SEARCH

SEARCH BY CITATION

Keywords:

  • children;
  • germ cell tumors;
  • cigarette smoke;
  • alcohol;
  • case–control study

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

BACKGROUND

The etiology of childhood germ cell tumors (GCT) is not well understood. The Children's Oncology Group conducted the largest case–control study of childhood GCT to investigate whether parental exposures to smoking and alcohol contributed to the disease.

METHODS

Cases included 274 children with GCT diagnosed between January 1, 1993 and December 31, 2001 who were age < 15 years. Controls (n = 421) were selected by random digit dialing and were frequency matched based on gender, age (±1 year), and geographic area. Exposure information was collected from subjects' parents using independent telephone interviews and self-administrated questionnaires.

RESULTS

No association was found between parental smoking or drinking alcohol and risk of childhood GCT (for smoking: odds ratio [OR] = 1.0, 95% confidence interval [95% CI], 0.8–1.3 and OR = 1.2, 95% CI, 0.9–1.5, for mothers and fathers, respectively; for drinking: OR = 0.9, 95% CI, 0.7–1.2 and OR = 1.0, 95% CI, 0.8–1.3, for mothers and fathers, respectively). No significant trend was observed for length of maternal exposure to passive smoking during the index pregnancy and GCT risk (for total subject: P = 0.77; boys: P = 0.52; girls: P = 0.93).

CONCLUSIONS

The authors found no evidence that childhood GCT was related to prenatal exposure to parental cigarette smoking, alcohol drinking, and maternal passive smoking. Cancer 2005. © 2005 American Cancer Society.

Germ cell tumors (GCT) originate from pluripotent germ cells of the embryonic yolk sac and are located primarily in the testes and ovaries. However, tumors also can arise in extragonadal sites, including the sacrococcygeal region, anterior mediastinum, neck, retroperitoneum, and brain.1, 2 Tumors of germ cell origin constitute a relatively small but important group of childhood tumors. Descriptive epidemiologic data have shown that the incidence rate of childhood GCT varies between countries and increased overall during the period 1962–1995.1–4 In the United States, approximately 5 children per million age < 15 years were affected with GCT.5 Because > 50% of children were age < 4 years when GCT was diagnosed, it has been speculated that prenatal and perinatal factors may play an important role in its etiology.1, 4, 6 To our knowledge, only 5 studies since 1982 have attempted to determine GCT risk factors in children. Two were conducted in the United States and included 73 and 105 GCT cases.7, 8 Two British studies included 41 cases and 87 cases, respectively,9, 10 and 1 Mexican study included 21 cases.11 Suggested risk factors for childhood GCT include maternal exogenous hormone use and hyperemesis during the index pregnancy, congenital malformations, radiation exposure (preconception, prenatal), infection during pregnancy, parental occupational exposures, and childhood virus (mumps, herpes virus) infections.7–11 However, all these risk factors were based on reports from a few small-scale studies.

Several lines of evidence support the potential role of estrogen in the pathogenesis of GCT.12–17 Smoking and alcohol consumption have been associated with alterations in endogenous estrogen levels.18–21 Parental smoking and drinking also have been linked to the risk of other childhood cancers.22–25 The Children's Oncology Group (COG) recently completed a case–control study of GCT in children age < 15 years, which to our knowledge is the largest epidemiologic study published to date. A range of prenatal, in utero, and postnatal exposures were evaluated. We report a comprehensive evaluation of the association between parental cigarette smoking and alcohol consumption and maternal passive smoking during index pregnancy with the risk of childhood GCT.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Cases were comprised of children age < 15 years with diagnoses of GCT (seminoma (dysgerminoma/seminoma/germinoma), embryonal carcinoma, yolk sac tumor, choriocarcinoma, immature teratoma, and mixed GCT) in all anatomic sites except encephalocoele. Encephalocoele GCTs are extremely rare and accurate diagnosis is difficult. Cases were recruited through COG. Eligible cases had to have newly diagnosed GCT between January 1, 1993 and December 31, 2001, and be registered with the COG data and statistics center. There had to be a telephone in the patient's residence, and the patient's biologic mother had to speak English and be available for interview. Permission to interview was obtained first from the child's physician, then from the parents. A total of 496 potentially eligible childhood GCT cases were registered with COG for the protocol during the study period. Of these, 344 cases were eligible. Telephone interviews were completed successfully with mothers of 278 cases (80.8%). Eight cases were deceased at the time of interview. Among nonparticipants, there were 44 refusals (12.8%) and 20 subjects had nonworking phone numbers (5.8%). Two subjects could not be interviewed because of the inability to schedule an interview. Table 1 shows the comparison between cases who participated and did not participate in the study. Participants and nonparticipants had a similar age distribution, but there were more boys and nonwhites among nonparticipants. Nonparticipants and participants also had different pathologic and anatomic disease presentations.

