42; 95% CI, 1.07-1.88; P = .0143). The median OS for the BRCA1/2 group was 18.4 months (95% CI, 16.7-20.3) vs 21.9 months (95% CI, 18.3-24.4) for the HRR non-BRCA group (HR, 1.40; 95% CI, 1.06-1.84; P = .0166).

“We may conclude that BRCA patients have significant worse outcomes than patients with mutations in HRR non-BRCA, and those with alterations in HRR non-BRCA genes have worse outcomes than non-mutated patients,” said Olmos.

An exploratory analysis within the BRCA1/2 cohort was performed to evaluate differences in response to novel hormonal therapy versus taxane, germline versus somatic mutation, biallelic versus monoallelic, and BRCA1 versus BRCA2. These analyses were largely not statistically significant due to the small sample sizes, but no difference was seen based on choice of first-line treatment or number of mutated alleles. Patients with germline mutations appeared to have shorter median rPFS but longer median OS. There was also a consistent trend showing longer rPFS, PFS2, and OS for BRCA1 versus BRCA2.

“Our study demonstrated that both germline and somatic BRCA1/2 alterations are associated with shorter rPFS, PFS2, and OS independent of the initial treatment of choice. [Patients with] BRCA1/2 [mutations] had worse outcomes than patients [in the] HRR non-BRCA subgroup, although we need to better understand the particular contribution of certain HRR genes that might also help prognosis,” concluded Olmos.

Reference:

Olmos D, Lorente D, Alameda D, et al. Presence of somatic/germline homologous recombination repair (HRR) mutations and outcomes in metastatic castration-resistant prostate cancer (mCRPC) patients (pts) receiving first-line (1L) treatment stratified by BRCA status. J Clin Oncol. 2023;41(suppl):5003. doi: 10.1200/JCO.2023.41.16_suppl.5003.