Epidemiology of testicular cancer: An overview
Testicular cancer is a rare disease, accounting for 1.1% of all malignant neoplasms in Canadian males. Despite the low overall incidence of testicular cancer, it is the most common malignancy among young men. The incidence rate of testicular cancer has been increasing since the middle of the 20th century in many western countries. However, the etiology of testicular cancer is not well understood. A search of the peer-reviewed literature was conducted to identify important articles for review and inclusion in this overview of the epidemiology of testicular cancer. Most of the established risk factors are related to early life events, including cryptorchidism, carcinoma in situ and in utero exposure to estrogens. Occupational, lifestyle, socioeconomic and other risk factors have demonstrated mixed associations with testicular cancer. Although there are few established risk factors for testicular cancer, some appear to be related to hormonal balance at various life stages. Lifestyle and occupational exposures occurring later in life may play a role in promoting the disease, although they are not likely involved in cancer initiation. In addition to summarizing the current epidemiologic evidence on risk factors for testicular cancer, we suggest future research directions that may elucidate the etiology of testicular cancer. © 2005 Wiley-Liss, Inc.
Over the last 30 years, many studies have sought to discover factors associated with the development and promotion of testicular cancer. At present, however, there are few well-established risk factors for this disease, most notably cryptorchidism and age. For the most part, the etiology of testicular cancer remains unknown.
This article provides an overview of risk factors that have been associated with testicular cancer to date. The purpose of this overview is to summarize the scientific literature on the epidemiology of testicular cancer, thereby providing a guide for future research that will advance our knowledge of testicular cancer etiology. Our overview is based on a detailed literature search dating to the late 1960s that was carried out through electronic databases and Medline. The publications considered in this overview are reliable scientific articles on the etiology of testicular cancer from different parts of the world.
The worldwide incidence of testicular cancer has doubled over the last 40 years.1 In most countries, testicular cancer is rare, with an age-standardized incidence rate ranging from about 1/100,000 in Asian and African/African-American populations to 9.2/100,000 in Denmark (Table I).2 The age distribution of testicular cancer is distinct from that for other cancers, with the majority of cases occurring between the ages of 25–35.3 A second, much smaller, peak occurs after 80 years of age.3, 4, 5 Testicular cancer incidence varies markedly by race, with blacks and other nonwhite races exhibiting extremely low rates in comparison to white populations.6
Table I. Age-adjusted (per 100,000) incidence ratesof tesicular cancer in selected countries*
|USA, SEER: White||5.4|
|UK, England and Wales||4.6|
|USA, SEER: Black||0.7|
The incidence of testicular cancer has been increasing since the middle of the 20th century in many western countries, including Canada,3 the United States,4 the Nordic countries7 and England,8 with the potential exception of children aged 14 years or less, where little evidence for temporal variation has been observed.9, 10, 11 In Canada, a 50% increase in the incidence rate of testicular cancer has been observed over the last 25 years.3 Testicular cancer accounts for 1.1% of all malignant neoplasms in Canadian males; it occurs primarily in young males, and it represents an important source of morbidity for this age group.3 Mortality rates for testicular cancer are very low, with the 5-year survival rate increasing from about 63% to more than 90% during the last 30 years.12
Several factors have been associated with the increasing trends in testicular cancer incidence, including a strong birth cohort effect. A study in Norway found that males born in the 5-year intervals before (1935–39) or after (1945–49) the Nazi occupation (1940–1944) had an increased risk of testicular cancer when compared to those born during the occupation.5 A similar trend was also found in a study of Danish data from the same time period.13 The occupation of Norway caused many lifestyle changes, including decreased use of polluting vehicles, and an increase in physical activity and consumption of vegetables and dietary fiber.5
Etiology of testicular cancer
