Several recent studies indicate that the majority of families with five or fewer cases of breast cancer and no cases of ovarian cancer are not due to BRCA1 or BRCA2. It has been proposed that a further breast cancer susceptibility gene that may account for some of these families is located on chromosome 8p12-p22. We have identified 31 site-specific breast cancer families that have a greater than 80% posterior probability of being due to genes other than BRCA1 or BRCA2. These families have been examined for linkage to 8p12-p22 using markers flanking the putative location of the gene. The overall multi-point LOD score is strongly negative across the whole 44 cM. The individual multi-point LOD score is negative in 23 families and only exceeds 0.5 in a single family (with a multi-point LOD score of 1.22). The maximum heterogeneity LOD score was 0.03 at marker D8S136 with estimated proportion linked (α) of 3% (95% CI 0–30%). These data do not lend support to the hypothesis that chromosome 8p12-p22 harbours a familial breast cancer susceptibility gene.
Two major familial breast cancer susceptibility genes have been identified, BRCA1 and BRCA2 (Miki et al., 1994; Wooster et al., 1995; reviewed in Rahman and Stratton, 1998). Mutations in these genes are responsible for almost all breast-ovarian families and for the majority of site-specific families with six or more cases of breast cancer (Ford et al., 1998). However, several recent studies indicate that the majority of families with five or fewer cases of breast cancer and no case of ovarian cancer are not due to BRCA1 or BRCA2 mutations (Hakansson et al., 1997; Schubert et al., 1997; Serova et al., 1997; Ford et al., 1998). The underlying molecular defect(s) in these smaller familial clusters of breast cancer, which are commoner in the population, has not been elucidated.
Loss of the short arm of chromosome 8 has been reported in several types of carcinoma including prostate, colorectal, bladder, liver, ovarian, oesophageal and breast (Spurr et al., 1995). Loss of heterozygosity at microsatellite markers on chromosome 8p is observed in more than 50% of sporadic breast cancers (Kerangueven et al., 1995; Imbert et al., 1996; Yaremko et al., 1995, 1996). Allele loss is predominantly detected in the chromosome region 8p12-p22, although the region of LOH on chromosome 8p in breast cancer varies between studies and a consensus region of loss has not been identified. The tumour suppressor gene(s) that are the presumed target(s) for the LOH, have not been identified.
Prompted by the observation of allele loss in sporadic breast cancer, linkage analyses at chromosome 8p12-p22 in a modest number of breast cancer families have been reported, and have been interpreted as strong evidence in favour of a familial breast cancer susceptibility gene in this region (Kerangueven et al., 1995; Seitz et al., 1997a,b). In an analysis of eight French families for linkage to chromosome 8p12-p22, positive three point LOD scores were generated in all eight families using the markers NEFL and D8S259, but no individual LOD score exceeded 0.5. Analyses at flanking markers on chromosome 8p generated negative LOD scores. Most of the families showed some evidence against linkage to BRCA1 and BRCA2, but mutation screening data was not presented (Kerangueven et al., 1995). Seitz et al. (1997a,b) presented results of an analysis of 12 German families (that showed evidence against linkage to BRCA1 and BRCA2) for linkage at chromosome 8p12-p22. One family generated a multi-point LOD score of 2.32 and some breast tumours from this family showed loss of the non-segregating allele at some markers. However, only three of the remaining 11 families yielded positive LOD scores and in each case the LOD score was less than 0.6.
To evaluate the contribution of a chromosome 8p gene to familial breast cancer predisposition, we performed linkage analyses with markers spanning the putative region in 31 breast cancer families that are highly unlikely to be due to mutations in BRCA1 or BRCA2.
Families with at least three breast cancer cases diagnosed under the age of 60 years were accrued from Britain, Germany, Holland and USA. Families with ovarian cancer were excluded. DNA was extracted from at least three likely gene carriers in each family using standard protocols.
