A case-control study of cryptorchidism and maternal hormone concentrations in early pregnancy.

Serum samples taken between 6 and 20 weeks of gestation were obtained from 28 mothers who gave birth to cryptorchid sons (cases) and from 108 control mothers. In comparison with controls the cases had 10% higher geometric mean oestradiol (95% CI -13% to +39%: P=0.42) and 10% lower geometric mean testosterone (95% CI -27% to +10%: P=0.30). Among the samples collected between 6 and 14 weeks of gestation geometric mean concentrations of oestradiol and testosterone were 5% lower (95% CI -32% to +31%: P=0.74) and 25% lower (95% CI -45% to +1%: P=0.06) respectively in cases than in controls. Among the samples collected between 15 and 20 weeks of gestation geometric mean concentrations of oestradiol and testosterone were 29% higher (95% CI -8% to +79%: P=0.14) and 21% higher (95% CI -8% to +60%: P=0.18) respectively in cases than in controls. The results do not support the hypothesis that cryptorchidism may be caused by high concentrations of oestradiol in the maternal blood during the first phase of testicular descent, but suggest that the possible association of cryptorchidism with low maternal testosterone during early gestation should be further investigated.

Candidate regions for a testicular cancer susceptibility gene.

Epidemiological data suggest the presence of a susceptibility gene for testicular cancer in some families. Families with multiple cases of testicular cancer are rare and almost all those reported have only two affected members. We have performed a sib-pair analysis on 35 families in which there are either two or three affected brothers. These families have been typed for 220 autosomal microsatellite markers spaced 10-20 cM throughout the genome. Six regions which gave a LOD score of more than 1.0 on formal linkage analysis or a P value of 0.05 or less using a non-parametric approach are considered as candidate regions for a susceptibility gene. Of particular interest is one region on chromosome 4. Two neighbouring probes in this region both scored positively with LOD score of 2.60 on multipoint analysis. An International Testis Cancer Linkage Consortium has been formed to pool resources and will investigate these findings further with the world-wide collection of families.

Familial testicular cancer: a report of the UK family register, estimation of risk and an HLA class 1 sib-pair analysis.

Forty-two families with two or more cases of testicular cancer have been reported to the UK Register for Familial Testicular Cancer, comprising two pairs of identical twins, 27 sets of other brothers (25 pairs, two triples), nine father-son pairs, two pairs of first cousins and two uncle-nephew pairs. In total 91 testicular tumours are described in 86 individuals (42 (46%) pure seminoma, 49 (54%) other germ cell tumours). The median age at diagnosis in these patients was significantly younger than that in a comparable series of non-familial patients (29 c.f. 32.5 years, P less than 0.01). In a case-control comparison of 794 testicular cancer patients, eight patients (1.0%) had a brother and four patients (0.5%) had a father with a previous diagnosis of testicular cancer at the time of their own diagnosis (and these families are all included in this report). Two out of 794 controls (0.3%) had a first degree relative with testicular cancer. The cumulative risk to a brother of a patient for developing testicular cancer by the age of 50 years was estimated to be 2.2% (95% C.I. 0.6-3.8%) which results in a relative risk of 9.8 (95% C.I. 2.8-16.7) in comparison with the general population. HLA Class I typing of 21 affected sib-pairs demonstrated four (19%) sharing two haplotypes, 13 pairs (62%) sharing one and four pairs (19%) sharing none. This did not differ significantly from the expected proportions of 25%/50%/25%. It is unlikely, therefore, that there is a major gene associated with testicular cancer predisposition within or closely linked to the major histocompatibility gene complex on chromosome 6.