In this study, we examined survival for men with prostate cancer and a BRCA2
mutation, and compared this to a similar group of men with a BRCA1
mutation. Prostate cancer is a clinical manifestation of the BRCA2
gene, but only a small proportion of prostate cancers are attributable to BRCA2
mutations. An excess risk of prostate cancer among men with BRCA2
mutations has been previously documented. Epidemiology studies have been of two types: (1) estimating the risk of prostate cancer in men from families with BRCA2
mutations, and (2) estimating the proportion of BRCA2
mutations among unselected men with prostate cancer. In a study of 173 families with BRCA2
mutations, the Breast Cancer Linkage Consortium reported in 1999 that there was a significantly increased risk of prostate cancer among male first-degree relatives of female carriers (Breast Cancer Linkage Consortium, 1999
). They estimated the odds ratio to be 4.7 (95% CI 3.5–56.2). In a study in the Netherlands of 139 families with BRCA2
mutations, Van Asperen et al (2005)
confirmed an excess risk of prostate cancer in first-degree relatives of carriers (OR=2.5; 95% CI 1.6–3.8). Given that only one-half of the first-degree relatives are expected to be gene carriers, this corresponds to a relative risk of prostate cancer, given a BRCA2
mutation, of approximately five.
Kirchoff et al (2004)
tested 251 unselected Ashkenazi Jewish men with prostate cancer from the New York area for the two founder mutations in BRCA1
and the one in BRCA2.
They also tested 1472 Ashkenazi controls. A mutation was found in 5.2% of cases and 1.9% of controls. The presence of a BRCA2
mutation was associated with a 4.8-fold increased risk of prostate cancer. In contrast, the BRCA1
carriers in that study were not at increased risk of prostate cancer. In a study from the United States, Agalliu et al (2007)
found a BRCA2
mutation in two of 290 men diagnosed with prostate cancer under the age of 55 (mixed ethnicities). The relative risk for early-onset prostate cancer was 7.8.
We observed that men with prostate cancer and a BRCA2 mutation experienced relatively poor survival, in comparison to men with prostate cancer and a BRCA1 mutation. The median survival for BRCA2 carriers was 4.0 years, and at 10 years post-diagnosis, 53% of the patients had died.
There are several limitations to our study. Ideally, one would follow a cohort of unselected patients with prostate cancer and a BRCA2 mutation, record details on the stage, grade and all treatments received, and then compare the outcome of the hereditary group with a similar group of patients without a mutation. We do not know the cause of death for these patients and expect that some will have died of causes other than prostate cancer. Not all of the patients in our study were proven carriers of a BRCA2 mutation; we included 67 carriers (or obligate carriers) and 116 men with prostate cancer who had not been tested for the presence of the mutation. The families in this study were those referred to various cancer genetics centres because of multiple cases of breast and/or ovarian cancer in the family, and were not selected on the basis of prostate cancer. The demographic and clinical information on the patients with prostate cancer were taken from a pedigree review and interview of the proband; we did not review the medical record or the pathology report of the prostate cancer patient. The ages of diagnosis and death were based on information provided by the female proband. It may be probable that not all cases of prostate cancer in male relatives were recorded. We do not know if the case was diagnosed clinically or through prostate cancer screening. We do not have information on the stage or grade at presentation, and we did not record details of treatment.
Despite these numerous limitations, we believe that the differences observed in survival of the BRCA1
carriers ( and ) can be attributed to the adverse effect of the BRCA2
mutation on prostate cancer survival. These data were recorded and collected using the same methods for the BRCA1
families. There is no reason to believe that carriers of BRCA2
mutations should have greater mortality than men with BRCA1
mutations from non-cancer causes, and men with other forms of cancer were excluded. The survival experience of the BRCA2
carriers is almost universally poor, as only 26% of the known carriers were alive 10 years after diagnosis and 4% were alive at 15 years. Compared to men with a BRCA1
mutation, men with a BRCA2
mutation in this study were more than two times as likely to die (of any cause) following the diagnosis of prostate cancer (hazard ratio 2.5), and it is likely that the relative mortality for prostate cancer-specific mortality would be even more extreme. These data support the conclusions of Tryggvadottir et al (2007)
who reported that men with prostate cancer and a BRCA2
mutation experienced an unexpectedly high rate of mortality. The basis for the aggressive behaviour of the BRCA2-associated prostate cancers is not known. Moro et al (2008)
show that downregulation of BRCA2 expression through the introduction of siRNA in prostate cancer cells promoted cancer cell migration and invasion. Mitra et al (2008)
found the average grade of prostate cancers among men with BRCA2-associated prostate cancers (Gleason score) to be higher than that of noncarrier control tumours.
We compared the survival of men with BRCA2
mutations with that of men with BRCA1
mutations. Ideally, we would have included a comparison group of noncarriers as well. However, it was not possible to identify a similar group of noncarrier patients from this database, given that we routinely collect pedigree information only on families with a mutation in one of the two genes. Although we do not have a comparable group of noncarriers in this study, it has been reported in a trial of patients undergoing watchful waiting vs
prostatectomy that 27% of men with prostatectomy died within 10 years of diagnosis (all-cause mortality) and 32% of men with watchful waiting died within 10 years of diagnosis (Bill-Axelson et al, 2005
). In comparison, 75% of the BRCA2 carriers in our study died within 10 years of diagnosis.
If survival in BRCA1
carriers is atypical, then our observed difference will not be representative of prostate cancer in general. To our knowledge, BRCA1
is not associated with an improved survival experience. Furthermore, the risk of prostate cancer in BRCA1
carriers is increased only to a small extent, if any (Kirchoff et al, 2004
; Cybulski et al, 2008
), and an increased risk is not associated with all mutations (Cybulski et al, 2008
). Therefore, we expect the survival experience of men with prostate cancer in BRCA1
families to be similar to that of the general population.
The data from the two studies suggest that men with BRCA2 mutations may not benefit from current therapies to the same extent as other men. We do not know which men in this cohort had screen-detected prostate cancer, and it will be important to study whether or not men who have subclinical, screen-detected prostate cancer also experience a high mortality rate. To address this issue, we plan to identify the BRCA2 carriers among a large sample of prostate cancer patients and to characterise them in terms of clinical presentation, pathology and response to treatment. Future studies should also address whether current surgical and non-surgical treatments improve survival from prostate cancer among BRCA2 carriers. Current treatments include surgery, hormonal-based therapies and radiotherapy. Given that the BRCA2 protein is involved in the repair of damaged DNA breaks, and that radiation induces double-strand breaks, it is possible that prostate cancer patients with a BRCA2 mutation are more sensitive to radiotherapy than patients without a mutation. Cytotoxic chemotherapy is not generally used in the early treatment of prostate cancer. Currently, there is interest in identifying directed treatments for breast cancer in BRCA1 and BRCA2 carriers. Drugs that induce DNA-strand breaks, such as cis-platinum, show increased sensitivity in pre-clinical models, and trials are now underway to study cis-platinum and PARP1 inhibitors in women with breast cancer and a BRCA mutation. A male patient with metastatic prostate cancer and a BRCA2 mutation has responded well to a PARP1 inhibitor (A Tutt, J De Bono, personal communication). It will be a matter of considerable interest to see whether or not the men with prostate cancer and a BRCA2 mutation benefit from targeted chemotherapy.