Objectives: The purpose of this study was to evaluate the effectiveness of acyclovir prophylaxis and preemptive ganciclovir treatment in preventing cytomegalovirus disease in children who underwent hematopoietic stem cell transplant.
Materials and Methods: We reviewed the clinical records of 66 children (36 boys, 30 girls; mean age, 9 ± 5 y; age range, 2-20 y) who underwent hema-topoietic stem cell transplant at Ankara Children’s Hematology and Oncology Hospital, Bone Marrow Transplantation Unit, between April 2010 and March 2012.
Results: In these 66 children, 61 children (92.4%) received allogeneic transplant; 50 children (76.9%) received a myeloablative regimen; and 14 children (21.2%) received anti-thymocyte globulin as part of the conditioning regimen. All children received acyclovir prophylaxis from the beginning of conditioning regimen until 100 days after transplant, and children received preemptive treatment with ganciclovir when cytomegalovirus DNAemia ≥ 400 copies/mL on 2 tests or ≥ 1000 copies/mL on 1 test. There were 19 children (28.8%) who had cytomegalovirus reactivation during median follow-up 381 days (range, 100-720 d). Cytomegalovirus disease was observed in only 2 patients (10.5%); 1 patient had cytomegalovirus hepatitis and 1 patient had cytomegalovirus gastrointestinal disease. Both patients were cured of cytomegalovirus with treatment for 1 month. There was no death attributable to cytomegalovirus reactivation and/or disease. Febrile neutropenia, acute graft-versus-host disease, and steroid use were more frequent in patients who had cytomegalovirus than did not have cytomegalovirus reactivation. The risk of cytomegalovirus reactivation was increased 5-fold in patients who used steroids.
Conclusions: Acyclovir prophylaxis and preemptive treatment with ganciclovir may be effective in preventing cytomegalovirus disease in most children who have hematopoietic stem cell transplant.
Key words : Antiviral, Febrile neutropenia, Graft-versus-host disease, Infection, Steroids
Infection caused by cytomegalovirus (CMV) is a major cause of morbidity and mortality after hematopoietic stem cell transplant (HSCT). After allogeneic HSCT, 60% to 70% CMV seropositive patients experience reactivation and 20% to 30% of these patients who experienced CMV reactivation develop end-organ disease when ganciclovir prophylaxis or preemptive therapy is not used.1 The risk of CMV antigenemia is higher in patients who receive allogeneic transplant, have grade 2 to 4 acute graft-versus-host disease (GVHD), have anti-thymocyte globulin treatment, receive HSCT from human leukocyte antigen (HLA) mismatch donor, or have CMV seropositivity before HSCT.2,3 In Turkey and at our bone marrow transplant unit, most recipients and donors are CMV seropositive.
During the past decade, there have been major advances in monitoring and treating CMV infections after HSCT. Prevention of CMV disease in patients undergoing HSCT includes (1) reducing the risk of CMV infection or reactivation by treating risk factors; (2) prophylaxis of all high-risk patients for a defined time, initiated at or soon after transplant; and (3) preemptive treatment with ganciclovir, guided by laboratory markers indicative of CMV antigenemia. Both acyclovir and ganciclovir are effective in the prophylaxis of patients who are at high risk for CMV disease. Although ganciclovir may be more effective than acyclovir, the tolerability profile of ganciclovir is less optimal. Adverse events with ganciclovir, especially myelosuppression, may limit its use in prophylaxis. In addition, prophylactic ganciclovir is not cost effective. Foscarnet may avoid myelo-suppression but is associated with major nephrotoxicity and is reserved for patients who are unable to tolerate ganciclovir or who have ganciclovir-resistant CMV disease. Newer drugs such as valacyclovir and valganciclovir are being evaluated, and preliminary experience is promising.4-7
Most factors for CMV infection have been defined in adult patients, and there are few studies about the appropriate extent of CMV monitoring and CMV preemptive therapy in children who have HSCT.4-7 The purpose of the present study was to evaluate the effectiveness of acyclovir prophylaxis used until 100 days after transplant and preemptive ganciclovir treatment to prevent CMV disease in children who underwent HSCT.