Table 1. Comparison between Cases Who Participated and Did not Participate in the Study
VariablesNo. of nonpartcipants (%)No. of participants (%)P
Gender    0.008
 Boys26(39.4)83(29.9) 
 Girls38(57.6)195(70.1) 
 Missing2(3.0)   
Child race    < 0.001
 White35(53.0)204(73.4) 
 Nonwhite24(36.4)74(26.6) 
 Missing7(10.6)   
Age at diagnosis (yrs)    0.18
 017(25.8)57(20.5) 
 1–218(27.3)77(27.7) 
 3–1010(15.2)67(24.1) 
 11–1520(30.3)77(27.7) 
 Missing1(1.5)   
Pathologic type    < 0.001
 Seminoma (seminoma, dysgerminoma, germinoma)4(6.1)45(16.2) 
 Yolk sac tumor (endodermal sinus tumor)21(31.8)125(45.0) 
 Other nonseminomas (embryonal carcinoma, Choirocarcinoma, and polyembryoma3(4.5)25(9.0) 
 Teratoma (malignant teratoma and immature teratoma)18(27.3)71(25.5) 
 Other (mixed germ cell tumor components and malignant tumor cells)2(3.0)10(3.6) 
 Not specified15(22.7)2(0.7) 
 Missing3(4.5)   
Anatomic locations    < 0.001
 Ovary20(30.3)98(35.3) 
 Testis6(9.1)46(16.5) 
 Extragonadal19(28.8)120(43.2) 
 Metastatic1(1.5)7(2.5) 
 Missing20(30.3)7(2.5) 

Controls were selected by random digit dialing and were frequency matched to cases based on the child's gender, date of birth (year of birth ± 1), and geographic area. The match ratio was approximately 1:2 for males and 1:1 for females. The different matching ratio by gender was designed to maximize study power because the incidence of GCT is much lower among boys than among girls. Before the study was implemented, a frequency matrix for control selection (by birth calendar year and gender) was generated based on the age and gender distribution of GCT cases using the information obtained from the Children's Cancer Group (now part of the COG) database. The random digit dialing used cases' telephone area code and exchange as the primary sampling unit (seed) and randomly modified the last four digits. A total of 17,292 phone numbers were selected randomly. Of these, 634 were from families with ≥ 1 eligible child. Telephone interviews with mothers were completed successfully for 422 of 634 potential controls (66.6%). Each index child's mother was given a self-administered questionnaire and telephone interview to collect information about exposure before, during, and after the pregnancy of the index child. The father was interviewed when available. Otherwise, the mother provided a surrogate interview. Interviews were conducted between January 1, 1997 and December 31, 2001.

In addition to questions about cigarette smoking and alcohol drinking, mothers' and fathers' questionnaires included questions about demographics, medication use, X-ray exposure, personal habits, household exposures, lifetime occupation history, family medical history, and pesticide exposures of the index child. Parents were asked whether they ever smoked regularly (e.g., 1 cigarette daily for a period of ≥ 3 months) and about their smoking practices 1) during the month before the index pregnancy, 2) during the first, second, and third trimesters, separately, and 3) during nursing. Information regarding the number of cigarettes smoked also was obtained. Mothers and fathers were asked about alcohol consumption during the month before index conception. Mothers were asked to describe alcohol consumption during each trimester of index pregnancy and during nursing. Detailed information was gathered concerning type (e.g., wine, beer, or liquor) and amount (e.g., number of glasses, cans, or ounces) consumed weekly. The total amount consumed was calculated by summing the number of drinks of three types of alcoholic beverages. One drink refers to an 8-oz glass of wine, a 12-oz can of beer, or 1.5 oz of liquor. Information regarding smoking and drinking was inadequate for four cases and one control. The final data set for the current analysis includes 274 cases and 421 controls.

We used the chi-square and Fisher exact tests for categoric comparisons of data. Relative risks were estimated using odds ratios (ORs) and corresponding 95% confidence intervals (95% CIs), calculated by unconditional logistics regression models with adjustments for age, gender, and relevant confounders.26 Confounders were factors that have been associated with GCT and were associated with GCT and the exposures under study in the study population. Tests for trend were performed by treating levels of categoric variables as a continuous variable in the logistic model.27 Analyses were performed considering all tumors together and then by gender. Analyses stratified by child's age also were performed. Statistical analyses were performed using the statistical package SAS (Version 8.0, SAS, Cary, NC). All tests were two-tailed.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Demographic characteristics of the study cases and controls are provided in Table 2. Boys (n = 83) and girls (n = 191) accounted for 30.3% and 69.7%, respectively, of total cases. Approximately 50% of case patients were age ≤ 2 years at diagnosis. Case and control children had similar distributions of birth weight and gestational age, parental age at index pregnancy, and education attainment. Case families tended to have a lower annual income than control families. Parental race distributions were different between cases and controls.

Table 2. Comparison of Demographics and Potential Confounders for Case and Control Subjects
VariablesNo. of cases (%)No. of controls (%)OR (95% CI)a
  • OR: Odds ratio; 95% CI: 95% confidence interval.

