A positive association between body fat and prostate cancer has been observed, though studies are inconsistent. Paradoxically, diabetes appears to be protective against prostate cancer. We examined the association of body fat distribution and diabetes with prostate cancer among Afro-Caribbean men on the Island of Tobago.

Participants comprised 226 men, aged 50-79 years, with screening-detected prostate cancer and 666 men with normal screening results (normal digital rectal examination (DRE) and serum prostate specific antigen (PSA)

≤ 4 ng/ml) ascertained in a population-based screening study. Standard protocols were used to collect anthropometric measures and self-reported history of diabetes. Regional body composition was measured by dual x-ray absorptiometry (DXA). Unconditional logistic regression was used to identify potential independent predictors of prostate cancer risk.

After controlling for age, prostate cancer cases and controls were similar in anthropometric measurements and in DXA estimates of lean and fat mass in the arm, leg, and trunk, except for percent trunk fat which was marginally higher in cases than controls (50.5 vs. 50.1, respectively, p = 0.085). History of diabetes was reported by 157 men. History of diabetes was marginally protective against prostate cancer (OR = 0.66, p = 0.065). However, there was a strong interaction between diabetes and percent trunk fat (p=0.008) for prostate cancer. Compared with men without diabetes and percent trunk fat below the median, age-adjusted risk was much lower for men with diabetes and percent trunk fat below the median (OR=0.24, 95%CI 0.10-0.58), while risk among men with percent trunk fat above the median was similarly elevated in men with diabetes (OR=1.32, 95%CI 0.92-1.90) and without diabetes (OR=1.30, 95%CI 0.76-2.22). No interactions were observed in similar analyses of body weight, body mass index, waist circumference or DXA estimated total body percent fat.

These results support an association between prostate cancer and central body fat deposition and diabetes which is modified by diabetes only in men with lower trunk fat. This interaction may explain some of the inconsistencies among previous studies. Further studies are needed to understand hormonal and other factors which may underlie this complex association.

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA