Abstract

(See the editorial commentary by Avery, on pages 322-324.)

Background. Valganciclovir (VGC) 900 mg is approved for CMV prophylaxis, but it has been associated with 10%—40% leucopenia rate. We hypothesize that VGC 450 mg daily may be as effective as and safer than 900 mg daily.

Methods. Studies evaluating valganciclovir 900 mg and 450 mg daily against controls were evaluated. Direct comparisons were performed by random-effects models and indirect comparisons by the Bucher method.

Results. Twelve trials with VGC 900 mg (1543 patients) and 8 trials with VGC 450 mg (1531 patients) were included. The risk of CMV disease with VGC 900 mg versus controls was 1.06 (95% confidence interval [CI], .64–1.76; P = .81; I2=29%) and with VGC 450 mg vs controls .77 (95%CI, .49–1.18; P = .23; I2=24%). The risk of leucopenia was 5.24 (2.09–13.15; P = .0004; I2=44%) for VGC 900 mg versus controls and 1.58 (.96–2.61; P = .07; I2=36%) for VGC 450 mg versus controls; the risk for acute allograft rejection was 1.71 (.45, –6.50; P = .43) for VGC 900 mg and .80 (.50–1.28; P = .34) for VGC 450 mg. Adjusted indirect comparison between VGC 900 mg and VGC 450mg: the risk for CMV disease was not significantly different: odds ratio (OR), 1.38 (.84–2.25); P = .19; the risk of leucopenia was significantly increased with VGC 900 mg: 3.32 (1.76–6.26); P = .0002; and the risk of rejection was significantly increased with VGC 900 mg: 2.56 (1.50–4.53); P = .0005. Results remained consistent after adjustments by allograft, CMV control strategy, and immunosuppression.

Conclusions. Valganciclovir 900 mg showed no superiority efficacy compared to controls (ganciclovir or preemptive) and equivalent efficacy to VGC 450 mg (statistical power: 94% and 97%, respectively) for CMV universal prophylaxis.VGC 900 mg was significantly associated with 3 times increase in the risk of leucopenia and 2 times increase in the risk of rejection compared with VGC 450 mg.

Solid organ transplantation evolved over the last 50 years, but infections remain the most significant cause of morbidity and mortality after transplantation. Cytomeglovirus (CMV) is the most frequent opportunistic infection in the first year post-transplant [1-3].

Valganciclovir is among the commonly used drugs for CMV prophylaxis after transplantation. Valganciclovir was developed to have higher oral bioavailability compared with ganciclovir [4]. Following the trial performed by Paya et al [5], valganciclovir was approved for the prevention of CMV infections in high-risk kidney, kidney-pancreas, and heart transplant recipients with the notable exception of liver transplant recipients [5, 6].

One of the significant side effects noted in the Paya trial was leucopenia and neutropenia, more frequent with valganciclovir (8.2%) than with ganciclovir (3.2%) [5]. Our group showed previously that valganciclovir has similar efficacy but significantly higher risk of absolute neutropenia, CMV late-onset disease, and CMV tissue-invasive disease compared to prophylaxis with ganciclovir, valacyclovir, high-dose acyclovir and preemptive therapy [3]. The risk of absolute neutropenia with valganciclovir was 3.63 higher compared with all other approaches, 2.88 compared with ganciclovir only, or 8.3 higher compared with non-ganciclovir therapies [3]. These results raised the question if the dose of valganciclovir could be decreased to minimize the side effects but still provide adequate prevention of CMV disease.

Pescovitz et al showed that drug exposure provided by 900 mg of valganciclovir is equivalent to that provided by a 5 mg/kg dose of intravenous ganciclovir, and the dose of 450 mg of valganciclovir provided an exposure equivalent to that provided by oral 3 g ganciclovir [4]. Several recent studies evaluating low-dose valganciclovir (450 mg daily) prophylaxis in SOT recipients have reported potential benefits [7-19].

The aim of our study was to assess the benefits and risks of 2 valganciclovir regimens (900 mg versus 450 mg daily) for preventing symptomatic CMV disease in SOT recipients of all ages and all allografts. The secondary aims were to evaluate the impact of each dosing regimen on acute rejection, graft loss, and death.

METHODS

Literature Search

The following databases were searched from their inception until March 2010: PubMed, Embase, Cochrane Library, Evidence-based Medicine BMJ, and American College of Physicians Journal Club. In addition, available abstracts from the American Transplantation Congress and the Infectious Disease Society of America from 2004 to 2009 were searched. The keywords used for the search were: valganciclovir, valcyte, organ transplantation, kidney, renal, heart, cardiac, lung, liver, hepatic, pancreas, small bowel, intestinal, cytomegalovirus, prophylaxis, randomized, cohort, and case-control-studies. No language restrictions were applied. Two authors (D.F. and C.M.) performed the literature search and the study selection separately. Any disagreement was resolved by review from a third author (A.K.) and a final consensus among all authors.

Study Selection

Inclusion criteria.—All studies evaluating the prevention of CMV disease after SOT that aimed to compare valganciclovir 900 mg or 450 mg daily for normal renal function with (1) ganciclovir 3 g per day, (2) valacyclovir 3-8 g per day, or (3) preemptive therapy were selected for the efficacy analyses. For the safety analyses, all studies that aimed to compare valganciclovir against the same controls were performed as above but with no restriction for dosing; this was done because all drug exposures should be accounted for in order to comprehensively evaluate all safety data with respect to these antiviral drugs.

Exclusion criteria. —Cytomegalovirus studies that did not evaluate valganciclovir or that had valganciclovir universal prophylaxis in both arms were excluded.

Data Extraction

The following variables were collected from all studies: authors, publication year, study design, type of allograft, sex, age, sample size, CMV sero-status, induction therapy, maintenance immunosuppressive therapy, CMV disease, CMV infection, valganciclovir dose, type of control, prophylaxis duration, length of follow up, acute rejection rate, allograft loss, and mortality.

Definitions

Cytomegalovirus disease. —CMV syndrome (fever or fatigue and cytopenia) and/or CMV organ disease (organ involvement by clinical or pathology findings).

Cytomegalovirus infection. —CMV disease and/or CMV viremia.

Leucopenia. —Less than 3500cells/mm3 during study follow-up.

