Abstract
Familial aggregation of testicular cancer has been reported consistently, but it is less clear if there is any association between risk of testicular cancer and other cancers in the family. We conducted a population based case-control study to examine the relationship between risk of testicular cancer and 22 different cancers in first-degree relatives. We included 3297 cases of testicular cancer notified to the Danish Cancer Registry between 1991 and 2003. 6594 matched controls were selected from the Danish Civil Registration System, which also provided the identity of 40,104 first-degree relatives of case and controls. Familial cancer was identified by linkage to the Danish Cancer Registry, and we used conditional logistic regression to analyse whether cancer among first-degree relatives was associated with higher risk of testicular cancer. Rate ratio (RR) for testicular cancer was 4.63 (95% CI: 2.41–8.87) when a father, 8.30(95% CI: 3.81–18.10) when a brother and 5.23 (95% CI: 1.35–20.26) when a son had testicular cancer compared with no familial testicular cancer. Results were similar when analyses were stratified by histologic subtypes of testicular cancer. Familial Non-Hodgkin lymphoma and oesophageal cancer were associated with testicular cancer; however these may be chance findings. The familial aggregation of testicular and possibly other cancers may be explained by shared genes and/or shared environmental factors, but the mutual importance of each of these is difficult to determine.
Keywords: Testicular cancer, familial aggregation, first-degree relatives, Denmark
Introduction
Testicular cancer is a rare disease, accounting for one percent of all incident male cancers worldwide 1. It is, however, the most common cancer among young men 15–34 years old 2, and Denmark has the highest incidence rate in the world with 9.7 cases per 100,000 inhabitants 3. Furthermore, incidence has increased over the last decades especially in the developed countries in Western Europe, North America, and Oceania 4. The few known risk factors for testicular cancer are cryptorchidism (undescended testes) and a family history of testicular cancer; early-life or in utero exposures are also suspected to play an important role 5;6. The increased risk of testicular cancer in male first-degree relatives of testicular cancer cases is well-documented 7–14. Some studies suggest that the risk is more pronounced in brothers of testicular cancer cases than in fathers and sons 8;9;11;13;14 and Swerdlow et al reported a relative risk as high as 37.5 in twin brothers of testicular cancer cases 15. Only few studies distinguish between histological subtypes of testicular cancer i.e. seminoma and nonseminoma providing no clear differences in risk patterns 8;12;16. Swedish data suggest that the association between testicular cancer in first-degree relatives is much stronger for pure than for mixed histological subtypes, i.e. the risk is higher for seminoma testicular cancer given seminoma testicular cancer in relatives and vice versa for nonseminoma testicular cancer 8;10.
While an association, thus, has been established between the risk of testicular cancer and testicular cancer in first-degree relatives, it is an open question whether an association also exists between risk of testicular cancer and non-testicular cancers in relatives. Such possible associations were explored in previous studies of which some indicated higher risk of testicular cancer in association with, among others, cancers of the pancreas, oesophagus, genitals and breast as well as melanomas and lymphomas, whereas other studies did not find such associations 7–12;16–19.
Thus, still little is known about the potential aggregation of cancers other than testicular cancer in families of testicular cancer cases, and large studies using registry data are useful for examining such possible associations. We used a large population-based case-control study from Denmark to explore if cancer among first-degree relatives was associated with the risk of testicular cancer.
Material and Methods
Study population
Cases
Cases were all males notified to the population-based Danish Cancer Registry with a testicular cancer diagnosis between 1991 and 2003. We included only first cancers with International Classification of Diseases for Oncology (ICD-O) morphology behaviour code 3 (malignant, primary site), however preceding diagnosis of non-melanoma skin cancer was allowed. In total, 3297 cases were included. Furthermore, we classified cases as either seminoma or nonseminoma testicular cancer. Seminoma testicular cancer was defined as ICD-O morphology codes: 9060/3, 9061/3, 9062/3 and 9063/3. The remaining testicular cancers were classified as nonseminoma. We identified 1871 seminoma and 1426 nonseminoma cases.
Controls
Eligible controls were all males born on the same day, month and year as cases and alive and living in Denmark at time of diagnosis. For cases 90 years of age or older eligible controls were males born in the same month and year. The wider time range was applied in this age group to ensure enough potential controls alive at the time of diagnosis. Further, eligible controls should have received no cancer diagnosis except from non-melanoma skin cancer at the date of diagnosis of the matched case. The pool of potential controls was identified in the Danish Civil Registration System, which was founded in 1968 and includes information on the whole Danish population. Two controls were randomly selected and matched individually with cases, thus the control group included 6594 individuals.