  • a

    Odds ratio values were adjusted for children age and gender.

  • b

    P value for chi-square test.

  • c

    P < 0.05.

Child age (yrs)    P = 0.02b
 057(20.8)95(22.6) 
 1–276(27.7)77(18.3) 
 3–1066(24.1)129(30.6) 
 11–1575(27.4)120(28.5) 
Gender    P < 0.001b
 Boys83(30.3)181(43.0) 
 Girls191(69.7)240(57.0) 
Birth weight (g)     
 ≤ 250022(8.0)26(6.2)1.3 (0.8–2.1)
 2501–300051(18.6)73(17.3)1.2 (0.8–1.7)
 3001–350081(29.6)156(37.1)1.0
 3501–400073(26.6)114(27.1)1.2 (0.9–1.6)
 ≥ 400147(17.2)52(12.4)1.5 (1.0–2.1)c
Gestational age (wks)     
 ≤ 3522(8.0)24(5.7)1.2 (0.8–1.9)
 36–3723(8.4)36(8.6)1.0 (0.7–1.6)
 38–41208(75.9)327(77.7)1.0
 ≥ 4221(7.7)34(8.1)1.0 (0.6–1.5)
Annual family income ($)     
 ≤ 20,00084(30.7)83(19.7)1.0
 20,001–30,00059(21.5)109(25.9)0.7 (0.5–1.0)c
 30,001–40,00040(14.6)81(19.2)0.7 (0.5–1.0)c
 40,001–60,00044(16.1)79(18.8)0.7 (0.5–1.0)
 ≥ 60,001 (0.6–1.1)47(17.2)69(16.4)0.8
Maternal age at index pregnancy (yrs)     
 ≤ 2479(28.8)114(27.1)1.0 (0.8–1.4)
 25–29103(37.6)152(36.1)1.0
 30–3457(20.80)110(26.1)0.8 (0.6–1.2)
 ≥ 3535(12.8)45(10.7)1.1 (0.7–1.6)
Maternal race     
 White213(77.7)355(84.3)1.0
 Nonwhite61(22.2)66(15.7)1.3 (1.0–1.8)c
Maternal education     
 ≤ 11 yrs26(9.5)23(5.5)1.0
 High school79(28.8)99(23.5)0.8 (0.5–1.2)
 Technical school83(30.3)141(33.5)0.6 (0.4–1.0)
 College and graduate school86(31.4)158(37.5)0.6 (0.4–1.0)c
Paternal age at index child (yrs)     
 ≤ 2447(18.2)59(15.5)1.1 (0.7–1.7)
 25–2982(31.8)119(31.2)1.0
 30–3465(25.2)131(34.3)0.7 (0.6–1.0)c
 ≥ 3564(24.8)73(19.1)1.1 (0.9–1.3)
Paternal race     
 White200(77.5)322(84.3)1.0
 Nonwhite58(22.5)60(15.7)1.3 (1.0–1.7)c
Paternal education     
 ≤ 11 yrs42(16.3)37(9.7)1.0
 High school71(27.5)115(30.1)0.7 (0.5–1.1)
 Technical school56(21.7)97(25.34)0.7 (0.5–1.0)
 College and graduate school89(34.5)133(34.8)0.8 (0.5–1.1)

Association of childhood GCT with parental cigarette smoking habits is shown in Table 3 and Table 4. A cigarette smoker was someone who smoked ≥ 1 cigarette daily for ≥ 3 months. Parents of cases were slightly more likely to smoke cigarettes at some point during their lifetimes than control parents (44.9% vs. 42.5% and 51.6% vs. 44.7% for mothers and fathers, respectively). However, after adjustment for child age, gender, paternal age at index pregnancy, education, race, and family income, no case–control differences were observed (OR = 1.0, 95% CI, 0.8–1.3 and OR = 1.2, 95% CI, 0.9–1.5, for mothers and fathers, respectively). Similar patterns were observed for girls and boys.

Table 3. Maternal Smoking and Risk of Malignant Germ Cell Tumors in Childrena
VariablesCases (%)Controls (%)ORb95% CI
  • OR: odds ratio; 95% CI: 95% confidence interval.

  • a

    The sample of mothers included 274 cases and 421 controls.

  • b

    All odds ratios were adjusted for children gender, age, maternal education, race, and family income.

