Role of CHEK2*1100delC in unselected series of non-BRCA1/2 male breast cancers
Article first published online: 21 OCT 2003
Copyright © 2003 Wiley-Liss, Inc.
International Journal of Cancer
Volume 108, Issue 3, pages 477–478, 20 January 2004
How to Cite
Neuhausen, S., Dunning, A., Steele, L., Yakumo, K., Hoffman, M., Szabo, C., Tee, L., Baines, C., Pharoah, P., Goldgar, D. and Easton, D. (2004), Role of CHEK2*1100delC in unselected series of non-BRCA1/2 male breast cancers. Int. J. Cancer, 108: 477–478. doi: 10.1002/ijc.11385
- Issue published online: 21 NOV 2003
- Article first published online: 21 OCT 2003
- Manuscript Accepted: 6 MAY 2003
- Manuscript Revised: 2 MAY 2003
- Manuscript Received: 5 DEC 2002
- Cancer Research U.K.
- NCI. Grant Number: NIH CA74415
- American Cancer Society. Grant Number: RPG-99-181-01CCE
- Kirby Foundation
- Utah Cancer Registry. Grant Number: NO1-CN-6700
- Utah State Department of Health
- SwissBridge Foundation
Breast cancer in men is rare compared to the high incidence of the disease in women, with approximately 1 male case occurring for every 100–150 female cases.1 Male breast cancer is associated with an increased familial risk of both female and male breast cancer.2, 3 At least part of this familial association is due to germline mutations in the 2 breast cancer susceptibility genes, BRCA1 and BRCA2, with BRCA2 being the more relevant of the 2 genes.4 Studies have shown that BRCA1/2 mutations account for approximately 6–15% of all cases,5, 6, 7, 8 with the higher values found in populations with founder BRCA2 mutations.
Recently, a frameshift mutation in the CHEK2 gene (1100delC) was shown to be associated with non-BRCA1/2 female breast cancer, particularly among cases with a family history of the disease.9, 10CHEK2 encodes a G2/M checkpoint kinase that can regulate BRCA1 activity,11 and the CHEK2*1100delC variant leads to a truncated protein that abolishes the CHEK2 kinase activity.12 Interestingly, in the CHEK2 Breast Cancer Consortium study,9 the authors also found an apparent stronger association in families with at least 1 case of male breast cancer, with 4 out of 33 (12%) index cases and 7/52 (14%) of all cases in these families carrying the CHEK2*1100delC variant. This compared to 4.2% of families without male breast cancer and 1.1% of cancer-free controls. Of the 8 male cases actually tested for the variant, 2 were positive. The authors estimated that the CHEK2*1100delC was associated with a 10-fold increased risk of male breast cancer. This observation raises the possibility that this variant could represent an important moderate risk susceptibility allele for male breast cancer, accounting for a significant fraction of all cases. Based on the estimated 1% carrier frequency and 10-fold risk, this variant would be predicted to be found in 9% of all male breast cancer cases. To evaluate the overall contribution of CHEK2*1100delC to male breast cancer incidence, we examined this variant in 2 population-based series of male breast cancer cases in the U.S. and U.K.
The U.S. cases were sampled as part of an ongoing study of male breast cancer and were recruited from Utah and the surrounding states of Colorado, Idaho and Wyoming, The majority of the cases were ascertained from the cancer registries in these 4 states and the remainder from patient support groups. A total of 126 male breast cancer cases were available; of these, 10 (8%) were excluded due to the identification of a deleterious mutation in BRCA2 and 7 were eliminated due to repeated genotyping failure. Age at diagnosis of breast cancer of the 109 genotyped cases ranged from 28 to 93 years with a mean of 60 years. One hundred thirty-eight males unaffected with cancer and with a similar age distribution as the cases were selected as controls. These individuals were controls from a population-based case control study of colon cancer in Utah. Over 95% of the cases and controls were Caucasian. Genotyping for cases and controls was performed using a restriction endonuclease gel-based assay. A positive control was run on each set of samples.
The U.K. cases were ascertained as part of a population-based series of male breast cancer in the areas covered by the East Anglia and West Midlands cancer registries. After exclusion of 5 cases with BRCA2 mutations, 83 cases were available for CHEK2 analysis, diagnosed between ages 36 and 89 years, with a mean of 68 years. Genotyping failed for 4 of these cases. Control subjects were females drawn from the European Prospective Investigation of Cancer (EPIC) study, a population-based cohort study of diet and health based in Norfolk, East Anglia. Genotypes were available on 3,749 controls. Genotyping was performed using the ABI PRISM 7700 sequence detection system (Applied Biosystems, Foster City, CA).
In the U.S. series, none of the 109 male breast cancer cases tested carried the CHEK2*1100delC allele, while 1 of the 138 controls carried the variant. In the U.K. series, none of the 79 male breast cancer cases tested carried the 1100delC variant, while 20 variants were found among the 3,749 U.K. controls. Thus in total, 0/188 of the male breast cancer cases were CHEK2*1100delC positive compared to 21/3887 (0.5%) controls (odds ratio 0, upper 95% confidence limit = 3.99). The 0.5% frequency of CHEK2*1100delC in our controls is lower than that in other studies (1.1% and 1.4% in references 9 and 10, respectively). This is unlikely to be due to lower sensitivity of the assays used in our study, since both of the methods utilised in the present study are of equal or higher sensitivity than the oligonucleotide hybridisation assay used in Meijers-Heijboer et al.9 In addition, all positive samples were sequenced to verify the presence of the variant. More likely the observed difference is related to differences in ethnic composition of the various studies. We note in this regard that the frequency of CHEK2*1100delC was higher in the Netherlands and in Finland than in other populations.9, 10 The frequency in controls from the UK and USA reported in the CHEK2 Consortium study was 9/976 (0.9%), which is not statistically significantly different from the 0.5% observed in the present study.
Of the 188 male cases tested, 47 (25%) had a family history of breast cancer in first-degree relatives (including 1 with a male relative with breast cancer). Of these 47, none was CHEK2 positive, significantly different from the 4/33 observed in Meijers-Heijboer et al.9 (p=0.026, Fisher's exact test). Of the family history positive cases in our study, 7 had at least 2 female relatives diagnosed under age 60.
The previous finding of an increased frequency of the CHEK2*1100delC variant among female breast cancer cases who also had a male relative affected with breast cancer indicate that this variant might account for a substantial fraction of male breast cancer cases in general. However, the results of our study show that this variant is unlikely to account for a significant fraction of male breast cancer cases. Based on our series, the upper 95% confidence limit for the proportion of male breast cancer cases carrying the CHEK2*1100delC allele is 2%; our study had more than 80% power to detect an odds ratio of 6 or larger at a significance level of 5%. Given that a number of control series have estimated the population frequency of this variant at about 1%, and our own data show an even lower frequency in the population, it is clear that the variant cannot explain a large fraction of male breast cancer cases. Even when considering the cases with family history, there is little evidence that CHEK2 plays a major role; the upper 95% confidence limit for the frequency of the 1100delC variant among male breast cancer cases with an affected female first-degree relative positive is 7.5%.
Although we did not find any evidence for an association between CHEK2*1100delC and male breast cancer, our results are not necessarily inconsistent with those previously published by the CHEK2 Consoritum.9 It is possible that the high frequency of CHEK2*1100delC in cases with a family history of male breast cancer found in that study is largely or even entirely due to chance—the significance of the difference between the 12% frequency in this group and the 4% frequency in families without a male case was only 0.04. Perhaps more likely is that CHEK2*1100delC is associated with a more modest increased risk of male breast cancer. The upper confidence limit from our study overlaps (albeit, barely) with that given by the segregation analysis of Meijers-Heijboer et al.9 (OR=10.28, 95%CI 3.54–29.87). Another explanation for the observed difference is that the families in the consortium study, which were selected on multiple cases of female breast cancer, are segregating an allele in an additional susceptibility locus that interacts with CHEK2*1100delC to increase risk for male breast cancer. This other allele could be sufficiently rare (or population specific) that it might not be present in our set of male cases, thus nullifying the effect of CHEK2*1100delC. This is analogous to the apparent absence of risk associated with CHEK2*1100delC in BRCA1 and BRCA2 carriers.
In summary, our results indicate that the relative risk of breast cancer in male carriers of CHEK2*1100delC is likely to be smaller than previously suggested and does not explain the familial aggregation of breast cancer in families of male cases.
Our study was supported by Cancer Research U.K., grant NIH CA74415 from NCI (to S.N.), grant RPG-99-181-01CCE from the American Cancer Society (to S.N.) with partial support by a generous contribution to the Society from the Kirby Foundation and by the Utah Cancer Registry (funded by Public Health Service Grant NO1-CN-6700) and the Utah State Department of Health. The authors also gratefully acknowledge the support of a SwissBridge Foundation grant to D.G. D.E. is a Principal Research Fellow of Cancer Research U.K.
- 1ParkinDM, WhelanSL, FerlayJ, RaymondL, YoungJ, eds. Cancer incidence in five continents. Vol. VII, IARC Sci Publ Vol. 143, Lyon: International Agency for Research on Cancer, 1997.