Objective: The aim is to explore the potential relationship between the virus load of Cytomegalovirus (CMV)infection and disease relapse in acute myeloid leukemia (AML)patients after Umbilical cord blood transplantation (UCBT), The mechanism of interaction in which lymphocyte subsets may be involved is also analyzed.

Design and method: All 126 patients with acute myeloid leukemia receiving a single UCB graft at the institution between Oct 2010 to Dec 2017 were enrolled in this retrospective study and engraft failure and death within 100 days excluded. All patients received common transplant processing: intensive myeloablative conditioning without ATG, Cyclosporine (CSA) and mycophenolate mofetil (MMF) for GVHD prophylaxis. All patients had negative plasma CMV DNA prior to the transplantation and examined twice a week for at least 100 days after transplantation. CMV were measured by Quantitative real-time PCR using TaqMan (ABI)-based method. Patients were divided into three groups according to the CMV viral load within 100 days: CMV-negative groups (no relevant special treatment), <1,000/mL CMV DNA copy group (reduced immunosuppressor and close follow up) and >= 1,000 / mL CMV DNA copy group (reduced immunosuppressor, antiviral treatment and close follow up), the first group was merged into the second group due to the 92% incidence of CMV infection after UCBT .To compare the two regimens which aimed to focus on disease relapse ,non-relapse mortality (NRM),overall survivor (OS),disease free survival(DFS) ,GVHD-free /relapse-free survivor (GRFS)and graft versus host disease (GVHD) after transplantation for two years. Also, retrospectively compared the lymphocyte ratio recovery at different time points after transplantation .

Results: Patients characteristic had no significant differences between high CMV virus load (>=10^3/mL) group and low CMV virus load(<10^3/mL) group. There were 63 males and 63 females with a median age of 13 years(range,1-52) and a median weight of 40 Kg (range,11-87) .The median follow-up time among survivors was 736 days (range, 100-2184).Univariate and multivariate analysis showed, high CMV load group has lower two-year cumulative incidence of relapse [(2.9% vs. 17.9%, P=0.017); adjusted HR 0.19, P=0.045], higher cumulative incidence of DFS [(88.3% vs. 63.2%, P=0.003), adjusted HR0.28, P=0.005],higher cumulative incidence of OS[(88.0% vs. 72.3%, P=0.037),adjusted HR0.35,P=0.027].

The immune reconstruction data of the two groups showed that the median value of CD8+ T cells in the high CMV load group at 1, 4, 6, 9, 12, 15, 18 and 24-36 months after transplantation was higher than the low CMV load Group (30.3% vs. 13.9%, 46.8% vs. 27.6%, 46.5% vs. 27.9%, 37.0% vs. 31.5%, 37.7% vs. 20.9%, 42.3% vs. 22.8%, 30.4% vs. 23.5%, 29.1% Vs.22.1%; P=0.032, P=0.014, P=0.01, P=0.621, P=0.005, P=0.008, P=0.088, P=0.371).

Conclusion: Higher CMV load within 100 days after UCBT presents lower relapse,improved OS and DFS in AML patients. At the same time, the proportion of CD8+ T cells increased significantly in patients with high CMV load after 1, 4, and 6 months after transplantation. The above data indicates that CMV virus infection may stimulates the activation of CD8+ T cells further induces a GVL effect.

keywords:Cytomegalovirus,acute myeloid leukemia,cord blood transplantation,virus load, relapse.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.