Autologous stem cell transplantation and allogeneic stem cell transplantation (SCT) are established treatment options for patients with acute myeloid leukemia (AML). There is a debate ongoing about the relative merits of each of these options in first complete remission (CR1). What is the preferred treatment and for whom? How does the choice of the SCT modality relate to the age and the considerable variability of prognostic risk of different subsets of leukemia among patients? Besides, in AML in CR2, treatment with autologous or allogeneic SCT appears to offer the best opportunities of salvage to the patient. Quite frequently a matched related sibling is not available to donate stem cells for an allogeneic transplant. In those circumstances, autologous transplants or, more recently, unrelated donor (URD) transplants offer alternative possibilities. The use of alternative donors is relatively new. As of today, few studies have been undertaken to critically evaluate URD transplantation.
The Center for International Blood and Marrow Transplant Research has reported on a large retrospective study dealing with the comparative value of autologous transplantation versus HLA-matched unrelated donor transplantation in AML. They analyzed the data of 668 autotransplants and 476 URD transplants that had been reported to their international registry. Autotransplants were associated with better three-year survival than URD transplants. This was true both in CR1 and CR2. The analysis reveals three-year adjusted survival rates of 57 percent (53-61) after autotransplants and 44 percent (37-51; p=0.002) after URD transplants for patients in CR1. Survival rates were 46 percent (39-53) and 33 percent (27-38; p=0.006) for patients in CR2, respectively. In univariate analysis, the relapse probability (at three years) was significantly reduced after URD transplantation as compared to autotransplantation (13 percent versus 40 percent, relapse at five years for CR1). However, the considerably higher treatment-related mortality (51 percent versus 10 percent at five years) reversed the superior antileukemia effect of URD transplantation. These trends were confirmed in multivariate analysis. The investigators report a profound effect of age. Recipients of URD transplants younger than 20 had not only a reduced risk of relapse but also a significantly reduced risk of treatment-related mortality after transplantation.
In Brief
The interesting point of this study is that it presents data of a large series of cases collected from various institutions with a long follow-up. There are obvious methodological limitations inherent to the retrospective design of the study. First, as the study is not based on a prospective comparison between autologous and alternative donor transplantations, it is quite likely that patients with different risk profiles were selected for each of the treatments. Indeed, the authors show that patients with unfavorable cytogenetics were more likely to be offered URD transplantation. In addition, URD recipients were more likely to have a reduced performance score and have leukemia with unfavorable cytogenetics, and they had more difficulty (i.e., required more time) attaining CR1. These adverse features of the URD patient group may have negatively influenced the URD transplant results. Furthermore, in this study important cytogenetic information was available only in a minority of cases. It would have been of clinical interest to know about outcome in the distinct subsets of patients with high-risk AML and intermediate prognostic risk. The numbers did not allow for such an analysis. Finally, in recent years molecular HLA matching has improved considerably. The degree of HLA match between unrelated donor-recipient pairs was insufficiently defined in this study. Nowadays, high-resolution HLA typing allowing for better donor selection appears to favorably affect the outcome of URD transplants.
What can we conclude? Both autologous and URD transplantation represent realistic and clinically meaningful therapeutic options. As a next step, additional studies involving (cyto)genetically characterized patient cohorts and using modern donor selection methods are warranted in order to more fully define the comparative values of autologous and URD transplantation modalities in patients with AML.
Competing Interests
Dr. Löwenberg indicated no relevant conflicts of interest.
