Randomized controlled trial of sequential intravenous and oral ganciclovir versus prolonged intravenous ganciclovir for long-term prophylaxis of cytomegalovirus disease in high-risk cytomegalovirus-seronegative liver transplant recipients with cytomegalovirus-seropositive donors

Abstract

Cytomegalovirus (CMV)-seronegative liver transplant recipients with CMV-seropositive donors have the greatest risk for CMV disease. We performed a randomized, controlled trial comparing sequential intravenous (IV) and oral ganciclovir with prolonged IV ganciclovir for long-term prophylaxis of CMV disease in these high-risk patients. Patients were initially given IV ganciclovir at a dose of 6 mg/kg per day from days 1 to 14 after transplantation. Patients then either received oral ganciclovir (1 g every 8 hr) or continued IV ganciclovir (6 mg/kg once per day on Monday–Friday of each week) from days 15 to 100 after transplantation. CMV disease occurred in 3 of 32 patients (9.3%) receiving oral ganciclovir and in 4 of 32 patients (12.5%) receiving IV ganciclovir within the first year after transplantation (P >0.2). All cases of CMV disease occurred more than 90 days after transplantation (median time of onset day +137 for oral ganciclovir and day +135 for IV ganciclovir). There were no deaths from CMV in either study group. Both oral and IV ganciclovir were generally well tolerated. These results indicate that, after induction with 14 days of IV ganciclovir, oral ganciclovir can be as effective as IV ganciclovir for long-term prophylaxis of CMV disease in high-risk CMV-seronegative liver transplant recipients with CMV-seropositive donors and eliminates the need for prolonged IV access.

In previous clinical trials at the University of California-Los Angeles (UCLA), we found that the most effective regimen for prevention of primary cytomegalovirus (CMV) disease in CMV-seronegative liver transplant recipients with CMV-seropositive donors is the administration of intravenous (IV) ganciclovir for 100 days after transplantation (^1,2). The incidence of CMV disease associated with 100 days of prophylactic IV ganciclovir is 5% to 10%, which is lower than the incidences reported with other prophylactic or preemptive therapy regimens in this high-risk population of patients (³). Nevertheless, the utility of this regimen has been limited by the need for prolonged IV access. Thus, we undertook this study to determine whether a sequential regimen of 14 days of IV ganciclovir followed by oral ganciclovir for 86 days could be as effective as 100 days of IV ganciclovir for prevention of primary CMV disease in high-risk CMV-seronegative liver transplant recipients receiving allografts from CMV-seropositive donors. We hypothesized that induction therapy with 14 days of IV ganciclovir immediately after transplantation, when initial viral exposure may be at its peak, would render subsequent oral ganciclovir more effective for prophylaxis of primary CMV disease.

PATIENTS AND METHODS

Patients undergoing orthotopic liver transplantation at the UCLA Medical Center were enrolled in the study if they were seronegative for CMV antibody before transplantation and received a liver allograft from a CMV-seropositive donor. Informed consent approved by the UCLA Human Subject Protection Committee was obtained from each patient or responsible relative. Operative procedures and posttransplant management used in liver transplant recipients at the UCLA Medical Center have been published (^4,5).

Patients were assigned randomly at the time of transplantation to receive prophylaxis with oral or IV ganciclovir. All patients initially received IV ganciclovir at a dose of 6 mg/kg of body weight once per day, starting on the first day after transplantation and continuing through day 14 after transplantation. Patients then received oral ganciclovir, at a dose of 1 g every 8 hr, or IV ganciclovir, at a dose of 6 mg/kg of body weight once per day on Monday through Friday of each week until day 100 after transplantation. If a patient assigned to oral ganciclovir became unable to take oral medications, IV ganciclovir (6 mg/kg of body weight once per day) could be given in place of oral therapy. The dosage of oral or IV ganciclovir was adjusted for renal dysfunction according to previously published guidelines (^1,6).

