Background: The peak incidence of central nervous system (CNS) tumors in childhood indicates that intrauterine or neonatal characteristics are potential risk factors or symptoms of early onset of disease.Methods: We conducted a registry-based case-control study nested in the childhood populations of Denmark, Finland, Sweden, and Norway on the association between indicators of fetal growth and neonatal stress and childhood CNS tumor risk diagnosed during the period . Each of the 3,443 cases was matched individually on date of birth, sex, and country to five controls sampled randomly from population registries. Information on birth characteristics was obtained from national birth registries. We estimated odds ratios (OR) and 95% confidence intervals (95% CI) by conditional logistic regression analyses.Results: We observed a U-shaped relation between risk for CNS tumors and birthweight, at >4.5 kg (OR, 1.27; 95% CI, 1.03-1.55) and <2.0 kg (OR, 1.50; 95% CI, 1.13-1.99), the latter being attenuated after adjustment for gestational age. Moreover, small-for-gestational age (OR, 1.28; 95% CI, 0.98-1.66) and large-forgestational age (OR, 1.26; 95% CI, 1.02-1.55) were both associated with CNS tumors. The OR for preterm births was increased per 1-week decrease in gestational age (OR, 1.58; 95% CI, 1.04-2.44). Increased ORs were also observed for head circumference >38 cm (1.80; 95% CI, 1.18-2.74), 5-minute Apgar score <7 (1.44; 95% CI, 0.98-2.12), and breech presentation (1.33; 95% CI, 1.04-1.69). The observed associations varied little by histologic subgroup.Conclusions: This study supports intrauterine or neonatal onset of childhood CNS tumors. The findings provide insight into the natural history of childhood CNS tumors indicating an early onset or, alternatively, potentially harmful exposures in the neonatal period that might be preventable. Cancer Epidemiol Biomarkers