Background: Prostate cancer is the most commonly diagnosed cancer among men in the United States, and disproportionately affects African‐American men. It has been suggested that diet plays a role in prostate cancer development and progression. Evidence regarding the effect of polyunsaturated fatty acids (PUFA) indicate that n‐3 PUFAs may suppress prostate cancer development while n‐6 PUFAs may enhance prostate cancer carcinogenesis. Epidemiologic studies, however, have been inconsistent and the association between PUFAs and prostate cancer has not been adequately studied in racially diverse samples.

Objectives: The objectives of this work were to 1) examine the association between dietary intake of PUFAs [n‐3 (ALA, EPA, DHA); n‐6(AA, LA)] and risk of prostate cancer, and 2) determine if these associations differ between Whites and African‐Americans.

Methods: Data are from an ongoing case‐control study among veterans at the Durham VA Medical Center that includes biopsy‐positive cases (N=66), biopsy‐negative controls (N=91), and healthy controls (N=101). All men had a PSA test done in the year prior to study enrollment. Data were collected on various demographic, lifestyle, and other risk factors for prostate cancer using self‐administered questionnaires. Diet was assessed using a food frequency questionnaire that asked participants to recall their diet in the past 12 months. Logistic regression analyses were used to estimate odds ratios (OR) and 95% confidence intervals (95% CI) among Whites (N=111) and African‐Americans (N=146).

Results: The median percent of energy from total n‐6 PUFAs was higher in cases compared to healthy controls (6.2% vs 5.7%, p=0.05). The ORs for cases relative to healthy controls were greater than one for high (i.e. ≥median) intakes of total n‐6 PUFAs (OR: 1.79, 95% CI 0.92–3.48), yet close to null for total n‐3 PUFAs (OR: 1.02, 95% CI 0.53–1.96). There were non‐statistically significant inverse associations for specific PUFAs such as EPA, DHA, and LA (ORs 0.77, 0.68, and 0.62, respectively; all p‐values >0.05). A higher n‐3:n‐6 ratio was suggestive of lower prostate cancer risk (OR: 0.80, 95% CI 0.42–1.54). These findings were relatively unchanged in analyses comparing cases to biopsy‐negative controls. In race‐specific analyses comparing cases to biopsy‐negative controls, there was evidence of effect modification by race for the relationship between LA intake and prostate cancer risk (OR: 0.78, 95% CI 0.30–2.04 in Whites; OR: 3.23, 95% CI 1.24–8.40 in African‐Americans; Pinteraction=0.04). We did not observe any other statistically significant differences in prostate cancer risk between Whites and African‐Americans.

Conclusion: Our overall findings do not suggest a strong role for n‐3 or n‐6 PUFAs in prostate cancer development, although racial differences in risk should be investigated in larger racially diverse samples.

Citation Information: Cancer Prev Res 2010;3(1 Suppl):B110.

Copyright © January 7, 2010, American Association for Cancer Research
2010