Materials and Methods
Sixty-six children (age, 2-20 y) who underwent HSCT for malignant (36 patients [55%]) or nonmalignant diseases (30 patients [45%]) at the Ankara Children’s Hematology and Oncology Hospital, Bone Marrow Transplantation Unit between April 2010 and March 2012 were included in this study (Tables 1 and 2). All children were ≥ 100 days after transplant. Blood products were not screened for CMV antibody but were filtered to remove leukocytes before transfusion. In all patients, low-resolution molecular typing of HLA-A, HLA-B, HLA-C, HLA-DRB1, and HLA-DQB1 were used. Written informed consent was obtained from all parents or subjects. The study was approved by the Ethical Review Committee of the institute. All protocols confirmed with the ethical guidelines of the 1975 Helsinki Declaration. The age and gender distribution, indications for HSCT, stem cell source, and transplant characteristics of our cohort are shown in Table 1.
Hematopoietic stem cell transplant
Conditioning regimens and GVHD prophylaxis were given according to current protocols (Table 1). Most patients (65 patients) received 14 different conditioning regimens; 1 patient who had severe combined immune deficiency was not given any conditioning regimen. Most patients (50 of 65 patients [77%]) received a myeloablative regimen. Anti-thymocyte globulin (20-40 mg/kg body weight) (ATG-Fresenius, Fresenius, Bad Homburg, Germany) was given to 14 patients (21%) as part of the conditioning regimen. All patients received intravenous immunoglobulin on day 1, 8, 15, and 22 after transplant, and intravenous immunoglobulin was given to all patients who had immunoglobulin G level < 400 mg/dL after discharge from the hospital. The preferred protocol (46 patients [70%]) for prevention of acute GVHD included cyclosporine and short-term methotrexate. Most patients (55 of 66 patients [83%]) had full-matched-related HSCT from siblings or other relatives, but the other patients had autologous HSCT (5 patients [8%]), haploidentical HSCT (3 patients [5%]), or 1-HLA-antigen-mismatched HSCT (3 patients [5%]). One patient received T-cell–depleted marrow at his second HSCT.
Severity of oral mucositis was evaluated and graded daily according to adapted common toxicity criteria for stomatitis.8 Total parenteral nutrition was started for mucositis (≥ grade 3) and/or insufficient oral intake. The presence and severity of acute GVHD were assessed daily by the attending HSCT physician and nurses and graded according to Glucksberg criteria.9 Steroids were used for patients who had acute and/or chronic GVHD and engraftment syndrome. Ciprofloxacin, fluconazole, and trimethoprim and sulfamethoxazole were used for antibacterial, antifungal, and Pneumocystis jiroveci prophylaxis. Infections were defined by the presence of positive cultures or specific molecular, histologic, or other diagnostic tests or pathognomonic radiographic findings. The presence of fever and neutropenia was sufficient to define febrile neutropenia. Neutrophil engraftment day was defined as the first of 3 sequential days in which the absolute neutrophil count was ≥ 0.5 × 109/L. Platelet engraftment day was defined as the first of 3 sequential days in which platelet count was ≥ 20 × 109/L without need for transfusion for more than 7 days. Assessment of regimen-related toxicities (hepatic, pulmonary, and renal) was made daily during hospitalization according to the National Cancer Institute common toxicity criteria.
Acyclovir prophylaxis was used against the risk of CMV reactivation (250 mg/m2/dose; 3 doses/d; intravenous [from conditioning regimen to neutrophil engraftment] or oral [after neutrophil engraftment]). Acyclovir (40-50 mg/kg/d, oral) was continued until 100 days after HSCT. In patients with chronic GVHD, monitoring for CMV antigenemia and prophylaxis for CMV reactivation was continued until resolution of chronic GVHD and cessation of immuno-suppressive treatment.
Risk-adapted preemptive therapy
Patients were monitored for CMV antigenemia with real-time polymerase chain reaction twice weekly during hospitalization after the conditioning regimen was started. After children had been discharged from the hospital, monitoring for CMV antigenemia was continued ≥ once weekly during the second month and at 2- to 4-week intervals after the second month at every visit until cessation of cyclosporine. When high and/or increasing levels of CMV antigenemia (≥ 1000 copies/mL once or ≥ 400 copies/mL twice or more) were detected, ganciclovir was started (5 mg/kg body weight, twice daily for 14 d). When antigenemia remained positive after a 14-day course of ganciclovir, the ganciclovir was continued until 2 consecutive negative results of CMV PCR were obtained. Granulocytopenia (absolute neutrophil count < 1 × 109/L for 2 consecutive days) during or after ganciclovir therapy was treated with filgrastim or lenograstim; when granulocytopenia persisted, a dosage reduction or discontinuation of ganciclovir was considered. The CMV disease was treated with gan-ciclovir at the same dose used for preemptive therapy.
Definition of cytomegalovirus disease
The DNAemia was defined as DNA detected in samples of plasma, whole blood, isolated peripheral blood leukocytes, or buffy coat with polymerase chain reaction.10 The CMV pneumonia was defined as interstitial infiltrate on chest radiography or computed tomography scan accompanied by histologic demonstration of CMV in a lung biopsy or a positive CMV culture from bronchoalveolar lavage fluid. The CMV gastroenteritis was defined as gastrointestinal symptoms accompanied by histologic demonstration of CMV and/or a positive CMV culture from a biopsy obtained by endoscopy. The CMV hepatitis was defined by elevated bilirubin and/or liver enzyme levels, absence of any other documented cause of hepatitis, and detection of CMV infection in a liver biopsy specimen (culture, histopathologic testing, immunohistochemical analysis, or in situ hybridization). The CMV retinitis was diagnosed from characteristic retinal changes. Late CMV disease was defined as disease occurring after 100 days after transplant.10
Data analysis was performed with a statistical program (SPSS for Windows, version 17.0, SPSS Inc., Chicago, IL, USA). Patients who had or did not have CMV antigenemia were compared with the chi-square test and the t test for age, sex, diagnosis (malignant or nonmalignant disease), conditioning regimen (myeloablative or nonmyeloablative regimen with or without anti-thymocyte globulin), stem cell source (bone marrow or peripheral stem cells), relapse of the malignancy, HSCT-related complications (acute GVHD, chronic GVHD, grade 3-4 oral mucositis, use of total parenteral nutrition, hemorrhagic cystitis, engraftment syndrome, veno-occlusive disease, and febrile neutropenia), steroid use, and outcome. Statistically significant parameters (febrile neutropenia, acute GVHD, and steroid use) were identified in the univariate analysis and other variables that were clinically important (age, sex, diagnosis, conditioning regimen, stem cell source, HSCT-related complications) were evaluated with logistic regression. Statistical significance was defined by P ≤ .05.
Patients and transplant course
Most children (61 patients [92.4%]) had allogeneic HSCT, and 5 children (7.6%) had autologous HSCT (Table 1). The mean number of CD34+ stem cells given to the patients was (5 ± 3) × 106/kg and the mean number of nucleated cells was (5 ± 2) × 108/kg. Neutrophil engraftment was achieved at median 15 days (9-22 d) and platelet engraftment was achieved at median 20 days (10-60 d) after HSCT. Most children (61 patients [92.4%]) had complications related to HSCT, most commonly febrile neutropenia, grade 3-4 oral mucositis, and acute GVHD (Table 1).
Most patients were alive without disease at follow-up, but 5 patients died (Table 1). There were 2 patients who had aplastic anemia, had engraftment failure, and died because of bacterial sepsis during the first 100 days after HSCT (1 patient had a transplant from an HLA 8/10 identical brother and received transfusion for > 4 years, and 1 patient had received transfusion for > 10 years). There were 3 children who died during follow-up (1 died because of a relapse of acute myeloid leukemia and 2 died because of acute and chronic GVHD).
Cytomegalovirus DNAemia and preemptive treatment
During HSCT, 1201 samples were sent for CMV polymerase chain reaction from 66 patients (samples per patient: mean, 18 ± 10; range, 12-64). Most donors (except 1 donor) and most patients (except 2 patients) were positive for CMV antibody before HSCT. In the 66 patients who were at high risk for CMV reactivation, 19 patients (28.8%) had CMV DNAemia at median follow-up 381 days (range, 100-720 d) (Table 2). The CMV polymerase chain reaction was positive for a median 3 times (range, 1-11 times) for the 19 patients who had CMV DNAemia, and CMV DNAemia was detected after a median 5 weeks (range, 2-9 wk) after HSCT. Late CMV DNAemia (> 100 days after HSCT) was not detected in any patient. Ganciclovir treatment was used for patients who had CMV DNAemia for a median 14 days (range, 4-21 d) until the DNAemia resolved. Foscarnet was used in only 2 patients for ganciclovir-resistant CMV antigenemia and/or disease (Table 1). The use of foscarnet for 14 days was sufficient to eliminate CMV antigenemia and/or disease in these patients.
The CMV disease was detected in 2 patients, including 1 patient who had HSCT for thalassemia major (class 3). This patient had CMV hepatitis, and steroid treatment was given for late hemolysis after engraftment; her liver function tests (aspartate aminotransferase, alanine transaminase, and gamma-glutamyl transferase levels) were increased and her CMV DNA reached a maximum 22 345 copies/mL at 5 weeks after HSCT. Liver biopsy could not be performed because her activated partial thrombo-plastin time and international normalized ratio were high. She received ganciclovir (2 wk) and valganciclovir (2 wk). The other patient received HSCT for Chédiak-Higashi syndrome and developed CMV gastrointestinal disease. This patient had received steroid treatment for acute GVHD (grade 2; skin and gastrointestinal involvement); CMV inclusion bodies and other findings compatible with acute GVHD were shown on an endoscopic biopsy of the gut. His CMV DNA reached a maximum 38 581 copies/mL at 2 weeks after HSCT. He received ganciclovir (2 wk) and foscarnet (2 wk), but his CMV DNA remained high and he had diarrhea. Both children were cured of CMV disease, and CMV DNA was negative after treatment for 1 month. They have been followed as outpatients and remain in good clinical condition. The other 17 children who had CMV antigenemia did not develop CMV disease. No patients developed posttransplant lympho-proliferative disorder or other life-threatening morbidity from CMV antigenemia and/or disease, possibly because of prompt initiation of preemptive ganciclovir treatment.
Probability and risk factors for cytomegalovirus antigenemia
The patient groups with CMV antigenemia present or absent were compared for multiple clinical factors, but only febrile neutropenia, acute GVHD, and steroid use were significantly different between CMV positive and negative groups (Table 2). Logistic regression showed that only steroid use was a significant risk factor for CMV reactivation; the risk of CMV reactivation was increased 5-fold with the use of steroids (odds ratio, 5.0; 95% confidence interval, 1.3-18.1; P ≤ .02).
Before 1990, anti-CMV therapy was started only when clinical symptoms of CMV infection appeared. During the past decade, major advances have been achieved in the treatment of CMV infection and disease. At present, antiviral prophylaxis with high-dose acyclovir and preemptive ganciclovir therapy, guided by viral monitoring, are the 2 main antiviral strategies used in preventing CMV infection and disease, regardless of the CMV serologic status of the donor and recipient.11 Acyclovir prophylaxis given to all patients having HSCT may decrease the risk of CMV infection and mortality and has low toxicity. Ganciclovir also has been used in CMV prophylaxis, but no survival advantage was demonstrated. Severe neutropenia has been observed with ganciclovir and may increase the risk of developing fungal or bacterial infection.12,13 In contrast, intravenous ganciclovir is the preferred drug for preemptive therapy. Although foscarnet and ganciclovir have similar efficacy, foscarnet is more commonly used as a second-line drug. Foscarnet has a different toxicity profile than ganciclovir, and foscarnet may cause impaired renal function and electrolyte abnormalities. Furthermore, cidofovir is safe and effective as second-line therapy in adults and children who have CMV reactivation and failure of typical antiviral treatments.14,15
The CMV prophylaxis and preemptive therapy used during the past few years for children having allogeneic HSCT has decreased the risk of CMV disease, CMV-associated death, and transplant-related mortality. However, there is controversy about whether prophylaxis and/or preemptive therapy are optimal treatment for prevention of CMV disease. Limited information is available about CMV reactivation and disease after HSCT in children.16-18 Therefore, we aimed to evaluate the incidence, risk factors, and outcome of CMV reactivation and disease in children after HSCT in our patients and to evaluate the effectiveness of acyclovir prophylaxis and preemptive ganciclovir treatment in preventing CMV reactivation and disease.
