Over the past two decades, much progress has occurred in clinical allogeneic hematopoietic cell transplantation (HCT). In a field where several logistical barriers complicate clinical trial design, the Blood and Marrow Transplant Clinical Trials Network (BMT CTN) has tenaciously and successfully conducted and published a multitude of randomized phase III clinical trials that have confirmed (or more commonly, refuted) single-arm phase II trial results and provided evidence-based, practice-guiding insights. For example, BMT CTN 0201, a randomized phaseIII study of myeloablative unrelated-donor transplantation comparing granulocyte colony-stimulating factor mobilized peripheral blood stem cells (PBSCs) to donor bone marrow, confirmed equivalent overall survival (OS) and relapse-free survival between the two cell sources, but importantly showed a significant increase in chronic graft-versus-host disease (GVHD) with the use of mobilized PBSC grafts.1 BMT CTN 0901, a randomized phase III trial of myeloablative versus reduced-intensity conditioning (RIC) in adults 65 years or younger with acute myeloid leukemia in complete remission, or controlled myelodysplastic syndrome, established myeloablative conditioning as the continued standard of care due to the excessively high relapse rate in patients receiving RIC.2 A more recent effort to clarify the most optimal “alternative” donor source in patients lacking a human leukocyte antigen (HLA) –matched donor led to BMT CTN 1101, a randomized phase III trial of RIC followed by either haploidentical or umbilical cord blood transplantation.3 In contrast to earlier phase II single-arm trials showing equipoise between these two alternative donor approaches, the results of BMT CTN 1101 clearly demonstrated the superiority of haploidentical transplantation for all patient outcomes and fueled the ongoing explosion of haploidentical donors in this setting.
It is therefore in keeping with the BMT CTN’s methods and mission of evaluating promising therapeutic approaches in multi-institutional trials to improve the outcomes of transplantation and cellular therapies (www.bmtctn.net), that BMT CTN 1301 was developed as a randomized phase III study to determine the preferred method of GVHD prevention in adults with acute leukemia and myelodysplastic syndrome who are undergoing myeloablative HLA-matched related or unrelated donor allogeneic HCT. The 1301 trial, led by co-chairs Drs. Leo Luznik, Marcelo Pasquini, and Miguel-Angel Perales, was established to compare the standard-of-care GVHD prophylaxis regimen of methotrexate (MTX) in combination with calcineurin inhibitor (CNI) to two investigational approaches that seemed to reduce GVHD rates in phase II clinical trials — donor T-cell depletion through ex vivo CD34+ selection or post-transplant cyclophosphamide (PTCy).4, 5 Of note, neither investigational arm included CNI; the PTCy was delivered as sole GVHD prophylaxis, and the CD34+ selected cells were administered after a conditioning regimen that included antithymocyte globulin with no additional GVHD prophylaxis. Importantly, patients randomized to either the standard MTX/CNI or the PTCy arm received a donor bone marrow graft, while patients randomized to receive CD34+ selection received mobilized PBSCs. The study used the relatively novel composite primary end point of chronic GVHD relapse-free survival (CRFS), in which success was defined as the absence of moderate or severe chronic GVHD, disease relapse, or death.
In Brief
Accrual began in September 2015, and 346 adults were enrolled across 26 centers by June 2017. Protocol-specified treatment was delivered to 300 patients, with protocol noncompliance highest in patients assigned to the CD34+ selection arm. In the intention-to-treat analysis, the primary endpoint of two-year CRFS was statistically equivalent among the three study arms and ranged from 41.0 percent to 50.6 percent. However, the key secondary end point of OS significantly differed across treatment groups, with two-year OS significantly lower in the CD34+ selection arm (60.1%; 95% CI, 50.1-68.8) relative to the PTCy (76.2%; 95% CI, 67.1-83.1) and MTX/CNI (76.1%; 95% CI, 67.1-83) arms. Although additional secondary end points of relapse and transplant-related mortality (TRM) did not significantly differ across treatment arms, two-year TRM was remarkably low following standard MTX/CNI (7.9%), and two-year relapse rates were relatively low across all arms, ranging from 13.9 to 29.6 percent. The results of the intention-to-treat analysis held in a post-hoc as-treated analysis.
The BMT CTN 1301 trial results are important for several reasons. First, this large, randomized controlled trial establishes the standard GVHD prophylaxis regimen of MTX in combination with CNI as an ongoing standard of care in myeloablative allogeneic HCT. Second, while not powered to show noninferiority, the CRFS and OS results suggest that PTCy can be considered a viable, but not superior, alternative to MTX/CNI in the HLA-matched donor setting. Third, the high rate of protocol treatment noncompliance in conjunction with inferior OS in the CD34+ selection arm suggests that center-specific expertise is likely required, and this approach should be limited to established centers that have demonstrated excellent outcomes with ex vivo T-cell depletion. Finally, the requirement for bone marrow grafts in the MTX/CNI and PTCy arms is noteworthy. While the BMT CTN 0201 trial demonstrating lower chronic GVHD with marrow relative to PBSC was limited to unrelated donor transplantation, in this trial, bone marrow was used as the stem cell source for both unrelated and matched sibling donors. It is unclear how the use of bone marrow grafts contributed to the lower-than-expected rates of TRM and GVHD in the standard MTX/CNI trial arm. Should we use these results to more seriously consider bone marrow grafts for all patients receiving myeloablative HLA-matched donor transplantation with standard GVHD prophylaxis, regardless of donor type? These questions can be answered, but to do so may require another BMT CTN–led clinical trial.
Competing Interests
Dr. Muffly indicated no relevant conflicts of interest.
