ESMO 2017: Germline Mutations in DNA Repair Genes Lowers Survival in mCRPC

Prospective analysis reveals patients with mCRPC and germline mutations in DNA repair genes have a trend to poorer outcomes

Patients with metastatic castration resistant prostate cancer (mCRPC) and germline mutations in the BRCA1, BRCA2, ATM, or PALB2 genes tended to be younger and have non-statistically significant trends towards shorter cause-specific (CSS) and progression-free survival (PFS) compared to similar patients without mutations. Only patients with BRCA2 mutations showed significantly worse outcomes in planned subgroup analyses, according to findings presented at ESMO 2017, the Annual Congress of the European Society for Medical Oncology in Madrid, Spain.

Elena Castro, Prostate Cancer Unit, CNIO - Spanish National Cancer Centre in Madrid, Spain explained that, although certain germline mutations in DNA repair genes have been associated with poor outcomes in prostate cancer, there were no conclusive data on the response to currently approved survival-prolonging therapies (SPT) and survival in mCRPC. Dr Castro and a multidisciplinary research team led by Dr Olmos conducted the prospective multicentre observational PROREPAIR-Bstudy (NCT03075735) of patients with newly diagnosed mCRPC and unknown germline mutation status at study entry.

The study enroled 419 eligible patients from January 2013 to April 2016 in 38 Spanish institutions who were treated with either abiraterone, enzalutamide, docetaxel, cabazitaxel or Ra-223 according to physician-choice.

The investigators looked for BRCA1, BRCA2, ATM, and PALB2 germline mutations in samples obtained from the patients and explored their impact on CSS from the time of diagnosis of mCRPC and first line therapy with a SPT. Enrolment of a minimum 408 patients was required and 171 deaths must have occurred to demonstrate a CSS hazard ratio (HR) of germline mutation carriers compared to non-carriers equal to 3 (α=0.05 & β=0.20). Secondary endpoints included the association of those mutations to the response to SPT.

A total of 26 (6.2%) patients were identified as carriers of germline mutations, including 14 BRCA2, 8 ATM and 4 BRCA1. The first SPT in carriers and non-carriers comprised a taxane for 63% versus 46% of patients, and a novel androgen receptor (AR) targeting therapies (ART) for 37% and 53%, respectively.

Further comparison of carriers to non-carriers revealed a non-significant trend for carriers to be younger with a median age of 66.5 versus 71.6 years, respectively (p = 0.16).

Nonsignificant trends between carriers versus non-carriers were also observed for other baseline characteristics at the start of SPT; carriers were more likely to have ECOG performance status of 0 versus 1, which was more often seen in non-carriers (92% versus 88%). In carriers versus non-carriers, median prostate-specific antigen (PSA) was 27.9 versus 31.0 ng/mL, and metastasis was observed to bone in 96% versus 86%, lymph nodes in 48% versus 52%, and viscera in 12% versus 16% of the respective groups.

Patients in the BRCA2 subgroup had the poorest outcomes overall

Carriers also demonstrated a shorter median time from initiation of ADT to development of mCRPC compared to non-carriers of 23.7 versus 26.7 months, respectively (p = 0.22); this was especially pronounced in patients in the BRCA2 subgroup where time to mCRPC was 18 months (p = 0.24).

Prostate-cancer specific deaths occurred in 207 patients after a median follow-up of 36 months. The median CSS from mCRPC was 28.5 months in carriers versus 36.0 months in non-carriers (p = 0.5); again the shortest CSS of 17.4 months was observed in patients in the BRCA2 subgroup (p = 0.02).

In patients receiving taxane treatment, median CSS in carriers versus non-carriers was 17.3 versus 24.5 months, respectively (p = 0.6) and 12.8 months in the BRCA2 subgroup (p < 0.01, as compared to non-carriers) and median PFS was 7.8 versus 7.1 months (p = 0.4), respectively, and 5.7 months in the BRCA2 subgroup (p = 0.3).

In the ART treated patients CSS from the beginning of the first ART was 25.4 months in carriers versus 26.6 months in non-carriers (p = 0.9), and 27.6 months in BRCA2 carriers (p = 0.5) and PFS was 8.2 versus 9.4 months, respectively (p = 0.8) and 5.8 months in the BRCA2 subgroup (p = 0.4).


The comparison of carriers to non-carriers of specific germline mutations showed non-significant trends to worse CSS from mCRPC from the start of the first taxane and the first ART.

Pre-planned subgroup analyses suggest that BRCA2 mutations associated with significantly poorer outcomes.

Commenting on the study findings, Dr Joaquin Mateo of The Institute of Cancer Research & The Royal Marsden, London, UK pointed out that it is about prospective study designed to assess the predictive value of a biomarker and we need more studies like it or even better randomised for testing, not for treatment. However, there are contradictory data among different studies in the last year. He questioned if treatment with PARP inhibitor, platinum can change outcome in this population. Based on the currently available data, he does not believe that patients with DDR germline mutations should be discouraged to receive standard therapy with taxanes, abiraterone, enzalutamide.


This study was funded by the Prostate Cancer Foundation, Fundación CRIS contra el cáncer, Fundación Obra Social La Caixa, Instituto de Salud Carlos III.


LBA32 – Castro Marcos E, et al. PROREPAIR-B: A prospective cohort study of DNA repair defects in metastatic castration resistant prostate cancer (mCRPC).