Etiology

The dramatically increased occurrence of testicular cancer in this century suggests that environmental causes have become more prevalent, and observed cohort effects suggest that some factors act early in life. Patterns of familial occurrence suggest that inherited genetic factors also predispose to testicular cancer. Accordingly, etiologic studies have addressed environmental, perinatal, and genetic factors.

Perinatal Factors

A series of epidemiologic studies found the risk of testicular cancer to be associated with a history of in utero exposure to exogenous steroid hormones.1,43,79

An individual can be exposed in utero if his mother takes oral contraceptives during an unrecognized pregnancy. In years past, estrogens and (sometimes) progestins were administered as a test for pregnancy and were included in preparations prescribed as "supplements" or were taken in an attempt to avert abortion during a recognized pregnancy. In a cohort of men known to have been exposed in utero to the estrogen analogue diethylstilbestrol (DES), testicular cancer occurred at two to three times the expected frequency (although this excess was not statistically significant).80 Case-control studies addressing the role of DES had additional limitations but reported essentially consistent findings.1,43'47,79

Associations with factors that are indicators of high levels of free estrogen early in a woman's pregnancy have also been reported. For example, risk is elevated among men who were firstborn children,73,81 and free estradiol is higher in a woman's first pregnancy than in her second.82,83 Risk is elevated among men whose mothers had high body weight,79 and heavy women are presumed to produce more estradiol by aromatization of testosterone in adipose tissue. Risk is also elevated among sons of women who experienced excessive nausea and vomiting during pregnancy,43,84 and elevated free estradiol is found in women with severe hyperemesis during pregnancy.

Several epidemiologic studies have found low birth weight to be associated with the occurrence of testicular cancer,20,85 cryptorchidism,65,69 and hypospadias.66,86

These observed associations of perinatal factors with the later occurrence of testicular cancer are consistent with the model proposed by Henderson and colleagues37 and underscore the etiologic importance of events in early life. However, epidemiologic studies can rarely establish the timing and dose of harmful hormone exposures, so the mechanistic inferences that can be drawn from these results are limited. However, experiments in which estradiol was administered to pregnant mice during carefully timed intervals not only demonstrated that in utero exposure can cause cryptorchidism but also identified a period of susceptibility in this species. Exposure on the thirteenth day of pregnancy (embryonic day 13 [E13], when sexual differentiation is beginning) causes cryptorchidism in immature male mouse pups,87 and in utero exposure was associated with increased frequency of testicular teratoma in one study.88

Exposures of Adolescence and Early Adulthood

A number of studies have explored the effects of hormones whose levels rise in puberty. In some of the studies, levels of testosterone, LH, and follicle-stimulating hormone (FSH) were measured. In others, factors that can be interpreted as proxies for androgens and gonadotropins were measured instead. In a study conducted among men with diagnosed unilateral testicular cancer, investigators examined the contralateral testicle for the presence of carcinoma in situ (CIS). Men with both testicular cancer and CIS at the time of diagnosis tended to have lower levels of testosterone and higher levels of both LH and FSH76 than men with only testicular cancer. However, the implications of this result are not clear. One possibility is that this hormone profile confers greater susceptibility to disease, indicated by carcinogenic processes that are under way in both testicles of men with CIS. Alternatively, this hormone profile may result from disease processes that are occurring in both testicles of men with CIS. To distinguish causes from effects of disease, proxy measures of hormone levels that can be recalled from the past are of value.

Possible associations with androgen levels have been explored, using proxies. Two presumed proxies for testosterone levels are a history of severe acne during puberty and male pattern baldness as both are associated with somewhat higher testosterone levels. A history of both conditions may be less frequent among men who develop testicular cancer.84 In addition, testicular germ cell tumors and CIS are observed in patients who are in a low-androgen state.89 While animal experiments show that androgen is required for testicular descent,90 a limited number of epidemiologic studies suggest that a high proportion of boys with cryptorchidism have normal androgen action and response.91,92 Some cases of hypospadias have been attributed to defects of androgen action or response, but this mechanism appears to account for only a small proportion of nonsyndromic cases.93,94

