CHEK2 c.1100delC mutation is associated with an increased risk for male breast cancer in Finnish patient population

Male breast cancer (MBC) is a rare disease comprising less than 1% of all cancer in men and less than 1% of all breast cancers, but the incidence is increasing [ 1]. Associated genetic risk factors for MBC are poorly understood. Mutations in BRCA1 and especially in BRCA2 are known risk factors, the prevalence varying in different MBC populations [ 1, 2]. According to a study in 2004, 7.8% of Finnish MBC patients carried a BRCA2 mutation [ 3]. In Finland, strong founder effects and enrichment of deleterious alleles are seen [ 4] and major founder mutations in BRCA1 and BRCA2 [ 5] as well as in other breast and ovarian cancer susceptibility genes account for the vast majority of the identified mutations.

Mutations in cell-cycle checkpoint kinase 2 ( CHEK2) are associated with an elevated risk for breast cancer [ 6, 7]. Approximately 4-fold elevated frequency of a protein truncating mutation, c.1100delC in exon 10, was found in Finnish breast cancer patients with a positive family history (5.5%) compared to population controls (1.4%) and a 6-fold in patients with bilateral breast cancer compared to patients with unilateral disease [ 8]. Among unselected female breast cancer patients, the mutation was observed at a 2.0% frequency. In a previous study of Finnish MBC patients, the frequency of the c.1100delC mutation was similar to that seen in population controls [ 9] while in a Dutch and an American study, the frequency was significantly elevated [ 10, 11]. Another variant, the missense I157T (c.470T>C) in the FHA domain of CHEK2, is associated with a milder elevation in the risk and was observed in 7.4% of unselected female breast cancer patients, in 5.5% of familial patients, and in 5.3% of population controls [ 12]. In a recent study [ 13], the incidence of germline mutations in genes mediating DNA-repair processes among men with metastatic prostate cancer was 11.8%, which was significantly higher than the incidence among men with localized cancer. The results indicated that a CHEK2 ( p < 0.001) or a BRCA2 ( p < 0.001) mutation may be associated with an elevated risk for a more aggressive type of prostate cancer.

Fanconi anemia is a genetically heterogeneous recessive disorder, which results from biallelic mutations in FA genes, a family of more than 20 genes involved in DNA repair [ 14]. An increased risk of breast or ovarian cancer is associated with monoallelic mutations in a subset of these genes ( BRCA1, BRCA2, BRIP1, PALB2, RAD51C) [ 15– 20]. In the Finnish population, founder mutations in BRCA2, PALB2, and RAD51C genes have been identified [ 5, 18, 21]. PALB2 c.1592delT mutation has been observed at 0.7–0.9% frequency among Finnish unselected breast cancer patients and at 2.0–2.7% frequency among familial patients and is associated with an aggressive breast tumour phenotype [ 21, 22]. RAD51C mutations c.93delG and c.837+1G>A were observed at a combined frequency of 1.0% among unselected ovarian cancer patients and at 2.1% frequency among breast and ovarian cancer families and associated with an increased risk of ovarian cancer [ 18]. Neither of the mutations was observed in unselected breast cancer patients. Mutations in RAD51D are associated with an increased risk of ovarian cancer [ 23] and a Finnish founder mutation c.576+1G>A in the gene was significantly more frequent among breast cancer patients with a family history of breast and ovarian cancer (2.9%) than among population controls (0.1%) [ 24]. Among Finnish unselected ovarian cancer patients, the frequency of the c.576+1G>A mutation was 0.6% whereas the frequency among unselected breast cancer patients (0.1%) was same as in controls [ 24]. Finally, a FANCM nonsense mutation c.5101C>T (p.Q1701X) has been identified as a susceptibility allele for triple-negative breast cancer in the Finnish population and was observed in 2.8% of unselected breast cancer patients and in 3.1% of breast cancer families [ 25].

Here we have genotyped the CHEK2 mutations I157T and c.1100delC, the FANCM mutation p.Q1701X, the PALB2 mutation c.1592delT, the RAD51C mutations c.837+1G>A and c.93delG, and the RAD51D mutation c.576+1G>A in 68 male breast cancer patients. These mutations explain the vast majority of all mutations observed in these genes in the Finnish population, with only a few other, very rare or unique mutations identified.

The median age at diagnosis of breast cancer was 64 years (range 24–90).

The occurrence of risk factors in the studied patients is presented in Additional file  2. Forty-one patients had been tested for BRCA mutations, and of these, 2 (4.9%) were carriers of BRCA1 mutations and 3 (7.3%) carried a BRCA2 mutation.

