Adverse Male Endpoints in Relation to Endogenous Hormone Exposure

While few data are available on pregnancy hormone levels in relation to adverse male endpoints, the literature supports relationships between these endpoints and several other pregnancy-related factors. There are also data suggesting that these factors, in turn, are reflective of the hormonal milieu of pregnancy. Birth order (and the correlated variable, maternal age) is quite consistently associated with adverse male reproductive outcomes, with risks of testicular cancer (particularly seminomas), cryptorchidism, and hypospadias greater among first birth. This elevated risk may be related to a higher level of estradiol [73-75], particularly unbound estradiol [76], in first compared to later pregnancies. The increase in risk was, in some studies, most marked among women experiencing their first birth at an older age. This finding may reflect a higher level of free estradiol in pregnancy among women delaying their first pregnancy until later in life. Bernstein et al. found free estradiol to be similar in young nulliparas and parous women, but to be higher among women having their first birth after age 34 [77]. Increased testosterone levels seen in the first trimester of first pregnancies (compared to later pregnancies) may also play a role in the increase in risk of these adverse male outcomes in first births [75].

Most studies also support an increased risk of testicular cancer, hypospadias, and cryptorchidism in association with low birth weight (and the correlated variable, prematurity). As discussed below, it is likely that this increase in risk reflects intrauterine growth retardation rather than preterm birth, per se.

We attempted to examine risks of these adverse male endpoints in relation to ethnicity. However, ethnic categories are themselves quite heterogeneous, and vary from one region to another, making it difficult to draw conclusions about ethnic variation. The strongest and most consistent association with ethnicity is seen for testicular cancer. The incidence of testicular cancer varies about tenfold, worldwide, from 0.7 per 1000 in African Americans to nearly 1 % (8.4 per 1000) in Denmark. This may be related to a lower serum testosterone level during the first trimester of pregnancy that has been reported in whites compared to African-Americans 78 which would suggest a role for reduced testosterone levels during early pregnancy in the etiology of testicular cancer and possibly cryptorchidism. The negative association for fetal serum testosterone and subsequent testicular cancer suggested by this comparison of testicular cancer rates in African-Americans and Caucasians appears inconsistent with the positive association suggested by the comparison of first with subsequent births described above. This suggests that any association between prenatal testosterone exposure and testicular cancer risk may be modified by ethnicity. In any case, the data presented here are inadequate to conclude that association between ethnicity and testicular cancer is, in whole or part, attributable to differences in endogenous hormone levels.

Testicular Cancer

First-born children have been found to be at increased risk of testicular cancer, though not consistently. The influences of birth order and maternal age appear to interact. The picture is further complicated by different patterns of association for seminomas and nonseminomas. M0ller and Skakkebaek [79] examined this question and found that for seminomas the increased risk conveyed by older maternal age was most marked for first births (OR=4.1, 95% CI 1.1-14.6). However, this pattern was not seen for nonseminomas. Consistent with this finding, Akre et al. [80] found the associations between maternal age and the risk of seminomas and nonseminomas to be opposite; risk increased 1.13 per 5 years of age for seminomas but decreased 0.86 per 5 years for nonseminomas (p = 0.054 for difference in trend). Prener et al. [81] found the risk of testicular cancer to be greater in first births; adjusted relative risk (ARR) for births of order four or higher relative to first births was 0.3 (95% CI 0.1-0.8). This association was particularly strong for seminomas (ARR = 0.1,95% CI 0.02-0.9).0n the other hand, Westergaard et al. [82] found the association between testicular cancer and birth order to be similar for seminomas and nonseminomas, while Moss et al. [83] found no association between testicular cancer risk and either birth order or maternal age.

Most studies that have examined testicular cancer risk in relation to low birth weight and prematurity found one or the other of these correlated variables to be associated with increased risk. Brown et al. [64] examined the joint effect of these variables and found a 12-fold increased risk for testicular cancer among infants weighing 5 lb. or less (95% confidence interval (CI 2.8-78.1)), a risk that was greater for shorter gestation (RR=16.9, 95% CI 2.3-346.7). However, fetal age at birth (gestational age) was not associated with risk of testis cancer independent of birth weight. Similarly, M0ller and Skakkebaek [79] found significantly increased risk for births less than 3000 and 2500 g (0R= 1.5 and 2.6 respectively) but no association with gestational age. Depue [69] found a threefold increased risk of testicular cancer among infants weighing less than 6 lb. (95% CI 1.2-8.4). This study did not examine prematurity. Gershman and Stolley [65] found a significant tenfold increase in risk of testicular cancer for premature births, but did not examine birth weight. Two studies that examined seminomas and nonseminomas separately [79, 80] found the association of these variables to be stronger with nonseminomas than seminomas.

Risk of testicular cancer has also been studied in twins compared to singleton pregnancies. While little difference in testicular cancer incidence was seen relative to the general population for monozygotic twins, among dizygotic twins there was a 60% excess incidence (p <0.05). The ratio of observed to expected cases was greatest in men less than 35 years old (O/E ratio = 2.6,95% CI 1.1-4.2) [84]. This may be related to the elevation in maternal hormonal levels in twin pregnancies relative to singleton pregnancies, with the greatest differences seen in dizygotic pregnancies. These increases, summarized by Braun et al. [84], include a doubling of estrogen, human chorionic gonadotropin, and human placental lactogen.