Family history of cancer
The Danish Civil Registration System is based on a10-digit personal identification number (PIN), which provides a unique identifier for every Danish citizen. Furthermore, the registry contains parental links between persons, which is utilised in the Danish Family Relations Database 20. 40,104 first-degree relatives (that is parents, siblings and children) of cases and controls were identified by linkage with this database. Subsequently, we identified cancer among these relatives by linkage with the Danish Cancer Registry, using their PIN. We only counted first primary cancers recorded in the cancer registry in 1943–2003 (ignoring non-melanoma skin cancers), and used the 7th Revision of the International Classification of Diseases (ICD-7) to group the cancer diagnoses of the fist-degree relatives into the following categories: 140–148: Buccal cavity and Pharynx, 150: Oesophagus, 151: Stomach, 153–154: Colorectal cancer, 155.0: Liver, 155.1: Gallbladder, 157: Pancreas, 161: Larynx, 162.0 and 162.1: Lung, 170: Female Breast, 171–176: Female genital organs, 178: Testis, 177 and 179: Other male genital organs, 178: Kidney, 181: Bladder, 190:Melanoma of skin, 193: Brain and nervous system, 197: Connective tissue, 200 and 202: Non-Hodgkin lymphoma (NHL), 201: Hodgkin’s disease, 203: Multiple myeloma, 204: Leukemia.
For each cancer site we counted the number of cases among the first-degree relatives. Only few persons had more than one relative with the same cancer type. Therefore, cancer in relatives was noted as ‘yes’ (for one or more cases) or ‘no’.
Statistical analyses
We used conditional logistic regression to analyse whether cancer in first-degree relatives was associated with higher risk of testicular cancer. All analyses were adjusted for number of known first-degree relatives (continuous variable). For sex-specific cancers, analyses were adjusted for number of known male and female relatives, respectively. We estimated rate ratios (RRs) for testicular cancer in association with each type of cancer in the first-degree relatives, and for testicular cancer separately for fathers, brothers and sons as well as for older and younger cases. Wald test was used to evaluate if estimates by age at diagnosis in family member and type of family member were statistically different. We also analysed risk of seminoma and nonseminoma testicular cancer separately. Wilcoxon tests were used to compare age in cases with and without testicular cancer in first-degree relatives and seminoma versus nonseminoma. Furthermore, we tested if age at diagnosis in the first-degree relatives of cases and controls differed. All analyses were performed using SAS 9.1, all tests were two-sided, and a 5 % significance level was used.
Results
In total, we included 3297 cases (57% seminomas and 43% nonseminomas), 6594 controls and 40,104 first-degree relatives. Sixty-nine cases and 26 controls had a male first-degree relative with testicular cancer, and the cases having a relative with testicular cancer were younger (mean: 33 years; median: 32 years) at date of diagnosis than cases with no testicular cancer in the family (mean: 37 years; median: 35 years) (Wilcoxon test: p=0.003). Table 1 shows characteristics of cases and controls. Due to the matching criteria, the age distribution of cases and controls was virtually identical. Furthermore, the number of known first-degree relatives was very similar for cases and controls, except that testicular cancer cases tended to have slightly fewer children than controls. Seminoma cases were older (mean: 40 years; median: 38 years) than nonseminoma cases (mean: 33 years; median: 30 years) at the time of diagnosis Wilcoxon test: p <0.001). Age at diagnosis in first-degree relatives was similar for cases and controls, both for all cancers except testicular cancer and when testicular cancer among fathers and brothers was considered (table 2). Sons of cases tended to be younger at date of diagnosis than sons of controls but the difference was not statistically significant (Wilcoxon test: p=0.49). Table 3 shows a RR of 5.97 (95% CI: 3.74–9.53) for testicular cancer in association with testicular cancer in any male first-degree relatives compared with no familial testicular cancer. Higher RRs for testicular cancer were observed when brothers and sons had testicular cancer compared to paternal testicular cancer, but the difference was not statistically significant (Wald test: p=0.49). RRs were similar for young and older cases (Wald test: p=0.82). We also tested the effect of age at disease onset in the relatives, which showed higher RR for testicular cancer when relatives were diagnosed at young ages (≤35 years) compared to relatives diagnosed at older ages (>35 years), however estimates were not significantly different (Wald test: p=0.35).
Table 1.