Possible effects of FSH levels have also been addressed, using proxy measures. Dizygous twinning may be a proxy for elevated FSH levels. Dizygous twinning is known to be associated with elevated maternal FSH levels9596 and is a familial trait.97,98 Risk of testicular cancer is reportedly elevated among dizygous twins.99-101 One explanation for the association is that these twins inherit from their mothers a tendency to have elevated FSH levels, which promotes testicular carcinogenesis after puberty. (An alternate explanation is that estrogen levels are especially high during pregnancies that produce twins.) Further suggestion of an effect of FSH comes from case reports in which tes-ticular cancer was diagnosed in men following gonadotropin treatment102 and from the observation that patients with Kallmann's syndrome (who have insufficient gonadotropin secretion) do not develop testicular cancer, in spite of frequent cryptorchidism.89

Although indirect, these observations are consistent with patterns of lower testosterone and high gonadotropin levels in individuals who develop tes-ticular cancer. Rajpert-De Meyts and Skakkebaek89 provided an elegant discussion of possible mechanisms whereby sex hormones may mediate germ cell carcinogenesis. The possibility that elevated gonadotropins may be a risk factor accords well with the effects of puberty that were hypothesized by Henderson and colleagues37 and suggests that FSH-responsive genes may play a role in those phases of testicular carcinogenesis that occur in adulthood.

Additional and nonhormonal exposures occurring in adulthood have also been implicated in tes-ticular cancer etiology. Numerous small studies have investigated infectious agents, but most of these studies were conducted with relatively small samples and are inconclusive. An adequately sized retrospective study that is under way at the time of this publication will explore both hormonal and viral causes of testicular cancer. This study uses the US Department of Defense serum depository, which includes samples collected before the onset of tes-ticular cancer and which may prove to be an exceptional resource for resolving these hypotheses.

A claim offered in support of a viral etiology is the increased occurrence of testicular cancer among men infected with human immunodeficiency virus (HIV), who are clearly at an elevated risk for cancers that involve viral infections, namely, Kaposi's sarcoma (which has been linked to human herpesvirus 8) and non-Hodgkin's lymphoma (which is linked to Epstein-Barr virus). However, the suggestion that men infected with HIV are at an elevated risk for testicular cancer is based on a small number of cases occurring in men with AIDS, among whom a doubling of the expected occurrence was seen only for seminoma.103 Even if occurrence is elevated, a relative risk of 2 suggests quite a different dynamic than do the reported AIDS-associated relative risks of 300 for Kaposi's sarcoma and < 100 for non-Hodgkin's lymphoma.

A single report indicates that antibodies to the endogenous retrovirus K10 can be detected in 50 to 60% of patients with testicular cancer and that titers resolve with treatment.104 Although further work is needed to learn whether expression of K10 is part of the pathophysiologic process and whether antibodies can be used to monitor disease processes, this report is intriguing because of the novelty of the agent.

Studies exploring the number of sexual partners and related variables have not shown a pattern of association with a measured or presumptive history of sexually transmitted disease.

Etiologies that involve elevated scrotal temperature and testicular trauma have been postulated. Although both associations seem plausible, it is hard to imagine how either of these factors could be measured retrospectively without introducing recall bias. Exercise and regular participation in specific sports activities have been suggested as proxies for both scrotal temperature and testicular trauma. However, a recent review of the eight epidemiologic studies of the effect of physical activity on the risk of testicular cancer indicated that there is currently insufficient evidence from which to draw any conclusions.105

A series of studies of occupation and investigations of testicular cancer clusters, recently reviewed by McGlynn,106 has not pointed to specific occupational hazards. However, numerous reports indicate that testicular cancer has for some time tended to occur disproportionately among men of high socioeconomic status and among sons of women of high socioeconomic status.1,85 These associations are not understood but are observed consistently enough to warrant further investigation.