Ever smoked ≥ 3 mos    
 Never55.157.51.0 
 Yes44.942.51.00.8–1.3
Ever smoked during perinatal period    
 1 mo before index pregnancy35.229.91.00.8–1.4
 First trimester30.724.61.10.8–1.5
 Second trimester24.520.11.10.7–1.5
 Third trimester22.119.11.00.7–1.5
 Nursing10.711.70.80.5–1.3
No. of cigarettes smoked per day during index pregnancy    
 None69.975.91.0 
 < 1017.114.11.10.7–1.5
 10–197.45.61.20.7–1.9
 ≥ 205.64.41.10.6–2.1
Trend test for P value  0.77 
Years smoking before the index pregnancy    
 Never57.657.91.0 
 ≤ 414.113.21.00.7–1.4
 5–914.514.41.00.7–1.4
 ≥ 1013.714.60.90.6–1.3
Trend test for P value  0.56 
Years smoking after the index pregnancy    
 Never63.768.61.0 
 ≤ 412.27.71.20.8–1.9
 5–910.611.60.90.5–1.3
 ≥ 1013.512.21.00.6–1.5
Trend test for P value  0.75 
Passive smoking during pregnancy and smoked    
 Never38.040.11.00.7–1.4
 Passive smoking only17.217.31.00.6–1.3
 Active smoking only16.118.10.90.7–1.4
Passive and active smoking28.824.51.0 
Hours exposure to passive smoking per day during pregnancy    
 Never45.249.01.00.8–1.5
 ≤ 121.718.81.10.7–1.4
 2–416.516.21.00.7–1.5
 ≥ 516.515.91.0 
Trend test for P value  0.77 
Table 4. Paternal Smoking and Risk of Malignant Germ Cell Tumors in Childrena
VariablesCases (%)Controls (%)ORb95% CI
  • OR: odds ratio; 95% CI: 95% confidence interval.

  • a

    The total number of fathers included 258 cases and 380 controls.

  • b

    The odds ratios were adjusted for children gender, paternal education, race, age at index pregnancy, and family income.

Ever smoked ≥ 3 mos    
 Never48.155.31.0 
 Yes51.944.71.20.9–1.5
Ever smoked during perinatal period    
 Never55.663.41.0 
 1 mo before wife index pregnancy44.436.61.20.9–1.6
 First trimester12.17.51.20.7–2.0
 Second trimester10.15.81.20.6–2.2
 Third trimester9.55.01.30.7–2.3
 The year after the index child was born12.76.31.30.8–2.2
Years smoking before the index pregnancy    
 Never54.465.41.0 
 ≤ 516.215.01.20.8–1.8
 6–1011.06.21.40.9–2.3
 11–1511.07.51.20.8–2.0
 ≥ 167.55.91.30.7–2.2
 Trend test for P value  0.13 
Years smoking after the index pregnancy    
 Never62.073.41.0 
 ≤ 517.09.81.40.9–2.1
 6–1010.07.31.30.8–2.2
 11–156.56.61.20.6–2.1
 ≥ 164.52.81.30.6–2.7
 Trend test P value  0.21 
No. of cigarettes smoked during 1 mo before index pregnancy to the yr the child was born    
 Never53.162.81.00.9–1.7
 < 1 cigarette per day39.831.21.20.7–1.9
 ≥ 1 cigarette per day7.16.01.1 
 Trend test P value  0.28 

Parental cigarette smoking during the month before the index pregnancy, during any trimester of pregnancy, during all three trimesters combined, or during nursing did not appear to alter ORs substantially from the null (Tables 3, 4). Similarly, no change in risk was observed when we analyzed the number of years of smoking during lifetime or number of years of smoking before and after the index birth. For example, ORs associated with paternal smoking for ≥ 16 years before the index pregnancy were 1.3 (95% CI, 0.7–2.2) for all subjects, 1.3 (95% CI, 0.5–3.2) for boys, and 1.1 (95% CI, 0.6–2.3) for girls. Further analyses were performed to examine the effect of maternal passive smoking during the index pregnancy on GCT risk among offspring. The definition of maternal exposure to passive smoke was mother exposed to cigarette smoke from spouse or partner or any other person who lived with her or mother exposed to cigarette smoke from anyone at work during the index pregnancy. No significant association was found (passive smoking: OR = 1.0, 95% CI, 0.7–1.4; active smoking only: OR = 0.9, 95% CI, 0.6–1.3; and active and passive smoking: OR = 1.0, 95% CI, 0.7–1.4). The length of maternal exposure to passive smoking during the index pregnancy also was found to be unrelated to GCT risk (trend tests for total subject: P = 0.77; boys: P = 0.52; and girls: P = 0.93).

Associations between GCT and maternal alcohol drinking are shown in Table 5. A drinker was someone who consumed ≥ 1 alcoholic drink weekly (e.g., beer, wine, or hard liquor) for ≥ 6 months. Slightly more control mothers (59.4%) drank than case mothers (53.9%) from 1 month before pregnancy to nursing, but the adjusted ORs did not depart appreciably from the null (OR = 0.9, 95% CI, 0.7–1.2). Because few mothers drank alcohol during the second and third trimesters and during nursing, we did not analyze the data by amount consumed during these periods. The drink rate of control fathers (72.3%) did not differ statistically from that of case fathers (69.5%) during the period 1 month before the index pregnancy to nursing. No aspect of drinking was associated with GCT risk. Analyses also were carried out for girls and boys, separately, and no significant associations were found (data not shown).

