Breast-feeding and risk of childhood acute lymphoblastic leukemia: Results and commentary from a case-control study

PURPOSE: To test the hypothesis that breast-feeding mediates the occurrence of childhood ALL as a result of a rare, abnormal response to a common infection in the Northern California Childhood Leukemia Study. METHODS: Incident cases of leukemia (age 0–14) were rapidly ascertained from 1995-1999. Controls were randomly selected from the statewide birth registry and were 1:1 matched to cases on date of birth, sex, maternal race, maternal “Hispanicity,” and maternal county of residence at birth. Breast-feeding data were obtained by self-administered questionnaire and in-home interviews. Using conditional logistic regression adjusting for household income and birth weight, 140 case-control pairs were analyzed. RESULTS: Compared to no breast-feeding, breast-feeding showed a non-significant, higher risk of ALL: ever breast-fed (odds ratio (OR) = 1.16; 95% confidence interval (CI): 0.58–2.33), breast-fed less-than-or-equals, slant6 months (OR = 1.04; 95% CI: 0.48–2.28), and breast-fed >6 months (OR = 1.31; 95% CI: 0.59-2.91). Compared to formula-fed only, breast-feeding exclusively produced no discernible pattern in ALL risk: less-than-or-equals, slant3 months (OR = 0.85; 95% CI: 0.37–1.96), 4–6 months (OR = 1.80; 95% CI: 0.68–4.75), 7–12 months (OR = 1.23; 95% CI: 0.49–3.05), and greater-or-equal, slanted3 months (OR = 0.81; 95% CI: 0.24–2.79). CONCLUSION: In the current literature on breast-feeding and its role in mediating childhood ALL, two recent large studies were conducted in the United States and the United Kingdom. Both studies drew differing conclusions on the effect of breast-feeding on childhood ALL. Although this study was comparatively smaller, the results do suggest further methods of elucidating the role of breast-feeding in the occurrence of childhood ALL via an infectious pathway.