We conducted a single-center retrospective analysis of 49 consecutive pediatric and AYA (adolescent and young adult) patients who underwent either T-cell receptor alpha/beta and CD19-depleted unrelated donor (URD) hematopoietic stem cell transplantation (HSCT) (TCR alpha beta/CD19-depleted URD HSCT; n = 22, including 13 matched unrelated donors [MUD] and 9 mismatched unrelated donors [MMUD]) or CD45RA(+)-depleted haploidentical (HAPLO) HSCT (CD45RA(+)-depleted HAPLO HSCT; n = 27) between 2018 and 2023 for malignant hematological disorders. Both groups showed comparable engraftment kinetics. Primary graft failure occurred in four patients (8.5 %), predominantly in the HAPLO group. Acute graft-versus-host disease (aGvHD) grade II-IV occurred in 72.9 % of patients, with the highest incidence in the MMUD subgroup (89 %; p = 0.03). Chronic GvHD (cGvHD) was observed in 42.6 % of patients, with a tendency toward a higher incidence in the MMUD group (67 % vs 38 % for MUD and 33 % for HAPLO; p = 0.08). The 2-year overall survival (OS) was 75.5 %, with no significant differences between groups (MUD 92.3 %, MMUD 77.7 %, HAPLO 66.6 %; p = 0.34). GvHD-free/relapse-free survival (GRFS) at 2 years favored MUD (51.9 %) and HAPLO (51.3 %) over MMUD (22.2 %), with significantly poorer GRFS in MMUD compared with HAPLO (p = 0.038). In multivariate analysis, disease status >= third complete remission (>= CR3) at HSCT was associated with significantly poorer OS (hazard ratio [HR] 13, 95 % confidence interval [CI] 1.2-130; p = 0.036), while sex mismatch (female donor to male recipient) was associated with inferior GRFS (HR 4.5, 95 % CI 1.6-12; p = 0.0035). Our results highlight the feasibility of both HAPLO and URD HSCT in children with malignancies, using distinct T-cell depletion strategies tailored to donor type. However, our findings also indicate that ex vivo T-cell depletion alone, without additional GvHD prophylaxis, is insufficient to prevent GvHD, underscoring the need for combined strategies to improve outcomes.