Reply

Abstract

In Response: We thank Lagiou and colleagues for their insightful letter in which they discuss indirect evidence in support of a possible negative association between circulating levels of insulin-like growth factor II (IGF-II) and breast cancer risk. There is insufficient direct evidence to determine whether or not IGF-II levels influence the risk of breast cancer. To our knowledge, only two small prospective studies have examined the role of IGF-II levels in breast cancer risk with inconsistent results (1, 2). Similarly, few studies have examined the role of IGF-II in mammographic density. Lagiou and colleagues discussed our finding of an unexpected negative association between IGF-II levels and density in premenopausal women, but, as we highlighted in our article, this finding should be interpreted with caution because the association disappeared on adjustment for body mass index. Moreover, no association between IGF-II levels and density was found in another study of premenopausal women (3), although findings from studies conducted among postmenopausal women are more consistent with a possible inverse association (3, 4).Lagiou and colleagues reported higher levels of umbilical cord blood IGF-II in Asians compared with Caucasians. Interestingly, cross-sectional studies suggest that such ethnic gradients may not persist into adulthood, with young adult Asians having significantly lower circulating levels of this growth factor than Caucasians (5), perhaps reflecting ethnic differences in body composition and other lifestyle variables. Similarly, Afro-Caribbean postmenopausal women, known to have a lower breast cancer risk than Caucasians, have been shown to have lower levels of IGF-II (4). In fact, ethnic difference in circulating levels of IGF-II, as well as in levels of IGF-I, IGF-binding protein 3 (IGFBP3), and sex steroid hormones, explained little of the ethnic gradient in mammographic density, which, as expected, was higher among Caucasian than Afro-Caribbean women (4). Unfortunately, no similar comparisons have been carried out among premenopausal women.We agree with Lagiou and colleagues that the associations between IGF-II, mammographic density, and breast cancer risk have thus far been vastly overlooked, and based on the findings of studies of IGF-I and IGFBP3, they should be investigated separately in pre- and postmenopausal women.No potential conflicts of interest were disclosed.