A prospective study of Trichomonas vaginalis and prostate cancer risk among African American men

Background African Americans (AA) have a higher prevalence of Trichomonas vaginalis (Tv) infection and a higher prostate (PC) risk. Past studies suggest an association between Tv seropositivity and PC, and therefore we prospectively investigated this association among AA men. Results Incident PC cases were individually matched to controls in a nested case–control study within the Southern Community Cohort Study (SCCS). Primary analysis included 296 PC cases and 497 race-matched controls. Levels of Tv antibody response were measured by ELISA in serum collected at baseline. Tv antibody response did not significantly differ between cases and controls overall or within AA participants (253 AA cases). There were no significant associations or trends between levels of Tv response and PC risk or the diagnosis of aggressive PC. Conclusion We found no evidence of a prospective association between baseline Tv infection and PC risk in AA men. Tv infection in men may have substantial health implications in HIV transmission and reproductive outcomes, but may not impact future PC risk in AA men at high-risk for PC. Further efforts need to define past vs. present Tv infection and to separate pathophysiology from PC detection.


Background
Trichomonas vaginalis (Tv) is a sexually transmitted protozoan parasite. In men, urethral infection may ascend to the prostatic urethra and glandular tissue. With asymptomatic infection common, Tv exposure may induce a sustained inflammatory response to advance prostate carcinogenesis [1]. Analysis of the Health Professionals Follow-up Study and the Physicians Health Study found that Tv antibody seropositivity was significantly associated with prostate cancer (PC) risk [2,3] or PC death [2]. In contrast, Tv seropositivity was not associated with PC risk in the Prostate Cancer Prevention Trial (PCPT) [4] or a population-based case-control analysis [5]. The purpose of this study was to determine the prospective relationship between Tv infection and PC risk among African American (AA) men. Past studies included few AA men, although there are data suggesting the prevalence of Tv infection and the risk of PC may be greater than among white men [6].

Methods
Details of the Southern Community Cohort Study (SCCS) have been published [7]. Incident PC cases among nearly 35,000 male participants enrolled at age 40-79 during 2002-2009 were identified through linkages through 2013 with state tumor registries and the National Death Index. All participants provided written informed consent, and all protocols were approved by Institutional Review Boards at Vanderbilt University and Meharry Medical College. All SCCS data can be requested through an online request (southerncommunitystudy.org). The request will be reviewed by the SCCS Data and Biospecimen Use Committee to ensure that it is scientifically justified and that participant confidentiality is preserved. Two controls per case were selected by incidence density sampling and were individually matched by age (5 years), race (self-reported black or white), site, and time of donation of blood samples at study entry. A blood sample was collected at baseline recruitment, and serum was frozen-stored at −80 °C . Serum was assayed in duplicate for antibodies against Tv by ELISA to detect the IgG antibodies against the purified recombinant Tv α-actinin protein and assessed spectrophotometrically [8]. We previously assigned scores from 0 to 4+ to serum based on the calculation of P/N values obtained using the protein ACT-P2 as the target [2][3][4]8]. For this study, cutoff points for seropositivity were obtained by dividing the average OD 405nm of the seropositive control serum by the corresponding seronegative control serum. The lowest seronegative control was assigned a score of 0 (zero). Values from 0 to the next lowest was given a score of 1+, and subsequent scores of 2+, 3+, and 4+ were assigned similarly as values increased. The sera with P/N scores ≤2 were negative as evidenced by lack of detection of any T. vaginalis proteins by immunoblot [8]. Scores ≥3 were positive and had antibody to α-actinin and other trichomonad proteins [8]. We used conditional logistic regression to compute odds ratios (OR) and 95 % confidence intervals while controlling for household income. Cancer stage or Gleason score from state tumor registries providing these data were also analyzed after controlling for age, race, and income.

Results
The majority of the study population was over 60 years of age at recruitment (n = 549, 62 %), and reported a household income less than $15,000/year (n = 512, 58 %).
Approximately 85 % of the study population was AA. Table 1 summarizes the association between Tv seropositive status and PC. Mean antibody response levels were  similar between cases and controls (all p > 0.05). There was no significant association between Tv and PC in the total study population, or when restricting to AA men. Furthermore, Tv exposure was not associated with PC Gleason score of 7 or more, or stage 2-4 PC at diagnosis, in a case-only analysis ( Table 2).

Discussion
Tv disproportionately affects minorities and lowincome populations [6]. However, unlike two past studies of predominately white professionals [2,3], we found no evidence of association within AA men. A post hoc power calculation based on a similar matched nested case-control design and assuming a 25 % exposure, Type I error of 5 %, and 80 % power, would be able to detect an OR = 1.42, suggesting a sufficient sample size to identify a moderate association. Residual confounding is unlikely to have led to a null association. As previously detailed, differences in patient characteristics and detection protocols could explain differences in results across studies [4]. Similarly, an inflammatory effect from Tv would likely increase PSA levels, and thus increase the likelihood of detecting an asymptomatic PC and lead to a positive association. No association between Tv seropositivity and PC was found in the PCPT, with rigorous control for PC screening including an end of study prostate biopsy [4]. Alternatively, analysis of the Prostate, Lung, Colorectal and Ovarian cancer screening trial found Tv seropositivity significantly associated with benign prostatic hyperplasia [9], suggesting Tv may exacerbate lower urinary tract symptom severity and possibly lead to a PC detection. [10] Tv seropositivity is relatively stable over time [10], limiting the ability to identify an etiologically relevant time of exposure. Our prospective analysis found no association between baseline Tv seropositivity and future PC risk, and further efforts are needed to define past vs. recent Tv infection and to separate pathophysiology from the detection of PC.