Ninety-five percent of all tumors of the testis are germ-cell neoplasms.14 The International Agency for Research on Cancer (IARC) recognizes 4 specific types of germ-cell tumours: seminomas, embryonal carcinomas, malignant teratomas and choriocarcinomas.15 These lesions can be combined into 2 histologic groups: seminomas and nonseminomas. Approximately 50% of all germ-cell tumors are seminomas, with the remaining 50% being nonseminomas.16 Nonseminomas often represent tumors of mixed histology and may include a variety of seminoma or nonseminoma histologic subtypes. A recent study evaluating the histology of mixed germ-cell tumors (MGCTs) reported a strong association between teratomas and yolk sac tumors.17 Further classification beyond the 2 main histologic categories (seminoma and nonseminoma) appears to have limited relevance in etiologic or clinical settings.18
The most established factor associated with testicular cancer is cryptorchidism (maldescendent testicle(s) or MDT).19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 It has been described that “the truly cryptochid testis is the testis that lies above the external inguinal ring, either within the inguinal canal or within the abdomen, and is nonpalpable in the unanesthetized patient.”30 The abnormality can be unilateral or bilateral. The association between MDT and testicular cancer was suggested as early as the beginning of the 19th century.26 MDT is associated with a 2–4-fold increase in the risk of testicular cancer, with a population attributable risk of 10%.19, 31 However, some studies have reported a relative risk (RR) for testicular cancer associated with MDT in the range of 5–10-fold.21, 25, 32, 33
A significant increase in the prevalence of MDT over the last 40 years has been observed in Denmark.34 Striking differences in testicular cancer incidence have previously been noted between nearby Denmark and Finland.35 Recently, 2 large studies have demonstrated large differences between the 2 countries in semen quality36 and in the prevalence of MDT34 at birth and at 3 months of age. Although the basis for the relationship between MDT and testicular cancer remains unclear,21 it has been suggested that common genetic, lifestyle and environmental factors may be involved in the etiology of these disorders.34 It is also possible that MDT may be part of a causal chain of events that can lead to testicular cancer or a less severe stage of testicular dysgenesis.
Carcinoma in situ
Carcinoma in situ (CIS) of the testis is a distinct histologic pattern preceding the development of seminomatous and nonseminomatous germ-cell tumours of the testis.37 The link between testicular cancer and CIS, which appears to precede the occurrence of testicular cancer, may be related to a condition referred to as testicular dysgenesis,38 which occurs in utero with varying levels of severity. Testicular cancer represents the most severe degree of dysgenesis, whereas CIS is a less severe form of the testicular dysgenesis syndrome.38
Fifty percent of patients diagnosed with CIS of the testis develop invasive testicular cancer within 5 years of diagnosis.26 It is thought that all patients who harbour CIS cells at puberty will eventually develop testicular cancer.39 The association between CIS and testicular cancer was first described in 1972.40 It is now generally agreed that nearly all occurrences of testicular cancer are preceded by the presence of CIS cells, with the exception of 2 rare tumour types: infantile germ-cell tumours and spermatocytic seminoma, which occurs in older men.39 The type of testicular cancer (seminoma or nonseminoma) is also hypothesized to be linked to CIS. There is a higher probability of developing nonseminomatous lesions if the transition from CIS to invasive cancer occurs at a young age.39 In contrast, seminomatous lesions are more likely to result from the progression of CIS at older ages.
Maternal risk factors
Several maternal factors have been associated with testicular cancer. These usually involve in utero exposures and surrogates. Exposures examined include those that the mother has control over (such as maternal smoking) and those that are beyond her control (such as exposure of the fetus to endogenous estrogens).
Characteristics of the newborn can act as surrogates for the maternal in utero environment and associated exposures. The length of the child at birth was measured in 2 studies but was found to be a nonsignificant risk factor for testicular cancer.41, 42 Another study examining testicular cancer risk factors in twins found no difference in the length of the affected twin as compared to the nonaffected twin.43 Seven studies have examined the effect of birth weight and testicular cancer: 4 found no association,20, 41, 44, 45 and 3 found an increased risk among low-birth weight children of approximately 1.5–2-fold.42, 46, 47 At this point, no clear conclusions can be drawn about the association between birth weight and testicular cancer risk, although the association may be mediated by intrauterine growth retardation.
