Cord blood transplant: the glass is half full—can we do better?

Goldstein et al., in this issue of Leukemia and Lymphoma, present their single-center results on umbilical cord blood stem cell transplant (UCB SCT) [1]. Over a 17-year period, 38 patients underwent transplant at a single institution in Israel. Twenty-nine were children and received a single cord, often from a matched sibling and often for non-malignant disorders. Nine adults received double UCB transplant from unrelated donors for hematologic malignancies. The results are illustrative of the problems and successes of UCB SCT in its first two decades. Of the 38 patients, more than a quarter failed to engraft. For the remainder, the average time to engraftment was 22 days, with some requiring up to 40 days. Such rates of non-engraftment and delayed engraftment lead to high rates of morbidity and mortality. Overall, 16 of the 38 transplants were considered failures because of graft failure with autologous recovery (n = 5), fatal graft failure (n = 5), or other treatment related deaths (n = 6). But for those who did engraft, relapse rates were low, and chronic graft versus host disease (GVHD) affected only one of 24 evaluable patients. About 60% of the patients were long-term survivors, and about half were cured of their disease. Few, if any, of the patients would have survived without a transplant.

Such outcomes are representative of the problems and potential of UCB SCT. Engraftment remains slow and erratic, and graft failure is common. Early treatment related mortality remains higher than that for unrelated cord blood transplant. For example, the Minnesota group finds that UBC SCT is associated with higher early death rate than transplant from adult related or unrelated donors [2]. However, for those who survive until engraftment, relapse rates are low, as is the incidence of chronic GVHD. Relapse rates may be even lower after double UCB SCT [3].

How then can we improve the time to engraftment? Increasing the cell dose of UCB cells is unlikely to have a huge impact. The correlation between cell dose and time to engrafment is rather weak, because nucleated cell count or CD34 counts are imperfect surrogates of stem cell content [4]. So, using three or four UCBs is not likely to have a major impact. Instead, efforts focus on increasing cell types with high proliferative content. In this regard, several fascinating studies of stem cell expansion are ongoing. Araki et al. were among the first to show that exposure of UCB stem cells to a chromatin modifying agent results in stem cell expansion [5,6]. More recently, Boitano et al. identified an agent labeled StemRegenin1 (SR1) that by antagonizing aryl hydrocarbon receptors promotes expansion of hematopoietic stem cells [7]. Delaney et al. used Notch ligand to expand UCB cells that then were coinfused with a non-manipulated UCB unit [8]. While not always successful, neutrophil recovery occurred extremely fast in those where the expanded unit predominated. Nevertheless, stem cell expansion, though rapidly evolving, still faces considerable hurdles before becoming widely applicable [9]. A simpler and more direct approach consists of the combination of adult and UCB cells. Here, T-cell depleted haploidentical cells are combined with UCB cells. The adult cells engraft rapidly and reliably, but are over time outcompeted by the UCB cells. This technique, originally pioneered by a Spanish group, is simple, can be rapidly adopted, and has yielded quite promising results [10]. In our own experience, it leads to rapid engraftment in the majority of patients and much reduced early toxicity [11].

Even with this type of haplo-cord transplant, SCT graft rejection remains a vexing and dangerous problem affecting a considerable proportion of patients. We have observed it in about 10% of our patients; it occurred in 10 of 38 in the present study. Graft rejection is a complex process mediated in part by host lymphocytes and influenced in part by human leukocyte antigen (HLA) mismatching [12]. It can be affected by increasing host immunosuppression or by increasing the stem cell dose [13]. Recently, the presence of donor specific antibodies (DSAs) in the transplant recipient has been shown to be an important risk factor for graft rejection. HLA antibodies acquired through pregnancy or previous transfusions are frequently detected in transplant candidates. In our own experience, they are present in about 20% of those with hematologic malignancies, and may be more common in patients with hemoglobinopathies. If such antibodies are directed against the donor (DSAs), the risk for graft rejection increases dramatically. Ironically, this was demonstrated almost 30 years ago in adult transplants [14], but has only recently been rediscovered in mismatched adult and UCB SCT [15,16]. It is debated whether graft rejection in this case is antibody mediated or whether the antibodies are merely an indicator of a more complex cell-mediated process [17]. In any case, if HLA antibodies are present in a potential transplant recipient, one should avoid UCB units with HLA types that are targeted by pre-existing HLA antibodies. This then requires complete typing of the UCB units, including typing for all class I (HLA A, B, and C) and all class II (HLA DR, DQ, and DP). In our experience, a non-targeted unit can be identified in the large majority of cases.

UCB SCT has come a long way since the pioneering reports from Gluckman and Broxmeyer et al. [18]. Much remains to be done, but the future is bright.

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 .