Effect of varying proportions of dietary menhaden and corn oil on experimental rat mammary tumor promotion

Dose-related effects of long-chain highly unsaturated n−3 fatty acids on the development ofN-nitrosomethylurea (NMU)-induced rat mammary tumors were assessed in female F344 rats. Four test groups (36 rats/group) were fed the following high-fat (HF) diets (23% fat, w/w): Group 1, 18% menhaden oil (MO) and 5% corn oil (CO); Group 2, 11% MO and 11.8% CO; Group 3,5% MO and 18% CO; Group 4, CO alone. A fifth group, serving as an internal control, was fed a low-fat diet containing 5% CO alone. Experimental diets were begun after initiation with NMU, and the experiment was terminated 31 wk later. Total tumor numbers in the five groups were 28, 16, 32, 26 and 11, respectively, indicating that the promotion phase of NMU-induced carcinogenesis was significantly suppressed only when equal parts of CO and MO (Group 2) were fed or when CO alone was fed at 5% (w/w). At high (Group 1) or low (Group 3) levels of MO, tumor numbers were indistinguishable from the HF CO group (Group 4). The same pattern was observed when assessed in terms of cumulative tumor incidence and multiplicity. However, when expressed in terms of final tumor incidence, dietary MO did not suppress tumor promotion in a statistically significant fashion at any concentration. Animals fed MO gained weight at the same rate as those fed CO, indicating that the presence of MO in the diet did not result in food avoidance behavior. Measurement of total serum cholesterol indicated an inverse trend with respect to the MO content of the diet. Analysis of serum fatty acid profiles indicated that the proportion of n−3 and n−6 polyun-saturated fatty acids (PUFA) in the serum reflected that of the diet. These results support the hypothesis that the relative proportions of dietary n−3/n−6 fatty acids play an important role in the suppression of experimental mammary tumorigenesis and suggest that changes in circulating cholesterol or n−3 PUFA levels, induced by dietary MO, are not directly related to tumor development. Presented in part at the 81st Annual Meeting of the American Association for Cancer Research, Washington, D.C., May 1990