All families were screened for germline mutations throughout the coding regions of BRCA1 and BRCA2, by PCR amplification of fragments smaller than 300 bp and subsequently by conformation sensitive gel electrophoresis, SSCP/heteroduplex analysis and/or direct sequencing. Families from The Netherlands were in addition typed for the large BRCA1 deletions known to account for a substantial proportion of mutations in this population (Petrij-Bosch et al., 1997). In addition, all families were genotyped at microsatellite markers in the vicinity of both BRCA1 and BRCA2 using a minimum of three markers at each locus. Linkage analysis at BRCA1 and BRCA2 was performed using the model derived by Claus et al. (1991) and modified by Easton et al. (1993). This assumes a gene frequency of 0.003 and a lifetime penetrance of 85% in carriers, and of 10% in non-carriers. All LOD scores were calculated using GENEHUNTER 1.1 (Kruglyak et al., 1996) or VITESSE 2.0 beta (O'Connell et al., 1995) according to the size of the family. Marker allele frequencies were estimated from unrelated individuals from each pedigree.
The posterior probability that each family was due to a mutation in a dominant susceptibility gene other than BRCA1 or BRCA2 was calculated assuming that the sensitivity of mutation testing was 70%, and that the prior probability of linkage to BRCA1 or BRCA2 is given by the number of breast cancers in the family (Ford et al., 1998). Families with a posterior probability >80% of being due to mutation of a gene other than BRCA1 or BRCA2 were examined for linkage to chromosome 8p12-p22 using microsatellite markers spanning the location of the putative susceptibility locus. At least four markers were analysed in each family. The order of markers and approximate genetic distances between them is D8S261–7 cM–D8S560–0 cM – D8S298 – 0 cM – D8S133 – 4 cM – D8S136 – 5 cM – NEFL – 8 cM – D8S339 – 5 cM – D8S505 – 0 cM – D8S259 – 0 cM – D8S283 – 15 cM – D8S260. Linkage analysis was performed as described above. The overall LOD score was then maximized allowing for heterogeneity (HLOD), and location of the gene within the fixed marker map. At each stepped location over the map of markers a maximum HLOD and a corresponding estimate of the proportion of families linked (α) was obtained.
Thirty-one families with a posterior probability greater than 80% of being due to a dominant susceptibility gene other than BRCA1 or BRCA2 were identified. At least one affected individual from each family was negative for mutations in both BRCA1 and BRCA2, and no family had a LOD score greater than 0.27 at either locus (Table 1).
The multi-point analyses at chromosome 8p in each family are shown in Table 1. The overall multi-point LOD score was strongly negative over the whole 44 cM examined (Figure 1). Of the 31 families, 23 generated negative LOD scores over the whole region and only one of the remaining families generated a LOD score >0.5 (F3568, LOD score=1.22). The heterogeneity analysis performed over the 44 cM generated a maximum HLOD of 0.03 at marker D8S136 with an estimated proportion of families linked (α) of 3% (95% CI 0–30%). The 95% confidence interval for α is compatible with no families being attributable to this locus and indicates that not more than 30% of families are likely to be due to a gene on chromosome 8p. These data do not support the hypothesis that there is a familial breast cancer gene at chromosome 8p12-p22.
Together with the previous analyses, a total of 51 breast cancer families have been evaluated for linkage at chromosome 8p12-p22. The majority of families are unlinked, and in only one family (MDC 60, Seitz et al., 1997a,b) has substantial evidence of linkage been reported. This family generates a multi-point LOD score of 2.32 and includes nine cases of breast cancer, one case with breast and ovarian cancer and one case with ovarian cancer (Seitz et al., 1997b). The overwhelming majority of breast-ovarian families are attributable to mutations in BRCA1 and BRCA2 (Ford et al., 1998). However, mutation analysis data at BRCA1 and BRCA2 has not been published in this family. Furthermore, the individual with only ovarian cancer does not share the haplotype of marker alleles segregating in the individuals with breast cancer and was not coded as affected in the analyses. If this ovarian cancer case had been coded as affected, the evidence in favour of linkage to chromosome 8p from this family would have been considerably weaker.
