In this issue of Blood Advances, through a multicenter analysis of 367 pediatric patients, Horgan et al have reported that T-replete umbilical cord stem cell (TRCB) transplants against myeloid disease provide superior disease-free survival and event-free survival (EFS) compared with that provided by all other donor sources, including matched unrelated donor, matched sibling donor, mismatched unrelated donor, haploidentical, and T-depleted cord.1 Similar to other reports, in young and adult patients from single institutions, the power of cord-derived T cells during transplantation appeared greatest in those children with measurable residual disease (MRD) before transplant.2 Horgan et al found similar EFS and overall survival in patients from all groups who had no MRD before transplant, including those who received transplants using other alternative donor sources, including haploidentical grafts. This last finding supports that of others in which the absence of this MRD positive cord-derived graft-versus-leukemia (GVL) benefit normalizes TRCB overall efficacy when compared with other donor stem cell sources.2,3 These results beg the question of why cord-derived T cells have such robust antitumor reactivity in patients with positive MRD and what are we willing to pay for it?
Transplanted T cells are critical to durable antileukemia immune surveillance and infection control early posttransplant but also lead to graft-versus-host disease (GVHD). The balance between GVHD and GVL may be different when using umbilical cord grafts compared with other stem cell sources. Historical studies using single- vs double-cord transplants found that 2 cords trended towards improved relapse-free survival and, despite a doubling of the overall T-cell dose, no increase in grade 2-4 acute GVHD (aGVHD).4 In this study, Horgan et al find that patients who underwent TRCB transplants had increased aGVHD supporting the age-old and bittersweet association between GVHD and GVL. Similar to other retrospective analyses, the increased aGVHD after TRCB transplantation dampens its transplant-related nonrelapse mortality.5 However, although children who received TRCB had significantly increased aGVHD, they also had significantly less chronic GVHD.1 Horgan et al once again focus the spotlight on important questions regarding cord transplants in general. What is the reason for the apparent immediate and durable GVL effect maintained by cord-derived T cells? Why do cord-derived cells induce increased aGVHD despite having numbers that are at least fivefold less than that of traditional bone marrow transplants?
Cord-derived T cells are different from those T cells that inhabit more “mature” graft sources, such as bone marrow or peripheral blood. In contrast to these sources, T cells from umbilical cord sources are primarily naïve.4 This naïve state initially raised concerns that they would not be as adept at providing GVL reactivity compared with that provided by more mature T cells. However, cord-derived T cells have been shown to undergo rapid CD4+ T-cell biased expansion after transplant in the absence of serotherapy, a phenomenon not observed after bone marrow or peripheral blood stem cell transplantation.6,7 This rapidly expanding T-cell compartment mediates significantly enhanced CD8+ T-cell–mediated GVL reactivity in preclinical models of human disease compared with adult T cells from peripheral blood or bone marrow.6 Based upon these inherent features, T cells from cord blood may be favorable over those from other stem cell sources, especially in patients with myeloid diseases known to be particularly responsive to allogeneic immune recognition.
Although promising, obstacles remain for the broad use of cord blood transplantation in patients of all ages, particularly for cases in which stem cell doses may be limiting and prolonged periods of early cytopenia can lead to infections.8 Limitations in cell dose can be overcome using double-cord transplants or through ex vivo cord expansion. In fact, the use of double cords in adults has shown similar success in patients with MRD positivity to those reported by Horgan et al.2 However, serious opportunistic infections and early viral reactivation remain major barriers to cord blood’s extended use, despite the absence of serotherapy and irrespective of single or double cord use.9 Ex vivo expansion can reduce risks of graft failure and lessen prolonged periods of cytopenia. Nevertheless, given the results by Horgan et al and those of others, such manipulation should also allow for the preservation of “natural” cord-derived T cells in an effort to retain their potent GVL effect.
Although limitations to this and other studies exist based on their retrospective nature that precludes random assignment of patients, there is mounting evidence across the globe that TRCB transplants for patients with measurable myeloid disease works best. For the time being, this must be weighed with increases in transplant-related morbidity and mortality including increased risk of aGVHD and infections. Perhaps individualized in vivo T-cell depletion with anti-thymocyte globulin (ATG) is 1 method of balancing GVHD and GVL in the setting of cord transplants while still allowing robust and timely T-cell reconstitution and infection control. Indeed, the timing and dose of in vivo T-cell depletion with ATG significantly affects early and late posttransplant immune reconstitution and rates of aGVHD.10 As management for GVHD and infections continue to advance, we may be nearing a time in which a randomized trial can answer these questions once and for all. Such a study would be particularly critical for patients who do not have matched donor options or for whom the time to transplant is short and alternative donor strategies are rapidly needed. We may be nearing a reality in which, at least when it comes to patients with MRD+ myeloid disease, TRCB transplants will be considered the best available therapy rather than an “alternative” strategy.
Conflict-of-interest disclosure: J.L.L. declares grant funding from AbbVie.