The tumour characteristics are described in Table  1. Nine percent of patients had a T4 tumour. One patient had a lobular carcinoma, one an adenocystic carcinoma, and one a ductal carcinoma in situ whereas all other cancers were ductal carcinomas. Most tumours were hormone receptor positive: 91% were estrogen receptor (ER) positive and 81% progesterone receptor (PR) positive.

The frequencies of mutations detected in this study and in the control populations are presented in Table  2. The CHEK2 c.1100delC mutation was found in 4 patients (5.9%) (odds ratio (OR): 4.47, 95% confidence interval (CI) 1.51–13.18, p = 0.021 compared to population controls). Median age of the CHEK2 c.1100delC carriers was 56 years and half of the patients were relatively young at the time of diagnosis, less than 50 years old (Table  3). All the carriers were BRCA1/2 negative and one patient had a family history of breast cancer. All carcinomas were ductal and estrogen receptor positive. The I157T variant was also detected in 4 patients ( p = 0.781). No RAD51C, FANCM, PALB2, or RAD51D mutations were found.

The CHEK2 c.1100delC mutation associated with an increased risk for MBC in the Finnish population of the present study. None of the patients with CHEK2 c.1100delC mutation had BRCA1/2 mutation. Mutations in CHEK2 are known risk factors for female breast cancer [ 7]. The 5.9% mutation frequency detected among the male breast cancer patients is higher than among the unselected female patients and comparable to that among female familial cases (5.5%) [ 8]. Similar to female breast cancer [ 27], the male CHEK2 c.1100delC carrier tumours were of ductal histology, ER positive, and poorly differentiated (grade 2–3). The CHEK2 c.1100delC mutation has been also linked to an increased risk for prostate cancer [ 28], and possibly indicates an elevated risk for metastatic prostate cancer [ 13].

The role of the CHEK2 c.1100delC mutation as a risk factor for MBC has been studied in several papers [ 6, 9– 11, 29– 32]. An association between CHEK2 c.1100delC mutation and MBC risk has been reported in three studies [ 6, 10, 11]. A wide variation in the population frequency of c.1100delC has been observed, with highest reported frequencies in the Netherlands (1.3–1.6%) and in Finland (1.1–1.4%) [ 7]. The rarity of MBC and the geographic variation in the frequency of the CHEK2 c.1100delC mutation might explain the differences between studies on the possible association between the CHEK2 c.1100delC mutation and MBC. In a previous Finnish study, the CHEK2 c.1100delC mutation frequency was not significantly elevated in MBC patients, and did not seem to affect the age on breast cancer onset [ 9]. Only 10.5% of patients in that study had a family history of female breast or ovarian cancer. In our study, however, the CHEK2 c.1100delC was significantly more frequent in the studied population compared to population controls, despite the smaller sample size. The median age of the patients with CHEK2 c.1100delC mutation was also younger than in the whole population studied. Of the 43 patients with known family history, 25% had a positive family history of female breast cancer. The population control frequency to which the previous study compared their results was the same as the one used in this study. Our finding is in line with the Dutch study where the frequency of the c.1100delC mutation was significantly elevated compared to that seen in population controls (OR: 4.1, 95% CI 1.2–14.3, p = 0.05) [ 10], both studies suggesting about four- to five-fold elevated risk for male breast cancer for the mutation carriers. Very recently, a large American multi-gene panel study with 715 MBC patients showed that CHEK2 c.1100delC was associated with moderately increased risks of MBC (OR: 3.8, 95% CI 1.7–7.8) [ 11].

Mutations in RAD51C and RAD51D are rare and have been primarily linked to an increased risk for ovarian cancer [ 19, 23] rather than breast cancer alone. Previously, no truncating RAD51C mutations were identified among 97 Italian MBC patients and the American multi-gene panel study also did not identify any RAD51C mutations whereas one RAD51D mutation was observed among the 715 MBC patients [ 11, 33]. PALB2 mutations among MBC patients are relatively rare (0.8%) in the US population but have been associated with a significantly increased risk of male breast cancer [ 11]. In this study, we did not detect the RAD51C, RAD51D, FANCM, or PALB2 mutations among male breast cancer patients. However, our sample size was small. Given the relatively low frequency of RAD51C, RAD51D, FANCM, and PALB2 mutations among unselected female breast cancer patients, the absence of the studied mutations in the MBC series was not unexpected. Larger studies are warranted to better evaluate the contribution of these genes to MBC susceptibility in the Finnish population.

In conclusion, the CHEK2 c.1100delC mutation is associated with an increased risk for MBC in the present study in Finnish population. In light of previous data from the Netherlands [ 10] and USA [ 11], our study thus suggests that inclusion of CHEK2 mutation analyses should be considered as a part of genetic testing of MBC patients, at least in populations with prevalent mutations.