Abstract
BACKGROUND: Administration of chemotherapy during pregnancy is often delayed, while preterm delivery is common. If in utero exposure to chemotherapy is associated with adverse pediatric outcomes, it is unknown whether that relationship is directly attributable to the chemotherapy or is mediated by preterm birth. METHODS: Patients were identified from Canadian cancer registries and administrative data in Alberta, British Columbia, and Ontario, 2003-2017, with follow-up until 2018. The primary exposure was receipt of chemotherapy during pregnancy. Severe neonatal morbidity and mortality (SNM-M), neurodevelopmental disorders and disabilities (NDDs), and pediatric complex chronic conditions (PCCC) reflected short- and long-term pediatric outcomes. Modified Poisson and Cox proportional hazard regression models generated adjusted risk ratios (RR) and hazard ratios (HR), respectively. The influence of preterm birth on the association between exposure to chemotherapy in pregnancy and each study outcome was explored using mediation analysis. RESULTS: Of the 1150 incident cases of cancer during pregnancy, 142 (12.3%) received chemotherapy during pregnancy. Exposure to chemotherapy in pregnancy was associated with a higher risk of SNM-M (RR = 1.67, 95% confidence interval [CI] = 1.13 to 2.46), but not NDD (HR = 0.93, 95% CI = 0.71 to 1.22) or PCCC (HR = 0.96, 95% CI = 0.80 to 1.16). Preterm birth less than 34 and less than 37 weeks mediated 75.8% and 100% of the observed association between chemotherapy and SNM-M, respectively. CONCLUSIONS: Most children born to people with cancer during pregnancy appear to have favorable long-term outcomes, even after exposure to chemotherapy in pregnancy. However, preterm birth is quite common and may contribute to increased rates of adverse neonatal outcomes.