Characteristics of testicular cancer cases and controls
| Variable | cases (n=3297) | controls (n=6594) |
|---|---|---|
| Age at diagnosis/censoring | ||
| Min | 0 | 0 |
| 5% | 21 | 21 |
| 25% | 28 | 28 |
| 50% (mean) | 35(36.7) | 35(36.6) |
| 75% | 43 | 43 |
| 95% | 59 | 59 |
| Max | 96 | 96 |
| Identified family members | ||
| Father | 2496(76%) | 5001 (76%) |
| Mother | 2547(77%) | 5118 (78%) |
| Brothers | ||
| 0 | 1957 (59%) | 3802(58%) |
| 1 | 977 (30%) | 1941(29%) |
| 2 | 288 (9%) | 660(10%) |
| 3 | 57 (2%) | 149(2%) |
| 4+ | 18 (1%) | 42(1%) |
| Sisters | ||
| 0 | 2015(61%) | 4011(61%) |
| 1 | 998(30%) | 1936(29%) |
| 2 | 226(7%) | 542(8%) |
| 3 | 53(2%) | 76(1%) |
| 4+ | 5(0%) | 29(0%) |
| Sons | ||
| 0 | 1739(53%) | 3141 (48%) |
| 1 | 979 (30%) | 2164 (33%) |
| 2 | 497 (15%) | 1058 (16%) |
| 3 | 67(2%) | 198 (3%) |
| 4+ | 15(0%) | 33 (1%) |
| Daughters | ||
| 0 | 1735 (53%) | 3237 (49%) |
| 1 | 1037 (31%) | 2177 (33%) |
| 2 | 428 (13%) | 949 (14%) |
| 3 | 81 (2%) | 199 (3%) |
| 4+ | 16(0%) | 32(0%) |
Table 2.
Age at cancer diagnosis in first-degree relatives of testicular cancer cases and controls.
| Cancer type in relatives | Age at cancer diagnosis in relatives (years) | ||||||
|---|---|---|---|---|---|---|---|
| Testicular cancer cases | Controls | Wilcoxon testa | |||||
| n | mean | median | n | mean | median | p value | |
| All cancer ex. testicular cancer | 864 | 57.8 | 59.5 | 1687 | 57.5 | 59.5 | 0.55 |
| Testis | 69 | 33.8 | 30.8 | 26 | 33.1 | 33.0 | 0.75 |
| Father | 30 | 41.0 | 40.0 | 13 | 37.4 | 37.4 | 0.32 |
| Brother | 32 | 28.5 | 27.8 | 10 | 26.9 | 25.6 | 0.45 |
| Son | 7 | 26.8 | 25.5 | 3 | 35.0 | 36.9 | 0.49 |
Test for difference between age at cancer diagnosis in first-degree relatives of cases and controls.
Table 3.
Rate ratio (RR) of testicular cancer according to site-specific familial cancer in first-degree relatives.
| Cancer in first-degree relative (ICD-7) | nca/nco | RR (95% CI) | p |
|---|---|---|---|
| Buccal cavity and Pharynx (140–148) | 20/40 | 1.02 (0.59–1.75) | 0.95 |
| Oesophagus (150) | 17/19 | 1.83 (0.94–3.58) | 0.08 |
| Stomach (151) | 19/35 | 1.10 (0.62–1.93) | 0.75 |
| Colorectal cancer (153–154) | 91/189 | 0.97 (0.75–1.26) | 0.83 |
| Liver (155.0) | 7/11 | 1.27 (0.49–3.27) | 0.63 |
| Gallbladder (155.1) | 3/8 | 0.81 (0.22–3.09) | 0.76 |
| Pancreas (157) | 31/45 | 1.38 (0.87–2.18) | 0.17 |
| Larynx (161) | 7/28 | 0.50 (0.22–1.14) | 0.10 |
| Lung (162.0,1) | 120/249 | 0.98 (0.78–1.22) | 0.84 |
| Breast, female (170)a | 135/258 | 1.06 (0.85–1.31) | 0.62 |
| Female genital organs (171–176)a | 78/177 | 0.89 (0.68–1.16) | 0.37 |
| Testis (178) | |||
| All male relativesb | 69/26 | 5.97 (3.74–9.53) | <0.0001 |
| Cases ≤ 35 years of ageb,c | 43/15 | 6.24 (3.41–11.43) | <0.0001 |
| Cases > 35 years of ageb,c | 26/11 | 5.59 (2.68–11.67) | <0.0001 |
| Relatives ≤ 35 years of ageb,d | 44/14 | 7.18 (3.86–13.36) | <0.0001 |
| Relatives > 35 years of ageb,d | 25/12 | 4.57 (2.24–9.32) | <0.0001 |
| Fatherse | 30/13 | 4.63 (2.41–8.87) | <0.0001 |
| Brothersf | 32/10 | 8.30 (3.81–18.10) | <0.0001 |
| Sonsg | 7/3 | 5.23 (1.35–20.26) | 0.02 |
| Other male genital organs (177,179) | 47/101 | 0.95 (0.67–1.34) | 0.75 |
| Kidney (180) | 17/49 | 0.70 (0.40–1.22) | 0.20 |
| Bladder (181) | 52/108 | 0.98 (0.70–1.37) | 0.91 |
| Melanoma of skin (190) | 36/70 | 1.06 (0.70–1.59) | 0.79 |
| Brain and nervous system (193) | 35/57 | 1.30 (0.84–1.99) | 0.24 |
| Connective tissue (197) | 8/14 | 1.15 (0.48–2.74) | 0.75 |
| Non-Hodgkin lymphoma (200,202) | 35/41 | 1.73 (1.10–2.73) | 0.02 |
| Hodgkin’s disease (201) | 6/19 | 0.65 (0.26–1.64) | 0.37 |
| Multiple myeloma (203) | 11/15 | 1.51 (0.69–3.30) | 0.30 |
