Hereditary non-polyposis colorectal cancer (HNPCC) is an autosomal dominant disease, characterized by the occurrence of predominantly colon and endometrial cancer and, less frequently, cancer of the small bowel, stomach, hepatobiliary tract, ureter, renal pelvis, ovaries and brain. The phenotypic diversity may partially be explained by allelic heterogeneity. The aim of this study was to investigate the frequency of extracolonic cancers in a cohort of females sharing the same c.C1528T disease-predisposing mutation in the hMLH1 gene. Data on cancer history were obtained from 87 mutation-positive females and 121 mutation-negative sisters, as a control group. Testing for microsatellite instability (MSI) and expression of the wild-type hMLH1 allele was performed on extra-colonic tumour tissue blocks of mutation-positive individuals. Extracolonic cancer occurred in 14% (12/87) of mutation-positive females vs. 7% (8/121) of mutation-negative females (P = 0.10). Multiple primary cancers occurred at a significantly higher incidence in the first group. Breast cancer, which was the most frequent extra-colonic cancer in mutation positive females (53%), occurred at a young age, and occurred bilaterally in two out of seven cases. Involvement of the hMLH1 gene was confirmed in five out of seven cases of breast cancer, two cases of endometrial cancer, one case of ovarian cancer and one case of renal cell carcinoma, by detecting immunohistochemical compromise of the gene product. Although the study might not have been adequately statistically powered (to provide a significant P value), the noteworthy findings in this study include the confirmation of a range of Lynch II type cancers in a cohort we previously thought was wholly predisposed to Lynch I features, and a confirmation of breast cancer as part of the spectrum of Lynch syndrome cancers affecting women.

Lynch Syndrome (LS) is a cancer susceptibility syndrome caused mostly by mutations in the mismatch repair genes, hMLH1, hMSH2 and hMSH6. Mutation carriers are at risk of colorectal and endometrial cancer and, less frequently, cancer of the ovaries, stomach, small bowel, hepatobiliary tract, ureter, renal pelvis and brain. The influence of environmental factors on extracolonic cancer risk in LS patients has not been investigated thus far. The aim of this study was to investigate some of these factors in South African females carrying the hMLH1 c.C1528T mutation and their mutation-negative relatives. Data were collected from 87 mutation-positive females and 121 mutation-negative female relatives regarding age, cancer history, hormonal contraceptive use, parity, duration of breast feeding, height, weight and age at first birth, last birth, menarche and menopause. Influence of these factors on cancer risk was analysed by mixed-effects generalised linear models. Extracolonic cancer occurred in 14% (12/ 87) of mutation-positive females versus 7% (8/121) of mutation-negative females, (P = 0.0279, adjusted for age and relatedness between women). Breast cancer was the most common extracolonic cancer. An association was found for oral contraceptive use and extracolonic cancer risk in mutation-negative females only. No association was found for any of the other risk factors investigated, when adjusted for age. This might be due to the scarcity of extracolonic cancers in our data. Future knowledge on the influence of additional environmental factors on cancer risk in LS females can lead to evidence-based lifestyle advice for mutation carriers, thereby complementing the prevention strategies available today. In addition, it can contribute to an integrated model of cancer aetiology. Therefore, this study should be taken as a thrust for further research.