Prevention of Posttransplant Cytomegalovirus Disease and Related Outcomes with Valganciclovir: A Systematic Review

Authors


* Corresponding author: Nina Singh, nis5@pitt.edu

Abstract

The precise impact of valganciclovir as preventive therapy for cytomegalovirus (CMV) in solid organ transplant (SOT) recipients is not fully defined. Data from studies using valganciclovir as preemptive therapy or prophylaxis for CMV in SOT recipients were synthesized for descriptive analysis. CMV disease occurred in 2.6% and 9.9% of the patients receiving valganciclovir as preemptive therapy and prophylaxis, respectively. Although the incidence of early-onset (≤90 days posttransplant) CMV disease was only 0.8% and 1.2% in all patients and R–/D+ patients receiving valganciclovir prophylaxis, the incidence of late-onset (>90 days posttransplant) CMV disease rose up to 8.9% and 17.7% in the prophylactic group, respectively. On the contrary, no patients developed late-onset CMV disease in preemptive group. Both approaches with valganciclovir have successfully decreased CMV disease in SOT recipients. Late-onset CMV disease is a complication observed uniquely with valganciclovir prophylaxis, particularly in R–/D+ patients, but not with preemptive therapy.

Introduction

Cytomegalovirus (CMV) is an important opportunistic pathogen in solid organ transplant (SOT) recipients. In the absence of any form of preventive therapy, CMV infection develops in 36–100% of the SOT recipients and symptomatic disease in 11–72% of the patients, most often during the first 100 days after transplantation (1). Current antiviral agents and preventive strategies have led to a decrease in the incidence of CMV disease. Compared to patients treated with placebo or no antiviral therapy, those receiving preemptive therapy and prophylaxis have a 72–80% lower likelihood for the development of CMV disease (2). Furthermore, reduced allograft rejection, opportunistic infections and mortality has been documented with the receipt of prophylaxis (2).

Despite these advances, a number of issues remain unresolved (3). Although prophylaxis has effectively prevented CMV disease in the early posttransplant period, the incidence of late-onset CMV disease occurring after discontinuation of prophylaxis has increased from 0–5% in the era of long-term use of acyclovir (4,5) to 2.6–7% in patient receiving prophylaxis with oral ganciclovir (6,7). It has been proposed that the optimal development of long-term protective immunity against CMV may be compromised with the prolonged use of a potent antiviral agent such as ganciclovir (8–11). In addition, indirect sequelae of CMV still complicate the posttransplant course (12–16).

Valganciclovir has emerged as the preferred antiviral agent for the prevention of CMV in SOT recipients because of its oral availability, convenient dosing schedule, and 10-fold higher bioavailability than oral ganciclovir (17). Although not approved by the FDA for prophylaxis in liver transplant recipients, surveys of CMV prevention strategies after liver and kidney transplantation show that valganciclovir is the most commonly prescribed antiviral agent for the prevention of CMV (18,19). However, existing studies have primarily comprised small sample sizes and previous meta-analyses have excluded valganciclovir (1,2,20,21). Thus, we performed a review and conducted a systemic analysis to examine the efficacy of valganciclovir as preemptive therapy versus prophylaxis for CMV disease and CMV-associated indirect outcomes in SOT recipients.

Methods

English-language reports of published studies utilizing valganciclovir for the prevention of CMV disease in SOT recipients were identified by cross-referencing the keywords ‘valganciclovir’ and ‘transplantation’ or ‘transplant’. Data bases searched from March, 2001 when valganciclovir was approved by the FDA to April, 2008 included PubMed (http://www.ncbi.nlm.nih.gov), ISI Web of Science (http://www.isiknowledge.com), ScienceDirect (http://www.sciencedirect.com), the Cochrance Central Register of Controlled Trials (http://www.mrw.interscience.wiley.com/cochrane/cochrane_clcentral_articles_fs.html), EMBASE (http://www.embase.com) and BIOSIS (http://www.biosis.org). Bibliographies of original articles were manually reviewed for additional studies. Studies were included if valganciclovir was used as preemptive therapy or prophylaxis in SOT recipients. Studies were excluded if (1) they were review articles not reporting original data; (2) valganciclovir was used as treatment of established CMV disease; (3) valganciclovir was used for CMV prevention in non- SOT recipients. Two of the authors independently extracted the data for analyses. Any discrepancies in data extraction were resolved by review and discussion. Authors of some original articles were contacted if additional or clarification of reported results was deemed necessary. Data from one institution with overlapping study cohorts in two reports were analyzed only once to avoid patient duplication (22,23).