| Leukemia (204) | 23/42 | 1.10 (0.66–1.83) | 0.72 |
| All cancer | 814/1513 | 1.13 (1.02–1.26) | 0.02 |
| All cancer excl. testis | 766/1492 | 1.06 (0.95–1.17) | 0.30 |
nca: number of cases with the particular cancer in the family
nco: number of controls with the particular cancer in the family
p: P-value
RR: rate ratio
95 % CI: 95 percent confidence interval
All analyses were adjusted for number of known first-degree relatives, unless otherwise stated. The numbers in the table do not add up, because cancer in relatives was only noted as ‘yes’ (for one or more cases) or ‘no’.
adjusted for number of known female first-degree relatives
adjusted for number of known male first-degree relatives
age at diagnosis for cases and the matched controls
age at diagnosis in relatives
adjusted for identification of father or not
adjusted for number of known brothers
adjusted for number of known sons
Further, the results showed a RR of 1.73 (95% CI: 1.10–2.73) for testicular cancer in association with NHL and a borderline statistically significant RR of 1.83 (95% CI: 0.94–3.58) in association with oesophageal cancer in first-degree relatives compared with no such familial cancer. The results showed no statistically significant association for any other type of cancer. For all cancers combined the risk of testicular cancer was slightly increased; however, this association disappeared when testicular cancer was not included in the total number of cancers. A further breakdown of this category by type of relative showed no statistically significant associations, and results were similar across type of relative (results not shown).
Separate risk estimates for seminoma and nonseminoma testicular cancer are shown in table 4. Cancer of the brain and nervous system in first-degree relatives increased the risk of nonseminoma testicular cancer with 91%, and a borderline significant association between NHL in relatives and risk for seminoma testicular cancer was observed. In general, however, the separate analyses for seminoma and non-seminama testicular cancer showed similar associations, although these estimates were based on relatively few cases and with wide confidence intervals.
Table 4.
Rate ratio (RR) of seminoma and non-seminoma testicular cancer according to site-specific familial cancer in first-degree relatives.
| Cancer in first-degree relative (ICD-7) | seminoma testicular cancer
|
non-seminoma testicular cancer
|
||||
|---|---|---|---|---|---|---|
| nca/nco | RR (95% CI) | p | nca/nco | RR (95% CI) | p | |
| Buccal cavity and Pharynx (140–148) | 9/23 | 0.81 (0.38–1.76) | 0.60 | 11/17 | 1.28 (0.60–2.75) | 0.52 |
| Oesophagus (150) | 12/13 | 1.87 (0.83–4.20) | 0.13 | 5/6 | 1.77 (0.54–5.82) | 0.35 |
| Stomach (151) | 12/22 | 1.11 (0.54–2.28) | 0.77 | 7/13 | 1.07 (0.42–2.68) | 0.89 |
| Colorectal cancer (153–154) | 55/117 | 0.95 (0.69–1.33) | 0.78 | 36/72 | 1.00(0.66–1.51) | 0.99 |
| Liver (155.0) | 2/5 | 0.79 (0.15–4.10) | 0.78 | 5/6 | 1.66 (0.51–5.47) | 0.40 |
| Gallbladder (155.1) | 2/4 | 1.09 (0.20–5.96) | 0.93 | 1/4 | 0.54(0.06–4.89) | 0.59 |
| Pancreas (157) | 19/28 | 1.37 (0.76–2.46) | 0.29 | 12/17 | 1.38 (0.66–2.90) | 0.39 |
| Larynx (161) | 3/15 | 0.41 (0.12–1.40) | 0.15 | 4/13 | 0.59(0.19–1.83) | 0.36 |
| Lung (162.0,1) | 68/141 | 0.99 (0.74–1.33) | 0.94 | 52/108 | 0.96 (0.68–1.35) | 0.80 |
| Breast, female (170)a | 74/147 | 1.02 (0.77–1.36) | 0.89 | 61/111 | 1.10(0.80–1.52) | 0.55 |
| Female genital organs (171–176)a | 46/106 | 0.87 (0.61–1.24) | 0.45 | 32/71 | 0.90 (0.59–1.37) | 0.63 |
| Testis (178) | ||||||
| All male relativesb | 39/13 | 6.43 (3.42–12.08) | <0.0001 | 30/13 | 5.40(2.69–10.83) | <0.0001 |
