Skip Navigation

Valacyclovir Provides Optimum Acyclovir Exposure for Prevention of Cytomegalovirus and Related Outcomes after Organ Transplantation

  1. Paul Fiddian,
  2. Caroline A. Sabin and
  3. Paul D. Griffiths
  1. Royal Free and University College Medical School (Royal Free Campus), London, United Kingdom
  1. Reprints or correspondence: Dr. Paul Fiddian, Dept. of Virology, Royal Free and University College Medical School, Rowland Hill St., London NW3 2PF, United Kingdom (paul.fiddian{at}which.net)

Abstract

A meta-analysis of 12 randomized trials (1574 patients) examined herpesvirus and related outcomes following organ transplantation over a range of acyclovir exposures (including valacyclovir). Overall, cytomegalovirus (CMV) infection (odds ratio [OR], 0.44; 95% confidence interval [CI], 0.34–0.57; P<.001), CMV disease (OR, 0.41; 95% CI, 0.31–0.54; P<.001), death (OR, 0.60; 95% CI, 0.40–0.90; P=.01), opportunistic infection (OR, 0.70; 95% CI, 0.53–0.91; P=.009), acute graft rejection (OR, 0.67; 95% CI, 0.52–0.86; P<.001), herpes simplex virus disease (OR, 0.17; 95% CI, 0.12–0.24; P<.001), and varicella-zoster virus disease (OR, 0.06; 95% CI, 0.01–0.25; P<.001) were significantly reduced. Increased acyclovir exposure influenced more end points: Maximum efficacy resulted from valacyclovir (8 g/day). Increasing acyclovir exposure to that achieved with valacyclovir extends benefits of prophylaxis to include impact on graft rejection and opportunistic infections

Cytomegalovirus (CMV) remains the leading infectious cause of morbidity in solid organ transplant recipients. The risk of CMV disease is greatest in the second month after transplant and is largely predicted by donor (D) and recipient (R) CMV serostatus and is determined by high CMV loads [1, 2]. The risk of disease may be twice as high in D+/R patients than in R+ patients and the disease is more severe [35]. Recipients of liver transplants are also at about twice the risk of CMV disease than renal allograft recipients [4]. CMV disease may manifest as a less severe CMV syndrome or as organ involvement such as pneumonitis, hepatitis, or nephritis. CMV is also associated with acute graft rejection [1, 6], development of opportunistic bacterial or fungal infections [2, 7], overall hospitalization [4], and decreased graft and patient survival [8, 9]. Other herpesviruses, including herpes simplex virus (HSV) and varicella-zoster virus (VZV), are also reactivated following organ transplantation [5] and may contribute to morbidity

Although effective therapy has virtually eliminated mortality from CMV disease, morbidity and increased overall costs of transplantation associated with CMV persist [1, 10]. Until recently, a lack of effective oral agents limited successful prophylaxis after organ transplantation. High-dose oral acyclovir was the first antiviral tested for its ability to prevent CMV disease in renal allograft recipients [11]. Despite impressive results in this study, the use of acyclovir did not gain wide acceptance, perhaps largely due to the misconception that acyclovir was not phosphorylated in CMV-infected cells. Although lacking a thymidine kinase, CMV possesses another gene product, UL97, which is responsible for phosphorylation of acyclovir to the monophosphate [12]

Over the past 10 years, a number of randomized controlled open studies with historic control groups and patient series have been reported, but the controversy over the value of acyclovir in CMV infection has persisted. Although more potent oral therapies are in development (e.g., ganciclovir in liver transplant recipients [13] and valganciclovir promises increased bioavailability of ganciclovir [14] and valacyclovir in renal allograft recipients [5]), it is necessary to consider both risk-benefit and cost-benefit ratios of the therapies before selecting the optimum regimen for each situation. Thus, it is important to utilize all available data to determine the effect of increasing acyclovir exposure for prophylaxis of CMV and other outcomes. In order to address this issue, we undertook a meta-analysis of all controlled randomized trial evidence on acyclovir or valacyclovir in solid organ transplantation. During preparation of this work, another group published a meta-analysis of antiviral prophylaxis in solid organ transplantation [15]. Findings in that report differ significantly from ours because the focus was on ganciclovir trials; only 3 of the acyclovir trials included in our report are discussed

