We have undertaken a study to examine whether the difference in breast cancer incidence between 2 populations of similar genetic background is reflected in a similar pattern of estrogen receptor ␣ expression in normal mammary gland. Study participants were 92 Japanese women from Sapporo, Japan (mean age 48.2 years) and 49 Japanese women from Honolulu, Hawaii (mean age 45.4 years), who underwent biopsy indicating normal breast tissue or benign, nonproliferative breast disease in hospitals in Sapporo, Japan and Honolulu, Hawaii. The breast tissue samples were formalin-fixed and paraffin-embedded. The estrogen receptor immunohistochemistry assays were conducted using Dako kits. Japanese women in Hawaii, who have a higher incidence of breast cancer compared with Japanese women in Sapporo, also had, as predicted, higher mean percentage of estrogen receptor ␣-positive normal mammary cells (2-tailed test, p ϳ 0.09). The results of our study are compatible with the hypothesis that estrogen receptor ␣ expression in normal mammary tissue increases breast cancer risk and they also indicate that the expression of these receptors is dependent, at least in part, on nongenetic factors. © 2002 Wiley-Liss, Inc. Key words: breast cancer; estrogen receptors; epidemiology; Japanese Endogenous estrogens are central in the etiology of breast cancer in women. [1][2][3][4] Nevertheless, no aspect of estrogen production, metabolism and physiologic effects can fully explain the higher incidence of the disease among Caucasian women in North America, northern Europe or Oceania in comparison with Japanese and Chinese women in their home countries. 5 Because estrogen receptors are mandatory for estrogen response, it has been hypothesized that expression of these receptors in normal breast tissue could be as important as estrogen levels for the determination of breast cancer risk. 6 -9 In a recent paper, we have reported that expression of estrogen receptor ␣ is much higher among Australian Caucasian women than among Japanese women in Japan, in parallel with the much higher breast cancer incidence among the former compared with the latter. 10 The incidence of breast cancer among Japanese women in Hawaii is at least twice as high as that among Japanese living in Japan, although not quite as high as the incidence of breast cancer among Caucasians. 5 This is a well-established pattern for migrants from low to high breast cancer risk countries and their descendants. 11-13 Indeed, this pattern has been used to argue that a large fraction of the variation of breast cancer incidence in women is of exogenous, rather than genetic, etiology. We undertook a study to examine whether the higher breast cancer incidence of Japanese in Hawaii in comparison with Japanese in Japan is reflected in a similar pattern of estrogen receptor ␣ expression in normal mammary gland. This, if detected, would support the hypothesis that the expression of these receptors in normal breast is a risk factor for breast cancer and would also indicate that the expression of t...