Cytomegalovirus infection following transplantation of autologous CD34-selected progenitor cells

In their report of autologous CD34-selected peripheral blood stem cell transplantation, Holmberg et al reported a very high incidence (22.6%) of cytomegalovirus (CMV) disease with 4 of the 7 patients who developed disease dying of CMV infection.1 This high incidence of infection after autologous transplantation is unusual, leading the authors to suggest that the process of CD34 selection, which results in 2- to 3-log T-cell depletion, may be responsible for susceptibility to CMV infection. Based on extensive experience with depletion technology, however, we think alternative explanations should be considered.

We recently completed a phase III trial that compared autologous CD34-selected and -unselected peripheral blood progenitor cell transplantation in 193 patients with chemotherapy-sensitive intermediate-to-advanced-stage multiple myeloma.2,3 Both groups underwent stem cell mobilization with cyclophosphamide/prednisone/G-CSF and a preparative regimen consisting of busulfan/cyclophosphamide; patients did not routinely receive prophylactic immunoglobulin therapy. Although CD34 selection achieved the same degree of T-cell depletion as described by Holmberg et al,1 we were unable to detect an increased susceptibility to infection. The incidence of overall infection was equivalent in patients receiving CD34-selected and -unselected transplants; specifically, CMV infection was documented in 2 patients and 1 patient, respectively. Interstitial pneumonitis occurred in 3 and 2 patients, respectively.

The results of this randomized study suggest to us that, although the process of CD34 selection does produce significant T-cell depletion, immune reconstitution is apparently sufficient to prevent life-threatening infection in a population of treated patients with multiple myeloma undergoing autologous transplantation.4 Alternative explanations for the relatively high incidence of CMV infection in Holmberg et al's study should focus on pretreatment characteristics. Their study used a population of more heavily pretreated patients, especially patients who may have received long-term glucocorticoid immunosuppression (autoimmune disease patients) or multiple cycles of pretransplant chemotherapy (oncology patients). A more intensive preparative regimen, including a greater percentage of patients in the CD34-selection arm receiving total body irradiation (TBI), may also have contributed to the high incidence of CMV-related complications and the poor outcome in their study.