LAUSR

BACKGROUND The likelihood of pregnancy and risk of obstetrical or perinatal complications is inadequately documented in female survivors of pediatric cancer. METHODS We assembled a population-based cohort of female survivors of cancer diagnosed at < 21 years in Ontario, Canada between 1985-2012. Survivors were matched 1:5 to women without prior cancer. Multivariable Cox proportional hazards and modified Poisson models assessed the likelihood of a recognized pregnancy and perinatal and maternal complications. RESULTS 4,062 survivors were matched to 20,308 comparisons. Median (interquartile range [IQR]) age was 11 (4-15) years at cancer diagnosis and 25 (19-31) years at follow-up. By age 30, the cumulative incidence of achieving a recognized pregnancy was 22.3% (95% confidence interval [CI] = 20.7%-23.9%) among survivors' vs 26.6% (95% CI = 25.6%-27.3%) among comparisons (hazard ratio [HR] = 0.80, 95% CI = 0.75-0.86). A lower likelihood of pregnancy was associated with a brain tumour, alkylator chemotherapy, cranial radiation, and hematopoietic stem cell transplantation (HSCT). Pregnant survivors were as likely as cancer-free women to carry a pregnancy >20 weeks' (relative risk [RR] = 1.01, 95% CI = 0.98-1.04). Survivors had a higher relative risk of severe maternal morbidity (RR = 2.31, 95% CI = 1.59-3.37), cardiac morbidity (RR = 4.18, 95% CI = 1.89-9.24) and preterm birth (RR = 1.57, 95% CI = 1.29-1.92). Preterm birth was more likely in survivors treated with HSCT (allogenic: RR = 8.37, 95% CI = 4.83-14.48; autologous: RR = 3.72, 95% CI = 1.66-8.35). CONCLUSION Survivors of childhood or adolescent cancer are less likely to achieve a pregnancy and once pregnant, are at higher risk for severe maternal morbidity and preterm birth.