Table 5. Maternal Drinking and Risk of Malignant Germ Cell Tumors in Childrena
VariablesCases (%)Controls (%)ORb95% CI
  • OR: odds ratio; 95% CI: 95% confidence interval.

  • a

    Mothers included 274 cases and 420 controls.

  • b

    All odds ratios were adjusted for children gender, age, maternal education, race, and family income.

Ever drank ≥ 6 mos    
 Never66.661.01.0 
 Yes33.439.10.90.7–1.2
Ever drank during 1 mo before pregnancy to nursing    
 Never46.140.61.0 
 Yes53.959.40.90.7–1.2
Type of alcoholic beverages drank during 1 mo before pregnancy    
 Wine40.950.20.80.6–1.1
 Beer40.042.51.00.7–1.3
 Liquor28.127.41.10.7–1.5
All types of alcohol combined    
 ≤ 7 drinks/wk44.849.10.90.7–1.2
 ≥ 8 drinks/wk7.79.00.90.6–1.4
Trend test for P value  0.49 
Drank first trimester    
 Wine22.627.40.90.6–1.3
 Beer21.720.21.10.7–1.6
 Liquor17.512.51.20.8–1.9
All types of alcohol combined    
 ≤ 2 drinks/wk21.625.21.00.7–1.4
 ≥ 3 drinks/wk11.911.21.00.7–1.6
Trend test for P value  0.96 
Alcohol drank second trimester    
 Never83.780.21.0 
 Yes16.319.80.90.6–1.4
 Wine11.515.40.80.5–1.4
 Beer7.59.90.70.4–1.4
 Liquor4.71.91.40.6–3.3
Alcohol drank third trimester    
 Never82.579.01.0 
 Yes17.521.00.90.6–1.4
 Wine14.015.91.00.6–1.6
 Beer5.49.90.70.3–1.4
 Liquor4.71.91.50.6–3.4
Alcohol drank during nursing    
Never79.974.81.0 
Yes20.125.20.90.6–1.3
Wine17.519.81.00.6–1.6
Beer16.924.10.80.5–1.3
Liquor6.13.11.50.7–3.1

Additional analyses stratified by age at diagnosis (≤ 2 years vs. > 2 years old), histologic type of tumors (mainly dysgerminoma, yolk sac tumor, and teratoma), and anatomic sites (mainly ovary, testis, and extragonadal) showed no significant association between parental smoking and alcohol consumption and childhood GCT risk (data not shown).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Previous studies suggested that GCT in children may be caused by exposures early in life or in utero,6 and most likely before birth.28 Our focus was to investigate the effects of parental smoking and drinking before pregnancy, during the index pregnancy, and during the nursing period on childhood GCT risk. In what to our knowledge is the largest case–control study of childhood GCT to date, we found no such association. Our findings were consistent with two other investigations on this topic.8, 10 A study from the United Kingdom of 41 children with GCT and 82 controls demonstrated no effect of maternal smoking and alcohol intake during the index pregnancy on GCT risk. A study of 105 GCT cases and 639 controls in the United States found that the number of cigarettes smoked and the time period of cigarette smoking were not related to childhood GCT risk.8

Tobacco contains several mutagenic and carcinogenic compounds that can cross the placenta.29–31 Smoking also has been shown to cause human germ cell mutations during spermatogenesis.32 Conversely, smoking also has been associated with low estrogen levels, whereas exposure to high estrogen levels in utero has been suggested to increase GCT risk.7, 17–19 Previous investigations on the association between smoking and a subsequent risk of childhood cancers have generated positive and negative results.22, 23, 33–36 The negative finding from our study, as well as an earlier study37 on GCT, might be a result of the finding that the carcinogenic effect of smoking is offset by its antiestrogenic effect.18, 19 Whereas alcohol and its metabolites have been shown to be teratogenic, mutagenic, and carcinogenic,38–40 the relation between alcohol consumption and childhood cancers has not always been consistent across studies.41, 42 Heavy drinking also has been linked to a high level of estrogen.20, 21 However, 53.9% of case mothers and 59.4% of control mothers reported alcohol consumption during the period 1 month before pregnancy to nursing, and 7.7% of case mothers and 9.0% of control mothers reported consumption of > 8 drinks per week. We observed no association of GCT with parental alcohol consumption, similar to an earlier study conducted by the Children's Cancer Group and the studies by Shu et al.8 and Johnson et al.10 In our study, only 4 case mothers and 1 control mother consumed > 14 cups per week during their first trimesters of pregnancy. Parents of the study subjects were the source by which data were collected in the current study. In case–control studies using self-reported data, it is conceivable that results may be biased because of exposure misclassifications. Because we are not reporting positive findings, we must consider whether misclassification biased our results toward the null. This can result with nondifferential misclassification of exposure, in which the information obtained from cases and controls is equally inaccurate. Bias toward the null also would arise if case parents were more inclined than control parents to underreport the amount of smoking and drinking during the index pregnancy. It is interesting to note that participant rates for control parents, particularly for control fathers, were lower than case parents. If nonparticipation were related to smoking and drinking, the results of the study would be biased. Although we could not verify whether participation was related to lifestyle factors, we found that the rates of smoking and drinking among control parents were similar to those reported by earlier studies.8, 10, 36