Neutropenia. —Less than 1500cells/mm3 during study follow-up.

Rejection.—Acute allograft rejection reported up to 12 months.

Mortality. —Death reported up to 12 months.

Statistical Analysis

The odds ratios were adjusted by the Mantel-Haenszel method, the Q statistic method was used to assess statistical heterogeneity, and the I-squared method was used to assess the magnitude of variation secondary to heterogeneity. All data were pooled by the use of random-effects model according to the DerSimonian and Laird methodology [20]. A metaregression was performed to evaluate the influence of CMV sero-mismatch (“D+/R- status”) on the risk of developing CMV disease. The software used was Comprehensive Meta-analysis Version 2 (Biostat). The adjusted indirect treatment effect was performed by the Bucher method [21]. This method has been validated by Song et al [22] and was used in our study to evaluate the differences in the risk of CMV disease, leucopenia, and rejection between valganciclovir 900 mg and 450 mg. Since the event rates were low across the treatment groups and subset analyses, the odds ratio (OR) was used to estimate the relative risk (RR). The Jadad score was used to evaluate the quality of studies, and the QUOROM criteria was used for the search method (Figure 1). Egger regression and the Begg and Mazumdar methods were used to evaluate publication bias [23–25]. Statistical power calculations were performed based on the comparison of 2 independent proportions using the software StatMate version 2.0 (GraphPad) and Power and Precision version 4.0 (Biostat).

Figure 1.

QUOROM trial flow.

Figure 2a.

Valganciclovir 900 mg: Risk of cytomegalovirus disease.

Figure 2b. Valganciclovir 900 mg: Risk of leucopenia.

Figure 3a.

Valganciclovir 450 mg: Risk of cytomegalovirus disease.

Figure 3b. Valganciclovir 450 mg: Risk of leucopenia.

RESULTS

Twelve studies (n = 1543) – 11 for efficacy and 7 for safety - were included in the valganciclovir 900 mg analysis [5, 26—36]. Eight studies (N = 1,531) – 8 for efficacy and 5 for safety - were included in the valganciclovir 450 mg analysis [8—10, 12—14, 17, 19]. See Table 1 for study characteristics.

Table 1.

Study Characteristics

Study, year, (Reference) Samplesize Mean age, years (treatment/control) Allograft type Maintenance regimen Induction Preventativestrategy Treatment arm Dose* Controlarm Duration of prophylaxis or preemptive therapy (months) Design:Randomized (R)/Cohort (C)/Case-Control (CC) 
Paya et al., 2004 [5364 45.7/45.3 heart, liver,kidney,kidney-pancreas NA Universal VGC 900 mg GCV 
Zamora et al. 2004 [37230 53.4/54 lung CsA, AZA, P. Universal VGC 900mg GCV+CMV IG and ACV 3-6 
Humar et al. 2005[2680 48.6/NA lung CsA, AZA, MMF, P.Induction (ALG). Universal VGC 900mg GCV 
Khoury et al. 2006[2798 51.9/47.5 kidney FK, CsA, MMF, AZA, P.Induction (thymoglobulin) Preemptive VGC 900mg Preemptive 3.3 
Said et al, 2007 [2864 42.5/38.2 kidney FK, CsA, MMF, P.Induction (ATG,baziliximab) Universal VGC 900mg GCV 0.5 
Guirado et al. 2008 [29150 NA kidney FK, MMF, P.Induction (ALG) Preemptive VGC 900mg Preemptive CC 
Moro et al. 2008 [3053 46/43.2 heart NA Universal VGC 900mg GCV NA CC 
Fernandez et al. 2009 [31100 48.1 total kidney FK, MMF, steroids.Induction (ATG, OKT3). Preemptive VGC 900mg Preemptive 4.3 
Monforte et al. 2009 [32163 51.1/48.5 lung FK, CsA, MMF, AZA.Induction (ATG, baziliximab) Universal VGC 900mg GCV 3.9 C C 
Parreira et al. 2009 [33135 NA kidney FK,CsA, MMF, P.Induction (ALG) Preemptive VGC 900mg Preemptive C C 
Potena et al. 2009 [3440 55/56 heart CsA, MMF, P.Induction (thymoglobulin) Preemptive VGC 900mg Preemptive 1.3 
Shiley et al., 2009 [3566 51/49.5 liver FK, CsA, MMF, P. Universal VGC 900mg GCV 3.3 
Akalin et al. 2003 [19115 NA Kidney-pancreas,kidney, FK, CsA, MMF, P. Induction(ATG) Universal VGC 450mg GCV CC 
Keven et al. 2004 [10176 48.5 total Kidney-pancreas, kidney FKs, P. Induction (alemtuzumab, thymoglobulin) Universal VGC 450mg GCV CC 
Park et al. 2006 [13109 49/50 liver FK, MMF, seroids.Induction (ATG, OKT3). Universal VGC 450mg GCV CC 
Manuel et al. 2007 [12143 44/48 Kidney FK, CsA, AZA, MMF, P. Induction (IL-2rec.a, ATG) Universal VGC 450mg GCV 
Weng et al. 2007 [8500 48.3/45.8 Kidney-pancreas;kidney FK, CsA, P, Sirolimus.Induction (ATG,IL2 antibodies) Universal VGC 450mg GCV 
Walker et al. 2007 [9203 48/49 kidney CsA, AZA, MMF, P.Induction (ALG, alemtuzumab). Universal VGC 450mg GCV 
Avidan et al. 2008 [14221 41.6/40.8 kidney FKTacrolimus, ALG.Induction (ATG, thymoglobulin) Universal VGC 450mg GCV 
Brady et al. 2008 [1764 51/50 liver Tacrolimus, MMF, P.Induction (ATG) Preemptive VGC 450mg Preemptive CC 
Study, year, (Reference) Samplesize Mean age, years (treatment/control) Allograft type Maintenance regimen Induction Preventativestrategy Treatment arm Dose* Controlarm Duration of prophylaxis or preemptive therapy (months) Design:Randomized (R)/Cohort (C)/Case-Control (CC) 
Paya et al., 2004 [5364 45.7/45.3 heart, liver,kidney,kidney-pancreas NA Universal VGC 900 mg GCV 
Zamora et al. 2004 [37230 53.4/54 lung CsA, AZA, P. Universal VGC 900mg GCV+CMV IG and ACV 3-6 
Humar et al. 2005[2680 48.6/NA lung CsA, AZA, MMF, P.Induction (ALG). Universal VGC 900mg GCV 
Khoury et al. 2006[2798 51.9/47.5 kidney FK, CsA, MMF, AZA, P.Induction (thymoglobulin) Preemptive VGC 900mg Preemptive 3.3 
Said et al, 2007 [2864 42.5/38.2 kidney FK, CsA, MMF, P.Induction (ATG,baziliximab) Universal VGC 900mg GCV 0.5 
Guirado et al. 2008 [29150 NA kidney FK, MMF, P.Induction (ALG) Preemptive VGC 900mg Preemptive CC 
Moro et al. 2008 [3053 46/43.2 heart NA Universal VGC 900mg GCV NA CC 
Fernandez et al. 2009 [31100 48.1 total kidney FK, MMF, steroids.Induction (ATG, OKT3). Preemptive VGC 900mg Preemptive 4.3 
Monforte et al. 2009 [32163 51.1/48.5 lung FK, CsA, MMF, AZA.Induction (ATG, baziliximab) Universal VGC 900mg GCV 3.9 C C 
Parreira et al. 2009 [33135 NA kidney FK,CsA, MMF, P.Induction (ALG) Preemptive VGC 900mg Preemptive C C 
Potena et al. 2009 [3440 55/56 heart CsA, MMF, P.Induction (thymoglobulin) Preemptive VGC 900mg Preemptive 1.3 
Shiley et al., 2009 [3566 51/49.5 liver FK, CsA, MMF, P. Universal VGC 900mg GCV 3.3 
Akalin et al. 2003 [19115 NA Kidney-pancreas,kidney, FK, CsA, MMF, P. Induction(ATG) Universal VGC 450mg GCV CC 
Keven et al. 2004 [10176 48.5 total Kidney-pancreas, kidney FKs, P. Induction (alemtuzumab, thymoglobulin) Universal VGC 450mg GCV CC 
Park et al. 2006 [13109 49/50 liver FK, MMF, seroids.Induction (ATG, OKT3). Universal VGC 450mg GCV CC 
Manuel et al. 2007 [12143 44/48 Kidney FK, CsA, AZA, MMF, P. Induction (IL-2rec.a, ATG) Universal VGC 450mg GCV 
Weng et al. 2007 [8500 48.3/45.8 Kidney-pancreas;kidney FK, CsA, P, Sirolimus.Induction (ATG,IL2 antibodies) Universal VGC 450mg GCV 
Walker et al. 2007 [9203 48/49 kidney CsA, AZA, MMF, P.Induction (ALG, alemtuzumab). Universal VGC 450mg GCV 
Avidan et al. 2008 [14221 41.6/40.8 kidney FKTacrolimus, ALG.Induction (ATG, thymoglobulin) Universal VGC 450mg GCV 
Brady et al. 2008 [1764 51/50 liver Tacrolimus, MMF, P.Induction (ATG) Preemptive VGC 450mg Preemptive CC 

