A retrospective chart review was performed in children with acute leukemia who received allogeneic HCT at Samsung Medical Center in Korea from October 1998 to December 2009. We excluded recipients who underwent HCT twice or more or who received two or more stem cell sources. In total, 134 patients were included in the analysis. All patients underwent a myeloablative conditioning regimen. Conditioning regimens were chosen depending on underlying diseases, and these were maintained consistently throughout the study period. We identified Samsung Medical Information System's clinical data which included information on age, sex, primary disease, type of stem cell graft, donor source, CMV serostatus of the donor and recipient, conditioning regimen, GVHD prophylaxis, antiviral prophylaxis, the use of anti-thymocyte globulin (ATG), and the presence of GVHD. All included recipients were sorted into the following three groups according to donor source and graft type: CB, related bone marrow or peripheral blood (PB) stem cell (RD), and unrelated bone marrow or peripheral blood stem cell (UD).

CMV infection was defined as isolation of CMV or detection of viral proteins or nucleic acid in any body fluid or tissue specimen.20 CMV antigenemia was defined as greater than or equal to 1 CMV pp65 positive cell per 2×10⁵ leukocytes. CMV disease was diagnosed by signs or symptoms and demonstration of CMV infection (upon culture, histopathologic confirmation, immunohistochemical analysis, or in situ hybridization) in organ biopsy specimens. In addition, CMV pneumonitis was diagnosed by interstitial pneumonitis accompanied by CMV antigenemia, unless other pathogens were confirmed. CMV retinitis was diagnosed by typical lesions confirmed by an ophthalmologist. CMV central nervous system (CNS) disease was defined by the identification of CNS symptoms together with the detection of CMV in cerebrospinal fluid samples upon a culture or polymerase chain reaction (PCR).

Recurrent CMV infection was defined as the development of a new positive CMV test occurring after a 50-day interval free of CMV infection. The duration of CMV antigenemia was defined as the number of days from the first detection to the last detection of CMV antigenemia. The total duration of CMV antigenemia was the sum of the duration of all episodes, when recurrent infections occurred.

Prior to conditioning, CMV serology of the recipient and donor was assessed. CMV immunoglobulin G (IgG) was checked by enzyme linked fluorescent assay. The test was considered positive when CMV IgG antibody level was greater than or equal to 4 AU/mL. The CMV serostatus of cord blood was considered as being negative.21 A CMV pp65 antigenemia test using immunofluorescence stain was performed one to three times weekly after peripheral white blood cell (WBC) count reached greater than or equal to 1.0×10⁹/L. The results of this test were presented as the number of positive cells per 2×10⁵ leukocytes.

Acyclovir was given as a herpes virus prophylactic drug for 4 weeks from the day before transplantation (5 mg/kg/dose q 8 hrs). For CMV prophylaxis, from the 2001 to 2002 period, ganciclovir was used in certain high risk recipients of unrelated donor grafts, but this ganciclovir-based prophylaxis for CMV infection was not used thereafter. Strategies for pre-emptive therapy were maintained throughout the study period, using antigenemia as a surrogate marker for CMV viremia. If CMV antigenemia was detected without evidence of CMV disease, preemptive first-line therapy was started with an induction dose of intravenous ganciclovir (5 mg/kg twice daily i.v.). In patients with a stable condition after post transplant day +30, antiviral therapy was started only when CMV antigenemia was detected above 5/2×10⁵ leukocytes or with increasing CMV antigenemia during follow-up. Foscarnet (180 mg/kg/day i.v. in divided doses) was used in patients with severe neutropenia in place of ganciclovir. A second-line treatment with intravenous foscarnet was generally started in recipients with prolonged CMV antigenemia. Cidofovir or CMV hyperimmunoglobulin was also used for patients who appeared to fail previous treatments. Tests for CMV antiviral resistance mutations were not available.

All packed red blood cells and platelets were transfused using leukocyte-depleting filters and radiation. Granulocyte-colony stimulating factor was given to maintain absolute neutrophil count above 1.0×10⁹/L and total WBC count above 2.0×10⁹/L.

GVHD prophylaxis consisted mainly of cyclosporine A (CsA) alone or CsA with other immunosuppressants (mycophenolate, steroid, methotrexate). Protocols for GVHD prophylaxis was consistently maintained, except for CB recipients. CB recipients received CsA and methylprednisolone until December 2004 and received CsA+mycophenolate beginning in January 2005. Most RD recipients and UD recipients received CsA only or CsA+methotrexate, respectively, as the GVHD prophylaxis throughout the study period.

Diagnosis of acute and chronic GVHD was based on standard clinical criteria and biopsy when available.22,23 Treatment of acute GVHD stage II or greater began with steroids, usually methylprednisolone.

