Risk for cytomegalovirus disease in patients receiving polymerase chain reaction–based preemptive antiviral therapy after allogeneic stem cell transplantation depends on transplantation modality

Recently, Holmberg et al reported on a strikingly high incidence of cytomegalovirus (CMV) infection and disease in a cohort of patients receiving high-dose chemotherapy followed by the infusion of autologous CD34-selected peripheral blood progenitor cell transplants (PBPCTs). In a multivariate logistic regression analysis, the use of CD34-selected autologous peripheral blood stem cells after high-dose therapy was associated with a marked increase in the incidence of CMV disease and CMV-associated deaths.1 

Here we report the incidence of CMV infection and disease following allogeneic stem cell transplantation comparing recipients of(a) unmanipulated allogeneic peripheral blood stem cells from HLA-identical siblings (group I), (b) CD34-selected allogeneic peripheral blood stem cells from HLA-identical (n = 15) or 1 Antigen-mismatch (n = 3) siblings (group II), and (c) a group of patients receiving in vivo T-cell depleted bone marrow from HLA-identical (n = 11) or 1 Antigen-mismatch (n = 4) unrelated donors (group III). Patient characteristics for the 3 groups are shown in Table 1. The Cellpro Ceprate System (Cellpro, Seattle, WA) was used in all cases of group II to select CD34⁺ cells. The median number of CD34⁺cells/kg body weight (bw) was not different between groups I and II (4.17 ± 0.53 × 10⁶ versus 3.96 ± 1.05 × 10⁶). CD34⁺ selection resulted in a reduction of CD3⁺ cells from 188.1 ± 76.9 × 10⁶ cells/kg bw (group I, n = 17) to 0.39 ± 0.24 × 10⁶ cells/kg bw (group II, n = 18). Graft-versus-host disease (GVHD)–prophylaxis consisted of cyclosporine A (CSA) (all patients) plus methotrexate (group I, n = 6; group II, n = 4) or antithymocyte globulin (ATG; Biomerieux, Paris, France) on days − 4 to − 2 at a dose of 2.5 mg/kg bw (group I, n = 10; group II, n = 2). Patients in group III received transplants from unrelated donors and received an intensified graft-versus-host prophylaxis regimen (German multicenter study) incorporating antithymocyte globulin (Biomerieux) at a dose of 3.5 mg/kg bw on days − 4 to − 1, CSA in a dose of 5 mg/kg bw (dose adjustment according to serum levels), methylprednisolone at 0.5 mg/kg bw on days +7 to +14 and at 1 mg/kg bw from days +14 to +28, and mycophenolate mofetil starting on day +7 after bone marrow transplantation (BMT).

All patients were monitored by a polymerase chain reaction (PCR)–assay once weekly and received preemptive antiviral therapy with ganciclovir at 2 × 5mg/kg bw or foscarnet at 2 × 60 mg/kg bw as reported previously.2 

Overall, 12 of 19 patients receiving unmanipulated peripheral blood stem cells developed CMV infection after transplantation, compared to 16 of 18 receiving CD34-selected stem cells and 15 of 15 after BMT from an unrelated donor (group I versus group II, P = .12; group I versus group III, P = .01) (Table 1). The time to the first positive PCR result after transplantation was observed at a median of 29.5 (3-44) days in group I, 24.5 (5-46) days in group II, and 21 (8-33) days in group III (for group I versus group III,P = 0.01, Wilcoxon rank test) (Figure1). The time to clearance of viral DNA in the blood after the initiation of antiviral therapy did not differ between groups I and II (3.3 [0-10] weeks and 3.3 [0-10] weeks, respectively), whereas it was significantly longer in patients after BMT from an unrelated donor receiving an intensified GVHD-prophylaxis regimen (median duration of DNAemia, 8.7 [range 2-18] weeks,P = .0002, Wilcoxon rank test). CMV disease was documented in 1 patient receiving unmanipulated allogeneic stem cells (fatal CMV pneumonia), in 2 patients receiving CD34-selected stem cells (fatal CMV pneumonia, n = 1; enteritis after day 100 after transplantation, n = 1), and in 3 patients following BMT from an unrelated donor accompanied by intensified GVHD prophylaxis (fatal IP, n = 1; hepatitis, n = 1; late onset retinitis, n = 1).

The median numbers of CD3⁺ and CD8⁺ lymphocytes per microliter of blood assessed at 3 and 6 months after transplantation did not differ among the 3 groups. But reconstitution of CD4⁺ T cells at 3 months was significantly delayed in patients receiving bone marrow from unrelated donors compared to patients receiving CD34⁺-selected allogeneic stem cells (median 26, range 2-770 CD4⁺ T cells/μL versus a median of 186, range 42-350 CD4⁺ T cells/μL,P = .01, Wilcoxon rank test; see Table2), and at 3 months in patients developing CMV disease (median 4, range 2-65 CD4⁺ T cells/μL, n = 3) compared to patients with asymptomatic CMV infection (median 70, range 5-770 CD4⁺ T cells/μL, n = 25) or without signs of CMV reactivation (median 100, range 49-350 CD4⁺ T cells/μL, n = 5) (P = .08, Wilcoxon rank test).

CMV infection was diagnosed by PCR in all patients prior to the onset of CMV-related clinical symptoms. But in 1 patient receiving CD34-selected stem cells, CMV-IP developed before a second positive PCR result was obtained 1 week later, indicating that the time from the first positive PCR result to the onset of clinical symptoms might be very short in patients receiving T-cell–depleted allografts. Thus patients receiving CD34-selected allogeneic stem cells from related donors may develop CMV infection quite soon after transplantation. But due to a rapid immune reconstitution and early initiation of antiviral therapy based on a PCR assay, no increase in CMV-related morbidity and mortality after CD34⁺-selected allogeneic stem cell transplantation was observed. Patients receiving bone marrow from unrelated donors followed by an intensified GVHD prophylaxis showed CMV PCR positivity even sooner after transplantation, but again PCR-based antiviral therapy was found to be safe with only 1 patient developing early fatal CMV disease as already reported previously.2But this group seemed to be at an increased risk for late onset CMV disease most likely due to a delayed reconstitution of CMV-specific T-cell responses.3-5 

In conclusion, as discussed by Holmberg et al and demonstrated in this study, the high incidence of CMV disease in recipients of CD34-selected stem cells can be reduced by the early initiation of preemptive antiviral therapy based on sensitive assays,1,6 but probably to the expense of an increased incidence of late-onset CMV disease, especially in patients with delayed immune reconstitution.7-9 

Supported by the Sonderforschungsbereich 510, project B3, and the IZKF Tübingen, project IIC2.