Portal de Periódicos da CAPES

Interaction of breast cancer and melanoma genotypes

1997; Elsevier BV; Volume: 350; Issue: 9082 Linguagem: Inglês

10.1016/s0140-6736(05)63267-9

1474-547X

Kari Hemminki, Pauli Vaittinen,

Cancer Genomics and Diagnostics

Breast cancer and melanoma are not known to share susceptibility genes.1Cannon-Albright LA Kamb A Skolnick M A review of inherited predisposition to melanoma.Semin Oncol. 1996; 23: 667-672PubMed Google Scholar We are not aware of any studies on the risk of melanoma in male breast-cancer patients but in families with BRCA1-linked breast and ovarian cancers melanomas have been reported, but no proven association with BRCA1.2Anderson DE Badzioch MD Breast cancer risks in relatives of male breast cancer patients.J Natl Cancer Inst. 1992; 84: 1114-1117Crossref PubMed Scopus (85) Google Scholar, 3Tonin P Moslehi R Green R et al.Linkage analysis of 26 Canadian breast and breast-ovarian cancer families.Hum Genet. 1995; 95: 545-550Crossref PubMed Scopus (33) Google Scholar Melanoma families are not known to have an excess of breast cancers although breast cancers occur in p16-linked families.1Cannon-Albright LA Kamb A Skolnick M A review of inherited predisposition to melanoma.Semin Oncol. 1996; 23: 667-672PubMed Google Scholar, 4Platz A Hansson J Månss on-Brahme E et al.Screening of germline mutations in the CDKN2A and CDKN2B genes in Swedish families with hereditary cutaneous melanoma.J Natl Cancer Inst. 1997; 89: 697-702Crossref PubMed Scopus (140) Google Scholar We used the recently established nationwide Swedish Family Cancer Database, unique both in size and population-based structure, which we think makes bias unlikely.5Hemminki K Vaittinen P Familial cancer in Sweden: population-based study.Int J Oncol. 1997; 11: 273-280PubMed Google Scholar The database consists of 6 million individuals born from 1941 on, with both biological parents and siblings organised in families. Cancer cases were taken from the nationwide Swedish Cancer Registry, and totalled 21 220 cases in adult offspring (15–51 years) born between 1941 and 1955. Five-year birth cohort specific relative risks (RRs) of cancer in the offspring were calculated and adjusted by the direct method, considering the RR of the offspring with cancer-free parents as referents. Confidence intervals (95% CI) were calculated on the basis of the Poisson distribution.5Hemminki K Vaittinen P Familial cancer in Sweden: population-based study.Int J Oncol. 1997; 11: 273-280PubMed Google Scholar Cancer risk in the offspring was analysed by parental breast cancer and melanoma (table). When both parents had breast cancer, RRs of all cancers and breast cancer in the offspring were 8·6 (95% CI 0·9–16) and 30 (95% CI 0–61), respectively. When the father had melanoma and the mother breast cancer, RR of all cancers in the offspring was 2·6. It may be notable that of 11 offspring five, including one male, had breast cancer (RR=5·1) and five had melanoma (RR=9·0). There was also increased RR of all cancers in the offspring, when the father had melanoma and the mother ovarian cancer (RR=4·0, n=3, 95% CI 0·9–8·9, data not shown). No cancers were found in the small groups of offspring in which these was paternal breast cancer, or melanoma and maternal melanoma.TableRelative risk (RR) of cancer in offspring of parents with cancerRR of cancers in offspring (n, 95% CI)Breast cancer in fatherMelanoma in fatherNo cancer in fatherBreast cancer in motherAll cancer8·6 (5, 0·9–16)2·6 (11, 1·0–4·1)1·1 (729, 1‚0–1·1)Breast30 (4, 0–61)5·1 (5, 0·5–9·8)1·5 (230, 1·3–1·7)Melanoma09·0 (5, 0·9–17)1·1 (93, 0·8–1·3)Melanoma in motherAll cancer001·3 (87, 1·0–1·6)Breast001·2 (18, 0·6–1·8)Melanoma002·8 (24, 1·7–4·0)No cancer in motherAll cancer1·2 (9, 0·4–2·0)1·0 (79, 0·8–1·3)1·0 (13640, 0·98–1·02)Breast1·1 (2, 20·5–2·7)0·7 (12, 0·3–1·1)1·0 (3041, 0·96–1·04)Melanoma02·0 (19, 1·1–2·9)1·0 (1763, 0·95–1·05)Offspring whose parents are cancer free are the referents (RR=1·0, right bottom corner); figures in bold indicate the 95% CI does not include 1·00, 0=no cancer cases. Open table in a new tab Offspring whose parents are cancer free are the referents (RR=1·0, right bottom corner); figures in bold indicate the 95% CI does not include 1·00, 0=no cancer cases. Offspring whose parents are cancer free are the referents (RR=1·0, right bottom corner); figures in bold indicate the 95% CI does not include 1·00, 0=no cancer cases. In family studies, inheritance is usually followed in one parental lineage but we could study joint effects from both parents. Numbers of cases were small, but they suggest interaction of genes inherited from both parents. Although breast-cancer risk was moderately increased in the offspring of the mothers with breast cancer (RR=l·5), it was high when the father also had breast cancer (RR=30, wide confidence intervals). Breast cancer and melanoma genes seemed to interact as melanoma in the father and breast cancer in the mother conferred a risk of breast cancer and melanoma in the offspring of 5·1 and 9·0, respectively. We suggest that both parents contribute mutated susceptibility genes to pathways partially shared in breast cancer and melanoma. We searched for germline mutations in the p16 gene by single-strand conformation polymorphisms in ten Swedish families with a history of breast cancer not linked to BRCA1 and presenting with at least one melanoma. We found none. The putative susceptibility genes shared by breast cancer and melanoma are yet to be identified.