NOTE. FK: tacrolimus, AZA: azathioprine, MMF: mycophenolate mofetil, ATG: antithymocyte globulin, ALG: antilymphocyte globulin, CMV: cytomegalovirus, CsA: cyclosporine, P: prednisone, IL-2ra: interleukin-2 receptor antagonist, IG: intravenous immunoglobulin, OKT3: anti-human CD3, GCV: ganciclovir, ACV: acyclovir, VGC: valganciclovir, NA: not available.

*

Daily dose of VGC for normal renal function.

Table 1.

Study Characteristics

Study, year, (Reference) Samplesize Mean age, years (treatment/control) Allograft type Maintenance regimen Induction Preventativestrategy Treatment arm Dose* Controlarm Duration of prophylaxis or preemptive therapy (months) Design:Randomized (R)/Cohort (C)/Case-Control (CC) 
Paya et al., 2004 [5364 45.7/45.3 heart, liver,kidney,kidney-pancreas NA Universal VGC 900 mg GCV 
Zamora et al. 2004 [37230 53.4/54 lung CsA, AZA, P. Universal VGC 900mg GCV+CMV IG and ACV 3-6 
Humar et al. 2005[2680 48.6/NA lung CsA, AZA, MMF, P.Induction (ALG). Universal VGC 900mg GCV 
Khoury et al. 2006[2798 51.9/47.5 kidney FK, CsA, MMF, AZA, P.Induction (thymoglobulin) Preemptive VGC 900mg Preemptive 3.3 
Said et al, 2007 [2864 42.5/38.2 kidney FK, CsA, MMF, P.Induction (ATG,baziliximab) Universal VGC 900mg GCV 0.5 
Guirado et al. 2008 [29150 NA kidney FK, MMF, P.Induction (ALG) Preemptive VGC 900mg Preemptive CC 
Moro et al. 2008 [3053 46/43.2 heart NA Universal VGC 900mg GCV NA CC 
Fernandez et al. 2009 [31100 48.1 total kidney FK, MMF, steroids.Induction (ATG, OKT3). Preemptive VGC 900mg Preemptive 4.3 
Monforte et al. 2009 [32163 51.1/48.5 lung FK, CsA, MMF, AZA.Induction (ATG, baziliximab) Universal VGC 900mg GCV 3.9 C C 
Parreira et al. 2009 [33135 NA kidney FK,CsA, MMF, P.Induction (ALG) Preemptive VGC 900mg Preemptive C C 
Potena et al. 2009 [3440 55/56 heart CsA, MMF, P.Induction (thymoglobulin) Preemptive VGC 900mg Preemptive 1.3 
Shiley et al., 2009 [3566 51/49.5 liver FK, CsA, MMF, P. Universal VGC 900mg GCV 3.3 
Akalin et al. 2003 [19115 NA Kidney-pancreas,kidney, FK, CsA, MMF, P. Induction(ATG) Universal VGC 450mg GCV CC 
Keven et al. 2004 [10176 48.5 total Kidney-pancreas, kidney FKs, P. Induction (alemtuzumab, thymoglobulin) Universal VGC 450mg GCV CC 
Park et al. 2006 [13109 49/50 liver FK, MMF, seroids.Induction (ATG, OKT3). Universal VGC 450mg GCV CC 
Manuel et al. 2007 [12143 44/48 Kidney FK, CsA, AZA, MMF, P. Induction (IL-2rec.a, ATG) Universal VGC 450mg GCV 
Weng et al. 2007 [8500 48.3/45.8 Kidney-pancreas;kidney FK, CsA, P, Sirolimus.Induction (ATG,IL2 antibodies) Universal VGC 450mg GCV 
Walker et al. 2007 [9203 48/49 kidney CsA, AZA, MMF, P.Induction (ALG, alemtuzumab). Universal VGC 450mg GCV 
Avidan et al. 2008 [14221 41.6/40.8 kidney FKTacrolimus, ALG.Induction (ATG, thymoglobulin) Universal VGC 450mg GCV 
Brady et al. 2008 [1764 51/50 liver Tacrolimus, MMF, P.Induction (ATG) Preemptive VGC 450mg Preemptive CC 
Study, year, (Reference) Samplesize Mean age, years (treatment/control) Allograft type Maintenance regimen Induction Preventativestrategy Treatment arm Dose* Controlarm Duration of prophylaxis or preemptive therapy (months) Design:Randomized (R)/Cohort (C)/Case-Control (CC) 
Paya et al., 2004 [5364 45.7/45.3 heart, liver,kidney,kidney-pancreas NA Universal VGC 900 mg GCV 
Zamora et al. 2004 [37230 53.4/54 lung CsA, AZA, P. Universal VGC 900mg GCV+CMV IG and ACV 3-6 
Humar et al. 2005[2680 48.6/NA lung CsA, AZA, MMF, P.Induction (ALG). Universal VGC 900mg GCV 