Preemptive therapy was defined as administration of anti-CMV agent upon detection of CMV viremia by pp65 antigen or polymerase chain reaction (PCR) assay (24), and prophylaxis as administration of a course of anti-CMV agents at the time of or soon after transplantation (7). Late-onset CMV disease was defined as CMV disease occurring after 90 days posttransplantation (25). Data regarding transplanted organ, donor and/or recipient CMV serostatus, CMV disease, late-onset CMV disease, time to CMV disease, T-cell-depleting antibody (alemtuzumab, OKT3 antibodies or antithymocyte globulin) use, recurrence of CMV viremia, antiviral resistance, rejection, graft loss, opportunistic infections and mortality were collected for analysis. Data from randomized or cohort studies, with or without a control group were analyzed and the results were reported in a descriptive fashion.

Results

Risk of CMV disease with preemptive therapy

Ten studies (two randomized controlled trials, four prospective studies, three retrospective studies one with concurrent control and one cohort study with historic control group) have described the use of valganciclovir as preemptive therapy (22,23,26–34) (Table 1). The follow-up duration was a minimum of 12 months after transplantation in all but three studies (29,30,33). In one report, the proportion of patients with treatment failure (8.3%, 7/84), defined as the necessity of valganciclovir change due to any other cause or recurrent viral detection during the 30 days after completion of antiviral course was employed as an end-point (26). Another study reported only the cumulative incidence of CMV disease in patients receiving valganciclovir as preemptive therapy or as prophylaxis (5.0%, 5/101) (29). The total number of patients receiving preemptive therapy with valganciclovir was not reported in one study, however, none developed CMV disease (33). Thus, the efficacy of preemptive therapy against CMV disease could be assessed in 7/10 studies (Table 2). Of 761 patients in these studies, 20 (2.6%) developed CMV disease. CMV disease comprised viral syndrome in 82.4–100% and tissue invasive disease in 0–17.6% of the patients (27,28,32). Exclusion of a study (28) where only 29 (26%) of the patients had surveillance monitoring for CMV DNA PCR as proposed yielded an incidence of CMV disease of 0.5% (3/651) in SOT recipients receiving valganciclovir as preemptive therapy.

Table 1.  Characteristics of studies with valganciclovir preemptive therapy for the prevention of CMV diseases
Study design1 (number of studies)Type of transplantationControl groups
  1. 1Numbers in the column represent reference numbers. Data from two references (22,23) with overlapping study cohorts were analyzed only once.

  2. 2VAC = valacyclovir; 3VGC = valganciclovir; 4GCV = ganciclovir; 5NA = not available; 6SOT = solid organ transplant.

Randomized controlled (2)
  (27)KidneyProphylaxis with VAC2
  (32)KidneyProphylaxis with VGC3
Prospective with historical controls (1)
  (22)LiverPreemptive therapy with GCV4
Prospective without controls (4)
  (23)LiverNA5
  (26)SOT6NA
  (31)SOTNA
  (30)SOTNA
Sequential cohort study (1)
  (28)KidneyProphylaxis with VGC
Retrospective with concurrent controls (1)
  (34)Liver or kidneyPreemptive therapy with GCV
Retrospective without controls (2)
  (29)KidneyNA
  (33)HeartNA
Table 2.  Outcomes with valganciclovir as preemptive therapy or prophylaxis
VariablesPreemptive therapyProphylaxis
No. of studies% (n/N)1Median (range)2No. of studies% (n/N)Median (range)
  1. CMV = cytomegalovirus.

  2. 1%= percentage; n = number of patients with positive results; N = total number of patients in the same category.

  3. 2Vales represent median and range of incidence for the reported variables.

  4. 3R–/D+= recipient-negative/donor-positive CMV serostatus and R+= recipient-positive CMV serostatus.

  5. 4NA = not available.