With respect to prostate cancer, there is no evidence that either birth order or maternal age is related to risk [85]. In relation to low birth weight and prematurity, the pattern for prostate cancer appears to be opposite to that seen for testicular cancer. Ekbom et al. [85] found no premature births among 80 cases, compared to 6% of 196 controls (p = 0.02) although mean birth weight was higher, though not significantly, among cases. Ross and Henderson [86] suggest that the dramatic (approximately 30-fold) difference in risk of prostate cancer between African-American and Japanese and Chinese men may be due, in part, to differences in androgen secretion and metabolism. For example, higher testosterone levels have been found in maternal serum of African-Americans [78], as well as in young adult males [87].


Depue [69] suggested that elevated levels of free estradiol, particularly early in the first trimester, may produce hypoplastic testis, a risk factor for crypt-orchidism [88]. Bernstein et al. [89] reported a significant 20% elevation in free estradiol in cryptorchid males compared to normal controls. Burton et al. [90], on the other hand, reported a 30% lower estradiol level in cases, though this was not statistically significant, and free estradiol was not examined, so these findings are not necessarily inconsistent. Key et al. [75] found serum testosterone in case mothers to be 25% lower in gestational weeks 6-14 (n = 18, p = 0.06), but 22 % higher in weeks 15-20, though this latter group was small (n =10, p = 0.18). This suggests that failure to control for gestational age at the time of serum sampling may account for some of the differences between study results.

Consistent with the findings on testicular cancer, a higher risk of crypt-orchidism has been seen for first births [91, 92] though this too has not been seen consistently. M0ller and Skakkebaek [79] reported a strongly decreased risk with higher birth order (A0R = 0.5 for birth order 4+, p-value for trend = 0.03), and Hjertqvist et al. [92] reported a significantly increased risk for first births (p<0.001). On the other hand, Key et al. [75] found no association between birth order and risk of cryptorchidism.

Low birth weight is consistently associated with an increased risk of cryptorchidism, independent of gestational age. M0ller and Skakkebaek [79] found the risk of cryptorchidism to be inversely correlated with birth weight, with odds ratios ranging from 0.4 for births of 4500+ g to 2.3 for births less than 2500 g (p-value for trend, 0.001). These authors found little association between cryptorchidism and gestational age. Similarly, Jones and colleagues [93] found low birth weight to be significantly associated with cryptorchidism, with no effect of gestational age after controlling for birth weight.

In one study, geometric mean testosterone during gestational weeks 6-14 was reduced 25% (p = 0.06) in whites compared to African-Americans [75]. A threefold greater risk of cryptorchidism in births to whites compared to African-Americans was reported in Los Angeles [78], a difference consistent with the far greater incidence of testicular cancer in whites in the United States. On the other hand, a cohort study in New York found no significant differences in ethnicity-specific prevalence rates of cryptorchidism at three months or one year of age [94].

We have focused this discussion on the maternal environment. However, there are two findings suggesting a possible role for paternal influence. Sweet and colleagues [95] reported a greater incidence of scrotal and testicular anomalies of the scrotum or testes in fathers of cases of hypospadias compared to normal controls (35% vs 3%, p < 0.001). Sperm concentration was also reduced among case fathers, but numbers were small. More recently, Fritz and Czeizel [96] reported poorer semen quality in fathers of cases of hypospadias compared to controls; p <0.02 for differences in sperm concentration, percent motile, and several morphological defects.


Hypospadias, which arises between weeks 6-14 of human embryonic development, can be regarded as a mild form of incomplete masculinization, a process that is androgen driven, and thus plausibly linked to endogenous hormone levels during pregnancy. Few studies on hypospadias have examined the association with birth order. Kallen et al. [97] found no association between birth order and hypospadias overall, but a higher risk among women over 40 at first birth, consistent with findings of Hay and Barbano [98], and consistent with the increase in risk of testicular cancer among women having their first birth at an older age.

Kallen et al. [97] reported an approximately doubled incidence of low birth weight (under 2500 g) in cases of hypospadias from seven national registries. An increased risk for premature birth was seen only among infants weighing less than 2500 g. An association between hypospadias and low birth weight is strengthened by the recent finding from a study of monozygotic twins discordant for hypospadias; in 16 out of 18 twins, the hypospadias occurred in the twin of lower birth weight (mean difference in birth weight 498 g; p <0.01) [99].

In the Collaborative Perinatal Project, which recorded diagnoses of hy-pospadias at birth hospital across the United States, rates in whites were somewhat elevated compared to African-Americans (8.1 per 1000 vs 6.9 per 1000) and Puerto Rican (5.2 per 1000) births. Based on data from The International Clearinghouse for Birth Defects Monitoring Systems (ICBDMS) rates of hypospadias in Asian populations (Tokyo and Szechwan) were reduced relative to those in the United States. However, comparisons across registries are highly problematic because of the range of diagnostic definitions that may be utilized.

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