To examine the relation between serum fatty acids and risk of colorectal cancer, the authors conducted a nested case-control study of 169 colorectal cancer cases and 481 controls matched by age and enrollment area as part of the Japan Collaborative Cohort Study. Serum samples were donated by subjects at baseline (between 1988 and 1990) and were stored at -80 degrees C until 2002. Serum fatty acid levels were measured by using gas chromatography and were expressed as the weight percentage of total lipids. Conditional logistic regression analyses adjusted for lifestyle factors revealed that total omega-3 polyunsaturated fatty acids (odds ratio = 0.24, 95% confidence interval: 0.08, 0.76), alpha-linolenic acid (odds ratio = 0.39, 95% confidence interval: 0.16, 0.91), docosapentaenoic acid (odds ratio = 0.30, 95% confidence interval: 0.11, 0.80), and docosahexaenoic acid (odds ratio = 0.23, 95% confidence interval: 0.07, 0.76) all showed a significantly decreased risk for the highest versus the lowest quartile levels for colorectal cancer in men. For women, a weak negative association was observed between docosapentaenoic acid and colorectal cancer risk, although it was not statistically significant. No adverse effects of high serum levels of omega-6 polyunsaturated fatty acids on colorectal cancer risk were detected.