Genetic Factors

Specific genes that confer an elevated risk of testicular cancer have not yet been found. However, numerous studies strongly suggest that inherited genetic predisposition plays an important role. A positive family history of germ cell carcinoma is invariably found to be a risk factor, and estimates of relative risk are large, ranging from 3 to over 12 for first-degree relatives.26 107-116 These estimates of familial relative risk are greater than for most other cancers117 and indicate that inherited genetic factors almost surely play an important role. Although family members tend to share environmental exposures, statistical arguments show that familial relative risks of this magnitude are unlikely to arise from shared environment alone.118119 Studies that distinguish between types of first-degree relatives reported higher relative risks for brothers than for fathers or sons.

A single genetic segregation analysis of testicular cancer120 supported a major gene model over models that incorporate a polygenic effect only. This result suggests the existence of one or more major susceptibility genes, defined as individual genes responsible for a substantial variation in risk.

Studies of genetic linkage are conducted in families to identify chromosomal regions that contain major susceptibility genes. In linkage studies that have been completed to date, the International Testicular Cancer Linkage Consortium has identified (1) four autosomal regions with preliminary suggestions of genetic linkage121122 and (2) a region on chromosome Xq27 with significant evidence of linkage among families in which one member has a history of bilateral testicular cancer.123 These studies are currently being extended to additional families, whose participation may clarify linkage relationships in the autosomal regions and identify a susceptibility locus on Xq27.

A small number of candidate gene studies that have been completed to date have not suggested particularly promising susceptibility loci.108 124 125 However, the finding of genetic linkage to chromosome Xq27 and elsewhere in the genome will help investigators focus future association studies on positional candidate genes that have a greater prior probability of involvement in testicular cancer.

Candidate genes for cryptorchidism were recently suggested by the observation of bilateral undescended testes in mice with targeted disruption in each of six genes. The products of these genes are (1) Leydig insulin-like hormone (Insl3), a presumed signaling molecule that is expressed in a sex-specific pattern126; (2) a novel G protein-coupled receptor (now called G protein-coupled receptor affecting testis descent [Great]) with an expression pattern in embryonic gonads and gubernacula127; (3) three abdominally expressed Hox proteins128-130; and (4) a regulatory molecule with a basic helix-loop-helix DNA-binding motif.131 Overbeek and colleagues postulated that Great is the Insl3 receptor.127 Nef and colleagues suggest that Insl3 mediates estrogen-induced cryptorchidism, based on their finding that in utero administration of estra-diol-17P or DES at E13 both causes cryptorchidism and specifically down-regulates Insl3 expression in embryonic Leydig's cells.87 Epidemiologic studies examining human homologues of these genes are very preliminary and have not yet found a pattern of involvement in human cryptorchidism.

Cytogenetic analyses have identified numerous karyotypic abnormalities in germ cell tumors. The most consistent finding is an isochromosome, i(12p), on the short arm of chromosome 12 that is described in over 80% of published germ cell tumor karyotypes. It is now recognized that virtually all germ cell tumors have an increased 12p copy number, present as either one or more copies of i(12p), tandem duplications, or elements transposed to other regions of the genome.132 Because increased 12p copy number is observed even in the presumptive precursor lesion (carcinoma in situ), it may be one of the earliest somatic gene changes in the genesis of testicular germ cell tumors.133

A series of candidate genes on 12p are actively being investigated. Based on a review of functional analyses, Chaganti and Houldsworth133 recently suggested that CCND2 is a 12p gene whose deregulated expression leads to germ cell transformation. This gene encodes cyclin D2, a protein that (with its catalytic partners CDK4 and CDK6) controls the G1-S cell cycle checkpoint by phosphorylating the retinoblastoma protein (pRB). Cyclin D2 shows suggestive patterns of expression in established germ cell tumor cell lines, germ cell tumors, and CIS. It is plausible that overexpression of CCND2 could mediate critical events in puberty because cyclin D2 protein appears to be required for the maturation of FSH-responsive gonadal cells during sexual maturation.

Although the significance of cytogenetic changes seen in germ cell carcinoma is not yet understood, more recent studies, employing molecular genetic techniques, have begun to describe corresponding patterns in gene expression134 and to identify additional promising candidate genes.135,136

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment