Baffoe-Bonnie Agnes B, Kiemeney Lambertus A L M, Beaty Terri H, Bailey-Wilson Joan E, Schnell Audrey H, Sigvaldason Helgi, Olafsdóttir Guriur, Tryggvadóttir Laufey, Tulinius Hrafn
Department of Epidemiology, Johns Hopkins School of Hygiene and Public Health, Baltimore, Maryland 21205, USA.
Genet Epidemiol. 2002 Nov;23(4):349-63. doi: 10.1002/gepi.10188.
Breast cancer and prostate cancer are the most commonly occurring cancers in females and males, respectively. The objective of this project was to test the hypothesis that breast cancer in females and prostate cancer in males represent homologous cancers that may be controlled by one or more common unidentified genes that may explain some of the observed familial aggregation. We modeled the transmission of a breast-prostate cancer phenotype in 389 pedigrees ascertained through a breast cancer proband drawn from the Icelandic Cancer Registry. Assuming that age at diagnosis of this combined phenotype followed a logistic distribution, segregation analyses were performed to evaluate residual parental effects, a sibship covariate, and a dichotomous cohort effect. The most parsimonious model was a Mendelian codominant model, which could partly explain the familial aggregation of both cancers. Inheritance of a putative high-risk allele (A) predicted gender-specific mean ages of onset for females as 53.8 years, 59.7 years, and 65.6 years for the putative AA, AB, and BB genotypes, respectively. Similarly, the predicted means were 73.7 years, 75.6 years, and 78.3 years, respectively, among males. Under this codominant model, the lifetime risk of a woman being affected was 19% by age 80 years. This implies that when prostate cancer among male relatives of breast cancer probands (unselected for family history or early-onset disease) is considered a pleiotrophic effect of the same gene that increases the risk for breast cancer, women are predicted to have a less than 1 in 5 risk of developing breast cancer when they carry the putative high-risk allele. However, this is a higher risk than in the general Icelandic population. Our results suggest that BRCA2 mutations alone are inadequate to explain all of the excess clustering of prostate cancer cases in families of breast cancer probands, and that additional genes conferring excess risk to both breast and prostate cancer may exist in this population.
乳腺癌和前列腺癌分别是女性和男性中最常见的癌症。本项目的目的是检验这样一个假设：女性乳腺癌和男性前列腺癌是同源性癌症，可能由一个或多个尚未确定的共同基因控制，这些基因或许可以解释部分观察到的家族聚集现象。我们对从冰岛癌症登记处选取的389个家系中通过乳腺癌先证者确定的乳腺-前列腺癌表型的传递进行了建模。假设该联合表型的诊断年龄遵循逻辑分布，进行了分离分析以评估残余的亲代效应、同胞协变量和二分队列效应。最简约的模型是孟德尔共显性模型，该模型可以部分解释两种癌症的家族聚集现象。假定的高危等位基因（A）的遗传预测，对于女性，假定的AA、AB和BB基因型的平均发病年龄分别为53.8岁、59.7岁和65.6岁。同样，男性中的预测均值分别为73.7岁、75.6岁和78.3岁。在这个共显性模型下，80岁时女性受影响的终生风险为19%。这意味着，当把乳腺癌先证者男性亲属中的前列腺癌（未根据家族史或早发疾病进行选择）视为增加乳腺癌风险的同一基因的多效性效应时，预计携带假定高危等位基因的女性患乳腺癌的风险低于五分之一。然而，这一风险高于冰岛普通人群。我们的结果表明，仅BRCA2突变不足以解释乳腺癌先证者家族中前列腺癌病例的所有额外聚集现象，并且在该人群中可能存在使乳腺癌和前列腺癌风险均增加的其他基因。