It is possible that a familial breast cancer predisposition gene at chromosome 8p12-p22 only accounts for an appreciable number of breast cancer families in certain populations. The families analysed in the studies postulating a familial breast cancer gene on chromosome 8p were from France and Germany (Kerangueven et al., 1995; Seitz et al., 1997a,b). However, we have analysed nine distinct families from Germany none of which yield data suggestive of a breast cancer predisposition gene at this location.
The results from this study suggest that a gene at chromosome 8p12-p22 is unlikely to be making a substantial contribution to familial breast cancer predisposition, and it is thus inappropriate to use the term BRCA3 to describe this locus on the basis of currently available data. Indeed, when the family phenotypes, results of BRCA1 and BRCA2 mutation and linkage screening, and chromosome 8p linkage analyses from this and previous studies are taken together, the evidence for the existence of a familial breast cancer susceptibility gene at this location is very weak.
Recent studies suggest that families with fewer than four cases of breast cancer are unlikely to be due to mutations in BRCA1 and BRCA2 (Ford et al., 1998). Furthermore, population based, direct estimates, suggest that BRCA1 and BRCA2 mutations do not account for much of the familial risk conferred by early onset breast cancer (Peto et al., 1999; Malone et al., 1998). We have thus far identified 31 families with three or more cases of breast cancer diagnosed before age 60 years that are highly unlikely to be due to BRCA1 or BRCA2 mutations, and that are informative for linkage. Using this resource, a genome-wide linkage screen is being performed in an effort to locate at least some of the additional genes involved in familial breast cancer predisposition.
Claus EB, Risch N and Thompson WD. . 1991 Am. J. Hum. Genet. 48: 232–242.
Easton DF, Bishop DT, Ford D, Crockford GP and the breast cancer linkage consortium. . 1993 Am. J. Hum. Genet. 52: 678–701.
Ford D, Easton DF, Stratton MR, Narod SA, Golgar D, Devilee P, Bishop D, Weber B, Lenoir G, Chang-Claude J, Sobol H, Teare MD, Struewing J, Arason A, Scherneck S, Peto J, Rebbeck T, Tonin P, Neuhausen S, Barkadottir R, Eyfjord J, Lynch H, Ponder B, Gayther S, Birch J, Lindblom A, Stoppa-Lyonnet D, Bignon Y, Borg A, Hamann U, Haites N, Scott R, Maugard C, Vasen H, Seitz S, Cannon-Albright L, Schofield A, Zelada-Hedman . and the breast cancer linkage consortium. 1998 Am. J. Hum. Genet. 62: 676–689.
Hakansson S, Johansson O, Johansson U, Sellberg G, Loman N, Gerdes AM and Holmberg E. . 1997 Am. J. Hum. Genet. 60: 1068–1078.
Imbert A, Chaffanet M, Essioux L, Noguchi T, Adelaide J, Kerangueven F, Le Paslier D, Bonaiti-Pellie C, Sobol H, Birnbaum D and Pebusque M. . 1996 Genomics 32: 29–38.
Kerangueven F, Essioux L, Dib A, Noguchi T, Allione F, Dib A, Noguchi T, Allione F, Geneix J, Longy M, Liderau, Eisinger F, Pebusque M, Jacquemier J, Bonaiti-Pellie C, Sobol H and Birnbaum D. . 1995 Oncogene 10: 1023–1026.
Kruglyak L, Daly MJ, Reeve-Daly MP and Lander ES. . 1996 Am. J. Hum. Genet. 58: 1347–1363.
Malone K, Dalin J, Thompson J, O'Brien C, Francisco L and Ostrander E. . 1998 JAMA 279: 922–929.