Khoury et al. 2006[2798 51.9/47.5 kidney FK, CsA, MMF, AZA, P.Induction (thymoglobulin) Preemptive VGC 900mg Preemptive 3.3 
Said et al, 2007 [2864 42.5/38.2 kidney FK, CsA, MMF, P.Induction (ATG,baziliximab) Universal VGC 900mg GCV 0.5 
Guirado et al. 2008 [29150 NA kidney FK, MMF, P.Induction (ALG) Preemptive VGC 900mg Preemptive CC 
Moro et al. 2008 [3053 46/43.2 heart NA Universal VGC 900mg GCV NA CC 
Fernandez et al. 2009 [31100 48.1 total kidney FK, MMF, steroids.Induction (ATG, OKT3). Preemptive VGC 900mg Preemptive 4.3 
Monforte et al. 2009 [32163 51.1/48.5 lung FK, CsA, MMF, AZA.Induction (ATG, baziliximab) Universal VGC 900mg GCV 3.9 C C 
Parreira et al. 2009 [33135 NA kidney FK,CsA, MMF, P.Induction (ALG) Preemptive VGC 900mg Preemptive C C 
Potena et al. 2009 [3440 55/56 heart CsA, MMF, P.Induction (thymoglobulin) Preemptive VGC 900mg Preemptive 1.3 
Shiley et al., 2009 [3566 51/49.5 liver FK, CsA, MMF, P. Universal VGC 900mg GCV 3.3 
Akalin et al. 2003 [19115 NA Kidney-pancreas,kidney, FK, CsA, MMF, P. Induction(ATG) Universal VGC 450mg GCV CC 
Keven et al. 2004 [10176 48.5 total Kidney-pancreas, kidney FKs, P. Induction (alemtuzumab, thymoglobulin) Universal VGC 450mg GCV CC 
Park et al. 2006 [13109 49/50 liver FK, MMF, seroids.Induction (ATG, OKT3). Universal VGC 450mg GCV CC 
Manuel et al. 2007 [12143 44/48 Kidney FK, CsA, AZA, MMF, P. Induction (IL-2rec.a, ATG) Universal VGC 450mg GCV 
Weng et al. 2007 [8500 48.3/45.8 Kidney-pancreas;kidney FK, CsA, P, Sirolimus.Induction (ATG,IL2 antibodies) Universal VGC 450mg GCV 
Walker et al. 2007 [9203 48/49 kidney CsA, AZA, MMF, P.Induction (ALG, alemtuzumab). Universal VGC 450mg GCV 
Avidan et al. 2008 [14221 41.6/40.8 kidney FKTacrolimus, ALG.Induction (ATG, thymoglobulin) Universal VGC 450mg GCV 
Brady et al. 2008 [1764 51/50 liver Tacrolimus, MMF, P.Induction (ATG) Preemptive VGC 450mg Preemptive CC 

NOTE. FK: tacrolimus, AZA: azathioprine, MMF: mycophenolate mofetil, ATG: antithymocyte globulin, ALG: antilymphocyte globulin, CMV: cytomegalovirus, CsA: cyclosporine, P: prednisone, IL-2ra: interleukin-2 receptor antagonist, IG: intravenous immunoglobulin, OKT3: anti-human CD3, GCV: ganciclovir, ACV: acyclovir, VGC: valganciclovir, NA: not available.

*

Daily dose of VGC for normal renal function.

Valgancicylovir 900 mg Daily versus Controls: Direct Comparison

Risk of cytomegalovirus disease.—The risk of developing CMV disease (11 studies; 1313 patients) with valganciclovir 900 mg was 1.06 (95% CI .64–1.76); P = .812; I2 = 29% (Figure 2a). When the analyses were performed separately based on the type of control, the risks were: for ganciclovir controls (6 studies; 790 patients): 1.14 (95% CI .63–2.06);P = .668; I2 = 42%; for preemptive controls (5 studies; 523 patients): .87 (95% CI .27–2.78); P = .816; I2 = 28%. The results by type of allograft are: kidney (5 studies; 547 patients): 1.80 (95% CI .80–4.02); P = .154; I2 = 0%; heart (2 studies; 93 patients): .34 (95% CI .05–2.44); P = .284; I2 = 79%; liver (1 study; 66 patients): 5.29 (95% CI .98–28.57); P = .053; I2 = 0%; lung (2 studies; 243 patients): .70 (95% CI .27–1.81); P = .466; I2 = 0%; multiple allografts (1 study; 364 patients): .92 (95% CI .52–1.61); P = .767; I2 = 0%. The metaregression evaluated the influence of the rate of D+/R- CMV mismatch over the treatment effect and showed no significant changes on therapy response after controlling for CMV mismatch: intercept:-.134; slope:0.003; P = .461.