| Fathersc | 16/6 | 5.36 (2.10–13.70) | 0.0005 | 14/7 | 4.00(1.62–9.92) | 0.003 |
| Brothersd | 17/5 | 7.54 (2.77–20.54) | <0.0001 | 15/5 | 9.58 (2.75–33.41) | 0.0004 |
| Sonse | 6/2 | 6.44 (1.30–31.93) | 0.02 | 1/1 | 2.62 (0.16–42.15) | 0.50 |
| Other male genital organs (177,179)b | 27/65 | 0.85 (0.54–1.33) | 0.47 | 20/36 | 1.12 (0.64–1.94) | 0.70 |
| Kidney (180) | 10/30 | 0.69 (0.33–1.41) | 0.30 | 7/19 | 0.71 (0.29–1.73) | 0.45 |
| Bladder (181) | 31/55 | 1.16 (0.74–1.82) | 0.51 | 21/53 | 0.79 (0.48–1.33) | 0.38 |
| Melanoma of skin (190) | 19/39 | 0.99 (0.57–1.73) | 0.98 | 17/31 | 1.14 (0.63–2.08) | 0.67 |
| Brain and nervous system (193) | 15/34 | 0.92(0.50–1.70) | 0.80 | 20/23 | 1.91 (1.01–3.58) | 0.05 |
| Connective tissue (197) | 3/8 | 0.78(0.21–2.93) | 0.71 | 5/6 | 1.61 (0.49–5.30) | 0.43 |
| Non-Hodgkin lymphoma (200,202) | 21/24 | 1.76 (0.97–3.18) | 0.06 | 14/17 | 1.70 (0.84–3.46) | 0.14 |
| Hodgkin’s disease (201) | 4/9 | 0.90 (0.28–2.92) | 0.85 | 2/10 | 0.43 (0.09–1.97) | 0.28 |
| Multiple myeloma (203) | 7/10 | 1.45 (0.55–3.83) | 0.45 | 4/5 | 1.62 (0.43–6.06) | 0.47 |
| Leukemia (204) | 14/30 | 0.96 (0.51–1.82) | 0.90 | 9/12 | 1.40 (0.59–3.33) | 0.45 |
nca: number of cases with the particular cancer in the family
nco: number of controls with the particular cancer in the family
p: P-value
RR: rate ratio
95 % CI: 95 percent confidence interval
All analyses were adjusted for number of known first-degree relatives, unless otherwise stated
The numbers in the table do not add up, because cancer in relatives was only noted as ‘yes’ (for one or more cases) or ‘no’.
adjusted for number of known female first-degree relatives
adjusted for number of known male first-degree relatives
adjusted for identification of father or not
adjusted for number of known brothers
adjusted for number of known sons
Table 5 shows risk of seminoma and nonseminoma testicular cancer in association with seminoma and nonseminoma testicular cancer in fathers, brothers and sons, respectively. Overall, associations were similar across the histologic subtypes and groups of male relatives, however each group counted only few cases and some too few to obtain an estimate.
Table 5.
Rate ratio (RR) of seminoma and non-seminoma testicular cancer according to familial testicular cancer by type of testicular cancer and type of relative.
| seminoma testicular cancer
|
non-seminoma testicular cancer
|
|||||
|---|---|---|---|---|---|---|
| Cancer Site (ICD-7) | nca/nco | RR (95% CI) | p | nca/nco | RR (95% CI) | p |
| All Seminomasa | 17/4 | 9.10(3.05–27.13) | <0.0001 | 20/6 | 9.63(3.27–28.37) | <0.0001 |
| Seminoma, fatherb | 8/3 | 5.35 (1.42–20.18) | 0.01 | 9/4 | 4.51(1.39–14.63) | 0.01 |
| Seminoma, brotherc | 5/0 | - - | - | 10/2e | - - | - |
| Seminoma, sond | 4/1 | 8.73(0.97–78.26) | 0.05 | 1/0 | - - | - |
| All Non-seminomaa | 22/9 | 5.23 (2.40–11.36) | <0.0001 | 10/7 | 2.97 (1.13–7.82) | 0.03 |
| Non-seminoma, fatherb | 8/3 | 5.33(1.42–20.08) | 0.01 | 5/3 | 3.33(0.80–13.92) | 0.10 |
| Non-seminoma, brotherc | 12/5 | 5.27(1.85–15.02) | 0.002 | 5/3 | 3.48(0.83–14.58) | 0.09 |
| Non-seminoma, sond | 2/1 | 4.18 (0.38–46.06) | 0.24 | 0/1 | - - | - |
nca: number of cases with the particular cancer in the family
nco: number of controls with the particular cancer in the family
p: P-value
RR: rate ratio
95 % CI: 95 percent confidence interval
adjusted for number of known male first-degree relatives
adjusted for identification of father or not
adjusted for number of known brothers
adjusted for number of known sons
both controls with a brother having seminoma testicular cancer were matched with a case also having an affected brother, consequently no estimates could be obtained.
