Haploidentical stem-cell transplant: the challenge of immune reconstitution

Allogeneic stem-cell transplant (SCT) is a potentially curative therapy for a variety of hematological malignancies. However, only about 70% of SCT candidates have a human leukocyte antigen (HLA)-matched sibling or a fully matched unrelated donor available, depending on ethnic background. The rest rely on an alternative stem-cell source such as a mismatched unrelated donor, umbilical cord blood or a haploidentical related donor [1].

Early trials of haploidentical transplants have failed due to unacceptably high rates of graft failure, graft-versus-host disease (GVHD) and non-relapse mortality (NRM) [2]. An intense bi-directional allogeneic response is exerted by a relatively high frequency of T-cells that recognize major HLA disparities between donors and recipients. The breakthrough came from a series of clinical trials in Perugia, Italy [3]. GVHD was almost completely prevented by extensive T-cell depletion of the donor graft, using CD34 + positive selection, and reducing the infused T-cell dose to less than 2 × 10⁴/kg, with no need for additional post-transplant pharmacological immune suppression. The HLA barrier to engraftment was crossed by using an intensive conditioning regimen and infusion of a mega-dose (> 10 × 10⁶/kg) CD34 + cell-containing graft. The high numbers of CD34 + cells possessed a veto effect against host anti-donor cytotoxic T-cells surviving the conditioning regimen, and constant engraftment was achieved. T-cell depletion is often associated with increased relapse rates after transplant due to loss of the graft-versus-leukemia (GVL) effect. However, in the HLA-mismatched setting, natural-killer (NK) cells can provide a strong anti-leukemia effect in the absence of GVHD. Selecting a donor with predicted NK alloreactivity in the donor-versus-host direction is feasible in a large proportion of transplants, allowing control of relapse. The major remaining obstacle to this extensively T-cell depleted mega-dose CD34 +approach is a very slow immune reconstitution after transplant, resulting in high rates of opportunistic infections such as viral (e.g. cytomegalovirus [CMV]) and fungal infections, resulting in high NRM. Clearly, enhancing immune reconstitution is the major challenge for improving outcomes after haploidentical SCT.

Several approaches have been made attempting to accelerate immune reconstitution [4]. One approach is to use a different method of T-cell depletion that will not deplete all immune cells. For example, negative selection by anti-CD3/CD19 beads depletes T-cells but retains NK cells, monocytes and dendritic cells that may facilitate immune reconstitution. Similarly, negative selection of αβ T-cells allows retaining γδ T-cells that do not cause GVHD but have immune effects. However, the experience with these methods is still limited. A second approach is to infuse after transplant pathogen-specific T-cells constructed against CMV, Aspergillus and other pathogens that do not cause GVHD. Of alloreactive cells have also been explored to improve immunity with no GVHD. Another option is infusion of T-cells engineered to express suicide genes that can be switched on should GVHD occur. More recently, researchers in Perugia co-infused regulatory T-cells together with conventional T-cells. This approach promoted immune reconstitution including pathogen specific immunity, with no GVHD and preserved GVL. However, all these methods are cumbersome, costly and require high expertise and are not widely applicable to most transplant centers.

Over the last few years renewed interest has emerged in T-cell replete haploidentical transplant. The first approach pioneered by Chinese researchers used granulocyte-colony stimulating factor primed non-T depleted bone marrow with intensive pre- and post-transplant immune suppression [5]. The second method uses high-dose cyclophosphamide given a few days after transplant to specifically deplete in vivo proliferating alloreactive T-cells and prevent GVHD and graft rejection while retaining immunity against infectious pathogens and malignant cells [6]. Both these methods have shown predictable engraftment and limited GVHD. The T-cell depleted and replete approaches have not been prospectively compared. Ciurea et al. showed in a small comparative series a better survival for the T-replete methods, with less NRM and better immune reconstitution [7].

In this issue of Leukemia and Lymphoma, Chang et al. report their experience with 78 haploidentical transplants for patients with myelodysplastic syndrome (MDS) and secondary acute myeloid leukemia (AML), using the T-replete Chinese approach [8]. They show that the absolute lymphocyte count on day 30 post-transplant (ALC-30) can predict outcome after transplant. Patients with rapid lymphocyte reconstitution had a lower incidence of relapse and bacterial infections and better survival. Seventy-nine percent of patients with ALC-30 > 300 cells/μL survived leukemia-free compared to 40% of patients with ALC-30 < 300 cells/μL. This retrospective analysis did not determine which lymphocyte subsets within the ALC are specifically associated with this observation. Prior studies suggested that early expansion of alloreactive NK cells may have a role, similarly to the T-depleted haploidentical approach, but other T-cell subsets may also be involved.

Testing immune reconstitution is difficult, and there are no widely accepted guidelines to the specific tests to be used. The ALC-30 has long been shown to be a simple test of immune recovery that can predict outcome after autologous SCT as well as allogeneic and haploidentical transplants [9]. It is easy, reproducible, requires no special expertise and can be analyzed retrospectively on a large scale. However, it is not clear whether a low ALC-30 is only a surrogate marker for patients who are going to have a worse outcome due to other patient or donor characteristics or whether manipulations to enhance ALC-30 or to intervene in patients with a low ALC-30 may be beneficial. It seems reasonable to prospectively explore graft engineering with various doses of T cells in the T-replete method and identify the dose associated with higher ALC-30 and best outcome. The measurement of ALC-30 has practical significance in applying interventions to facilitate immune reconstitution as described above for T-depleted transplants. Results for patients with high ALC-30 are very favorable, and probably no further intervention is needed, as the risk may overweigh the benefit. However, it is reasonable to explore these novel interventions in the high-risk group with poor lymphocyte recovery. Recent advances in haploidentical transplant make it a valid option with overall outcome that is similar to those with the other alternative stem-cell sources [1]. Haploidentical transplants have the advantage of rapid availability for almost every patient, the opportunity to proceed to transplant with no delay of donor search, and the option of post-transplant cellular therapies. The non-T-depleted approach is promising mainly due to simplicity and reduced costs, and for been widely feasible in non-specialized transplant centers; however, promoting immune reconstitution remains the major challenge.

Potential conflict of interest:

A disclosure form provided by the author is available with the full text of this article at www.informahealthcare.com/lal.