Miki Y, Swensen J, Shattuck Eidens D, Futreal PA, Harshman K, Tavtigian S, Lui Q, Chochran C, Bennett L, Ding W, Bell R, Rosenthal J, Hussey C, Tran T, McClure M, Frye C, Hattier T, Phelps R, Haugen-Strano A, Katcher H, Yakumo K, Gholami Z, Shaffer D, Stone S, Bayer S, Wray C, Bogden R, Dayananth P, Ward J, Tonin P, Narod S, Bristow P, Norris F, Helvering L, Morrison P, Rosteck P, Lai M, Barrett J, Lewis C, Neuhausen S, Cannon-Allbright L, Goldgar D, Wiseman R, Kamb A and Skolnick M. . 1994 Science 266: 66–71.
O'Connell JR and Weeks DE. . 1995 Nat. Genet. 11: 402–408.
Peto J, Collins N, Barfoot R, Seal S, Warren W, Rahman N, Easton D, Evans C, Deacon J and Stratton M. . 1999 J. Natl. Cancer Inst. 91: 943–949.
Petrij-Bosch A, Peelen T, van Vliet M, van Eijk R, Olmer R, Drusedau M, Hogervorst F, Hageman S, Arts P, Ligtenberg M, Meijers-Heijboer H, Klijn J, Vasen H, Cornelisse C, Veer L, Bakker E, Ommen G and Devilee P. . 1997 Nat. Genet. 17: 341–345.
Rahman N and Stratton MR. . 1998 Ann. Rev. Genet. 32: 95–121.
Schubert E, Lee M, Mefford H, Argonza R, Morrow J, Hull J, Dann J. . 1997 Am. J. Hum. Genet. 60: 1031–1040.
Seitz S, Rohde K, Bender E, Nothangel A, Kolble K, Schlag P and Scherneck S. . 1997a Oncogene 14: 741–743.
Seitz S, Rohde K, Bender E, Nothangel A, Pidde H, Ullrich O-M, El-Zehairy A, Haensch W, Jandrig B, Koble K, Schlag PM and Scherneck S. . 1997b Br. J. Cancer 76: 983–991.
Serova O, Mazoyer S, Puget N, Dubois V, Tonin P, Schugart Y, Goldgar D, Narod S, Lynch H and Lenoir G. . 1997 Am. J. Hum. Genet. 60: 486–495.
Spurr NK, Blanton S, Bookstein R, Clarke R, Cottingham R, Daiger S, Drayna D, Faber P, Horrigan S, Kas K, Kirchgessner C, Kumar S, Leach R, Luedecke H, Sim E, Sullivan L, Takle L, Vance J, Wagner M, Wells D, Westbrook C, Yaremko L, Zaletayev D, Zuffardi and Wood S. . 1995 Cytogenet. Cell Genet. 68: 148–155.
Wooster R, Bignell G, Lancaster J, Swift S, Seal S, Mangion J, Collins N, Gregory S, Gumbs C, Micklem G, Barfoot R, Hamoudi R, Patel S, Rice C, Biggs P, Hashim Y, Smith A, Connor F, Arason A, Gudmundsson J, Ficence D, Kelsell D, Ford D, Tonin P, Bishop D, Spurr N, Ponder B, Eeles R, Peto J, Devilee P, Cornelisse C, Lynch H, Narod S, Lenoir G, Egilsson V, Barkardottir R, Easton D, Bentley D, Futreal P, Ashworth A and Stratton MR. . 1995 Nature 378: 789–791.
Yaremko M, Recant W and Westbrook C. . 1995 Genes Chrom. Cancer 13: 186–191.
Yaremko M, Kutza C, Lyzak J, Mick R, Recant W and Westbrook C . 1996 Genes Chrom. Cancer 16: 189–195.
We are grateful to all the participating families. This research was supported by the Cancer Research Campaign, the Institute of Cancer Research, the Association for International Research (via grant MKHT 18 to D Goldgar) and the Medical Research Council (UK).
About this article
Breast Cancer Research and Treatment (2009)
Proceedings of the National Academy of Sciences (2008)
Critical Reviews in Oncology/Hematology (2007)
Immunohistochemical classification of non-BRCA1/2 tumors identifies different groups that demonstrate the heterogeneity of BRCAX families
Modern Pathology (2007)
Genetic Epidemiology (2006)