Estimating copy number to determine BRCA2 deletion status and to expect prognosis in localized prostate cancer.

ABSTRACT

BACKGROUND: The significance of BRCA alterations has been implicated in the development of metastatic castration-resistant prostate cancer (PC). The details of the frequency and significance of BRCA alterations in localized PC remain unknown. In this study, we investigated the frequency and clinical significance of BRCA alterations in localized PCs using an in-house next-generation sequencer (NGS) system. METHODS: DNA was extracted from formalin-fixed paraffin-embedded tissues of surgical specimens from 126 patients with clinically localized PC who underwent radical prostatectomy. The mutation information of 164 cancer genes was analyzed using the PleSSision-Rapid test. Both copy number (CN) variation and loss of heterozygosity of various genes, such as BRCA1 and BRCA2, were estimated and reported. RESULTS: Next-generation sequencer analyses revealed that the BRCA2 CN was decreased in 17 patients (13.5%) and the BRCA1 CN in six (4.8%) patients. NGS-based CN values were shown to be highly correlated with droplet digital PCR-based CN values. Tissue-specific BRCA expression investigated using the Human Protein Atlas showed that the decreased CN of BRCA2, but not BRCA1, is responsible for the decreased BRCA activity in PC. Ten of the 22 patients with decreased BRCA2 CN were presumed to have somatic heterozygous deletion. There were no observed associations between the heterozygous deletion of BRCA2 and various clinicopathological parameters. Furthermore, three of 10 patients developed biochemical recurrence within 3 months after surgery. Multivariate analyses revealed that the initial prostate-specific antigen levels and BRCA2 CN were independent factors for biochemical recurrence. CONCLUSION: Our results suggest that a decrease in BRCA2 CN may be used as a biomarker for predicting recurrence after surgery in localized PC. Early screening for somatic alterations in BRCA2 using NGS may help to broadly predict the risk of PC progression.