Risk of leucopenia and neutropenia. —The risk of developing leucopenia (7 studies; 1134 patients) with valganciclovir 900 mg daily was 5.24 (95% CI 2.09–13.15); P = .0004; I2 = 43% (Figure 2b). When the analysis was performed separately based on the type of control, the following results were observed: for ganciclovir controls (4 studies; 671 patients): 2.99 (95%CI 1.62–5.55); P = .0005; I2 = 0%; for preemptive controls (2 studies; 233 patients): 12.39 (95% CI .25–608); P = .205; I2 = 77%; for acyclovir controls (1 study; 230) 21.6 (95% CI 1.20–388); P = .037; I2 = 0%. The results by type of allograft are: kidney (3 studies; 297 patients): 8.48 (95% CI 1.07–66.85); P = .042; I2 = 59%; lungs (3 studies; 473 patients): 7.40 (95% CI 2.30–23.79); P = .0008; I2 = 0%; multiple allografts (1 study; 364 patients): 2.06 (95% CI .95–4.47); P = .065; I2 = 0%. The risk of developing leucopenia was also analyzed by use of induction therapy (present: 3.90 [95% CI .95–16.0]; P = .058; I2 = 0%); and absent: 10.51 (95% CI 1.97–55.99); P = .005; I2 = 44%); and by use of MMF (present: 4.02 (95% CI .94–17.24); P = .061; I2 = 0%; and absent: no studies were available. The risk of neutropenia with valganciclovir 900 mg daily (5 studies; 907 patients) was 3.72 (1.61–8.60); P = .002; I2 = 0%.

Risk of an Allograft rejection, allograft loss, and death. —After controlling for type of allograft, the risk of acute rejection was 1.71 (95% CI .45—6.50); P = .434; I2 = 75%; risk of allograft loss: 1.23 (95% CI .45—3.36); P = .683; I2 = 14%; and risk of death: 1.17 (95% CI .56—2.43); P = .678; I2 = 0%.

Valganciclovir 450 mg Daily versusControls: Direct Comparison

Risk of cytomegalovirus disease. —The risk of developing CMV disease (8 studies; 1531 patients) with valganciclovir 450 mg daily was .77 (95% CI .49–1.18); P = .230; I2 = 24% (Figure 3a). When the analyses were performed separately based on the type of control, the following results were observed: for ganciclovir controls (6 studies; 1220 patients): .92 (95%CI .57,1.48);P = .723;I2 = 0%; for preemptive controls (2 studies; 346 patients): .53 (95% CI .18–1.53); P = .239; I2 = 69%. The results by type of allograft are: kidney (3 studies; 567 patients): .77 (95% CI .29–2.05); P = .598; I2 = 71%; kidney and pancreas (3 studies; 826 patients): .73 (95% CI .41–1.28); P = .270; I2 = 0%; liver (2 studies; 173 patients): 1.22 (95% CI .27–5.44); P = .796; I2 = 0%. The metaregression evaluated the influence of the rate of D+/R- CMV mismatch over the treatment effect and showed no significant changes on therapy response after controlling for CMV mismatch: intercept:-.0346; slope:-.003; P = .817.

Risk of leucopenia and neutropenia. —The risk of developing leucopenia (5 studies; 781 patients) with valganciclovir 450 mg was 1.58 (95% CI .96–2.61);P = .073;I2 = 36% (Figure 3b). When the analyses were performed separately based on the type of control, the following results were observed: for ganciclovir controls (3 studies; 435 patients): 1.17 (95% CI .75–1.81); —P = .494; I2 = 0%; for preemptive controls (2 studies; 346 patients): 2.72 (95% CI 1.44–5.16); P = .002; I2 = 0%. The results by type of allograft are: kidney (2 studies; 346 patients): 2.72 (95% CI 1.44–5.16); P = .002; I2 = 0%; kidney and pancreas (2 studies; 326 patients): 1.12 (95% CI .71–1.76); P = .638; I2 = 0%; liver (1 study; 109 patients): 2.14 (95% CI .40–11.52); P = .377; I2 = 0%. The risk of developing leucopenia was also analyzed by the use of induction therapy (present: 1.92 [95% CI .92–4.01]; P = .082; I2 = 52%; absent: 1.09 (95% CI .55–2.16); P = .80; I2 = 0%); and by use of MMF (present: 3.13 (95% CI .72–13.49); P = .127; I2 = 0%; absent: 1.59 (95% CI .61–4.16); P = .343; I2 = 73%. The risk of neutropenia with valganciclovir 450 mg daily (3 studies; 367 patients) was 2.92 (.32–27.03); P = .346; I2 = 36%).

Risk of acute rejection, allograft loss, and death. —After controlling for type of allograft, the risk of acute rejection was .80 (95% CI .50–1.28); P = .346; –I2 = 87%; risk of allograft loss: .67 (95% CI .27–1.64); P = .376; I2 = 14%; and risk of death: 1.12 (95% CI .47–2.71); P = .799; I2 = 13%.

Valganciclovir 900 mg Daily versus Valganciclovir 450 mg Daily: Adjusted Indirect Treatment Comparison

Risk of cytomegalovirus disease. —The risk of CMV disease with valganciclovir 900 mg daily was associated with: similar risk of CMV disease: OR=1.38 (.84 –2.25); P = .19, compared to valganciclovir 450 mg daily. For ganciclovir controls only the risk was 1.24 (95%CI .69–2.24; P = .477) and for preemptive controls only the risk was 1.64 (95% CI .59–4.58; P = .343).

Risk of leucopenia and neutropenia. —The risk of leucopenia with valganciclovir 900 mg daily was significantly increased by 3.32 times (95% CI 1.76–6.26; P = .0002) compared to valganciclovir 450 mg daily. The risk of leucopenia with valganciclovir 900 mg versus 450 mg for ganciclovir controls only was 2.56 (95% CI 1.24–5.29; P = .01) and for preemptive controls only was 4.56 (95% CI .57–36.26; P = .151). The risk of neutropenia could not be evaluated because the valganciclovir 900 mg trials used 1500 cells/mm3 cutoff, while the 450 mg trials used 500 cells/mm3 cutoff.