In one study, a significant trend was found with timing of delivery.47 A 50% increase in risk was observed in males born at least 2 weeks early, whereas those born late were seen to be at reduced risk. A similar finding was also reported by Weir et al.20 It has been suggested that the relation between preterm birth and testicular cancer risk may involve elevated exposure to estrogens in utero.20
Excess of endogenous hormones (likely estrogen) can cause nausea during pregnancy.27 Nausea in pregnancy has been associated with a 4-fold increase in risk of testicular cancer.26 However, the most recent studies examining nausea in pregnancy found no excess risk27, 33, 48 or a protective effect.42, 47 The measurement of nausea in pregnancy is difficult and usually requires the mother to remember an event that occurred 30 years ago. Other differences in women with nausea such as dietary preferences may also exist.47 At this time, no definitive conclusions can be drawn about the association of nausea in pregnancy and testicular cancer.
Several studies have indicated that children of mothers with high parity have a decreased risk of testicular cancer when compared to children of nulliparous mothers.41, 44, 49, 50 However, not all studies have demonstrated such a protective effect.22, 42, 47 It has been previously suggested that maternal endogenous estrogen levels are higher in first pregnancies as compared to subsequent ones. Since elevated levels of estrogens in utero have been found to increase testicular cancer risk, firstborn sons may be at increased risk as a result of their increased exposure to estrogens in utero.44 More recently, a Swedish study reported a significant trend of testicular cancer risk decreasing both with increasing birth order and sidship size in men born before 1960.51 It was suggested that parental subfertility in testicular cancer cases (discussed further in “Reproductive health”) and not elevated maternal hormone levels alone likely explain the results.51
Several twin studies have examined the effect of endogenous hormones on testicular cancer risk.52, 53, 54 It is thought that dizygotic twin pregnancies have higher maternal hormone levels than monozygotic pregnancies.55, 56 Swerdlow et al.52 noted an odds ratio (OR) of 1.5 (95% confidence interval [CI] 1.1–2.2) for testicular cancer in dizygotic twins compared to monozygotic twins. Braun et al.54 reported a significant 2-fold increase in risk among dizygotic twins compared to Swedish population levels. The other twin study, which looked at maternal hormones in cancer, did not consider zygoticity but examined birth order and same/opposite sex twin pairs.53 Prenatal estrogen levels might be higher in males with a female co-twin as compared to males with a male co-twin.53 However, the finding of elevated testicular cancer risk in male/female twins was restricted to instances where the male twin was born after his female twin.
It has been suggested that exposure to the synthetic estrogen diethylstilbestrol (DES) during pregnancy might lead to an increased risk of testicular cancer in male offspring. DES was prescribed to about 4 million pregnant women worldwide from the late 1940s to the early 1970s to prevent abortions and pregnancy complications.57 Subsequently, DES has been banned as a result of a high incidence of clear-cell adenocarcinoma of the vagina in pubertal girls exposed to DES in utero.58 There is insufficient evidence to support an increased risk of testicular cancer in men exposed to DES in utero,31, 57, 59 although a recent study performed in the U.S. reported a nonsignificant 3-fold increase in risk in men exposed to DES in utero compared to unexposed men.60 This study involved only 7 cases of testicular cancer in the DES-exposed group and 2 in the nonexposed group, preventing definitive conclusions.