To our knowledge the current study represents the most recent available information regarding childhood GCT in the United States and the largest epidemiologic study ever conducted for childhood GCT. Our study has sufficient statistical power (80%) to detect a moderate effect (OR > 1.6) of parental smoking and drinking on GCT. Exposure information was comprehensive and covered all relevant periods. However, we did not find an association between parental smoking and drinking and GCT, as suggested by several previous studies. GCTs have a number of histologic subtypes. If different histologic types have different etiologies, this would restrict the ability of a single study to detect significant case–control differences. We were only able to analyze the data by histologic type of major GCTs (i.e., pure dysgerminoma, yolk sac tumor, and teratoma) given the small sample of other tumor types. We did not find a strong indication that parental smoking and drinking were related to any specific type of GCT. The small sample involved in these stratified analyses limited our ability to detect a moderate risk. Considering the inherent difficulty of studying rare tumors, our study remains a valuable contribution to the limited body of literature regarding GCT etiology. Future studies with sufficient power and central pathology review are required to better understand any possible association between peripartum exposures and development of childhood tumors.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The authors thank the following institutions, which participated in this study: University of South Alabama, Mobile, AL; University of Arkansas, Little Rock, AR; Phoenix Children's Hospital, Phoenix, AZ; British Columbia's Children's Hospital, Vancouver BC, Canada; Children's Hospital of Los Angeles, Los Angeles, CA; Children's Hospital of Oakland, Oakland, CA; Children's Hospital of Orange County, Orange, CA; City of Hope National Medical Center, Duarte, CA; Kaiser Permanente Medical Group, Inc., Northern CA, Oakland, CA; Loma Linda University Cancer Institute, Loma Linda, CA; Santa Barbara Cottage Hospital, Santa Barbara, CA; Southern California Permanente Medical Group, Downey, CA; University of California, San Francisco School of Medicine, San Francisco, CA; University of California, Los Angeles School of Medicine, Los Angeles, CA; Harbor/University of California, Medical Center, Torrance, CA; The Children's Hospital, Denver, Denver, CO; Children's National Medical Center, D.C. Washington, DC; Children's Healthcare of Atlanta at Egleston, Atlanta, GA; Raymond Blank Children's Hospital, Des Moines, IA; University of Iowa Hospitals & Clinics, Iowa City, IA; Southern Illinois University School of Medicine, Springfield, IL; University of Illinois, Chicago, IL; Clarian Health, Indianapolis, IN; Indiana University, Riley Children's Hospital, Indianapolis, IN; Kosair Children's Hospital, Louisville, KY; Dana Farber Cancer Institute/Children's Hospital of Boston, Boston, MA; C.S. Mott Children's Hospital, Ann Arbor, MI; DeVos Children's Hospital, Grand Rapids, MI; Kalamazoo Community Clinical Oncology Program, Kalamazoo, MI; Michigan State University, Lansing, MI; Mayo Clinic and Foundation, Rochester, MN; The Children's Mercy Hospital, Kansas City, MO; University of North Carolina at Chapel Hill, Chapel Hill, NC; Janeway Child Health Center, St Johns NL, Canada; University of Medical and Dentistry of New Jersey, New Brunswick, NJ; Izaak Walton Killam Children's Hospital, Halifax NS, Canada; Sunrise Children's Hospital, Sunrise Hospital and Medical Center, Las Vegas, NV; Albany Medical Center, Albany, NY; Montefiore Medical Center, New York, NY; New York University Medical Center, New York, NY; SUNY Health Science Center at Brooklyn, Brooklyn, NY; Children's Hospital Medical Center, Akron, Akron, OH; Children's Hospital of Columbus, Columbus, OH; Children's Medical Center, Dayton, Dayton, OH; Cincinnati Children's Hospital Medical Center, Cincinnati, OH; Mercy Children's Hospital, Toledo, OH; Rainbow Babies and Children's Hospital, Cleveland OH; Children's East Ontario, Ottawa ON, Canada; Children's Hospital of Western Ontario, London ON, Canada; Emanuel Hospital-Health Care, Portland, OR; Children's Hospital of Philadelphia, Philadelphia, PA; Children's Hospital of Pittsburgh, Pittsburgh, PA; Penn State Children's