Risk of allograft rejection, allograft loss, and death. —The risk of acute rejection with valganciclovir 900 mg daily was 2.56 (1.50–4.53); P = .0005, compared to valganciclovir 450 mg daily. For ganciclovir controls only: 1.79 (95% CI .96–3.35; P = .067); and for preemptive controls only: 3.76 (95% CI .85–16.72; P = .081). The risk of allograft loss was 1.85 (.15–1.91; P = .34) for all controls; the risk was not evaluable by control type. The risk of death was .90 (.31–2.63); P = .85 for all controls; for ganciclovir controls only: .81 (95% CI .42–3.70; P = .700); and for preemptive controls only: 2.33 (95% CI .01–14.87; P = .642).

Sensitivity Analysis

All analyses were evaluated based on the study design. The risk of CMV disease for valganciclovir 900 mg daily was: randomized: 1.53 (95% CI .63–3.73); P = .349; I2 = 43%; cohort: 1.13 (95% CI .25–5.01); P = .875; I2 = 59%; case-controls: .62 (95% CI .30–1.28); P = .198; I2 = 0%. The risk of CMV disease for valganciclovir 450 mg daily was: cohort: .69 (95% CI .34–1.42); P = .318; I2 = 58%; case-controls: .92 (95%CI .49–1.74); P = .803; I2 = 0%. The risk of CMV disease was also examined by the type of immunosuppression (tacrolimus or cyclosporine; presence or absence of induction, mycophenolate, and steroids) all of which showed no significant differences on the effects of either valganciclovir dose versus controls. The risk of developing CMV disease remained similar independent of the length of CMV prophylaxis and the length of study follow up (data not shown).

Power Analysis

Valganciclovir 900 mg versus controls.—Based on the CMV disease rate of 10.4% found in the control arm of our main analyses (Figures 2a), and on the expected CMV disease rate of 5.2% with valganciclovir 900 mg, a sample size of 657 in each group has a 94% power to detect a 5% decrease with a significance level (alpha) of .05 (two-tailed).

Valganciclovir 450 mg versus controls. —Based on the CMV disease rate of 10.2% found in the control arm of our main analyses (Figures 3a), and on the expected CMV disease rate of 5.1% with valganciclovir 450 mg, a sample size of 766 in each group has a 96% power to detect a 5% decrease with a significance level (alpha) of .05 (two-tailed).

Valganciclovir 900 mg versus 450 mg. —Based on the CMV disease rate of 12.7% with VGC 900 mg (Figure 2a), and a equivalence margin (delta) of 6.3%, a sample size of 670 in each group has a 97% power to show that the CMV disease rate is the same in both groups and excludes differences greater than delta in either direction.

Publication Bias Analysis

For the valganciclovir 900mg: No publication bias was detected by Egger regression (intercept = .395; standard error = .781; P = .625), or by Begg and Mazumdar rank correlation (Kendall's tau = .127; P = .640). For the valganciclovir 450mg: No publication bias was detected by Egger regression (intercept = 1.281; standard error = 1.424; P = .402), or by Begg and Mazumdar rank correlation (Kendall's tau = .392; P = .173).

DISCUSSION

The direct comparison of valganciclovir 900 mg and 450 mg daily showed similar efficacy for preventing CMV disease independent of the type of control, allograft, CMV mismatch, prophylaxis duration, or length of follow up. Neither dose was superior to controls (ganciclovir and preemptive) despite adequate statistical power. The adjusted indirect comparison demonstrated that there were no statistical differences between the 450 and 900 mg daily dosing with respect to the prevention of CMV disease after transplantation. In fact, based on our study sample size, VGC 900 and 450 mg showed equivalent efficacy with 97% statistical power.

The risk of CMV disease was consistently numerically higher with the 900 mg dose, independent of the type of controls. This suggests that the 450 mg dose could be associated with better CMV disease prevention. Paradoxical as it seems, a biological explanation could be that patients receiving 900 mg would be more prone to late-onset CMV disease than the ones receiving 450 mg. Singh et al [37] have suggested that valganciclovir 900 mg may not allow the host immune system to be exposed to low-level CMV antigenemia. Thus, once the prophylaxis is completed, the patient would be naïve from the immunological standpoint and consequently more vulnerable to CMV infections. We hypothesize that the lower rate of CMV disease observed in our study with 450 mg daily may be due to the fact that 450 mg may not lead to this naïve immunological state, so when valganciclovir 450 mg prophylaxis is completed, the patient would still be able to mount the specific T cell response against cytomegalovirus.

The risk of developing leucopenia was 5 times statistically significantly higher with valganciclovir 900 mg compared to controls in the direct comparison. This higher risk remained remarkably similar independent of the type of control or allograft. Further, the risk was not confounded by other drugs known for causing leucopenia (mycophenolate and induction). Similarly, the risk of absolute neutropenia was also high (3.7 times) with 900 mg. Interestingly, the risk of both leucopenia and neutropenia with valganciclovir 450 mg were not significantly higher compared to controls and adjusted for allograft or other drugs. The adjusted indirect comparison showed that the use of valganciclovir 900 mg was associated with 3.3 times higher risk of leucopenia than valganciclovir 450 mg (P = .0002). This is a very important finding since it is very unlikely that the drug sponsor will ever perform a large trial comparing these 2 doses due to the costs and logistics that would be needed.

Allograft rejection was similar in the direct comparisons for both regimens; however, the adjusted indirect comparison demonstrated a statistically significant higher risk of allograft rejection with 900 mg compared to 450 mg. This was an unexpected finding, but one possible hypothesis is that if the 450 mg dose would be associated with less late-onset CMV disease; thus, the lower rejection rate would be a consequence of the lower rate of CMV disease, which is well known to be associated with rejection [1]. Another possibility is that the occurrence of leucopenia often leads to a temporary decrease or discontinuation of immunosuppressive agents, which might leave the transplant recipient more vulnerable to rejection. It is possible that our results could be confounded by different immunosuppressive regimens used in the trials, however, this is unlikely because all trials included in our analysis used the same regimen for both arms. One last explanation is that this was found just by chance, but based on the P = .0005, we believe that it is less likely.