Several studies have found an increased risk of testicular cancer of 5–8-fold in relation to hormone use during pregnancy.20, 27, 46 The hormones studied have been varied and were not specified in most studies. One study found that the relationship remained significant even when stratifying by histologic subtype of testicular cancer,20 whereas another study found no increased risk of testicular cancer with exogenous hormone use during pregnancy.33
Other maternal factors
There is conflicting evidence on whether maternal age is a risk factor for testicular cancer. Some studies have found older maternal age to be associated with a 2-fold increase in testicular cancer risk,41, 42, 44 but other studies found no association.20, 61 A recent case-control study reported a doubling in risk in men whose mothers were aged 15–19 years at conception compared to older mothers.47 The effect of maternal age on testicular cancer risk remains unclear at this time, although it is possible that an impact of maternal age may be present only in women with no prior children.44
The association between maternal smoking and testicular cancer is inconsistent. A study performed in Canada20 found that sons of mothers who smoked more than 12 cigarettes per day during pregnancy had a lower risk of developing testicular cancer (OR = 0.6; 95% CI 0.4–0.9). Both Brown62 and Coupland47 found that maternal smoking had no effect on testicular cancer risk. Recently, 2 studies have found evidence for an increase in testicular cancer risk and maternal smoking, one a significant ecologic correlation in Nordic countries63 and the other a 2-fold increase in testicular cancer among children whose mothers developed lung cancer.64 These inconsistencies make definitive conclusions on the effect of maternal smoking on testicular cancer risk difficult. However, it appears likely that the effect may be mediated by the anti-estrogenic effect of cigarette smoke in the pregnant woman.63, 65
Few studies have examined the effect of maternal or paternal occupation on testicular cancer risk. It is plausible that the occupation of the mother might affect germ cells or the fetus if she were working while pregnant (due to exposures to pesticides or other chemicals). To date, no studies have found a statistically significant association between testicular cancer and maternal occupation.66, 67 Testicular cancer has been associated with paternal occupations such as wood processors, metalworkers, stationary engineers, food product workers, metal product workers, and food and beverage service workers.66
Other maternal health characteristics have only been sparsely examined with respect to testicular cancer. Retained placenta, presence of Rh-antibodies and fetal presentation have been associated with testicular cancer risk, although, other than a dysfunctional in utero environment, the reason for this is unclear.42, 44
Personal risk factors
The age distribution of testicular cancer is distinct. Peak incidence occurs between the ages of 25–35.3 A second, much smaller, peak occurs after 80 years of age.3, 4, 5 This age distribution is different from that for most other cancers, which normally peak much later in life. The incidence of testicular cancer appears to be related to sex hormone activity. The steep increase in testicular cancer incidence observed following puberty is thought to reflect the promotion of cells initiated during gestation to CIS by adult sex hormones.13, 68 CIS is thought to represent the precursor to most types of invasive testicular cancer (see “Carcinoma in situ”).
Testicular cancer incidence varies with race. Blacks and other nonwhite races have extremely low rates of testicular cancer in comparison to Caucasian populations.2, 6 Although the reasons for the differing rates are unknown, certain hormonally related racial differences have been observed. Pregnancy levels of androstenedione and testosterone were seen to be higher in black mothers compared to white mothers.69 Higher concentrations of free testosterone have also been recorded in young black men compared to young white men.70, 71 However, the possibility that other genetic, lifestyle, or environmental influences may contribute to this association should not be excluded.
Month of birth
In the late 1980s, several studies were performed examining the temporality of testicular cancer, particularly month of birth.72, 73, 74, 75 Testicular cancer risk has been observed to increase both on a 4-month cycle75 and on a 2-month cycle.72 One study found some evidence of an annual cycle, peaking in August.74 There is currently no explanation as to why testicular cancer risk may vary by month of birth.
In the early 1970s, it was hypothesized that there might be a relationship between androgen levels and testicular cancer risk, as testicular cancer incidence increases during the time of changing androgen levels.76 Because specific androgen levels are difficult to measure in population-based studies, surrogates are usually used. Because high androgen levels in males result in baldness at an early age, male baldness has been used as a surrogate for elevated androgen levels. One study48 assessed the level of baldness of subjects at the time of interview and found that an increasing level of baldness was associated with a 20% reduction in the risk of testicular cancer.
Age at puberty
Although age at puberty is difficult to measure, it would appear that late age at puberty is associated with a marked decrease (up to 50%) in the risk of testicular cancer.22, 77, 78, 79, 80 This association is based on results from studies that created a summary variable for age at puberty and supports a hormonal influence in the etiology of this cancer. Other studies that have used an indicator of puberty (such as shaving, body hair, nocturnal emissions or voice breaking) rather than defining a single age at puberty are more difficult to interpret in evaluating the effect of age at puberty.