CMV disease
 Overall rate7 2.6 (20/761)  0 (0–15.5)21   9.9 (175/1,767)8.2 (0–85.7)
 R–/D+ patients335.5 (3/55) 7.7 (0–33.3)1720.1 (123/613)18.8 (0–85.7) 
 R+ patients36 2.6 (17/650)  0 (0–15.5)144.0 (28/699)2.6 (0–15.9)
 Late-onset
 Overall3  0 (0/195)0 (NA4) 16   8.9 (104/1,164)7.5 (0–85.7)
 R–/D+ patients3 0 (0/55)0 (NA)  1317.7 (89/504) 16.3 (0–85.7) 
 Tissue-invasive315.0 (3/20)    0 (0–17.6)1233.9 (41/121) 31.7 (0–100)  
Rejection
 Incidence410.8 (41/379) 8.7 (6.4–36.1)1017.6 (173/985)15.5 (2.0–42.9)
Graft loss
 Incidence4 3.9 (15/380)2.4 (1.8–5.9) 82.5 (22/870)2.1 (0–14.3)
Opportunistic infections
 Incidence328.5 (77/270) 24.3 (10.2–75.0) 37.8 (27/347)7.5 (0–18.4)
Mortality
 Incidence4 8.2 (31/380)0.9 (0–15.7)14  4.4 (50/1,128)1.9 (0–15.3)

The proportion of patients with late-onset CMV disease was reported in three studies; none (0%, 0/20) had late-onset CMV disease (27,28,32). CMV disease occurred within 30 days after transplantation in five patients (25%), and between 30 and 90 days in the other 15 (75%). The incidence of CMV disease in R–/D+ patients base on three studies was 0% (0/36), 7.7% (1/13) and 33.3% (2/6), respectively (23,27,32). None of these high-risk patients developed late-onset CMV disease. Of a total of 650 R+ patients in six studies, 17 (2.6%) developed CMV disease (27,28,30–32). Only two studies explicitly reported receipts of T-cell-depleting agents (23,28). CMV disease developed in 15.5% (17/110) of the patients receiving these agents and 0% (0/187) of those not receiving (23,28).

Preemptive therapy tool utilized to initiate valganciclovir was CMV pp65 antigenemia in six studies (54.5%) and CMV DNA PCR in seven (63.6%). The cut-offs for these assays to initiate preemptive therapy varied for different studies. Some studies set criteria of antigenemia for the initiation of preemptive therapy as ≥1 cell with characteristic immunofluorescence per 2 × 105 polymorphonuclear cells (22,23) while others employed ≥20 (29) or >25 of 2 × 105 polymorphonuclear cells to initiate such therapy (30,31). Similarly, the threshold of DNAemia by CMV DNA PCR to initiate preemptive treatment varied from ≥15 to ≥2000 copies/mL (23,27–29,32–34). In all, three studies monitored for antigenemia or DNAemia for 3 months, 2 for 6 months, 2 for 12 months, and 1 for 4 months posttransplant (22,23,26–32). Six studies monitored weekly and two every 2 weeks.

Regardless, 329 (35.5%) of 926 patients in preemptive groups in nine studies had detectable viremia during surveillance monitoring as per study criteria (22,23,27–34). Six of the nine studies described explicitly the relationship between preemptive therapy and subsequent development of CMV disease (22,23,27,30,31,33,34). No CMV disease occurred after receipt of as preemptive therapy in five studies. In the sixth study, one patient developed CMV disease 1 day after preemptive therapy with valganciclovir; the other patient had CMV syndrome 11 days after the diagnosis of CMV DNAemia with subsequent noncompliance and missing the next PCR assessment (27). Forty-eight (19.4%) of 248 patients in eight studies who received valganciclovir as preemptive therapy for CMV infection had recurrent viremia (22,23,26,27,29–33). Five studies reported subsequent follow-up for 39 patients with recurrence after preemptive valganciclovir use for their first episodes of viremia (23,27,30–32). Two patients developed CMV disease (27); one of two did not have CMV PCR monitoring. Therefore, CMV disease developed later while DNAemia progressed to CMV disease in the other patient despite valganciclovir and reduced immunosuppression. However, analysis of UL97-gene sequencing failed to detect ganciclovir-resistant mutations and viremia resolved with intravenous ganciclovir (27). Only one study performed sequencing of the CMV UL97 gene in patients with recurrence, and none had mutations known to cause ganciclovir resistance (32). Most patients with recurrent viremia were successfully treated with a repeat course of valganciclovir.