Patient characteristics, GVHD, and clinical features of CMV infection and disease were compared across donor sources and types of grafts using the Chi-square test and Fisher's exact test with a permutation method for multiple testing of categorical data. Continuous variables were compared using the Kruskal-Wallis test along with a Tukey test using ranks. The cumulative incidences of CMV antigenemia (for 365 days) and disease (for 730 days) were evaluated using a Fine-Gray regression model.24,25 All statistical analyses were performed using SAS software version 9.1.3 (SAS Institute Inc., Cary, NC, USA) and Stata software version 11.0 (StataCorp LP, 4905 Lakeway Drive, College Station, TX, USA).

The characteristics of the patients are shown in Table 1. All included recipients were classified into three groups: CB (n=36), RD (n=41), and UD (n=57). These groups were similar with respect to sex, age at HCT, primary disease, and conditioning regimen. The median age of the recipients was 8.2 years. The male to female ratio was 1.6 : 1. ATG was used most frequently in CB (34/36) than the other two groups (p<0.01). The proportion of positive CMV serology recipients was high in all three groups (CB 91.7%, RD 97.5%, UD 96.5%). In addition, the proportion of positive CMV serology donors was high (RD 73.1%, UD 87.7%). All recipients, except for one in the RD group, received only CsA as GVHD prophylaxis, whereas patients in the other groups received CsA plus another immunosuppressant (p<0.01). One of the UD recipients received tacrolimus. Acyclovir was used as herpes virus prophylaxis in all groups. Seven patients in the UD group received ganciclovir in addition to acyclovir. Fewer patients developed grade II to IV acute GVHD in the RD group than other patients (p=0.01). Extensive chronic GVHD occurred similarly among the groups (p=0.53).

The 365-day cumulative incidence of CMV antigenemia is shown in Fig. 1. The clinical features of CMV antigenemia are shown in Table 2. The cumulative incidence of CMV antigenemia was 67% in CB, 49% in RD, and 65% in the UD group. There was no significant differences among these groups in cumulative incidence of CMV antigenemia. First, CMV antigenemia was detected earlier in CB recipients (p=0.05). CB recipients also had the highest median value of peak CMV antigenemia (CB 160/2×10⁵ leukocytes vs. RD 7/2×10⁵ leukocytes vs. UD 19/2×10⁵ leukocytes, p<0.01). In addition, the duration of CMV antigenemia was the longest in the CB group (CB 87 days vs. RD 17 days vs. UD 28 days, p<0.01). However, the recurrence of CMV antigenemia occurred similarly among all three groups (p=0.77).

The 730-day cumulative incidence of CMV disease is shown in Fig. 2. The clinical features of CMV disease are shown in Table 3. CB recipients had the highest cumulative incidence of CMV disease (CB 36% vs. RD 2% vs. UD 5%, p<0.01). All of the recipients with CMV disease had CMV antigenemia before the diagnosis of organ involvement. Among all of the groups, 17 patients developed CMV disease at 24 organ sites (colitis in 9 patients, retinitis in 7 patients, pneumonia in 5 patients, CNS disease in 2 patients, and hepatitis in 1 patient). While HCT recipients in the RD and UD groups had only one organ involvement during CMV disease, five of the 13 CB recipients (38.5%) developed multi-organ CMV disease. Three patients had pneumonia plus retinitis. Of which, two patients had three-organ involvements; one patient had retinitis, colitis, and CNS disease; and the other had pneumonia, retinitis, and CNS disease. Among the 17 patients with CMV disease, two patients died of interstitial CMV pneumonitis. Both of them were CB recipients. One was a 12-year old male who received CBT for acute myeloid leukemia. He received multiple immunosuppressive agents to control chronic GVHD. He developed CMV reactivation after the use of alemtuzumab, which eventually progressed to CMV pneumonitis and was not controlled despite aggressive antiviral treatment. The other patient was a 10-month old infant who received CBT for infantile acute lymphocytic leukemia (ALL). He received steroid for GVHD and developed CMV antigenemia at day +28. CMV disease (pneumonitis, retinitis, and CNS disease) was uncontrollable despite treatment. One patient in the RD group developed organ involvement, which was CMV colitis. This patient was a 16 year old male with ALL who received a sibling matched allogeneic bone marrow transplantation. He developed chronic GVHD, which required immunosuppressants including cyclosporine, azathioprine, and mycophenolate. He developed hematochezia for which a colonoscopy was performed and pathologic confirmation was made for CMV colitis. The patient received ganciclovir; however, the general condition was very poor presenting multiple other complications, and he died.