The CMV antibody status of patients and liver donors was determined by latex agglutination (CMV SCAN, Becton-Dickinson, Sparks, MD). Patients were evaluated for CMV viremia and disease whenever any of the following clinical or laboratory abnormalities developed: fever (temperature ≥38°C), leukopenia (leukocyte count <3.0×10⁹/L), thrombocytopenia (platelet count <75×10⁹/L), generalized wasting, liver graft dysfunction, gastrointestinal complaints (nausea, vomiting, abdominal pain, diarrhea), dyspnea and pulmonary infiltrates on chest roentgenogram, or any other organ dysfunction suggestive of CMV disease. For patients with suspected CMV disease, blood for quantitation of CMV DNA (Digene Hybrid Capture System, Digene Corporation, Gaithersburg, MD) and cultures of the blood and urine for CMV were obtained. Whenever indicated by clinical findings, viral cultures of bronchoalveolar lavage, biopsy material, or any suspicious viral lesion were performed. Biopsy material and bronchoalveolar lavage were also examined histologically for typical viral inclusions and then stained immunohistochemically by indirect immunofluorescence using CMV monoclonal antibodies. Complete blood counts with differential white blood cell counts and platelet counts, serum creatinine and electrolyte determinations, and liver function studies were obtained at study entry, at least weekly during periods of hospitalization and at each posttransplant clinic visit to assess for effects related to the study drugs.

The primary endpoint of the study was development of CMV disease. Documentation of CMV disease required the presence of CMV infection and appropriate clinical manifestations (⁷). Fisher’s exact two-tail test was used to compare differences in proportions. Ninety-five percent confidence intervals were determined for differences by using the normal approximation. All patients were included in the efficacy analysis (intent-to-treat analysis) and the safety analysis. Patients were followed until 1 year after transplantation or time of death.

RESULTS

Sixty-four patients were enrolled in the study from June 1997 to December 2000. Two groups of 32 patients each were randomized to receive oral or IV ganciclovir. The characteristics of the patients in the two study groups were similar, except that more patients in the oral ganciclovir group (56%) than in the IV ganciclovir group (25%) received the regimen of tacrolimus plus corticosteroids for initial immunosuppression (P =0.02) (Table 1).

Table 1: Patient characteristics

Three patients in the oral ganciclovir group (9.3%) and four patients in the IV ganciclovir group (12.5%) developed CMV disease within 1 year after transplantation (difference −3.2% [95% confidence interval −21.5% to 15.3%]; P >0.2) (Table 2). There were two cases of CMV syndrome and one case of CMV hepatitis in the oral ganciclovir group, whereas three cases of CMV syndrome and one case of CMV hepatitis with colitis occurred in the IV ganciclovir group. The median times of onset of CMV disease after transplantation were day 137 (range, days 123–162) in the oral ganciclovir group and day 135 (range, days 92–223) in the IV ganciclovir group. Except for one case of CMV syndrome on day 92 after transplantation in the IV ganciclovir group, all cases of CMV disease during the study occurred after day 100 posttransplantation when patients were no longer taking the prophylactic study drug. All patients with CMV disease in both the oral and IV ganciclovir groups improved after treatment with therapeutic doses of IV ganciclovir (5 mg/kg of body weight every 12 hr) for 2 to 3 weeks. There were no deaths from CMV infection. Diseases associated with herpes simplex virus, varicella-zoster virus, Epstein-Barr virus, or human herpes virus 6 were not observed during the study.

Table 2: Incidence of cytomegalovirus disease

No clinical adverse events could be attributed to either oral or IV ganciclovir. Leukopenia (white blood cell count <3×10⁹/L) occurred in 9 of 32 patients (28%) receiving oral ganciclovir and in 13 of 32 patients (41%) receiving only IV ganciclovir (P =0.43). Four patients (13%) receiving oral ganciclovir and four patients (13%) receiving only IV ganciclovir had their prophylactic ganciclovir discontinued because of leukopenia. Except for one patient receiving oral ganciclovir who developed graft-versus-host disease associated with bone marrow aplasia, all the patients with leukopenia causing discontinuation of study drug resumed prophylactic oral or IV ganciclovir at the original dose when the leukopenia resolved. Granulocyte colony-stimulating factor was used for treatment of leukopenia in four patients (13%) receiving oral ganciclovir and in six patients (19%) receiving IV ganciclovir. There was no increased incidence of infections associated with these episodes of leukopenia.

