Authors
Lennart Hardell, Bert Van Bavel, Gunilla Lindstrom, Michael Carlberg, Mikael Eriksson, Ann Charlotte Dreifaldt, Hans Wijkstrom, Hans Starkhammar, Arne Hallquist, Torgny Kolmert.

Title:
Concentrations of polychlorinated biphenyls in blood and the risk for testicular cancer.

Journal:
International Journal of Andrology, 2004; 27:282-290.

Summary:
The incidence of testicular cancer has been increasing in western countries. According to current thought, both seminoma and non-seminoma histopathological sub-types are thought to begin as carcinoma-in-situ during the fetal period. Prenatal exposure to environmental pollutants including endocrine disrupting chemicals with estrogenic and anti-estrogenic effects has been implicated in the etiology of testicular cancer. Polychlorinated biphenyls have been of particular concern. Polychlorinated biphenyls (PCBs) are fat soluble chemicals with a prolonged half life that bio-accumulate in the human body.

The authors, in their previous study, reported a significant increase in blood concentrations of PCBs, hexachlorobenzene (HCB), and trans- & cis-nonachlordanes in mothers of testicular cancer cases. In this study, the authors further analyzed the data for 37 PCB congeners. A total of 61 testicular cancer cases were recruited by physicians at university hospitals in Sweden from 1997 to 2000. Age matched controls (5 year age strata) were selected for each case from the Swedish population registry. The mothers of the controls that were identified from the population registry acted as controls for case-mothers. Blood samples were obtained from all the subjects participating in the study during the same time period (1997-2000). Information on other factors such as weight, length and reproductive history of the mothers was also obtained by questionnaire. Because of technical reasons, such as sample losses in the laboratory, 58 cases and 61 controls were available for analysis. 44 case-mothers and 45 control-mothers were available for the analysis, after excluding the subjects unwilling to participate in the study. Blood samples were assessed for PCB congeners using high-resolution gas chromatography and coupled mass spectrometry.

Unconditional logistic regression analysis was performed, adjusting for age and body mass index (BMI), to determine the relationship between PCBs and testicular cancer. Median concentration of PCBs in controls was taken as the reference for the estimation of odds ratios and confidence intervals. The Wilcoxon rank sum test was used to compare the concentration of PCBs in cases and controls using ranks, in addition to unconditional logistic regression analysis. In addition to analyzing the case-control status of individual PCB congeners, an analysis was also performed for PCB congeners that were grouped according to structural and biological activity. PCB congeners were grouped into 1) estrogenic & weak phenobarbital inducers, and 2) enzyme inducing PCBs. Furthermore, dioxin-like toxic equivalents (TEQ) were estimated using weighting factors proposed by the World Health Organization (WHO).

The analysis of individual PCB congeners and grouped PCB congeners revealed no significant differences in serum concentration of PCBs in cases and controls. However, analysis of mothers indicated a significant increase in the concentration of 19 PCB congeners as well as the sum of all PCBs in case-mothers. Moreover, significantly increased serum concentrations of enzyme inducing PCBs and TEQs were also observed in case-mothers. The concentration of estrogen-inducing PCBs was found to be increased in case-mothers using the Wilcoxon test but not using unconditional logistic regression. Adjusting for age and BMI in unconditional logistic regression and using median concentration of PCBs in the controls as reference for odds ratio estimates could be the reasons for the differences observed above. Odds ratios were highest in mothers of non-seminoma testicular cancer cases compared to mothers of seminoma cases.

In summary, this study found significantly increased concentrations of PCBs in case-mothers but not in cases, indicating some etiologic role of PCB exposure during the fetal period. Moreover, this finding is consistent with current hypotheses that testicular cancer is initiated in the fetal period. The collection of blood from all subjects during the same time period (1997-2000) and before initiation of treatment in cases eliminated potential biases due to changes of concentrations of PCBs in the population over time and to the influence of treatment, adding strength to the findings in this study. The estrogen receptor binding capacity of metabolites of certain PCB congeners supports their estrogen-like characteristics. Overall, this study demonstrated that exposure to PCB congeners during the fetal period may be implicated in the etiology