One hundred patients with prostate cancer and two different control series [100 benign prostatic hyperplasia (BPH) patients and 100 general hospital patients] were matched to each other upon hospital admittance, age (±3 years) and date of admission (±3 months), and directly interviewed during admission from 1981 to 1984 in Kyoto, Japan. Major dietary findings derived from a quantitative food frequency technique for estimating usual diet are as follows. (a) The smaller the dietary intake of β-carotene and vitamin A as well, the higher the risk, with a highly significant linear trend. From the β-carotene analyses, the relative risk (95% confidence interval) for the lowest intake quartile relative to the highest was 2.10 (0.98–4.47) for the uncorrected intake, 2.35 (1.08–5.12) for the intake per kg, and 2.94 (1.34–6.44) for the intake per kcal in the comparison with BPH patients; 2.88 (1.31–6.32), 2.56 (1.14–5.76), and 3.50 (1.52–8.06), respectively, in the comparison with hospital controls. The corresponding relative risk obtained from the vitamin A analyses was 2.82 (1.30–6.14), 2.64 (1.24–5.60), and 3.29 (1.47–7.35) in due order in the comparison with BPH patients; 2.69 (1.22–5.94), 4.78 (1.98–11.52), and 3.50 (1.52–8.06) in the comparison with hospital controls. (b) β-Carotene as well as vitamin A contained in green/yellow vegetables were significantly protective, and those in seaweeds and kelp suggestively protective. But those in fruits appeared to enhance the risk. (c) The risk reduction by dietary β-carotene and vitamin A was significant in the older men (70–79 years), but not in the younger men (50–69 years). (d) Total energy intake and the dietary intake of fat, protein, carbohydrate, water, fiber, ash, such vitamins as retinol, B1, B2, C, and niacin, and such minerals as calcium, potassium, sodium, phosphorus, and iron were not linked with prostate cancer risk. (e) A protective effect of dietary β-carotene and vitamin A against prostate cancer could be related to the low overall fat intake in Japan.
Supported by a Grant-in-Aid (60010013) for Cancer Research from the Ministry of Education, Science and Culture, Japan, and grants from the Toyota Foundation; the Dutch Cancer Foundation, the Netherlands; and the International Agency for Research on Cancer, Lyon, France. Presented at the Fifth Symposium on Epidemiology and Cancer Registries in the Pacific Basin, November 16–21, 1986, Kauai, HI.