Abstract B101: Phytoestrogen intake in relation to prostate cancer incidence and survival

Abstract

Abstract Background: Phytoestrogens are plant-based compounds derived mainly from soy, oilseeds, legumes, and grains. Several studies support an inverse association between soy intake and prostate cancer risk, though few have investigated phytoestrogens from dietary sources that are more common to Western diet patterns. Purpose: The objective was to assess whether dietary intake of phytoestrogens was associated with prostate cancer risk and survival in a large, prospective cohort study in Sweden. Design: The Cohort of Swedish Men is a prospective study of 45,906 men from central Sweden, born 1918–1948. During enrollment in 1997, 44,974 (98%) participants completed a self-administered food frequency questionnaire, and we calculated energy-adjusted dietary intakes of lignans (secoisolariciresinol, matairesinol, lariciresinol, pinoresinol, syringaresinol, and medioresinol) and isoflavonoids (genistein and diadzein). Incident prostate cancer was histologically confirmed in 2721 participants from 1997–2008, and 215 prostate cancer-specific deaths were identified using the Swedish Death Registry. We used Cox proportional hazards regression to calculate multivariable hazard ratios (HR) and 95% confidence intervals (CI). Results: Mean intake was 2295 µg/day for total lignans and 10.2 µg/day for total isoflavonoids. Neither total lignan nor total isoflavonoid intake was associated with risk of prostate cancer. The HRs for increasing quintiles of total lignans were all 1.0, p-trend=0.72; for total isoflavonoids, the HRs were 1.0 (referent), 1.1, 1.1, 1.0 and 1.0, p-trend=0.75, adjusting for age at entry and total energy intake. Results were similarly null for the individual phytoestrogen subtypes and prostate cancer incidence. However, we found a weak positive association between the newly-identified lignan, medioresinol, and prostate cancer-specific death, with HRs of 1.0 (referent), 1.5, 1.5, 1.7, and 1.7 for increasing quintiles of intake (p-trend=0.03). The estimates were similar (p-trend=0.04) after further adjustment for alcohol intake and clinical measures. Dietary intakes of the remaining phytoestrogen subtypes were not associated with prostate cancer survival. Conclusions: In this large, prospective study, our findings do not support a major role for dietary phytoestrogens in the incidence or progression of prostate cancer. As phytoestrogen intake is relatively low in the Swedish population, we do not exclude the possibility of significant associations at higher levels of intake. Citation Information: Cancer Prev Res 2010;3(1 Suppl):B101.