Aim: To explore the validity and strength of evidence on the association between environmental factors and risk of developing childhood (0-14 years) cancer. Methods: An umbrella review was conducted including systematic reviews and meta-analyses of observational epidemiological studies that examined the association of any environmental exposure of either parent or child with any type of childhood cancer. PubMed and Scopus databases were searched until April 2020. Based on predefined criteria, the evidence was graded into strong, highly suggestive, suggestive or weak. Results: 509 meta-analyses explored environmental exposures and risk of developing 10 different types of childhood cancer. Only 2.4% of the associations were considered to have highly suggestive evidence inferred by strongly statistically significant results. These associations were confined to increased risk of overall leukemia, especially acute lymphoblastic leukemia (ALL), in relation to high birthweight, paternal smoking and exposure to pesticides, particularly insecticides. By contrast, maternal multivitamin supplementation during pregnancy (summary odds ratio [OR]: 0.64, 95% confidence intervals [CI]: 0.52, 0.80) and breastfeeding for more than 6 months (summary OR: 0.76, 95%CI: 0.68, 0.84) were supported by highly suggestive evidence for decreased risk of ALL. There was also highly suggestive evidence for increased risk of central nervous system (CNS) tumors, especially astrocytoma, in relation to high birthweight, as well as increased risk of testicular cancer in relation to testicular microlithiasis and isolated cryptorchidism. Conclusions: The present findings provide evidence that exposure to seven maternal/neonatal factors significantly affects the risk of childhood leukemia, CNS tumors and testicular cancer. Further evidence from sufficiently powered studies and large consortia with uniform reporting of analyses is needed to allow firmer conclusions to be drawn.

BACKGROUND Despite improvements in overall survival for pediatric cancers, treatment disparities remain for racial/ethnic minorities compared to non-Hispanic white; however, the impact of race on treatment outcomes for pediatric brain and central nervous system (CNS) tumors in the United States is not well known. METHODS We included 8713 children aged 0 – 19 years with newly diagnosed primary brain and CNS tumors between 2000 – 2015 from the Census Tract-level SES and Rurality Database developed by Surveillance, Epidemiology and End Results Program. We used Chi-square tests to assess differences in sociodemographic, cancer, and treatment characteristics by race/ethnicity and Kaplan–Meier curves and Cox proportional hazards models to examine differences in 10-year survival, adjusting for these characteristics. RESULTS Among 8,713 patients, 56.75% were non-Hispanic white, 9.59% non-Hispanic black, 25.46% Hispanic, and 8.19% from “other” racial/ethnic groups. Median unadjusted survival for all pediatric brain tumors was 53 months but varied significantly by race/ethnicity with a median survival of 62 months for Non-Hispanic whites, 41 months for Non-Hispanic blacks, and 40 months for Hispanic and Other. Multivariable analyses demonstrated minority racial groups still had significantly higher hazard of death than non-Hispanic whites; Hispanic [aHR 1.25 (1.18 - 1.31)]; non-Hispanic black [aHR 1.12(1.04 - 1.21)]; Other [aHR 1.22(1.12 - 1.32)]. Results were consistent when stratified by tumor histology. CONCLUSION We identified disparities in survival among racial/ethnic minorities with pediatric brain and CNS tumors, with Hispanic patients having the highest risk of mortality. Eliminating these disparities requires commitment towards promoting heath equity and personalized cancer treatment.