Soy Formula and Breast Cancer Risk

To the Editor: Infant soy formula has been available since 1929,1 and up to 36% of infants consume it in their first year.1,2 Like other soyfoods,3 it contains high levels of isoflavones,4 a class of phytoestrogens whose actions may reduce breast cancer risk.5–8 No epidemiologic study has examined infant soy consumption and breast cancer risk, despite suggestions of an inverse association with soy intake in adolescence,6–8 and in rodents.9,10 We examined this association in a population-based case-control study evaluating phytoestrogen intake during the life course.8 Registry-identified women aged 25–74 years were diagnosed with breast cancer between June 2002 and April 2003; controls were matched within 5-year age groups. Subjects were mailed questionnaires in which maternal contact was requested. Mothers were mailed questionnaires with 2 items querying what the subjects were fed during their first 4 months and months 5–12. Feeding options were: only breast milk; only cow's milk formula; only soy formula; breast milk and cow's milk formula; breast milk and soy formula; cow's milk formula and soy formula; other. The University of Toronto Research Ethics Board granted study approval. Since isoflavones are substantially higher in soy formula than other feeding options,4 “only soy formula” was retained as a separate variable. Options excluding soy were combined, and combinations including soy were assigned to the “other” category because the extent of soy intake was unknown. Final multivariate models, adjusted for identified confounders and age group, were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs). Of the 75% (3099/4109) of cases and 85% (3474/4098) of controls returning questionnaires, 60% (cases) and 54% (controls) reported their mothers were dead, and 13% consented to maternal contact. Among the mothers contacted, 89% (372/418) of case and 81% (356/437) of control mothers completed questionnaires. Feeding only soy formula during the first 4 months was associated with reduced odds of breast cancer (OR = 0.42; 95% CI = 0.13–1.40). The odds associated with soy formula during months 5–12 were also reduced (OR = 0.59; 95% CI = 0.18–1.90) (Table 1). However, both of these estimates lacked precision, and the CI around each included 1.0.TABLE 1: Distribution of Breast Cancer Cases and Controls, Multivariate Odds Ratio (OR) Estimates and 95% Confidence Intervals (CI) Associated With Type of Milk Consumed During the First 4 or 5–12 Months of InfancyAs the first epidemiologic study to examine this association, suggestions of reduced risk are intriguing. A recent meta-analysis of breast cancer and adult soy/isoflavone intake suggested reduced risk,5 as did 3 studies reporting adolescent exposure.6–8 Soy intake by infants has never been examined, although rodent studies suggest neonatal exposure reduces carcinogen susceptibility by inducing mammary gland differentiation.9,10 No study has examined breast cancer risk and aspects of infant feeding other than breast-feeding, for which findings are inconsistent.11 Breast-feeding of our subjects did not affect risk, and likely provided negligible isoflavone exposure4: probably few mothers ate soy since fewer than 5% of their daughters consumed soy during adolescence.8 Subjects were born in 1930–1977; soy formula was available, but its limited use suggests that higher rates reported elsewhere1,2 have developed more recently. The study is limited by inadequate statistical power (low soy formula use, few mothers recruited). Younger women may have had more opportunity for exposure, as soy formula use has increased over time. Response bias is unlikely as maternal response was high, and recall bias is also unlikely since infant feeding is not a known cancer risk factor. Few studies have explored validation of infant feeding by maternal recall beyond 2 decades, or of practices other than breast-feeding.12,13 Breast-feeding recall may be valid up to 50 years later; formula recall may be less so, although no studies have validated formula feeding duration or soy formula specifically.12,13 Our findings warrant investigation in larger breast cancer studies that include measurement of soy formula feeding. Recent reports1,2 of higher prevalence of soy formula use may also improve the power to detect an association, if one indeed exists. ACKNOWLEDGMENTS Supported by grant 13572 from the Canadian Breast Cancer Research Alliance with special funding support of the Canadian Breast Cancer Foundation Ontario Chapter. We thank the study coordinator, Noori Chowdhury, and staff, Leah Palma and Razia Sultana, for their dedication, and Catherine Mills for conducting the multivariate statistical analysis. Beatrice A. Boucher Division of Preventive Oncology Cancer Care Ontario Toronto, Canada [email protected] Michelle Cotterchio Division of Preventive Oncology Cancer Care Ontario Department of Public Health Sciences University of Toronto Toronto, Canada Nancy Kreiger Division of Preventive Oncology Cancer Care Ontario Departments of Public Health Sciences and Nutritional Sciences University of Toronto Toronto, Canada Lilian U. Thompson Department of Nutritional Sciences University of Toronto Toronto, Canada

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.