Methods

Identification and selection of studiesTwelve trials were identified in total, 11 through MEDLINE searches and review of references listed in original articles (1539 patients) and 1 as yet unpublished study available to the authors (35 patients; Coates K, personal communication, GlaxoSmithKline). We considered all prospective randomized efficacy trials of oral acyclovir or valacyclovir versus no treatment or placebo plus 1 trial that used as a control, short-duration ganciclovir therapy plus human immune globulin for only 7 days after transplant. Trials that used more prolonged intravenous ganciclovir prophylaxis or preemptive therapy were excluded as such approaches are known to be effective and may not represent untreated controls. Nonrandomized studies with historic control groups and open-patient series were not included in the meta-analysis

End pointsWe defined 7 end points for formal analysis as these were considered directly or indirectly related to herpesviruses: CMV infection, CMV disease, death, opportunistic infection (OI), acute graft rejection, HSV disease, and VZV disease. Data were extracted from original articles and from a small as yet unpublished trial. Additional information collected included acyclovir exposure (based on intended dose of acyclovir or valacyclovir), CMV serostatus of recipient and donor if known, type of organ transplanted, and whether cause of death was reported as associated with an infection

Statistical analysisMantel-Haenszel estimates of the odds ratios (ORs) associated with acyclovir exposure compared with the controls were calculated for each end point by using the FREQ procedure in the Statistical Analysis System statistical package. We used the Breslow-Day method to test for homogeneity of ORs across the different studies. Because there was little evidence of heterogeneity, these data are not included. The analyses were repeated by stratifying by dose of acyclovir or valacyclovir received (valacyclovir 8 g/day; oral acyclovir 3.2, 2.0, and <1 g/day) and by type of transplant (kidney or liver)

Results

Individual trial characteristicsWe identified 12 trials ([4, 5, 11, 1623]; Coats K, personal communication, GlaxoSmithKline) for inclusion in the meta-analysis (table 1). These included 3 trials of low-dose acyclovir (600–800 mg/day; daily area under the curve [AUC], <36 mM/h), 2 trials of middose acyclovir (2 g/day; daily AUC, 65 mM/h), 6 trials of high-dose acyclovir (3.2 g/day; daily AUC, 90 mM/h), and 1 large trial of high-dose valacyclovir (8 g/day; daily AUC, 480 mM/h). Organs transplanted were kidney, 7 trials; liver, 4 trials; and liver, kidney, and/or pancreas, remaining study. Seven trials employed a placebo control group, 4 trials had an observation group, and the control group in the other study involved had ineffective therapy (short-course ganciclovir and immune globulin). In total, 1574 patients were included in the analysis: 793 in the active treatment arm and 781 in the control arm. On average, 72% of patients contributed data for each end point but this ranged from only 56% for OIs to 84% for death and 91% for CMV disease. Of the 1393 patients with known CMV serostatus, 68% were seropositive and 32% were seronegative before transplant. The donor CMV serostatus was known for 95% of the latter patients

Figure 1

Odds ratios (OR) and 95% confidence intervals for 11 trials with high-dose (HD) and mid-dose (MD) acyclovir (ACV) exposures and for total overall populations for CMV disease end point. OR for an effect of low-dose acyclovir exposure on CMV disease could not be derived (too few events). VACV, valacyclovir. †, Coats K, personal communication, GlaxoSmithKline

Table 1

Design of trials of acyclovir (Acy) and valacyclovir (Val) included in the meta-analysis

CMV infection and diseaseOverall, there was a 56% decrease in the risk of CMV infection associated with prophylactic therapy (OR, 0.44; 95% confidence interval [CI], 0.34–0.57; P<.001; table 2). CMV disease was also significantly reduced, by 59%, as a result of acyclovir exposure (OR, 0.41; 95% CI, 0.31–0.54; P<.001). Figure 1 shows individual trial results and the overall CMV disease outcome. Where CMV serostatus of the recipients was known, significant effects on CMV disease could be determined for both high- and medium-risk patients. For D+/R patients, the incidence was reduced from 46% to 21% (OR, 0.30; 95% CI, 0.19–0.48; P<.001) and for R+ patients from 20% to 11% (OR, 0.48; 95% CI, 0.33–0.72; P<.001)

