Germline BRCA2 Status, Other Aberrations Independently Linked With Cause-Specific Survival in Prostate Cancer

Patients with prostate cancer who harbor germline BRCA2 mutations (gBRCA2) were found to have significantly higher rates of somatic BRCA loss, somatic RB1 loss, and MYC amplification compared with non-carriers—factors that were independently associated with poorer survival outcomes, according to results of the case-control PROREPAIR-A study that were presented during the virtual 2020 ESMO Congress.

At a median follow-up of 12 years, the median cause-specific survival (CSS) was 9.1 years in gBRCA2 carriers compared with 17.6 years in non-carriers and 13.5 years in gBRCA1 carriers (log-rank test P = .004). Additional factors were independently associated with poorer CSS, such as gBRCA2 carriers (HR, 3.70; 95% CI, 1.41-9.68; P = .008), somatic BRCA2-RB1 codeletion (HR, 4.13; 95% CI, 1.58-10.77; P = .004), MYC amplification (HR, 2.27; 95% CI, 1.07-4.82; P = .033), a Gleason score of 8 or higher (HR, 3.91; 95% CI, 1.67-9.15; P = .002), and M1 stage at diagnosis (HR, 15.57; 95% CI, 5.50-43.33; P <.001).

“To our knowledge, PROREPAIR-A is the largest series of gBRCA2 carriers assembled to date evaluating the association between somatic aberrations and clinical outcomes in prostate cancer,” said lead study author Rebeca Lozano Mejorada, MD, of the Spanish National Cancer Research Centre, in a virtual presentation during the meeting. “Our study confirmed the independent prognostic value of gBRCA2 mutations for cause-specific survival in prostate cancer.”

In prostate cancer, gBRCA2 mutations have been associated with more aggressive disease and poor clinical outcomes. Moreover, data have shown that prostate tumors in germline BRCA2 carriers harbor more copy number alterations versus non-carriers, including a higher frequency of BRCA2 loss of heterozygosity, RB1 loss, and MYC amplification/gain. Investigators theorize that these additional aberrations may contribute to the poor prognosis of patients with germline BRCA2-mutated prostate cancer. These copy number alterations involve chromosomal segments that are larger than the genes themselves and could be detected via fluorescence in situ hybridization (FISH).

In the multicenter, international, matched case-control PROREPAIR-A study, investigators sought to validate the prognostic value of germline BRCA2 mutations and to also evaluate the role of somatic BRCA2 loss, RB1 loss, MYC amplification/gain, and other prostate cancer–related events.

The study comprised those with germline BRCA2 mutations (n = 73) who were matched 1:2 to a control group of non-carriers (n = 127), and an exploratory cohort of germline BRCA1 carriers (gBRCA1; n = 14).

The primary end point was to confirm the independent prognostic value of gBRCA2 mutations on CSS from a prostate cancer diagnosis. Secondary end points were to compare the molecular aberrations in gBRCA2 carriers versus non-carriers via FISH in BRCA2, RB1, MYC, PTEN, and TMPRSS2-ERG, as well as the clinical impact of these somatic aberrations in patients with and without previously known germline BRCA1/2 mutations.

Baseline characteristics were mostly similar between the gBRCA2 carriers and non-carriers, except for age (median 62.6 years in gBRCA2 vs 64.5 years in non-carriers; P = .028). Additionally, gBRCA2 carriers were more likely to have T3/T4 disease (31.5%) than non-carriers (9.4%; P <.001). In gBRCA1 carriers, the median age was 67.6 years (P = .164) and 42.9% of patients had T3/T4 disease.

In gBRCA2 versus non-carriers, the median prostate-specific antigen (PSA) at diagnosis was 9.0 ng/mL versus 12.9 (P = .077), the rate of N1 stage at diagnosis was 11% versus 3.9% (P = .073), and the rate of M1 stage at diagnosis was 28.8% versus 22% (P = .287). A total 57.5% of gBRCA2 carriers had a Gleason grade between 8 and 10 compared with 56.7% of those who were non-carriers (P = .908). Prostatectomies were done in 50.7% of gBRCA2 carriers and in 66.1% of non-carriers.

In the exploratory gBRCA1 cohort, the median PSA at diagnosis was 10.1 (P = .634), more than half of patients (57.1%) had T1/T2 disease (P = .003), 14.3% had N1 stage disease (P = .144), 28.6% had M1 stage at diagnosis (P = .522), and 50% of patients had a Gleason score between 8 and 10 (P = .632). A total 57.1% of patients underwent prostatectomy.

Results showed that gBRCA2 carriers presented with more copy number alterations compared with non-carriers, including somatic BRCA2 loss (42.5% vs 11.8%; P <.001), somatic RB1 loss (54.8% vs 21.3%; P <.001), and BRCA2-RB1 codeletion, which included BRCA2 loss only (1.4% vs 0%), RB1 loss only (11.0% vs 9.4%), B2-RB1 codeletion (41.1% vs 11.8%), and cannot be assessed (2.7% vs 0%; all P <.001).

Additionally, gBRCA2 carriers versus non-carriers had more MYC amplification (47.9% vs 9.4%; P <.001) but lower MYC gain (5.5% vs 9.4%); there was also a higher percentage of gBRCA2 carriers with PTEN loss (34.2% vs 26.8%; P = .213) and TMPRSS2-ERG rearrangements (41.1% vs 15.0%), but not in those that couldn’t be assessed (38.4% vs 73.2%; P = .328).

In the gBRCA1 cohort, 35.7% of patients had somatic BRCA2 loss (P = .030) as well as somatic RB1 loss (P = .309); 7.1% of patients each had only BRCA2 loss or RB1 loss, while 28.6% had B2-RB1 codeletions. Two patients (14.3%) had MYC amplification, and 1 patient (7.1%) had MYC gain (P = .827). Fifty percent (n = 7) of patients had PTEN loss (P = .116) and an TMPRSS2-ERG rearrangement, but 28.6% (n = 4) could not be assessed (P = .489).

When CSS was examined by gBRCA2 carriers and gBRCA2 carriers with BRCA2-RB1 codeletions, the median CSS was 11.3 years and 6.3 years, respectively (log-rank test P = .041). Similarly, the median CSS was 13.4 years in gBRCA2 carriers compared with 6.0 years in those who had gBRCA2 mutations and MYC amplification (log-rank test P <.001).

Additionally, results showed that somatic BRCA2 and somatic RB1 status led to a Pearson correlation of 0.96 (P = .001), with a concordance Kappa index of 0.74. Most patients with somatic BRCA2 loss (n = 49/51) also presented with somatic RB1 loss, and 49 out of 65 samples with somatic RB1 loss also showed somatic BRCA2 loss.

Similar outcomes were reported for non-BRCA2 carriers versus non-BRCA2 carriers with BRCA2-RB1 codeletion; the median CSS was 17.6 years and 9.8 years, respectively (P <.001). In non-BRCA2 carriers versus non-BRCA2 carriers with MYC amplification, the median CSS was 17.6 years and 4.8 years, respectively (P <.001).

Lozano Mejorada concluded that somatic BRCA2-RB1 codeletion and MYC amplification define an aggressive prostate cancer subtype with poor clinical outcomes in both germline carriers and non-carriers.

References

Lozano R, Castro E, Aragon IM, et al. Clinical impact of somatic alterations in prostate cancer patients with and without previously known germline BRCA1/2 mutations: results from PROREPAIR-A study. Presented at: 2020 ESMO Congress; September 19-21, 2020; virtual. Abstract 612MO.