High Survival in Patients with Myelofibrosis Undergoing a Two-Step Approach to Hematopoietic Stem Cell Transplantation (HSCT): Effects of Donor Source and Pre-HSCT Splenic Therapy on Outcomes

Introduction: Hematopoietic stem cell transplant (HSCT) remains the only curative therapy for myelofibrosis (MF). However as compared to other hematologic malignancies, HSCT for MF is associated with an increased risk of graft failure due to a limited marrow niche and splenic sequestration of allogeneic progenitors. We report on engraftment and other outcomes in 12 consecutive patients with MF undergoing matched (M-HSCT) and haplo-identical (HI-HSCT) utilizing a two-step T cell-tolerization approach.

Methods: Regardless of donor source, all patients were conditioned with our institution’s two-step approach. After reduced intensity conditioning consisting of fludarabine, 2 Gy total body irradiation, thiotepa (n=3) OR busulfan (n=8), and in 1 case myeloablative conditioning with 12 Gy total body irradiation, patients received an unmanipulated donor lymphocyte product (DLI) containing 2 x 10⁸/kg CD3+ cells. After 2 days, cyclophosphamide (CY) 60 mg/kg was administered daily x 2 for bidirectional T cell tolerization. One day after CY completion, a CD34-selected stem cell product was infused. All patients received mycophenolate mofetil and tacrolimus beginning on d-1. Pre-conditioning splenic radiation was added starting in 2017 to decrease splenic sequestration of donor cells. This change affected the last six patients in the analysis.

Results: Six patients had primary MF and six patients had secondary MF. Donor source was a HLA matched (n=5, median age 54y, range 44-65) or haploidentical (n=7, median age 66y, range 47-67) relative. Eight patients received pre-HSCT radiation or splenectomy. Outcome data is summarized in the table. Median follow-up is 29 (range 6-95.6) months. Median time to neutrophil and platelet engraftment was 12 (range 9-13) days and 19 (13-40) days respectively. Patients undergoing pre-HSCT splenic therapy recovered platelets more rapidly than those without, median 19 vs 27 days, (p=0.059). All patients ultimately achieved 100% donor chimerism. While there was a trend in the M-HSCT group for higher donor T cell chimerism at d+28 in patients undergoing pre-HSCT splenic therapy, (p=0.083), donor source had the most significant impact on the pace of donor chimerism recovery. All 4 DLIs given in this group for fluctuating donor T cell chimerism occurred in M-HSCT recipients. At d+28 median donor T cell chimerism was 100% (range 91-100) in HI-HSCT versus 92% (range 67-100%) in M-HSCT recipients, (p=0.037). Probability of OS at 3 years was 90%. No patient receiving splenic therapy experienced disease progression. Two patients who received DLI experienced graft versus host disease (GVHD), grade 3 acute GVHD in one and extensive chronic GVHD in the other.

Conclusion: HSCT using the two-step approach is associated with rapid engraftment and excellent survival in patients with MF. Pre-HSCT splenic therapy was associated with more rapid platelet recovery. Unlike HI-HSCT, M-HSCT patients receiving RIC conditioning experienced initial fluctuating donor T-cell chimerism, a finding worth further exploration in larger trials.