Table 2

Effect of therapy on herpesvirus and related end points

Related outcomesThere was a 40% overall reduction in the risk of death from all causes as a result of prophylaxis (OR, 0.60; 95% CI, 0.40–0.90; P=.01). When we examined the reported causes of death, the only notable difference between treatment groups was a 54% decrease in infection-associated deaths in the group exposed to acyclovir after transplant (OR, 0.46; 95% CI, 0.23–0.91; P=.03). In particular, there were fewer cases of bacterial and/or fungal sepsis (17 vs. 8) and fewer CMV-associated deaths (5 vs. 1)

Overall, there was a 30% reduction in the incidence of OIs associated with prophylactic therapy (OR, 0.70; 95% CI, 0.53–0.91; P=.009). Where specific information was available, this could also be explained by a reduction in major bacterial and fungal infections. There was also a 33% reduction in patients who experienced acute graft rejection in the group exposed to acyclovir (OR, 0.67; 95% CI, 0.52–0.86; P<.001)

Other herpesvirus infectionsThe incidence of both HSV and VZV disease was dramatically reduced by exposure to acyclovir. For HSV the rate of affected patients decreased by 83% (OR, 0.17; 95% CI, 0.12–0.24; P<.001), while for VZV the rate declined by 94% (OR, 0.06; 95% CI, 0.01–0.25; P<.001). There were insufficient data for the remaining herpesviruses to allow any analyses

Effects of increasing acyclovir exposureWith increasing acyclovir exposure, an enhanced beneficial effect was detected for an increasing number of end points (table 3). The most sensitive herpesvirus, HSV, was similarly suppressed across the whole dose range. Low-dose acyclovir appeared to have no effect on the remaining outcomes. As the acyclovir exposure increased, an effect on CMV infection and disease emerged, the latter becoming maximal at the highest exposure resulting from valacyclovir administration. These data therefore confirm that high-dose oral acyclovir significantly reduces CMV disease (OR, 0.43; 95% CI, 0.29–0.64; P<.001) but indicate that this can be further improved by valacyclovir (OR, 0.27; 95% CI, 0.16–0.48; P<.001). During prophylactic treatment (data not shown), valacyclovir prevented 95% of CMV disease [5], whereas high-dose acyclovir prevented only 56% of episodes [4, 11, 2123]. Survival differences at each dose level failed to achieve significance, probably because of the small numbers of deaths in each exposure group. Finally, both OIs and acute graft rejection were positively influenced only at the highest acyclovir exposure possible with valacyclovir

Table 3

Effects of increasing acyclovir (Acy) exposure on herpesvirus and related end points shown as odds ratio (95% confidence interval), P value

Analysis of renal and liver transplant recipientsAll end points for renal transplant recipients, apart from death, remained significantly influenced by acyclovir exposure (table 4). For the liver transplant recipients (just 24% of the population), the results are remarkably consistent with the overall findings when allowance is made for the level of acyclovir exposure. In particular, there were significant effects on CMV infection and disease, although the effect on survival in favor of acyclovir exposure just failed to achieve significance. When only infection-associated deaths were considered, significance was evident (OR, 0.20; 95% CI, 0.06–0.66; P=.008). There were no discernable effects on OIs or acute graft rejection, but no liver transplant recipients received valacyclovir. Consistent results were found for CMV disease independent of whether the CMV serostatus was D+/R or R+ (table 5)

Table 4

Effect of acyclovir and valacyclovir on herpesvirus and related end points in kidney and liver transplant recipients

Table 5

Effect of acyclovir (Acy) or valacyclovir (Val) therapy on cytomegalovirus (CMV) disease stratified by donor and recipient CMV serostatus

Discussion

This meta-analysis confirms an effect of acyclovir on CMV infection and disease. Although greater efficacy results from the higher acyclovir exposure achieved with high-dose valacyclovir, these data clearly demonstrate that high-dose oral acyclovir prophylaxis significantly reduces CMV disease following solid organ transplantation. By subgroup analysis, this effect is evident for both renal and liver transplantation