Body mass index
Studies of body mass index (BMI) and testicular cancer risk are mixed, with some finding no association,45, 81, 82, 83 others reporting a 2-fold increase in testicular cancer risk with a low BMI48, 84 and one study reporting a 3-fold increase in risk of testicular cancer with high BMI.85 Studies examining weight separately from BMI found no association between increased body weight and testicular cancer risk.80, 81 A clear link between BMI and testicular cancer remains to be defined in future studies.
Four studies have examined height as a risk factor for testicular cancer:80, 82, 83, 84 3 studies have found a statistically significant elevation (OR approx. 1.5) in testicular cancer risk with increasing height,82, 83, 84 whereas the other found an elevated but nonsignificant risk.80 Several potential mechanisms have been proposed to explain a positive association of height with testicular cancer risk, including childhood nutrition,83 increased levels of insulin-like growth factor86, 87 or an earlier age at puberty in those with a greater height.88, 89
Although many studies have found positive associations between occupation and testicular cancer,6, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104 no occupation has emerged as a clear risk factor for this lesion. Occupational exposures to several chemical substances have also been associated with testicular cancer. Dimethylformamide (DMF) was implicated as a risk factor in case-reports of testicular cancer in leather tanners105 and aircraft repairmen.106 In 1999, however, a review of the literature on DMF conducted by IARC concluded that inadequate evidence exists to implicate DMF as a human carcinogen.107 Other chemical exposures of potential concern include: fertilizers high in nitrogen,108 nonspecified fertilizers,109 fumes or smoke109 and pesticides.93, 96 All of these studies focused on occupational exposures, except for fertilizers high in nitrogen, where exposure occurred as a result of nutrient runoff.108 It has also been suggested that childhood residence in an area with high nitrate concentrations in ground water might be associated with increased risk.110
Several studies have examined male infertility as a risk factor for testicular cancer.80, 109, 111, 112 Infertility is defined as the failure to conceive after 1 year of regular unprotected intercourse with the same partner.113 Since infertility conveys a complete failure to conceive, “subfertility” has become a more contemporary way of classifying couples unable to conceive, leaving the possibility of conception open. All 3 studies addressing the link between subfertility and testicular cancer found an increased risk in subfertile men.80, 109, 111, 112 It is unclear whether subfertility is a risk factor for testicular cancer, whether it is associated with testicular cancer through sharing common etiological factors or whether it is part of the causal chain leading to testicular cancer.
A linkage between testicular cancer and male reproductive health is supported by several investigations suggesting that male reproductive health has also been declining since World War II,57 following which the incidence of testicular cancer increased markedly. A meta-analysis of data from 61 studies has suggested a substantial decrease in sperm count and semen volume from 1938 to 1990.114 There are several plausible hypotheses to explain this decline, including in utero exposure to DES57 and exposure to environmental chemicals disrupting endocrine function.115 Personal factors such as change in dietary intake, which could be associated with increasing prosperity,116 and an increase in sedentary lifestyles117 may also have contributed to this trend. These risk factors indicate that male hormonal balance is important in reducing the risk of testicular cancer.
Studies in the late 1980s indicated that there might be an increased risk of testicular cancer after vasectomy.118, 119 Given the widespread use of vasectomy throughout the world, the population health implications of a causal link with testicular cancer would be appreciable. As a result, a number of studies have been undertaken to determine if this association could be confirmed. Two large-scale studies (1 cohort study and 1 case-control study) found no association between vasectomy and testicular cancer.120, 121 A review of 8 articles122 found only 1 study supportive of a positive association.