Risk of CMV disease with prophylaxis

A total of 2094 patients in 25 studies reported valganciclovir use for CMV prophylaxis (three randomized controlled trials, five prospective studies two with historic controls, sixteen retrospective two with concurrent and five with historic control and one cohort study with historic control group) (28,31,32,35–56) (Table 3). One study where valganciclovir use was preceded by intravenous ganciclovir for 30 days in all patients and for 90 days in R–/D+ patients was excluded (49). Treatment duration varied from 14 to 240 days, but most of the patients received valganciclovir for at least 90–100 days. The efficacy of valganciclovir for the prevention of CMV disease was reported in the 21 studies comprising a total of 1767 patients. CMV disease developed in 175 (9.9%) with follow-up duration over 1 year (28,31,32,35–48,50–52,56) (Table 2). Of 148 patients with CMV disease in 16 studies, 117 (79%) were R–/D+ (28,32,35,37,38,40–47,50,52,56). CMV disease consisted of viral syndrome in 66.1% (80/121) and tissue invasive disease in 33.9% (41/121) in 12 studies with data available (28,32,38,40–44,46,47,52,56). All episodes of CMV disease were due to viral syndrome in 4/12 studies and due to tissue invasive disease in one. In the remaining seven studies, 40–82.8% of CMV disease was due to viral syndrome and 17.2–60% due to tissue invasive disease.

Table 3.  Characteristics of studies with valganciclovir prophylaxis for the prevention of CMV diseases
Study design1 (number of studies)Type of transplantationControl groups
  1. 1Numbers in the column represent reference numbers.

  2. 2GCV = ganciclovir; 3VGC = valganciclovir; 4NA = not available.

Randomized controlled (3)
  (39)KidneyProphylaxis with GCV2
  (32)KidneyPreemptive therapy with VGC3
  (44)Solid organ transplantProphylaxis with GCV
Prospective with historical controls (1)
  (42)LungProphylaxis with GCV
Prospective without controls (3)
  (31)Solid organ transplantNA4
  (37)Solid organ transplantNA
  (47)Solid organ transplantNA
Sequential cohort study (1)
  (28)KidneyPreemptive therapy with VGC
Retrospective with concurrent controls (2)
  (54)LiverProphylaxis with GCV
  (45)Kidney; kidney–pancreasProphylaxis with GCV
Retrospective with historical controls (5)
  (35)LiverProphylaxis with GCV
  (55)HeartProphylaxis with GCV
  (36)Kidney or/and pancreasProphylaxis with GCV
  (40)LiverProphylaxis with GCV
  (48)Kidney or pancreasProphylaxis with GCV
Retrospective without controls (9)
  (56)HeartNA
  (53)KidneyNA
  (38)LiverNA
  (41)LiverNA
  (43)KidneyNA
  (46)Kidney or/and pancreasNA
  (50)Kidney–pancreasNA
  (51)LiverNA
  (52)KidneyNA

CMV disease developed a median of 151 days posttransplant (range, 20–312 days) in 20 patients with data available (32,38,40,43,46,52). Three of the 20 patients developed CMV disease within 90 days, that is, on 20, 60 and 82 days; CMV occurred after 90 days in the remaining 17 patients (38,43). In all, 104/113 (92.0%) patients with CMV disease developed late-onset CMV disease in 16 studies with data available (28,31,32,35,37,38,40,42–47,50,51,56). The incidence of early-onset CMV disease was 0.8% (9/1164), and that of late-onset CMV disease was 8.9% (104/1164). Excluding the two studies in which prophylaxis was used for more than 90–100 days, the incidence of late-onset CMV was 10.2%. Four of the 16 studies had cases of breakthrough CMV disease under prophylaxis use, but none contributed to the cases of late-onset CMV disease (28,38,43,44). The duration from transplantation to development of late-onset CMV disease was a median of 152 days (range 114–312 days).

In the 10 studies that considered ganciclovir-resistance, five assessed if resistant virus existed by clinical observation and none was documented (35,45,48,52,53). PV16000 trial detected ganciclovir-resistant virus by UL97-gene sequencing in none of the patients receiving valganciclovir on day 100 posttransplantation (0%, 0/198) and also in none of those receiving valganciclovir and with suspicion of CMV disease (0%, 0/55) (44,57). One study performed genotypic resistance testing in all patients with detectable viremia after prophylaxis, and two patients receiving valganciclovir had ganciclovir-resistant virus (42). In the remaining three studies (36,54,55), patients were assessed for ganciclovir-resistant CMV as clinical indicated and one patient had documented resistant virus (54).

