Pregnancy outcomes in women conceiving after HSCT for childhood or adolescent leukemia

Abstract Background: Myeloablative conditioning delivered before childhood or adolescent hematopoietic stem-cell transplantation (HSCT) reduces adult uterine volume, yet the real-world consequences for later fertility and obstetric health have remained uncertain. To clarify these effects, we assessed pregnancy, obstetric and neonatal outcomes in female leukemia survivors, distinguishing between those who had received total-body irradiation (TBI) or high-dose alkylating chemotherapy as conditioning, and we compared their results with outcomes after conventional chemotherapy without HSCT. Methods: This was a multicenter study, nested within the LEA cohort, which if the French long-term leukemia survivorship program. All women aged at least 18 years who reported at least one clinical pregnancy were eligible. Forty-one women conceived after HSCT and were matched 1:2 by age with 76 women treated only with chemotherapy. Altogether 117 women contributed 266 pregnancies: 86 after HSCT—55 following TBI and 31 after alkylating conditioning—and 180 after chemotherapy alone. Primary endpoints were live-birth rate and a composite measure of obstetric or perinatal morbidity; secondary end-points included mode of conception, the full spectrum of pregnancy losses and maternal complications. Standard Chi2/Fisher and t-tests were applied with a two-sided α of 0.05. Results: Despite a high prevalence of premature ovarian insufficiency, 70 % of post-HSCT pregnancies occurred spontaneously and 29 % followed oocyte donation. Eight per cent of HSCT pregnancies were unintended and ended in elective termination. Among intended pregnancies, the live-birth rate was 48 % after HSCT versus 82 % after chemotherapy alone. The deficit was driven by TBI: live-birth rate reached only 37 % after TBI but 70 % after alkylating conditioning. TBI was also associated with markedly higher late-pregnancy loss (24 % vs 3 %), preterm delivery (50 % vs 5 %), low-birth-weight infants (33 % vs 0 %) and postpartum hemorrhage (37 % vs 5 %). Women exposed to TBI required hospitalization during 84 % of pregnancies, compared with 5 % after alkylators and 10 % after chemotherapy alone. By contrast, pregnancies after alkylating conditioning closely resembled those after conventional chemotherapy in live-birth rate (70 % vs 82 %), distribution of pregnancy losses, obstetric complications and neonatal outcomes. Conclusions: High-dose alkylating-agent conditioning yielded maternal and perinatal outcomes similar to conventional chemotherapy, countering prevailing concerns derived from reduced uterine volume seen on imaging. In contrast, TBI markedly compromises later reproductive success and amplifies obstetric and perinatal risks, so pregnancies in these survivors should be managed in tertiary-level high-risk units. Routine reproductive guidance remains essential, because nearly one in six HSCT pregnancies was unintended, demonstrating that residual fertility—although unpredictable—persists even in women with prior ovarian failure. These data may inform reproductive counseling and obstetric surveillance in female leukemia survivors treated with HSCT during childhood and adolescence.