Preemptive and Therapeutic Donor Lymphocyte Infusion for Acute Leukemia or Myelodysplastic Syndromes: A Real-World...

Allogeneic hematopoietic cell transplantation (HCT) remains a potentially curative therapy for a broad range of hematologic malignancies. Early attempts to reduce graft-versus-host disease (GVHD), particularly through in vitro T-cell depletion, underscored the central role of donor immune cells in disease control but also increased infectious complications and compromised the graft-versus-leukemia (GvL) effect. Donor lymphocyte infusion (DLI) is a strategy to enhance the GvL effect by adoptively transferring donor-derived immune effector cells. This single-center, retrospective study at Hospital Israelita Albert Einstein in São Paulo, Brazil, evaluated clinical outcomes among patients receiving preemptive or therapeutic DLI after HCT. We included patients with acute myeloid leukemia (AML), myelodysplastic syndrome (MDS), or acute lymphoblastic leukemia (ALL) who received DLI after HCT between January 2010 and December 2025 for MRD+ or relapse. A formal comparison between preemptive and therapeutic DLI was not performed. Forty-five patients were included, with a median age of 47 years and a median follow-up of 18 months. Of these, 34 received therapeutic DLI, and 11 received preemptive DLI. The mean initial dose was 5 × 106 CD3/kg for therapeutic DLI and 1 × 106 CD3/kg for preemptive DLI. Patients with MSD received 1 × 107 CD3/kg; URD, 5 × 106 CD3/kg; and Haplo, 1 × 106 CD3/kg. Seventeen patients were receiving concurrent immunosuppression at the time of DLI. Overall, 34 of 45 patients (76%) also received additional disease-directed pharmacologic therapy. The response rate (RR) after the first DLI dose was 67% (30 of 45), with 82% (9 of 11) in the preemptive group and 62% (21 of 34) in the therapeutic group, including 23 MRD-negative responses (51%). Progression-free survival (PFS) at 1 year was 73% (95% CI 51% to 100%) with preemptive therapy and 38% (24% to 58%) with therapeutic strategy. Overall survival rates were 73% (51% to 100%) and 51% (36% to 72%), respectively. One-year relapse rates were 18% (5% to 64%) for preemptive therapy and 53% (39% to 73%) for therapeutic therapy. Nonrelapse mortality was 9% for both groups (4% to 23%). All patients were free of active GVHD. The cumulative incidence of clinically significant GVHD (grade III-IV acute GVHD or moderate-to-severe chronic GVHD) was 9% (1% to 59%) with preemptive treatment and 18% (9% to 37%) with therapeutic treatment. Of these, 58% had steroid-refractory disease. In conclusion, our findings support the effectiveness of preemptive, MRD-based DLI strategies. However, the durability of response remains strongly influenced by underlying disease biology. Therapeutic DLI for overt relapse continues to yield suboptimal results, reinforcing the need for clinical trials testing combined approaches. The role of prophylactic DLI was not addressed in the current study and can only be reliably assessed in randomized controlled trials. Future studies should focus on identifying patients who might benefit from more intensive disease monitoring.