With the development of new diagnostic tests and greater understanding of the disease, the definitions of CMV infection and disease have altered over the course of the studies included in this meta-analysis. Each study also measured different end points and although CMV disease was measured in 10 of the 12 studies, no single end point was measured in all studies. For these reasons, the results obtained must be interpreted with some caution; however, the methods used for the meta-analysis may be considered somewhat conservative because proportions of patients were used for analysis rather than time to event. This takes no account of the delay in onset of most end points found in the active treatment groups of individual trials in addition to the reduced incidence. As a result, the findings can be considered strong evidence of real treatment effects but the ORs will tend to underestimate the actual impact of treatment. Nevertheless, the best available evidence from the meta-analysis indicates that high-dose acyclovir can reduce CMV disease incidence by up to 57% and high-dose valacyclovir by ⩾73% over a 6-month period. During therapy, CMV disease was reduced by 58% and 95% with high-dose acyclovir and valacyclovir, respectively [5]

Clearly, both HSV and VZV can effectively be prevented by all of the doses used and the data in the patient populations studied do not discriminate between the differing acyclovir exposures that result. Although it is not possible to rule out association with other herpesviruses (e.g., Epstein-Barr or human herpesvirus type 6), considerable data link CMV with the remaining outcomes. In this case, again we would expect higher acyclovir exposures to result in improved efficacy. For OIs and graft rejection, there is no clear evidence of an effect until the highest exposure is reached. Indeed, high-dose valacyclovir is the only prophylactic antiviral therapy shown to influence these end points. In the case of death, a positive trend was evident with mid- or high-dose acyclovir or valacyclovir and seemed to correlate with similar effects against CMV infection

Of interest, for liver transplantation where death is a more frequent outcome, there appeared to be greater evidence of a survival benefit. Also of note is that the overall survival benefit and impact on OIs were both apparently driven by a reduction in major bacterial and fungal infections. Association of these outcomes with CMV is reinforced by the fact that the greatest evidence of efficacy from acyclovir exposure was in patients at highest risk of CMV, namely liver transplant recipients and D+/R renal transplant recipients, respectively. Also, the overall effect on acute graft rejection was driven by the data from D+/R renal allograft patients receiving valacyclovir (4 of the 6 high-dose acyclovir studies measured acute graft rejection). In other words, both a high risk and a good anti-CMV effect might be needed to demonstrate efficacy for these indirect outcomes. It would seem reasonable to conclude that optimal prevention of CMV by maximal acyclovir exposure would be associated with the most favorable outcome for the more important end points across the different organ transplant types. Indeed, valacyclovir therapy was also associated with notable reductions in in-patient medical resource use, including rehospitalization, as a likely consequence of such effects [5, 24]

Although ganciclovir also prevents CMV infection and disease after liver or heart transplantation [13, 25, 26], there is no evidence for an effect on survival, OIs, or acute graft rejection—even with the highest exposures after intravenous therapy [15, 25, 26]. High acyclovir exposure also reduces deaths after bone marrow transplantation [27, 28] but ganciclovir does not [29, 30], even though both drugs prevent CMV viremia in this setting. In addition, high-dose oral acyclovir was recently confirmed by meta-analysis to have a survival benefit in human immunodeficiency virus infection, although this was evident in the absence of an effect against CMV disease [31]

Because high-dose valacyclovir provided the greatest benefit overall, it may be the optimal drug for prophylaxis after organ transplantation. Nevertheless, the resulting average acyclovir exposure is >5 times that of high-dose oral acyclovir (480 vs. 90 μM/h), suggesting that lower doses of valacyclovir might be worth exploring

In summary, these data provide confirmatory evidence that acyclovir and its prodrug valacyclovir have anti-CMV activity and provide increasing efficacy as exposure is increased. We believe these results are the first to show that antiherpes prophylaxis is associated with a survival benefit in organ transplantation. In conjunction with the effects on acute rejection and OIs found with valacyclovir, the findings illustrate the importance of safely preventing CMV replication in order to improve outcome

Footnotes

  • Presented in part: 38th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Diego, September 1998 (abstract H105); 21st International Congress on Chemotherapy, Birmingham, United Kingdom, July 1999 (abstract P18)

    Institutional review boards approved all trials in this analysis and all patients gave written informed consent

    Financial support: GlaxoSmithKline (P.F. and P.D.G., periodic consultants; no financial support received for this meta-analysis)

References

| Table of Contents