Background

Findings from epidemiologic studies examining associations of serum retinol and carotenoids with prostate cancer risk have been inconsistent. This case-control study nested in the Prostate Cancer Prevention Trial evaluated associations of serum retinol and carotenoids with total, low-, and high-grade prostate cancer risk in a highly screened study population.

Methods

We used logistic regression adjusting for age, family history of prostate cancer, race, body mass index, and serum cholesterol to estimate ORs and 95% confidence intervals (CI) of prostate cancer by quartiles of serum retinol and carotenoids, separately in the placebo (975 cases/1,009 frequency-matched controls) and finasteride (708 cases/743 frequency-matched controls) arms of the trial.

Results

Serum retinol concentrations were associated with increased risk of total prostate cancer [OR (95% CI) comparing the highest quartile of serum retinol with the lowest: 1.30 (1.00-1.68)] and high-grade prostate cancer [OR (95% CI), 1.74 (1.14-2.68)] in the placebo arm of the trial only. Also in the placebo arm, there was a moderate positive association of α-carotene with risk of total prostate cancer [OR (95% CI), 1.32 (1.01-1.73)]. None of the other carotenoids was associated with prostate cancer risk in the placebo arm. No associations were observed for retinol and carotenoids in the finasteride arm.

Conclusion

In the placebo arm of this prospective study, high serum retinol and α-carotene concentrations were associated with increased risk of total and high-grade prostate cancers.

Impact

Men with higher levels of serum retinol and α-carotene may be at increased risk for prostate cancer.

The nitrogen isotope ratio (δ(15)N) of RBCs has been proposed as a biomarker of marine food intake in Yup'ik people based on strong associations with RBC eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). However, EPA and DHA derive from marine fats, whereas elevated δ(15)N derives from marine protein, and these dietary components may have different biologic effects. Whether δ(15)N is similarly associated with chronic disease risk factors compared with RBC EPA and DHA is not known. We used covariate-adjusted linear models to describe biomarker associations with chronic disease risk factors in Yup'ik people, first in a smaller (n = 363) cross-sectional study population using RBC EPA, DHA, and δ(15)N, and then in a larger (n = 772) cross-sectional study population using δ(15)N only. In the smaller sample, associations of RBC EPA, DHA, and δ(15)N with obesity and chronic disease risk factors were similar in direction and significance: δ(15)N was positively associated with total, HDL, and LDL cholesterol, apolipoprotein A-I, and insulin-like growth factor binding protein-3 (IGFBP-3), and inversely associated with triglycerides. Based on comparisons between covariate-adjusted β-coefficients, EPA was more strongly associated with circulating lipids and lipoproteins, whereas δ(15)N was more strongly associated with adipokines, the inflammatory marker interleukin-6, and IGFBP-3. In the larger sample there were new findings for this population: δ(15)N was inversely associated with blood pressure and there was a significant association (with inverse linear and positive quadratic terms) with adiponectin. In conclusion, δ(15)N is a valid measure for evaluating associations between EPA and DHA intake and chronic disease risk in Yup'ik people and may be used in larger studies. By measuring δ(15)N, we report beneficial associations of marine food intake with blood pressure and adiponectin, which may contribute to a lower incidence of some chronic diseases in Yup'ik people.

Inflammation may play an etiologic role in prostate cancer. Several dietary factors influence inflammation; studies have shown that long-chain n-3 polyunsaturated fatty acids are anti-inflammatory, whereas n-6 and trans fatty acids are proinflammatory. We evaluated whether serum phospholipid n-3, n-6, and trans fatty acids were associated with intraprostatic inflammation, separately in 191 prostate cancer cases and 247 controls from the placebo arm of the Prostate Cancer Prevention Trial (PCPT). Men without a prostate cancer diagnosis underwent prostate biopsy at trial end, and benign prostate tissue inflammation was evaluated in approximately three biopsy cores per man; this was expressed as no, some, or all cores with inflammation. In controls, serum eicosapentaenoic acid [OR of all cores with inflammation versus none (95% CI), 0.35 (0.14-0.89)] and docosahexaenoic acid [OR (95% CI), 0.42 (0.17-1.02)] were inversely associated with, whereas linoleic acid [OR (95% CI), 3.85 (1.41-10.55)] was positively associated with intraprostatic inflammation. Serum trans fatty acids were not associated with intraprostatic inflammation. No significant associations were observed in cases; however, we could not rule out a positive association with linoleic acid and an inverse association with arachidonic acid. Thus, in the PCPT, we found that serum n-3 fatty acids were inversely, n-6 fatty acids were positively, and trans fatty acids were not associated with intraprostatic inflammation in controls. Although, in theory, inflammation could mediate associations of serum fatty acids with prostate cancer risk, our findings cannot explain the epidemiologic associations observed with n-3 and n-6 fatty acids.