We previously reported results of a French randomized clinical trial (RCT) comparing the risk of transplantation failure (including transplant-related mortality [TRM], engraftment failure, and autologous recovery) in single and double unrelated cord blood (UCB) transplantation in children and young adults with hematologic malignancies. We concluded that single-UCB transplantation with an adequate cell dose is the standard of care, leading to a 70% two-year overall survival (OS). It remains unclear, however, whether RCT participants have better outcomes than comparable patients not treated in the setting of a clinical trial. We compared the characteristics and outcomes of RCT participants (n = 137) to a Francophone population-based registry of patients (real-world [RW] group) fulfilling the eligibility criteria used in our RCT and transplanted with 1 or 2 UCB units after a myeloablative conditioning (MAC) regimen between March 2015 (end of inclusion in the RCT) and February 2019 (n = 141). The primary endpoint was the 2-year cumulative incidence (CI) of transplantation strategy failure as defined in our RCT. The 2 groups were comparable in terms of age, disease distribution, hematologic status at transplantation, follow-up, and HLA compatibility. Patients in the RW group were more likely to be transplanted with a single-unit UCB (87.9% versus 49.6%, P< .001) and to receive a radiation-free regimen (39.0% versus 60.6%, P< .001). The 2-year CI of transplantation strategy failure, TRM, and the 2-year probability of OS were similar between the 2 groups, although the relapse risk was higher in the RW group (31.2% ± 7.7% versus 20.4% ± 6.8%, P= .01), resulting in a significantly lower disease-free survival (DFS) (59.2% ± 8.4% versus 69.3% ± 8.0%, P= .047). This difference remained statistically significant only in the group of patients with acute lymphoid leukemia (ALL) who did not receive the conditioning regimen recommended by the RCT (fludarabine 75 mg/m², total body irradiation 12 Gy, cyclophosphamide 120 mg/kg). The results of our RCT appear to be reproducible in real-world conditions, provided that the same cord blood selection criteria and conditioning regimen are used.