Hospital, Hershey, PA; South Carolina Cancer Center, Columbia, SC; Dakota Midwest Cancer Institute, Sioux Falls, SD; Allan Blair Cancer Centre, Regina SK, Canada; Saskatoon Cancer Center, Saskatoon SK, Canada; East Tennessee Children's Hospital, Knoxville, TN; Vanderbilt University, Nashville, TN; Columbia Medical Center, West El Paso, TX; M.D. Anderson Cancer Center, Houston, TX; Scott & White, Temple, TX; Southwest Texas Methodist Hospital, San Antonio, TX; Texas Tech University, Health Science Center, Amarillo, TX; Primary Children's Medical Center, Salt Lake City, UT; Children's Hospital, King's Daughters, Norfolk, VA; Children's Hospital & Regional Medical Center, Seattle, Seattle, WA; Deaconess Medical Center, Spokane, WA; Group Health Cooperative of Puget Sound, Redmond, WA; Bellin Memorial Hospital, Green Bay, WI; Marshfield Clinic, Marshfield, WI; Oregon Health Sciences University, Portland, OR; Medical Center of Delaware-Alfred I. Dupont Institute, Wilmington, DE; Atlantic Health System, Morristown, NJ; University of Minnesota, Minneapolis, MN; Childrens Hospital & Clinics Minneapolis & St Paul, Minneapolis & St Paul, MN; Children's Hospital Central California, Madera, CA; Sioux Valley Children's Specialty Clinics, Sioux Falls, SD; The Children's Hospital at The Cleveland Clinic, Cleveland, OH; MeritCare Medical Group DBA Roger Maris Cancer Center, Fargo, ND.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  • 1
    Muir KR, Parkes SE, Lawson S, Thomas AK, Cameron AH, Mann JR. Changing incidence and geographical distribution of malignant paediatric germ cells tumours in the West Midlands health authority region 1957-92. Br J Cancer. 1995; 72: 219223.
  • 2
    Pinkerton CR. Malignant germ cell tumours in childhood. Eur J Cancer. 1997; 33: 895902.
  • 3
    Dos Santos Silva I, Swerdlow AJ, Stiller CA, Reid A. Incidence of testicular germ-cell malignancies in England and Wales: trends in children compared with adults. Int J Cancer. 1999; 83: 630634.
  • 4
    Glazer ER, Perkins CI, Young JL Jr., Schlag RD, Campleman SL, Wright WE. Cancer among Hispanic children in California, 1988-1994:comparison with non-Hispanic white children. Cancer. 1999; 86: 10701079.
  • 5
    United States Cancer Statistics, 2000 incidence. Washington, DC: Department of Health and Human Services, 2003.
  • 6
    Moller H. Work in agriculture, childhood residence, nitrate exposure, and testicular cancer risk: a case-control study in Denmark. Cancer Epidemiol Biomarkers Prev. 1997; 6: 141144.
  • 7
    Walker AH, Ross RK, Haile RW, Henderson BE. Hormonal factors and risk of ovarian germ cell cancer in young women. Br J Cancer. 1988; 57: 418422.
  • 8
    Shu XO, Nesbit ME, Buckley JD, Krailo MD, Robinson LL. An exploratory analysis of risk factors for childhood malignant germ-cell tumors: report from the Children's Cancer Group (Canada, United States). Cancer Causes Control. 1995; 6: 187198.
  • 9
    Swerdlow AJ, Stiller CA, Wilson LM. Prenatal factors in the aetiology of testicular cancer: an epidemiological study of childhood testicular cancer deaths in Great Britain 1953-73. J Epidemiol Community Health. 1982; 36: 96101.
  • 10
    Johnston HE, Mann JR, Williams J, et al. The inter-regional, epidemiological study of childhood cancer (IRESCC): case-control study in children with germ cell tumours. Carcinogenesis. 1986; 7: 717722.
  • 11
    Fajardo-Gutierrez A, Gomez-Gomez M, Danglot-Banck C, Alvarez-Contreras JJ, Yamamoto-Kimura L. Risk factors for the development of germ cell tumors in children. Gac Med Mex. 1998; 134: 273281.
  • 12
    Nomura T, Kanzaki T. Induction of urogenital anomalies and some tumors in the progeny of mice receiving diethylstilbestrol during pregnancy. Cancer Res. 1977; 37: 10991104.
  • 13
    McLachlan JA, Newbold RR, Bullock B. Reproductive tract lesions in male mice exposed prenatally to diethylstilbestrol. Science. 1975; 190: 991992.
  • 14
    Coscrove MD, Benton B, Henderson BE. Male genitourinary abnormalities and maternal diethylstilbestrol. J Urol. 1977; 117: 220222.
  • 15
    Chilvers C, Pike MC. Epidemiology of undescended testis. In: OliverRTD, editor. Urological and genital cancer. Oxford: Blackwell, 1989: 306321.
  • 16
    Rothman KJ, Louik C. Oral contraceptives and birth defects. N Engl J Med. 1978; 299: 522524.
  • 17
