99P - Prevalence of homologous recombination repair deficiency-associated variants in non-selected Portuguese patients with metastatic prostate cancer

Background

International guidelines recommend the use of PARP inhibitors (PARPi) in patients with germline and somatic variants of genes related to homologous recombination. At our center we have implemented universal testing of these genes for all patients with metastatic prostate cancer. International data estimate the prevalence of somatic or germline mutations of BRCA1/2 between 9-11%, and mutations or deletions in ATM and PALB2 of 6% and 1%, respectively. The clinical impact of universal testing in the Portuguese population is unknown. We aimed to determine the prevalence of somatic and germline gene variants with clinical impact in Portuguese patients who underwent universal testing.

Methods

We included all patients with metastatic prostate cancer followed at our center who were candidates for systemic treatment and who signed the informed consent form for universal testing between 01 Jan 2022 and 30 May 2024. We performed testing by NGS in tumor tissue and/or peripheral blood. The gene panel includes ATM, BRCA1, BRCA2, CHEK2, PALB2, HOXB13, MLH1, MSH2, MSH6, PMS2. We excluded patients in which testing was guided by clinical criteria. Clinico-demographic characteristics were gathered from medical records. Mutation prevalence is reported as 95% Bayesian credible intervals (95% CdI), assuming a binomial model with uniform prior.

Results

We identified 71 patients, with median age 72 years old (min 51, max 96). Somatic testing was performed in 36 patients and germline testing was performed in 55 patients. No variants predictive of response were found. The prevalence was 0% (95% CdI 0–8%) for somatic mutations and 0% (95% CdI 0–5%) for germline mutations.

Conclusions

In our population, universal testing didn’t lead to changes in clinical management. The probability of actionable variants was estimated to be < 8% for somatic mutations and < 5% for germline mutations, lower than in international case series. This suggests a low yield for universal testing. In order to determine the real clinical benefit of universal testing, larger studies should be performed on the Portuguese population. In case the low yield is confirmed, further research can help determine which sub-populations should be tested.

Editorial acknowledgement

Clinical trial identification

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

A.B. Mansinho, A.R. Dias Teixeira Sousa: Financial Interests, Personal, Invited Speaker: AstraZeneca. All other authors have declared no conflicts of interest.