Other diseases and conditions
First primary testicular cancer
Males with an initial malignant testicular neoplasm have a greatly increased risk of a second tumor in the contralateral testicle.123 Two studies in Denmark and New Zealand have shown that having testicular cancer leads to an increased risk of a contralateral tumour (RRs ranging from 24.8–27.5).124, 125
Inguinal hernia has been implicated as a risk factor for testicular cancer, although the findings are mixed. Inguinal hernias are intestinal bulges that have pushed through a weak spot in the inguinal canal, a triangle-shaped opening between layers of abdominal muscle near the groin.126 Some studies have reported significant associations between testicular cancer and inguinal hernia,79, 109, 127 whereas other studies have found no association.23, 25, 31, 33 In one study,77 inguinal hernia was associated with an increased risk of nonseminomas (OR = 2.4, 95% CI 1.3–4.5) but not seminomas (OR = 1.6, 95% CI 0.9–2.9). The lack of histologic stratification in most studies may have attenuated the potential association between inguinal hernia and nonseminomas. It is also possible that recall bias and confusion between “hernia operations” and operations to correct MDT may have obfuscated the assessment of this risk factor.128 Hernias diagnosed before the age of 15 have also been associated with an increased risk of testicular cancer.77, 79
Many diseases have been examined in relation to testicular cancer, including mumps orchitis,18, 24, 48, 100, 112, 129, 130 spina bifida,131 neonatal jaundice,44 hydrocele,100 sexually transmitted diseases,129 polythelia132 and mononucleosis.33 Although mumps orchitis has been studied most frequently, the majority of these studies have failed to demonstrate a significant association with testicular cancer.18, 24, 48, 112, 129, 130
Smoking has not been thoroughly examined as a risk factor for testicular cancer. Because testicular cancer is primarily a disease of young men, smoking duration may be insufficient to enhance testicular carcinogenesis. Although a recent study by Srivastava and Kreiger133 reported a significant positive association between smoking and testicular cancer risk, they did not explicitly present data in support of their conclusion. Another study by Moller and Skakkebaek22 found no increased risk of testicular cancer in relation to smoking. The association between smoking and testicular cancer requires further study before definitive conclusions can be drawn.
Over the last 15–20 years, there has been concern about the possibility that 60 Hz electromagnetic fields (EMF) produced by electric blankets might increase the risk of cancer.134 Studies examining this hypothesis have encountered problems related to exposure assessment, as EMF arise from many more sources than just electric blankets.135 The one study examining testicular cancer risk and electric blanket use found no increase in risk.136
Trauma to the scrotum or testicle has been associated with an elevated risk of testicular cancer.24, 25, 109, 129, 130 As well, increased risk is associated with activities that can cause persistent low-level trauma, such as bicycle or motorcycle riding.109 Trauma could be hypothesized to increase mitotic activity in an already malignant testicle,129 or it may cause the victim to become aware of an already existing tumor.
Several studies have examined the relationship between high testicular temperature, due to occupational and nonoccupational exposures, and testicular cancer risk. Two general occupational studies found mixed results,109, 137 whereas nonoccupational studies have examined bathing as opposed to showering109 and the type of underwear worn129 without identifying an association with testicular cancer.
Male use of exogenous hormones prior to testicular cancer diagnosis was reported to have a protective effect in 1 case-control study,33 although the specific hormone used by these men was not specified.
A wide variety of lifestyle factors have been examined in the literature in the search for clues about the etiology of testicular cancer. The results have been mixed, with strong associations found with diet but weak and varied associations for sociodemographic indicators.