    Torfs CP, Milkovich L, van den Berg BJ. The relationship between hormonal pregnancy tests and congenital anomalies: a prospective study. Am J Epidemiol. 1981; 113: 563574.
  • 18
    MacMahon B, Trichopoulos D, Cole P, Brown J. Cigarette smoking and urinary estrogens. N Engl J Med. 1982; 207: 10621065.
  • 19
    United Kingdom Testicular Cancer Study Group. Aetiology of testicular cancer: association with congenital abnormalities, age at puberty, infertility, and exercise. Br Med J. 1994; 308: 13931399.
  • 20
    Gill J. The effects of moderate alcohol consumption on female hormone levels and reproductive function. Alcohol Alcohol. 2000; 35: 417423.
  • 21
    Reichman ME, Judd JT, Longcope C, et al. Effects of alcohol consumption on plasma and urinary hormone concentrations in premenopausal women. J Natl Cancer Inst. 1993; 85: 722727.
  • 22
    Sorahan T, McKinney PA, Mann JR, et al. Childhood cancer and parental use of tobacco: findings from the inter-regional epidemiological study of childhood cancer (IRESCC). Br J Cancer. 2001; 84: 141146.
  • 23
    Ji BT, Shu XO, Linet MS, et al. Paternal cigarette smoking and the risk of childhood cancer among offspring of nonsmoking mothers. J Natl Cancer Inst. 1997; 89: 238244.
  • 24
    Moller H. Clues to the aetiology of testicular germ cell tumours from descriptive epidemiology. Eur Urol. 1993; 23: 815.
  • 25
    Everson RB, Randerath E, Santella RM, Cefalo RC, Avitts TA, Randerath K. Detection of smoking-related covalent DNA adducts in human placenta. Science. 1986; 231: 5457.
  • 26
    Schlesselman JJ. Case-control studies: design, conduct, analysis. New York: Oxford University Press, 1982.
  • 27
    Breslow NE, Day NE. Statistical methods in cancer research. Vol. I: the analysis of case-control studies. Lyon: International Agency for Research on Cancer, 1980.
  • 28
    Infante-Rivard C, Krajinovic M, Labuda D, Sinnett D. Childhood acute lymphoblastic leukemia associated with parental alcohol consumption and polymorphisms of carcinogen-metabolizing genes. Epidemiology. 2002; 13: 277281.
  • 29
    Cocco P, Rapallo M, Targhetta R, Biddau PF, Fadda D. Analysis of risk factors in a cluster of childhood acute lymphoblastic leukemia. Arch Environ Health. 1996; 51: 242244.
  • 30
    Hansen C, Asmussen I, Autrup H. Detection of carcinogen-DNA adducts in human fetal tissues by 32P-postlabeling procedure. Environ Health Perspect. 1993; 99: 229231.
  • 31
    Finette BA, O'Neill JP, Vacek PM, Albertini RJ. Gene mutations with characteristic deletions in cord blood T lymphocytes associated with passive maternal exposure to tobacco smoke. Nat Med. 1998; 4: 11441151.
  • 32
    Schmidt F. Smoking damages male fertility. Andrologia. 1986; 18: 445454.
  • 33
    Olshan AF, Breslow NE, Ealletta JM, et al. Risk factors for Wilms tumor. Report from the National Wilms Tumor Study. Cancer. 1993; 72: 938944.
  • 34
    Brondum J, Shu XO, Steinbuch M, Severson RK, Potter JD, Robinson LL. Parental cigarette smoking and the risk of acute leukemia in children. Cancer. 1999; 85: 13801388.
  • 35
    Yang Q, Olshan AF, Bondy ML, et al. Parental smoking and alcohol consumption and risk of neuroblastoma. Cancer Epidemiol Biomarkers Prev. 2000; 9: 967972.
  • 36
    Pang D, McNally R, Birch JM. Parental smoking and childhood cancer: results from the United Kingdom Childhood Cancer Study. Br J Cancer. 2003; 88: 373381.
  • 37
    Burguet A, Agnani G. Smoking, fertility and very preterm birth. J Gynecol Obstet Biol Reprod (Paris). 2003; 32: 1S91S16.
  • 38
    Ivankovic S. Teratogenic and carcinogenic effects of some chemicals during prenatal life in rats, Syrian golden hamsters, and minipigs. Natl Cancer Inst Monogr. 1979; 51: 103115.
  • 39
    Rice JM, Ward JM. Age dependence of susceptibility to carcinogenesis in the nervous system. Ann N Y Acad Sci. 1982; 381: 274289.
  • 40
    Olshan AF, Faustman EM. Male-mediated developmental toxicity. Annu Rev Public Health. 1993; 14: 159181.
  • 41
    Severson RK, Buckley JD, Woods WG, Benjamin D, Robison LL. Cigarette smoking and alcohol consumption by parents of children with acute myeloid leukemia: an analysis within morphological subgroups—a report from the Children's Cancer Group. Cancer Epidemiol Biomarkers Prev. 1993; 2: 433439.
  • 42
    Sorahan T, Lancashire R, Prior P, Peck I, Stewart A. Childhood cancer and parental use of alcohol and tobacco. Ann Epidemiol. 1995; 5: 354359.