AUTHORS
Boula Gattas, MD, Xia Bi, MD2, Joanne Filicko-O'Hara, MD3, Joseph Van Galen, MD, MS4, William O'Hara, PharmD5, John Wagner6, Dolores Grosso7, Neal Flomenberg, MD8, Usama Gergis, MD, MBA9
AFFILIATIONS
1 Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, United States
2 Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, United States
3 Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, United States
4 Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, United States
5 Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, United States
6 Metrohealth, Cleveland, Ohio, United States
7 Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, United States
8 Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, United States
9 Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, United States
ABSTRACT
Background
The two-step allogeneic hematopoietic stem cell transplantation (HSCT) platform separates donor lymphocyte infusion (DLI) from CD34-selected stem cell infusion, enabling administration of a fixed donor T-cell dose while protecting hematopoietic stem cells from post-DLI cyclophosphamide. Patients receive DLI (2 × 10⁸ CD3+ T cells/kg) on day −6, cyclophosphamide (50–60 mg/kg/day) on days −3 and −2 to induce bidirectional donor–recipient immune tolerance, followed by infusion of CD34-selected stem cells 24 hours after the second cyclophosphamide dose. This platform has been successfully implemented using myeloablative (MAC), reduced-intensity (RIC), and non-myeloablative (NMA) conditioning. However, the impact of conditioning intensity on transplant outcomes within this platform remains unknown.
Methods
We retrospectively analyzed 387 consecutive patients who underwent two-step HSCT between 2006 and 2025 on IRB-approved clinical trials. All patients received mobilized peripheral blood stem cell grafts. Patients were grouped according to conditioning intensity: MAC (n=230), RIC (n=135), and NMA (n=22). MAC consisted of 12 Gy total body irradiation (TBI) delivered in six fractions. NMA consisted of fludarabine 30 mg/m² for three days plus 4 Gy TBI in two fractions. Most RIC patients received fludarabine 30 mg/m² for four days, 2 Gy TBI, and either busulfan 3.2 mg/kg for two days or melphalan 50 mg/m² for two days. Overall survival (OS), progression-free survival (PFS), relapse, non-relapse mortality (NRM), and graft-versus-host disease (GVHD) were compared across conditioning groups.
Results
The median age of the cohort was 56 years (MAC, 52; RIC, 60; NMA, 68), and the median HCT-CI was 3 (MAC, 2; RIC, 3; NMA, 4), reflecting the expected selection of older and more medically complex patients for lower-intensity conditioning. Acute myeloid leukemia (42%) and myelodysplastic syndromes (13%) were the most common indications for transplantation. Eighty-one percent of patients underwent haploidentical transplantation, while 19% received matched related donor grafts.
For the entire cohort, 1- and 5-year OS were 67.2% (95% CI, 62.6–72.1%) and 54.1% (95% CI, 49.2–59.5%), respectively. Corresponding PFS rates were 60.9% (95% CI, 56.2–66.0%) and 48.5% (95% CI, 43.7–54.0%). The 5-year cumulative incidences of NRM and relapse were 23.6% and 27.8%, respectively. Grade II–IV acute GVHD occurred in 29% of patients, while moderate-to-severe chronic GVHD occurred in 11%.
Despite significant differences in baseline age and comorbidity, conditioning intensity was not associated with differences in outcomes. Three-year OS was 59.0% for MAC, 53.6% for RIC, and 69.3% for NMA (p=0.41), while 3-year PFS was 52.5%, 45.5%, and 69.3%, respectively (p=0.14). NRM was comparable across all conditioning groups. Although relapse was numerically lowest in the NMA cohort (5% at 3 years), this difference did not reach statistical significance (p=0.17), likely reflecting the limited sample size (n=22).
On multivariable analysis, younger age (p=0.003) and MRD transplantation (p=0.003) were independently associated with improved overall survival. Conditioning intensity was not independently associated with OS, PFS, relapse, or NRM.
Conclusions
Within the two-step HSCT platform, de-escalation of conditioning intensity did not compromise OS, PFS, relapse, or non-relapse mortality despite treatment of substantially older and more medically complex patients. These findings support the concept that the efficacy of the two-step platform is driven predominantly by its controlled cellular immunotherapy rather than by conditioning intensity. Prospective studies are ongoing to validate these findings
DISCLOSURES
Usama Gergis, MD, MBA: BioNTech - Stock (publicly traded); IOVANCE Biotherapeutics, Inc. - Stock (publicly traded); Kite, a Gilead Company - Stock (publicly traded); Moderna, Inc. - Stock (publicly traded); Kyverna Therapeutics, Inc. - Honoraria; Autolus Therapeutics PLC - Honoraria; Kite, a Gilead Company - Honoraria; Bristol-Myers Squibb Company - Honoraria; Kite, a Gilead Company - Consultancy (Includes expert testimony); Autolus Therapeutics PLC - Consultancy (Includes expert testimony); Kyverna Therapeutics, Inc. - Consultancy (Includes expert testimony); Bristol-Myers Squibb Company - Consultancy (Includes expert testimony)