The relationship between marital status and testicular cancer risk is unclear. Studies have found both increased138 and decreased risk95 for single men. However, most studies have found nonsignificant results.27, 28, 95, 139, 140
The majority of the studies on the effect of socioeconomic status (SES) on testicular cancer risk have used SES indicators based on social class or education. Testicular cancer was associated with higher social class in 2 studies,102, 141 perhaps due to other factors such as diet or occupation. Other studies have observed no such effect.22, 50 Education has also been used as an indicator of SES. One study133 found a positive association between testicular cancer and increasing educational attainment, although the data were not explicitly presented. A recent study in Finland revealed that the once large disparity in testicular cancer risk between high and low social class, with higher incidence among men with higher SES, has diminished over time.142
The effect of physical activity on testicular cancer risk is unclear. Some studies have found that higher levels of physical activity have a protective effect,79, 129 whereas others have found an adverse effect133 or no effect.143 The studies finding the protective effect for physical activity failed to control for diet, whereas the study finding an adverse effect did control for some dietary factors.133
Diet has been associated with testicular cancer in several studies.43, 80, 85, 144, 145, 146, 147, 148 Despite the inconclusive nature of these findings, a recent article states: “The timing of the testicular cancer trend is consistent with a dietary origin, and the search for candidates should extend beyond hormonal agents to include those capable of causing genetic damage.”116. High intake of fat has been associated with increased risk of testicular cancer,145, 147 a finding consistent with many reports for cancers at other hormonally influenced sites such as breast, prostate, colon and ovary.147, 149, 150, 151 Dairy products, particularly milk and cheese,85, 144, 146, 148, 152 have been linked to testicular cancer. These foods contain the female sex hormones estrogen and progesterone.153 It is reasonable to hypothesize that estrogens or progesterone in milk and dairy products may be associated with the development of testicular cancer.146 The strength of this association is further supported by the recent increasing trend in testicular cancer incidence and the increased consumption of dairy products in developed countries starting in the 1940s and 1950s.154 Another factor contributing to the level of estrogens in dairy products may be the treatment of cattle with hormones and antibiotics, although the impact of this practice on carcinogenesis is unknown. A recent study examined the relationship between dietary intake of phytoestrogens and testicular cancer but did not observe an association.155 Other foods found to have an association with testicular cancer are meat,85, 144, 147, 156 low intake of fruits and vegetables156 and low intake of dietary calcium.147
There have been several conflicting studies on rural and urban patterns of testicular cancer.157 An article published in 1974 suggested a rural preponderance of testicular cancer;157 however, more recent articles have revealed no rural/urban differences in testicular cancer rates.158, 159
Family aggregation and genetics
There is strong evidence to support genetic susceptibility to testicular cancer. There is an increased risk of testicular cancer in brothers,18, 160, 161, 162, 163 fathers161, 162 and twin brothers52, 163 of testicular cancer cases. All of these findings support the existence of a testicular cancer susceptibility gene or genes.
The genetic characteristics of testicular germ-cell tumors in adults are well described in the literature.164 Testicular tumors have a chromosome number in the triploid range and are characterized by specific chromosomal gains at chromosomes 7, 8, 12, 21 and X, and by specific chromosomal losses at chromosomes 11, 13 and 18.164 As early as 1982, a nonrandom genetic alteration was localized to i(12p).165 Studies have shown that 80% of testicular tumors have one or more copies of i(12p).166 It would seem that isochromosome 12p is the recurrent structural chromosomal abnormality of these tumors.167 Recently, a susceptibility gene has been localized to Xq27.168 One study has estimated that 33.4% of all cases of testicular cancer are in individuals with the malignant genotype,162 assuming it is a recessive trait. However, the responsible genotype is not clear. There are many other potential genetic factors involved in the development or predisposition to testicular cancer;166 however, these are beyond the scope of this overview.
Detection, prevention and treatment
Testicular cancer does not have early symptoms. Cases usually present with nonspecific complaints, such as a heavy feeling in the testicle.169 Early detection can be accomplished by monthly self-examination, which requires manual examination of the testicles for lumps or other abnormalities. Although ultrasound scans can assist in the diagnosis of testicular tumors, confirmation requires a biopsy and pathologic examination of tissue.18 At diagnosis, seminomas are more often localized and less often metastatic than nonseminomas.13 The identification and monitoring of individuals with CIS can lead to early detection of testicular cancer, which is the key to successful treatment. It is possible that detection of testicular cancer could eventually be aided by the identification of CIS in young or adolescent males, as it is thought all cases of CIS of the testis progress to cancer. Thus, mitigating the factors that cause CIS and preventing the progression of the CIS to invasive cancer may help in reducing testicular cancer incidence and increasing survival.
