From Pan-T-Cell Depletion to Selective Graft Engineering: Progress, Limitations, and Open Questions

Abstract

The success of allogeneic haematopoietic stem cell transplantation is limited by three linked problems: graft-versus-host disease (GVHD), relapse, and infection. Pan-T-cell depletion, introduced in the 1980s, controlled GVHD but at the cost of relapse and impaired immune reconstitution. This trade-off has driven the field toward selective graft engineering, manipulating the donor graft’s cellular composition to separate GVHD-causing cells from those responsible for graft-versus-leukaemia (GVL) activity and infection control. This review traces that evolution, from CD34+ selection through αβ/CD19 depletion, naïve (CD45RA+) T-cell depletion, adoptive add-back, and gene-edited donor grafts, evaluated against current in vivo prophylaxis, principally post-transplant cyclophosphamide (PTCy). Selective graft-engineering platforms have achieved low rates of chronic GVHD and non-relapse mortality in defined settings. More recently, TREGZI (developed as Orca-T) received U.S. Food and Drug Administration (FDA) approval in June 2026 for adults with haematological malignancies undergoing matched-donor myeloablative allo-HSCT, following a randomised phase III trial against tacrolimus/methotrexate rather than PTCy. However, evidence for most other platforms remains dominated by single-arm, paediatric, or single-centre series; adult data are sparse, and no randomised trial has directly compared ex vivo graft engineering with optimised PTCy-based prophylaxis. Manufacturing infrastructure, cost, and access further limit adoption relative to established prophylaxis. Randomised trials are needed to determine whether, and for which patients, selective graft engineering offers a survival advantage.