Are dietary influences on the risk of prostate cancer mediated through the insulin‐like growth factor system?

Abstract

Objectives To investigate whether dietary factors that appear to affect the risk of prostate cancer may be similarly associated with serum levels of insulin-like growth factor 1 (IGF-1). Patients and methods In the context of a case-control study, 112 men were admitted to three teaching hospitals in Athens, Greece, for disorders other than cancer. Sociodemographic data and detailed histories of smoking, alcohol and coffee consumption were recorded. A validated food-frequency questionnaire was administered by an interviewer and serological measurements of IGF-1 and its binding protein-3 conducted. Results IGF-1 declined signi®cantly by almost 25% among men aged >75 years and there was a small reduction in IGF-1 levels with increased alcohol intake, with a mean (95% con®dence interval, CI) change of x1.6 (x2.2 to x0.9)% for an increment of one drink per day. There was no evidence for an effect of either smoking or coffee consumption on IGF-1 level. Among foods, the consumption of cooked tomatoes was substantially and signi®cantly inversely associated with IGF-1 levels, with a mean (95% CI) change of x31.5 (x49.1 to x7.9)% for an increment of one serving per day. Conclusions The strongest known dietary risk factor for prostate cancer (lycopene de®cit, as re¯ected in a reduced intake of cooked tomatoes) and an important endocrine factor in the aetiology of this disease (IGF-1) seem to be related in a way that suggests that at least one, and perhaps more, exogenous factors in the development of prostate cancer may be mediated through the IGF-1 system.

Dietary Determinants of Circulating Insulin-like Growth Factor (IGF)-I and IGF Binding Proteins 1, -2 and -3 in Women in the Netherlands

“…Both energy and protein are needed to restore IGF-I after fasting [11]. Our results, in a study population with adequate intake of energy and protein, are consistent with most previous cross-sectional studies, in which no association between energy and protein intake and circulating IGF-I and IGFBP-3 was observed [12][13][14][15][16][17][18][19][20]. The two largest cross-sectional studies did show a positive association between IGF-I and protein intake within the normal range [22,23], as is also found by two smaller studies [21,24].…”

“…We also found that higher alcohol intake was associated with decreased IGFBP-1 concentrations in postmenopausal women. Cross-sectional studies on the association between alcohol intake and the IGF-system published to date have not investigated IGFBP-1, and have been very inconsistent with respect to IGF-I and IGFBP-3 [14,15,17,22,26]. This may be partly explained by the different ranges of intakes studied.…”

“…In our study, no association between either tomatoes or lycopene and IGF-I or IGFBP-3 was observed. Four cross-sectional studies have investigated the relationship between tomato consumption [14,15,18] or lycopene intake [22] and IGF-I and IGFBP-3. In three of these studies, higher intake of cooked or processed tomatoes or lycopene was associated with either lower IGF-I levels [14], higher IGFBP-3 levels [22], or a lower IGF-I/IGFBP-3 molar ratio [18].…”

“…Results of cross-sectional studies investigating the relation between the normal range of energy and protein intake and serum IGF-I levels are conflicting [12][13][14][15][16][17][18][19][20], with only few studies showing a significant positive association [21][22][23][24]. Alcohol consumption can possibly increase IGF-I production by the liver [25], but has been inconsistently related to IGF-I and IGFBP-3 levels in cross-sectional studies [14,15,17,22,26].…”

“…Lycopene is an anti-oxidant mainly found in cooked or processed tomatoes, and has been suggested to interfere with IGF signaling in experimental studies [34][35][36]. Also in cross-sectional studies lycopene [22] or cooked or processed tomatoes [14,18] were associated with IGF-I or IGFBP-3.…”