Discussion
This population-based case control study found increased risk of testicular cancer in association with testicular cancer among first-degree relatives. Similar associations were found when risk of histological subtypes of testicular cancer was evaluated. Furthermore, our findings suggested an association between testicular cancer and familial NHL and oesophageal cancer.
Cases were identified in the virtually complete, high-quality population-based Danish Cancer Registry 21, thus the study has very reliable case ascertainment. Furthermore, the the Danish Civil Registration System provided an ideal frame for control selection as well as identification of first-degree relatives of cases and controls.
The Danish Civil Registration System has some historical limitations, as the identity of parents can (only) be found for persons born since the beginning of the 1950s, and the information is first regarded complete for persons born in 1960 and onwards 6;9. Consequently, our database had some gaps in information on first-degree relatives among the older cases and controls. To address this issue we adjusted all analyses for number of known first-degree relatives. Furthermore, as testicular cancer is rare and the specificity of the registration of the cancer is very high, the misclassification will only induce a very small downward bias.
Previous studies consistently reported familial aggregation of testicular cancer in agreement with our findings. Most studies found a 4-fold increased risk of testicular cancer, when the father had testicular cancer, and estimates ranged from 8–13-fold increased risk for brothers and 2–5-fold increased risk when a son had testicular cancer8;9;11;13;14. Thus, our results agree with these previous findings. Higher risk among brothers than fathers was also consistently reported; however the difference between fathers and brothers that we observed was less pronounced and not statistically significant.
Early onset of disease is considered one of the established factors that characterize genetic predisposition for cancer in families 22. In agreement with such genetic predisposition, we found that cases with testicular cancer in the family were on average four years younger at time of diagnosis than cases with no history of familial testicular cancer, which is in agreement with a an age difference of 3.5 years previously reported 23. Two other studies did not observe such age differences; these studies were, however, based on few cases with testicular cancer in the family 9;24. Further, risk of testicular cancer in association with testicular cancer in first-degree relatives tended to be higher if relatives were diagnosed at young ages compared to relatives, who were older at time of diagnosis, but the difference was not statistically significant.
We found that a family history of NLH was associated with a73% higher risk of testicular cancer. Results from a Swedish study population similarly suggested a slightly increased risk of testicular cancer in association with familial NHL 10;11, which to our knowledge is the only previous investigation of this association. Furthermore, we observed a borderline statistically significant association between familial oesophageal cancer and testicular cancer, which is in agreement with prior studies 10;17. A similar tendency was found in a previous Danish study of cases from an earlier time period, however statistically insignificant and based on very few cases 9. Others have not been able to detect such association 16. In addition, we found that cancer of the brain and nervous system in first-degree relatives increased the risk of nonseminoma testicular cancer. Only few others have investigated and found such association, however Dong et al found a similar association between seminoma testicular cancer and cancer of the nervous system 8, and Hemminki et al between testicular cancer as a whole and paternal nervous system cancer 10. Thus little evidence exists to support our finding of associations between testicular cancer and NHL and cancer of the oesophagus and brain and nervous system, and it cannot be ruled out that these results are chance findings, due to multiple testing.
Explanations for familial aggregation of testicular (and other cancers) involve genetic and environmental components, however the mutual importance of each of these is difficult to determine. Some interpret the consistent findings of familial aggregation of testicular cancer as evidence for a genetic predisposition, while others emphasize environmental factors.
It has been pointed out that the familial risk of testicular cancer is much higher than the familial risk of other cancers, which suggest the involvement of some genetic mechanisms, as it is considered unlikely that environmental factors alone can explain such strong familial aggregation 14;23;24. Consistent with a genetic element in testicular carcinogenesis, a higher risk of testicular cancer was found in monozygotic twins of men with testicular cancer compared to dizygotic twins 15. On the other hand, specific heritable genes have not been identified 22, and the fast increase in the incidence of testicular cancer observed in the industrialized countries indicates that environmental factors are of primary importance, because genetic factors change at a much slower rate. Migrant studies of testicular cancer also argue for a substantial environmental influence, because individuals who migrate from low-risk areas to areas of high risk adopt the rate in the host country within few generations6.
