Phase II Trial of Reduced-Intensity Fludarabine, Melphalan, and Total Body Irradiation Conditioning Results in Favorable Outcomes with Haploidentical Donor Peripheral Blood Stem Cell Transplant

Introduction

High risk hematologic malignancies are curable with allogeneic hematopoietic cell transplant (HCT). Haploidentical (haplo) HCT with post-transplant cyclophosphamide (PTCy) has improved access to transplant in patients lacking an HLA matched donor. However, traditional reduced-intensity conditioning (RIC) with fludarabine/cyclophosphamide/total body irradiation (TBI) for haplo HCT with PTCy is marred by high rates of relapse. This study seeks to optimize the RIC haplo HCT with PTCy platform in older adults with a slight intensification in conditioning and the use of peripheral blood stem cell (PBSCT) grafts in lieu of bone marrow grafts (BMT). Additionally, tacrolimus was substituted by sirolimus based on prior phase II data suggesting favorable rates of graft-versus-host disease (GVHD) after haplo PBSCT with PTCy (Bejanyan, Blood Adv. 2021). We hypothesized that this approach could successfully reduce the risk of relapse and improve disease-free survival (DFS) without substantially increasing non-relapse mortality (NRM).

Methods:

This single arm, single center phase II trial of haplo PBSCT used a RIC regimen consisting of Melphalan 70 mg/m² IV on day -6, fludarabine 30mg/m² IV daily on days -6 to -2, and TBI 200 cGy on day -1 (NCT04191187). Subjects received haplo PBSCT with a target dose of 5 x 10⁶ CD34+ cells/kg and capped at 7 x 10⁶ CD34+ cells/kg. GVHD prophylaxis consisted of PTCy 50 mg/kg on days +3 and +4, mycophenolate mofetfil (MMF) 1 gram three times daily from days +5 to +35, and sirolimus initiated at a target trough of 8-14 ng/mL on day +5 and tapered off between days +100 and +180. Eligibility criteria required a hematologic malignancy as the indication for allogeneic HCT and either age >55 years or a hematopoietic cell transplant comorbidity index (HCT-CI) > 3. The primary endpoint of this study is DFS with a hypothesis that the experimental regimen would have an 18 month DFS of 60% compared to a historic DFS of 40% and 90% power to reject the null hypothesis.

Results:

A total of 34 subjects were treated on the trial. The median age was 66 years (range: 31-74 years). The HCT-CI score was > 3 in 41%. Acute myeloid leukemia (AML, 44%) was the predominant indication for transplant followed by myelodysplastic syndromes (MDS, 29%). Nine patients (26%) were from underrepresented racial/ethnic minorities (those other than non-Hispanic white). Five patients (15%) received tacrolimus for GVHD prophylaxis before a protocol amendment changed the regimen to sirolimus for the remaining 29 patients (85%). The median follow-up time for 25 surviving patients was 28.5 months . The trial met the primary endpoint with 18-month DFS of 70.5% (90% CI ­> 59.1%, p=0.002). We observed no primary graft failure events. Neutrophil engraftment by day 30 was achieved in 30 patients (88%), and platelet engraftment by day 60 was achieved in 28 patients (82%). The cumulative incidence of grade II-IV acute GVHD at day 100 was 11.8% (95% CI: 3.6-25.1%). The cumulative incidence of moderate to severe chronic GVHD at 18 months was 9.1% (95% CI: 2.2-22.0%). At 18 months, the cumulative incidence of NRM was 17.7% (95% CI: 7.0-32.2%) and of relapse was 11.9% (3.7-25.4%). GVHD-free relapse-free survival at 18 months was 61.4% (95% CI: 42.9-75.5%) and the overall survival at 18 months was 73.3% (95% CI: 54.9-85.1%).

Conclusion:

In this completed phase II trial of fludarabine, melphalan 70 mg/m², TBI with haplo PBSCT and PTCy/sirolimus/MMF, the primary endpoint was met with an 18-month DFS that was superior to historical control. This was attributable to low rates of relapse without excessive NRM. Acute and chronic GVHD rates were favorable. Pharmacogenomic and pharmacokinetic correlative data is pending and will be included in the final presentation. These promising results merit further investigation to potentially establish this platform as a new standard approach to RIC haplo PBSCT.