We note the following limitations: First, neutropenia was poorly reported in most studies. In addition, the use of different cutoffs for neutropenia preclude the adjusted indirect comparison between both valganciclovir doses. Second, no systematic collection and reporting of bacterial and fungal infections were seen in most studies; this did not allow us to evaluate the direct consequences of leucopenia and neutropenia. Third, adjusted indirect comparisons may not reflect direct comparisons. However, based on the fact that the controls for both doses were quite similar, and the results remained highly consistent after adjustments for several variables, we believe that our findings are a robust representation of the effects of these 2 valganciclovir doses. Fourth, CMV resistance data could not be analyzed because of the lack of reporting, but based on the very low rate (<2%) of CMV resistance observed in studies that evaluated ganciclovir 3 grams a day [38] (similar pharmacological exposure to valganciclovir 450mg once a day), it is expected that valganciclovir 450mg should not be associated with more resistance than the 900mg dose. Last, it is important to note that our study was specifically aimed at valganciclovir universal prophylaxis, so our results may not be applicable to preemptive approach for CMV, in which higher valganciclovir dose may still be necessary since viremia is already present.

The performance of a collaborative clinical trial among transplant centers could bring additional light to this issue. A randomized trial directly comparing 900 mg versus 450 mg valganciclovir for CMV prophylaxis would potentially evaluate the incidence of CMV disease and viremia, rate and timing of CMV seroconversion of D+/R- patients, magnitude and timing of CMV-specific cellular immune function, allograft outcomes, and drug side effects.

In conclusion, the prevention of CMV disease was equivalent with both doses. Valganciclovir 900 mg daily was significantly associated with increase in the risk of leucopenia and rejection compared to valganciclovir 450 mg daily. Valganciclovir 450 mg appears to be as beneficial as and safer than valganciclovir 900 mg for CMV prophylaxis in SOT.

The authors thank Ashley Calhoon for the preparation of this manuscript.

Potential conflict of interest. All authors: no conflicts.

References

1.
Kalil
AC
Levitsky
J
Lyden
E
Stoner
J
Freifeld
AG
Meta-analysis: the efficacy of strategies to prevent organ disease by cytomegalovirus in solid organ transplant recipients
Ann Intern Med
2005
, vol. 
143
 (pg. 
870
-
80
)
2.
Fishman
JA
Infection in solid-organ transplant recipients
N Engl J Med
2007
, vol. 
357
 (pg. 
2601
-
14
)
3.
Kalil
AC
Freifeld
AG
Lyden
ER
Stoner
JA
Valganciclovir for cytomegalovirus prevention in solid organ transplant patients: an evidence-based reassessment of safety and efficacy
PLoS One
2009
, vol. 
4
 pg. 
e5512
 
4.
Pescovitz
MD
Rabkin
J
Merion
RM
, et al. 
Valganciclovir results in improved oral absorption of ganciclovir in liver transplant recipients
Antimicrob Agents Chemother
2000
, vol. 
44
 (pg. 
2811
-
5
)
5.
Paya
C
Humar
A
Dominguez
E
, et al. 
Efficacy and safety of valganciclovir vs. oral ganciclovir for prevention of cytomegalovirus disease in solid organ transplant recipients
Am J Transpl
2004
, vol. 
4
 (pg. 
611
-
20
)
6.
Valcyte (valganciclovir HCL tablets) dear healthcare professional letter sep 2003 (letter from roche pharmaceuticals)
 
[homepage on the Internet] Accessed at August 17, 2010 Available at: http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm169502.htm. Accessed 17 August 2010
7.
Abate
MT
Triolo
D
Nord
EP
CMV infection in renal transplant recipients: Five year experience with campath-1H induction and early steroid withdrawal
Am J Transpl
2009
, vol. 
9
 pg. 
680
 
8.
Weng
FL
Patel
AM
Wanchoo
R
, et al. 
Oral ganciclovir versus low-dose valganciclovir for prevention of cytomegalovirus disease in recipients of kidney and pancreas transplants
Transplantation
2007
, vol. 
83
 (pg. 
290
-
6
)
9.
Walker
JK
Scholz
LM
Scheetz
MH
, et al. 
Leukopenia complicates cytomegalovirus prevention after renal transplantation with alemtuzumab induction
Transplantation
2007
, vol. 
83
 (pg. 
874
-
82
)
10.
Keven
K
Basu
A
Tan
HP
, et al. 
Cytomegalovirus prophylaxis using oral ganciclovir or valganciclovir in kidney and pancreas-kidney transplantation under antibody preconditioning
Transpl Proc
2004
, vol. 
36
 (pg. 
3107
-
12
)
11.
Gruber
SA
Garnick
J
Morawski
K
, et al. 
Cytomegalovirus prophylaxis with valganciclovir in african-american renal allograft recipients based on donor/recipient serostatus
Clin Transpl
2005
, vol. 
19
 (pg. 
273
-
8
)
12.
Manuel
O
Venetz
JP
Fellay
J
, et al. 
Efficacy and safety of universal valganciclovir prophylaxis combined with a tacrolimus/mycophenolate-based regimen in kidney transplantation
Swiss Med Wkly
2007
, vol. 
137
 (pg. 
669
-
76
)
13.
Park
JM
Lake
KD
Arenas
JD
Fontana
RJ
Efficacy and safety of low-dose valganciclovir in the prevention of cytomegalovirus disease in adult liver transplant recipients
Liver Transpl
2006
, vol. 
12
 (pg. 
112
-
6
)
14.
Avidan
YP
Paul
M
Rahamimov
R
, et al. 
Selective low-dose valganciclovir for prevention of cytomegalovirus disease following kidney transplantation
J Infect
2008
, vol. 
57
 (pg. 
236
-
40
)
15.
Luan
FL
Stuckey
LJ
Park
JM
Kaul
D
Cibrik
D
Ojo
A
Six-month prophylaxis is cost effective in transplant patients at high risk for cytomegalovirus infection
J Am Soc Nephrol
2009
, vol. 
20
 (pg. 
2449
-
58
)
16.
Dupuis
R
Harris
M
Gillis
K
, et al. 
Experience with low-dose valganciclovir prophylaxis in adult liver transplant recipients
Transpl Proc
2007
, vol. 
39
 (pg. 
3266
-
70
)
17.
Brady
RL
Green
K
Frei
C
Maxwell
P
Oral ganciclovir versus valganciclovir for cytomegalovirus prophylaxis in high-risk liver transplant recipients
Transpl Infect Dis
2009
, vol. 
11
 (pg. 
106
-
11
)
18.
Gabardi
S
Magee
CC
Baroletti
SA
Powelson
JA
Cina
JL
Chandraker
AK
Efficacy and safety of low-dose valganciclovir for prevention of cytomegalovirus disease in renal transplant recipients: a single-center, retrospective analysis
Pharmacotherapy
2004
, vol. 
24
 (pg. 
1323
-
30
)
19.
Akalin
E
Sehgal
V
Ames
S
, et al. 
Cytomegalovirus disease in high-risk transplant recipients despite ganciclovir or valganciclovir prophylaxis
Am J Transpl
2003
, vol. 
3
 (pg. 
731
-
5
)
20.
DerSimonian
R
Laird
N
Meta-analysis in clinical trials
Control Clin Trials
1986
, vol. 
7
 (pg. 
177
-
88
)
21.
Bucher
HC
Guyatt
GH
Griffith
LE
Walter
SD
The results of direct and indirect treatment comparisons in meta-analysis of randomized controlled trials
J Clin Epidemiol
1997
, vol. 
50
 (pg. 
683
-
91
)
22.
Song
F
Altman
DG
Glenny
AM
Deeks
JJ
Validity of indirect comparison for estimating efficacy of competing interventions: empirical evidence from published meta-analyses
BMJ
2003
, vol. 
326
 pg. 
472
 