Twins were found to have larger risk of testicular cancer than the general population, which could be due to higher levels of maternal hormones associated with twin-pregnancies compared to singleton-pregnancies 15;25, which is referred to as the maternal-hormone aetiology hypothesis. Such shared in utero environment has also been used to explain the higher risk of testicular cancer among brothers than fathers of cases of testicular cancer 8;10. One study found higher risk among brothers close in age compared to those further apart, indicating that shared childhood environment may also play an important role 10. In summary, is seems plausible that multiple factors causes familial clustering of testicular cancer. The existing literature offers reliable arguments for both genetic and environmental influence, but these have not convincingly been separated into more specific contributions, suggesting that the aggregation is perhaps rather a product of complex gene-environment interactions. Except for a few known risk factors the aetiology of testicular cancer still remains unclear.
This large population-based study confirmed familial aggregation of testicular cancer and suggested an association between testicular cancer and familial NHL and oesophageal cancer.
Novelty and impact.
The study is among the largest case-control studies of familial cancer and risk of testicular cancer, and it is an important contribution to the limited number of population-based studies that examine association between testicular cancer and familial cancer other that testicular cancer.
Acknowledgments
This research was funded by the US National Institute of Environmental Health Sciences, R21 ES015501, the Danish Cancer Society and the Danish Agency for Science, Technology and Innovation. The perspectives are those of the authors and do not necessarily represent the official position of the funding agencies.
Abbrevations
- RR
Rate Ratio
- CI
Confidence Interval
- ICD-O
International Classification of Diseases for Oncology
- ICD-7
7th Revision of the International Classification of Diseases
- PIN
Personal Identification Number
- NHL
Non-Hodgkin lymphoma
References
- 1.Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000. The global picture. Eur J Cancer. 2001;37 (Suppl 8):S4–66. doi: 10.1016/s0959-8049(01)00267-2. [DOI] [PubMed] [Google Scholar]
- 2.Bosl GJ, Motzer RJ. Testicular germ-cell cancer. N Engl J Med. 1997;337:242–53. doi: 10.1056/NEJM199707243370406. [DOI] [PubMed] [Google Scholar]
- 3.Engholm G, Ferlay J, Christensen N, Bray F, Gjerstorff ML, Klint Å, Køtlum JE, Ólafsdóttir E, Pukkala E, Storm HH. NORDCAN: Cancer Incidence, Mortality, Prevalence and Prediction in the Nordic Countries, Version 3.6. Association of Nordic Cancer Registries. Danish Cancer Society; 2010. ( http://www.ancr.nu) [Google Scholar]
- 4.Huyghe E, Matsuda T, Thonneau P. Increasing incidence of testicular cancer worldwide: a review. J Urol. 2003;170:5–11. doi: 10.1097/01.ju.0000053866.68623.da. [DOI] [PubMed] [Google Scholar]
- 5.Richiardi L, Pettersson A, Akre O. Genetic and environmental risk factors for testicular cancer. Int J Androl. 2007;30:230–40. doi: 10.1111/j.1365-2605.2007.00760.x. [DOI] [PubMed] [Google Scholar]
- 6.Myrup C, Westergaard T, Schnack T, Oudin A, Ritz C, Wohlfahrt J, Melbye M. Testicular cancer risk in first-and second-generation immigrants to Denmark. J Natl Cancer Inst. 2008;100:41–7. doi: 10.1093/jnci/djm276. [DOI] [PubMed] [Google Scholar]
- 7.Gundy S, Babosa M, Baki M, Bodrogi I. Increased predisposition to cancer in brothers and offspring of testicular tumor patients. Pathol Oncol Res. 2004;10:197–203. doi: 10.1007/BF03033760. [DOI] [PubMed] [Google Scholar]
- 8.Dong C, Lonnstedt I, Hemminki K. Familial testicular cancer and second primary cancers in testicular cancer patients by histological type. Eur J Cancer. 2001;37:1878–85. doi: 10.1016/s0959-8049(01)00172-1. [DOI] [PubMed] [Google Scholar]
