Abstract Background Maternal prenatal infections may affect fetal development, increasing the immunological vulnerability of offspring to childhood leukaemia (CL). The role of maternal infections in CL is unclear and might vary by subtype (lymphoblastic, ALL; myeloid, AML) or other characteristics. Understanding this potentially modifiable risk factor could inform prevention strategies. Methods Seventeen hospital- and population-based case–control studies of children born in 1972–2019 within the Childhood Cancer and Leukemia International Consortium with self-questionnaires or health-registry data on maternal infection were included (13 638 cases; 26 870 controls). Meta-analyses assessed CL and maternal infection (overall, viral, bacterial, respiratory, influenza/cold, urinary, genital) stratified by subtype, infection timing, race and ethnicity, and diagnosis age. Results The adjusted meta-analysis odds ratio (OR) for any maternal prenatal infection was 1.13 95% confidence interval (CI) 0.91–1.40, with similar estimates for ALL and AML. Infection-specific estimates varied. ORs for first-trimester infections were highest for CL and ALL, but not AML, although all CIs contained one. We found modest risk differences between White and Hispanic/Latino children, most notably for CL diagnosed at 2 years (White children: OR 1.25, 95% CI 1.02–1.53; Hispanic/Latino children: OR 0.79, 95% CI 0.34–1.81, subgroup difference P = .05), with similar differences for viral and respiratory/influenza/cold infections. Conclusion Although findings only modestly support an association between maternal prenatal infections and CL, some infections might increase the risk more markedly in young White children compared with Hispanic/Latino children. Risk patterns across race and ethnicity, type, and timing of maternal prenatal infection merit further investigation, as do studies with documented exposure information.
Piedvache et al. (Mon,) studied this question.