At present, a surgical testicular biopsy is the most common method of diagnosing testicular CIS. Because of the invasive nature of this test, it is not suitable as a screening tool for the general population.39 There are several new noninvasive techniques, including scrotal ultrasonography (in which the testicles are examined for irregular echo patterns) and the analysis of semen for CIS cells.39 Due to the difficulty in diagnosing CIS in apparently healthy men, it is not possible to obtain a valid estimate of the prevalence of CIS in the general population at this time.
Treatment of testicular cancer involves surgery, usually an orchiectomy, but sometimes including a retroperitoneal lymph node dissection, depending on the stage of the cancer. After surgery, the patient usually undergoes radiation and/or chemotherapy. However, if the cancer is metastatic, surgery is sometimes delayed until after chemotherapy or radiation.169 Advances in treatment have led to an increase in the 5-year survival for testicular cancer during the last 30 years from 63% to more than 90%.12 Current research on treatment focuses on how to individualize therapy for specific patients so that survival is not compromised, while minimizing morbidity and maintaining quality of life.12
Testicular cancer has been increasing among men in many countries during the last 3 decades and is now the most common malignancy among young males. Testicular cancer may also represent an important indicator for male reproductive health. Despite expanding knowledge on the risk factors for testicular cancer, the etiology of this disease and the reasons for its increasing incidence remain poorly understood.
Current information on the causes of testicular cancer is limited to a few established and suspected risk factors including age, race and MDT. MDT, a major risk factor, is associated with a 2–4-fold increase in the risk of testicular cancer, although some studies have reported much higher risk. Significant geographic variation in the prevalence of this important risk factor has also been recently reported in Finland and Denmark, 2 countries with strikingly different testicular cancer incidence rates. The prevalence of MDT has also been recently reported to have greatly increased in recent decades in Denmark, a country with one of the highest age-standardized incidence rates of testicular cancer in the world.
Male subfertility may increase the risk of this disease, but the mechanism by which this occurs is not known. A declining trend in male reproductive health has also been observed in recent decades, as reflected in indicators such as sperm count and semen volume.
Maternal risk factors, such as age at pregnancy, maternal in utero exposures including maternal endogenous estrogen levels and maternal smoking, may play an important role in the etiology of this disease. A number of changes have been observed over recent decades in the prevalence of several potential maternal risk factors including increasing age at first birth,170 decreasing parity170 and improvements in the survival of premature infants.171
Early age at puberty and decreased levels of androgen have been associated with an increased risk of testicular cancer. The potential effect of BMI and physical activity remains unclear; however, height appears to be positively associated with testicular cancer risk. Secular trends involving both age at puberty and height have been observed with age at puberty decreasing172 and adult height increasing.173
The results of studies of the possible role of vasectomy and inguinal hernia in the etiology of testicular cancer are inconsistent and require clarification. It appears that testicular trauma increases the risk of testicular cancer.
Although testicular cancer appears to be more prevalent among higher social classes, the role of education in the etiology of this disease is not consistent. Food habits and dietary patterns may be related to the occurrence of the disease. A high intake of dairy products, particularly milk consumption, may lead to an increased risk of testicular cancer, possibly because of the estrogen and progesterone content of such products. Consumption of dairy products has also been observed to have increased since the mid-1900s.154 Risk is elevated among male relatives of testicular cancer cases and may be partly due to a testicular cancer susceptibility gene and common environmental risk factors.
Geographic and temporal differences in testicular cancer incidence rates have led investigators to suspect a number of potential lifestyle and environmental risk factors. Changes observed over time in the prevalence of many candidate risk factors may help to provide etiologic clues about the increasing incidence of this cancer. In particular, the trend of declining male reproductive health, coupled with the hormonal basis of several of the established and postulated risk factors for testicular cancer, suggests a need for more studies focused on elucidating a possible hormonal mechanism of testicular carcinogenesis. It is hoped that through future study, the incidence of this increasingly important source of morbidity for young males may be reduced.
We are grateful to the referees for helpful comments, which served to improve the original draft of this article. D.K. is the NSERC/SSHRC/McLaughlin Chair in Population Health Risk Assessment at the University of Ottawa.