- 9.Westergaard T, Olsen JH, Frisch M, Kroman N, Nielsen JW, Melbye M. Cancer risk in fathers and brothers of testicular cancer patients in Denmark. A population-based study. Int J Cancer. 1996;66:627–31. doi: 10.1002/(SICI)1097-0215(19960529)66:5<627::AID-IJC8>3.0.CO;2-V. [DOI] [PubMed] [Google Scholar]
- 10.Hemminki K, Li X. Familial risk in testicular cancer as a clue to a heritable and environmental aetiology. Br J Cancer. 2004;90:1765–70. doi: 10.1038/sj.bjc.6601714. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hemminki K, Chen B. Familial risks in testicular cancer as aetiological clues. Int J Androl. 2006;29:205–10. doi: 10.1111/j.1365-2605.2005.00599.x. [DOI] [PubMed] [Google Scholar]
- 12.Bromen K, Stang A, Baumgardt-Elms C, Stegmaier C, Ahrens W, Metz KA, Jockel KH. Testicular, other genital, and breast cancers in first-degree relatives of testicular cancer patients and controls. Cancer Epidemiol Biomarkers Prev. 2004;13:1316–24. [PubMed] [Google Scholar]
- 13.Sonneveld DJ, Sleijfer DT, Schrafford KH, Sijmons RH, van der Graaf WT, Sluiter WJ, Hoekstra HJ. Familial testicular cancer in a single-centre population. Eur J Cancer. 1999;35:1368–73. doi: 10.1016/s0959-8049(99)00140-9. [DOI] [PubMed] [Google Scholar]
- 14.Heimdal K, Olsson H, Tretli S, Flodgren P, Borresen AL, Fossa SD. Familial testicular cancer in Norway and southern Sweden. Br J Cancer. 1996;73:964–9. doi: 10.1038/bjc.1996.173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Swerdlow AJ, De Stavola BL, Swanwick MA, Maconochie NE. Risks of breast and testicular cancers in young adult twins in England and Wales: evidence on prenatal and genetic aetiology. Lancet. 1997;350:1723–8. doi: 10.1016/s0140-6736(97)05526-8. [DOI] [PubMed] [Google Scholar]
- 16.Spermon JR, Witjes JA, Nap M, Kiemeney LA. Cancer incidence in relatives of patients with testicular cancer in the eastern part of The Netherlands. Urology. 2001;57:747–52. doi: 10.1016/s0090-4295(00)01058-x. [DOI] [PubMed] [Google Scholar]
- 17.Chia VM, Li Y, Goldin LR, Graubard BI, Greene MH, Korde L, Rubertone MV, Erickson RL, McGlynn KA. Risk of cancer in first-and second-degree relatives of testicular germ cell tumor cases and controls. Int J Cancer. 2009;124:952–7. doi: 10.1002/ijc.23971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Heimdal K, Olsson H, Tretli S, Flodgren P, Borresen AL, Fossa SD. Risk of cancer in relatives of testicular cancer patients. Br J Cancer. 1996;73:970–3. doi: 10.1038/bjc.1996.174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Walschaerts M, Muller A, Auger J, Bujan L, Guerin JF, Le LD, Clavert A, Spira A, Jouannet P, Thonneau P. Environmental, occupational and familial risks for testicular cancer: a hospital-based case-control study. Int J Androl. 2007;30:222–9. doi: 10.1111/j.1365-2605.2007.00805.x. [DOI] [PubMed] [Google Scholar]
- 20.Boyd HA, Poulsen G, Wohlfahrt J, Murray JC, Feenstra B, Melbye M. Maternal contributions to preterm delivery. Am J Epidemiol. 2009;170:1358–64. doi: 10.1093/aje/kwp324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Storm HH, Michelsen EV, Clemmensen IH, Pihl J. The Danish Cancer Registry -history, content, quality and use. Dan Med Bull. 1997;44:535–9. [PubMed] [Google Scholar]
- 22.Lindor NM, McMaster ML, Lindor CJ, Greene MH. J Natl Cancer Inst Monogr. 2. 2008. Concise handbook of familial cancer susceptibility syndromes; pp. 1–93. [DOI] [PubMed] [Google Scholar]
- 23.Forman D, Oliver RT, Brett AR, Marsh SG, Moses JH, Bodmer JG, Chilvers CE, Pike MC. Familial testicular cancer: a report of the UK family register, estimation of risk and an HLA class 1 sib-pair analysis. Br J Cancer. 1992;65:255–62. doi: 10.1038/bjc.1992.51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Dieckmann KP, Pichlmeier U. The prevalence of familial testicular cancer: an analysis of two patient populations and a review of the literature. Cancer. 1997;80:1954–60. [PubMed] [Google Scholar]
- 25.Braun MM, Ahlbom A, Floderus B, Brinton LA, Hoover RN. Effect of twinship on incidence of cancer of the testis, breast, and other sites (Sweden) Cancer Causes Control. 1995;6:519–24. doi: 10.1007/BF00054160. [DOI] [PubMed] [Google Scholar]