Prostate cancer in native Japanese and Japanese-American men: Effects of dietary differences on prostatic tissue

Objectives. To investigate the relationship between diet and prostate cancer (CaP) among native Japanese (NJ) and second-generation or third-generation Japanese-American (J-A) men - focusing on the effects of animal fat and soy on prostatic tissues. The subjects were 50 Japanese men undergoing radical prostatectomy, 25 NJ living in Nagoya, Japan and 25 U.S.-born J-A men, living in Los Angeles, California. A priori, the NJ men were believed to be a low-fat, high-soy group and the J-A men, a high-fat, low-soy group. The studies included postoperative measurements of diet (Block questionnaire), body fat (bioimpedance), blood, urine, and prostatic biomarkers in malignant and adjacent normal tissue, using a tissue microarray made from the original paraffin blocks. The NJ and J-A men were similar in age (65 to 70 years old; P <0.05), prostate-specific antigen level (7.1 to 8.6 ng/mL), prostate volume (35 to 38 cm ³), and Gleason score (5.6 to 6.6), but their body composition differed. J-A men had more body fat (24% versus 19%), higher serum triglyceride levels (245 versus 106 mg/dL), lower estradiol levels (27 versus 31 ng/mL), and much lower urinary soy-metabolite levels (1:3) than NJ men (P <0.02). In both NJ and J-A groups, expression of numerous tissue biomarkers separated normal from CaP tissue, including markers for apoptosis (Bcl-2, caspase-3), growth factor receptors (epidermal growth factor receptor), racemase, 5-lipoxygenase, kinase inhibition (p27), and cell proliferation (Ki-67; all P <0.02). Furthermore, within both normal and CaP tissues, caspase-3 and 5-lipoxygenase were expressed more in NJ than in J-A men (P <0.01). Nuclear morphometry showed that the chromatin in each of the four groups (normal versus CaP, NJ versus J-A) was different (area under the curve 85% to 94%, P <0.01), despite fundamental genetic homogeneity. NJ and J-A men, products of similar genetics but differing environments, were shown to have differences in body composition that could influence CaP evolution. The CaP specimens from the NJ and J-A men were histologically similar, but tissue biomarker expression, especially of lipoxygenase and the caspase family, suggested differing mechanisms of carcinogenesis. Differences in nuclear morphometry suggested the additional possibility of gene-nutrient interactions.