The number of patients who experienced thrombocytopenia (decrease in platelet count to <100,000×10⁹/L posttransplantation) was similar in the oral ganciclovir group (6/32 patients or 19%) and the IV ganciclovir group (8/32 patients or 25%). An increase in serum creatinine concentration (≥1.5 mg/dL) during administration of the study drug posttransplantation occurred in 9 of 32 patients (28%) receiving oral ganciclovir and in 11 of 32 patients (34%) receiving IV ganciclovir. However, no cases of thrombocytopenia or renal dysfunction were attributable to either oral or IV ganciclovir. No patients were removed from the study or had the study drug stopped because of thrombocytopenia or renal failure. There were no significant differences in liver function tests between patients receiving oral or IV ganciclovir.

A total of 27 of 32 patients (84%) in the oral ganciclovir group and 31 of 32 patients (97%) in the IV ganciclovir group were alive at 1 year posttransplantation. The causes of death in the five patients receiving oral ganciclovir who died within 1 year posttransplantation were bacterial or fungal infection (three cases), chronic renal failure (one case), and graft-versus-host disease (one case). The only death in the IV ganciclovir group was in a patient with central pontine myelinolysis. No deaths from CMV infection occurred in either study group.

DISCUSSION

The risk of developing CMV disease after liver transplantation and other types of solid-organ transplantation depends on the CMV serologic status of both the donor and recipient. CMV-seronegative recipients who receive an allograft from a CMV-seropositive donor have the greatest risk for symptomatic CMV disease. Without effective prophylaxis, 50% to 75% of these patients develop CMV disease (³). Higher and more rapid increases in CMV viral load are associated with this increased risk for disease (⁸), which may require more potent and prolonged levels of antiviral activity in body fluids and tissues for prevention.

In this prospective, randomized study, 14 days of IV ganciclovir followed by 86 days of oral ganciclovir was as effective as 100 days of IV ganciclovir for CMV prophylaxis in CMV-seronegative liver transplant recipients with CMV-seropositive donors. The patients receiving oral ganciclovir had a 9.3% incidence of CMV disease during the first year after transplantation, which was similar to the 12.5% incidence among patients given 100 days of IV prophylactic ganciclovir. These results indicate that oral ganciclovir after a short course of induction therapy with IV ganciclovir can provide similar protection against primary CMV disease as prolonged IV ganciclovir and eliminates the need for extended IV access. Induction therapy with IV ganciclovir immediately after transplantation may be crucial in preventing early viral replication at a time of peak viral exposure and may strengthen the prophylactic efficacy of subsequent oral ganciclovir.

Two other randomized, controlled trials have evaluated oral ganciclovir for CMV prophylaxis in high-risk CMV-seronegative solid-organ transplant recipients with CMV-seropositive donors. In a single-center study involving renal transplant recipients, patients were randomized to receive antiviral prophylaxis with either oral ganciclovir plus CMV hyperimmune globulin or high-dose oral acyclovir plus CMV hyperimmune globulin for 3 months (⁹). CMV disease occurred in none of the patients receiving oral ganciclovir compared with 38% of the patients receiving oral acyclovir. The combination of ganciclovir plus CMV hyperimmune globulin, which is considerably more expensive than ganciclovir alone, has been shown to have synergistic activity in a murine model of CMV infection (¹⁰). In a second multicenter study, CMV-seronegative kidney, heart, or liver transplant recipients with CMV-seropositive donors initially received induction therapy with IV ganciclovir (5 mg/kg once per day) for 5 to 10 days followed by either oral ganciclovir or low-dose oral acyclovir for an additional 84 days (¹¹). Symptomatic CMV disease developed in 19% of the patients receiving oral ganciclovir and 27% of the patients receiving oral acyclovir.

Induction therapy with a higher dose of IV ganciclovir (5 mg/kg twice daily) or extending the length of oral ganciclovir prophylaxis beyond day 100 after transplantation might improve the prophylactic efficacy of sequential IV and oral ganciclovir. Of note, five of the seven cases of CMV disease in this study occurred within 40 days after discontinuation of ganciclovir prophylaxis. Thus, we are currently evaluating the efficacy of extending oral ganciclovir prophylaxis to day 150 after transplantation in high-risk CMV-seronegative liver transplant recipients with CMV-seropositive donors.

Surveillance tests for CMV infection were not performed during this study. Because of logistic and financial reasons, surveillance testing for CMV is not part of standard care of UCLA liver transplant recipients. Because the primary endpoint of the study was development of CMV disease, the sponsor of the study did not provide funding for CMV surveillance tests. Nonetheless, all UCLA liver transplant recipients are closely followed by clinical nurse specialists who maintain frequent contact with patients by phone. All patients are also instructed to call the transplant center for any fever, gastrointestinal symptoms, shortness of breath, or sense of ill-feeling. Thus, although cases of asymptomatic CMV infection could have been missed, it is unlikely that cases of CMV disease were overlooked.

The most common toxicity limiting the use of ganciclovir is myelosuppression, which is more common in bone marrow transplant recipients than in solid-organ transplant recipients (^1,2,6). In this study, leukopenia (decrease in white blood cell count to <3.0×10⁹/L) occurred in 28% and 41% of the patients receiving oral or IV ganciclovir, respectively, and led to temporary interruption of ganciclovir prophylaxis in 13% of patients in both study groups. These incidences of leukopenia are higher than the incidences of leukopenia previously reported by us with the administration of 100 days of prophylactic IV ganciclovir (^1,2) and may be related to the greater use of mycophenolate in this study. Mycophenolate commonly causes leukopenia but may also potentiate the anti-herpesvirus activity of ganciclovir (¹²).

Another potential limitation of oral ganciclovir is its low bioavailability (6%), which requires the administration of a relatively large number of capsules in divided doses to maintain adequate concentrations in the blood (^6,13). Recently, oral valganciclovir, the valine ester of ganciclovir, has become available. Valganciclovir has more favorable bioavailability (60%) than oral ganciclovir and at a dose of 900 mg (2 tablets) provides blood levels of ganciclovir in liver transplant recipients comparable to IV ganciclovir (¹³). The relative efficacy of oral valganciclovir and oral ganciclovir for prevention of primary CMV disease in CMV-seronegative solid-organ transplant recipients with CMV-seropositive donors is currently being investigated.

A concern associated with the use of ganciclovir for universal prophylaxis is the emergence of resistant organisms. The emergence of ganciclovir-resistant CMV isolates has been reported mostly in patients with acquired immunodeficiency syndrome and less frequently in solid-organ transplant recipients. Nevertheless, a recent report highlighted the emergence of ganciclovir-resistant CMV disease after prolonged oral ganciclovir prophylaxis in transplant recipients (¹⁴). This resistance was restricted to CMV-seronegative patients receiving simultaneous kidney and pancreas allografts or pancreas allografts alone from CMV-seropositive donors. It was suggested that the high CMV loads associated with the CMV-seronegative recipient and CMV-seropositive donor transplant group in the presence of relatively low serum antiviral concentrations achieved with oral ganciclovir for a long period may lead to the selection of ganciclovir-resistant mutants. Of note, there were no cases of ganciclovir-resistant CMV disease among the liver transplant recipients in this report (¹⁴). Although we did not perform antiviral susceptibility testing or use molecular techniques to detect ganciclovir-resistant CMV isolates, ganciclovir resistance was not a clinically apparent problem during our study. All seven patients who developed CMV disease improved after treatment with therapeutic doses of IV ganciclovir. Furthermore, despite the routine use of ganciclovir for universal prophylaxis in liver transplant recipients at UCLA for the last 7 years, we have not found any clinical or laboratory evidence of increased ganciclovir resistance.

Another strategy for preventing CMV disease after transplantation is preemptive therapy or the initiation of ganciclovir in an asymptomatic patient with a surveillance test (culture, antigenemia, polymerase chain reaction) positive for CMV and predictive of CMV disease. This strategy is designed to restrict prophylaxis to high-risk patients and thereby reduce drug costs and the risk for drug toxicity. Preemptive therapy has been successful in low-risk CMV-seropositive transplant recipients but has been studied in only a limited number of CMV-seronegative transplant recipients with CMV-seropositive donors (¹⁵). Many of these high-risk patients ultimately develop CMV infection requiring treatment with repeated courses of IV or oral ganciclovir (¹⁵). A clinical trial comparing the relative efficacy, toxicity, cost, and convenience of preemptive therapy with universal prophylaxis is needed.

Acknowledgments.

The authors thank the following study coordinators for their valuable assistance during this study: Anna Kroeber, RN, Janet Mooney, RN, and Joan Krause, RN. We are also grateful to Ed Arriola and the staff of the UCLA pharmacy for administrative assistance; Dr. David Bruckner and the staff of the UCLA clinical microbiology laboratory for technical assistance; Katharine Fry for preparation of the article; and the UCLA liver transplant physicians and nurses for their help and care of the patients during the study.