With acyclovir prophylaxis, CMV reactivation occurred in 19 of 66 patients (28.8%) at median 381 days (range, 100-720 d) in the present study, and CMV reactivation was detected after a median 5 weeks (range, 2-9 wk) after HSCT. Preemptive ganciclovir treatment was used for median 14 days in patients who had CMV reactivation, and foscarnet was used successfully in 2 patients who had resistance to ganciclovir. The present results with acyclovir prophylaxis and preemptive ganciclovir and/or foscarnet were favorable, and there was no death caused by CMV reactivation and/or disease.
In a study from Chile, CMV reactivation occurred in 26 of 97 children (26.8%) who had HSCT; 3 of 26 children (11.5%) had CMV disease (pneumonia, 2 patients; hemorrhagic cystitis, 1 patient) and 1 patient died because of pneumonia.16 In a study from Denmark, CMV reactivation occurred in 24 of 110 patients (21.8%) and CMV disease occurred in 3 of 24 children (12.5%) (pneumonia, 2 patients; gastrointestinal disease, 1 patient).17 The frequency of CMV reactivation and disease was similar between the present and previous studies from other countries.
In the present study, acyclovir prophylaxis was used for 100 days after HSCT but not beyond 100 days because 15.1% patients had an increase in urea nitrogen and creatinine levels when acyclovir was used with cyclosporine. After acyclovir prophylaxis was stopped, the urea nitrogen and creatinine levels returned to normal. Therefore, we recommend acyclovir prophylaxis for 100 days after HSCT. Ganciclovir was used (5 mg/kg body weight, twice daily) for 14 days in most patients and was continued until the antigenemia resolved; only 3 patients who had CMV reactivation needed therapy for 21 days.
In the present study, febrile neutropenia, acute GVHD, and steroid use were significantly associated with antigenemia in univariate analysis. Logistic regression model showed that the risk of CMV reactivation had increased 5-fold with the use of steroids in children who had HSCT. Either donor or recipient was CMV seropositive before all HSCT in this study, and all patients were at high risk for developing CMV reactivation. In the previous study from Chile, risk factors for CMV reactivation included the pretransplant serologic status (positive recipient), acute and chronic GVHD, GVHD prophylaxis, and anti-thymocyte globulin used in GVHD treatment.16 In the previous study from Denmark, risk factors for developing CMV infection included the use of unrelated-donor transplant.17
In conclusion, high-dose acyclovir prophylaxis and preemptive treatment with ganciclovir were effective in preventing CMV disease in children who had HSCT. A risk-adapted approach for CMV preemptive treatment, based on the presence of acute GVHD and steroid use, may be useful in children having allogeneic HSCT.
Volume : 12
Issue : 5
Pages : 462 - 468
DOI : 10.6002/ect.2013.0238
From the Departments of 1Pediatric Hematology and
2Pediatric Infectious Diseases,
Ankara Children’s Hematology and Oncology Hospital, Ankara, Turkey
Acknowledgements: The authors have no conflicts of interest to disclose, and there was no funding for this study.
Corresponding author: Betul Tavil, Ankara Children’s Hematology and Oncology Hospital, Department of Pediatric Hematology, 06110-Diskapi, Ankara, Turkey
Phone: +90 312 596 9872
Fax: +90 312 596 9870
Table 1. Clinical Characteristics of Patients Who Had Hematopoietic Stem Cell Transplant*
Table 2. Relation Between Cytomegalovirus Antigen Status and Characteristics of Patients Who Had Hematopoietic Stem Cell Transplant*