23.
Mulrow
CD
Cook
D
Systematic reviews: synthesis of best evidence for health care decisions
Philadelphia, Pennsylvania: the American College of Physicians,
1998
pg. 
117
 
24.
Sterne
JA
Gavaghan
D
Egger
M
Publication and related bias in meta-analysis: power of statistical tests and prevalence in the literature
J Clin Epidemiol
2000
, vol. 
53
 (pg. 
1119
-
29
)
25.
Begg
CB
Mazumdar
M
Operating characteristics of a rank correlation test for publication bias
Biometrics
1994
, vol. 
50
 (pg. 
1088
-
101
)
26.
Humar
A
Kumar
D
Preiksaitis
J
, et al. 
A trial of valganciclovir prophylaxis for cytomegalovirus prevention in lung transplant recipients
Am J Transpl
2005
, vol. 
5
 (pg. 
1462
-
8
)
27.
Khoury
JA
Storch
GA
Bohl
DL
, et al. 
Prophylactic versus preemptive oral valganciclovir for the management of cytomegalovirus infection in adult renal transplant recipients
Am J Transpl
2006
, vol. 
6
 (pg. 
2134
-
43
)
28.
Said
T
Nampoory
MR
Pacsa
AS
, et al. 
Oral valgancyclovir versus intravenous gancyclovir for cytomegalovirus prophylaxis in kidney transplant recipients
Transpl Proc
2007
, vol. 
39
 (pg. 
997
-
9
)
29.
Guirado
L
Rabella
N
Diaz
JM
, et al. 
Prophylactic and pre-emptive therapy for cytomegalovirus infection in kidney transplant patients using oral valganciclovir
Nefrologia
2008
, vol. 
28
 (pg. 
293
-
300
)
30.
Moro
JA
Almenar
L
Martinez-Dolz
L
, et al. 
Utility of oral valganciclovir for cytomegalovirus prophylaxis: Does it improve treatment compliance?
Transpl Proc
2008
, vol. 
40
 (pg. 
3063
-
4
)
31.
Fernandez
A
Amezquita
Y
Fernandez-Tagarro
E
, et al. 
Prophylaxis and treatment of cytomegalovirus infection postrenal transplantation in two madrid units
Transpl Proc
2009
, vol. 
41
 (pg. 
2416
-
8
)
32.
Monforte
V
Lopez
C
Santos
F
, et al. 
A multicenter study of valganciclovir prophylaxis up to day 120 in CMV-seropositive lung transplant recipients
Am J Transpl
2009
, vol. 
9
 (pg. 
1134
-
41
)
33.
Parreira
L
Bruges
M
Gaspar
A
Weigert
A
Machado
D
Prevention of cytomegalovirus disease in renal transplantation: single-center experience
Transpl Proc
2009
, vol. 
41
 (pg. 
877
-
9
)
34.
Potena
L
Grigioni
F
Magnani
G
, et al. 
Prophylaxis versus preemptive anti-cytomegalovirus approach for prevention of allograft vasculopathy in heart transplant recipients
J Heart Lung Transpl
2009
, vol. 
28
 (pg. 
461
-
7
)
35.
Shiley
KT
Gasink
LB
Barton
TD
Pfeiffenberger
P
Olthoff
KM
Blumberg
EA
Increased incidence of cytomegalovirus infection in high-risk liver transplant recipients receiving valganciclovir prophylaxis versus ganciclovir prophylaxis
Liver Transpl
2009
, vol. 
15
 (pg. 
963
-
7
)
36.
Zamora
MR
Nicolls
MR
Hodges
TN
, et al. 
Following universal prophylaxis with intravenous ganciclovir and cytomegalovirus immune globulin, valganciclovir is safe and effective for prevention of CMV infection following lung transplantation
Am J Transpl
2004
, vol. 
4
 (pg. 
1635
-
42
)
37.
Singh
N
Late-onset cytomegalovirus disease as a significant complication in solid organ transplant recipients receiving antiviral prophylaxis: a call to heed the mounting evidence
Clin Infect Dis
2005
, vol. 
40
 (pg. 
704
-
8
)
38.
Boivin
G
Goyette
N
Gilbert
C
, et al. 
Absence of cytomegalovirus-resistance mutations after valganciclovir prophylaxis, in a prospective multicenter study of solid-organ transplant recipients
J Infect Dis
2004
, vol. 
189
 (pg. 
1615
-
8
)

Comments

0 Comments
Submit a comment
You have entered an invalid code
Thank you for submitting a comment on this article. Your comment will be reviewed and published at the journal's discretion. Please check for further notifications by email.