Re: plasma insulin-like growth factor-I, insulin-like growth factor-binding proteins, and prostate cancer risk: a prospective study.

The important article by Stattin et al. (1) provides further evidence that insulin-like growth factor-I (IGF-I) may play a key role in the development of prostate cancer. Prediagnostic levels of serum IGF-I were, on average, 7% higher in men who subsequently developed prostate cancer compared with healthy individuals, a result similar to that found in the Physicians’ Health Study (2). Circulating IGF-I levels have been shown to be sensitive to nutritional factors: A diet severely restricted in energy and/or protein intake substantially reduces serum IGF-I concentrations in both animals and humans (3). Stattin et al. (1) suggest that elevated IGF-I levels could be a potential mechanism through which a Western high-energy diet may increase prostate cancer risk. However, they acknowledge that obesity is not related to an increase in IGF-I concentration and that a high body mass index is not a strong risk factor for prostate cancer. An alternative explanation is that protein intake may be the critical determinant of IGF-I serum concentrations that ultimately influence prostate cancer risk. The rationale is that a diet low in protein and essential amino acids will reduce IGF-I production and thereby reduce the growth of prostate cancer from a latent to a clinically manifest form. One test of this hypothesis is to examine IGF-I levels among men who consume a vegan diet, which contains only plant protein and, therefore, fewer essential amino acids than diets containing primarily animal protein (4). We have recently shown that the mean serum IGF-I concentration was 9% (95% confidence interval 6% to 13%) lower among 233 vegan men who consumed no animal products than it was among either 237 lacto-ovo-vegetarians, who consumed no meat or fish but did consume either dairy and/or egg products, or 226 meat eaters (5). Among the 696 men studied, consumption of a diet rich in essential amino acids (i.e., containing protein derived from animal and soy products) was positively correlated with serum IGF-I concentrations. Serum IGF-I concentration increased 20% from 17.4 nmol/L in the lowest decile of intake (<9.7 g/day of animal and soy protein) to 20.8 nmol/L in the highest decile of intake ( 74.5 g/day of animal and soy protein after adjustment for age, body mass index, smoking, vigorous exercise, and factors related to blood collection (test for linear trend; P .007). These results suggest that the low essential amino acid intake among vegan men may explain their low serum IGF-I concentrations. It is important to note that total protein, total energy intake, and body mass index were not statistically significantly associated with IGF-I concentration in this population. An epidemiologic study (6) has shown that mortality due to prostate cancer is correlated with per capita consumption of both animal fat and animal protein. However, studies on the role of animal fat in prostate cancer development [reviewed in (7)] have produced inconsistent results. The increasing evidence that small differences in serum IGF-I concentrations are associated with prostate cancer risk and the sensitivity of IGF-I concentration to protein intake suggest that differences in dietary intake of animal protein may partly explain the wide international variation in prostate cancer mortality.