Seventeen studies enrolled a total of 613 R–/D+ patients receiving valganciclovir as prophylaxis. Of there, 123 (20.1%) developed CMV disease (28,32,35,37,38, 40–47,50,52,53,56) (Table 2). In 13 studies with data available, the incidence of early-onset CMV disease in R–/D+ patients was 1.2% (6/504) and that of late-onset was 17.7% (89/504). Additionally, 14 studies evaluated a total of 699 R+ patients, and 28 (4.0%) of them had CMV disease (28,31,32,38–40,42,43,45–47,50–52). Thirteen studies reported the onset of CMV disease in R+ patient; the incidence of early-onset CMV disease was 0.5% (3/630) and that of late-onset was 2.2% (14/630). CMV disease developed in 4.7% (16/399) of the patients who received T-cell-depleting agents in five studies with data available (28,31,45,47,50) and in 16.5% (54/328) of those not receiving these agents in four studies (35,40,41,56).

Rejection

Four studies reported the overall rejection rate of 10.8% (41 patients) in a total of 379 patients receiving preemptive therapy (23,27,28,32). The overall rejection was 17.6% (173/985) in the 10 prophylactic studies with data available (28,31,32,36,39,40,44,46,50,56).

Graft loss

Fifteen patients (3.9%) of 380 patients in preemptive group had graft loss in four studies with data available (23,27,28,32). Eight prophylactic studies described the number of patients with graft loss, and the overall rate was 2.5% (22/870) (28,32,36,39,44,46,48,56).

Opportunistic infections

Three preemptive studies reported the proportion of study population that developed opportunistic infections (23,27,32). A total of 270 patients were enrolled in these three studies, and 77 (28.5%) developed opportunistic infections. Three prophylactic studies reported the proportion of patients with opportunistic infections, and the overall rate was 7.8% (27/347) (32,44,53).

Mortality

The overall mortality rate was 8.2% (31/380) from four preemptive studies with reported data (23,27,28,32). The mortality in 14 prophylactic studies with reported data was 4.4% (50/1,128) (28,31,32,36–38,40,42,44,46,48,53,55,56).

Discussion

Evidence-based systematic synthesis of published data in our study shows that the overall incidence of CMV disease in patients receiving valganciclovir prophylaxis was 9.9%. Ninety-two percent of the patients with CMV disease in the prophylactic group had late-onset disease. The incidence of CMV disease in patients receiving valganciclovir as preemptive therapy was 2.6%. Late-onset CMV disease was not observed in this group. Compared with the preemptive therapy and prophylaxis with ganciclovir in a meta-analysis (21), the prevention of CMV disease with valganciclovir effectively decreased the incidence of CMV disease in SOT. However, the characteristics of CMV disease have evolved in the current era. Most CMV disease with the use of antiviral prophylaxis presents as late-onset CMV disease and occurs a median of 152 days after transplantation.

Optimal prevention of CMV is particularly relevant in R–/D+ patients given substantial morbidity associated with CMV in these patients (58,59). Existing guidelines recommend antiviral prophylaxis for the prevention of CMV disease in R–/D+ patients (60,61). The overall incidence of CMV disease, and of late-onset CMV disease in R–/D+ patients receiving prophylaxis was 20% and 18%, respectively. In contrast, all CMV disease in R–/D+ subgroups receiving preemptive therapy occurred within 90 days of transplantation. Thus, the data show that R–/D+ patients receiving prophylaxis are uniquely susceptible to the development of CMV disease in the late posttransplant period. Prior studies have documented late-onset CMV disease in 18–26% of R–/D+ SOT patients receiving ganciclovir prophylaxis (6,40,44,62,63). Furthermore, late-onset CMV disease has been shown to be an independently significant predictor of mortality during the first posttransplant year (6,54). Thus, the use of valganciclovir as prophylactic strategy may not be wholly adequate for preventing CMV disease in R–/D+ patients.

Compared with the patients receiving preemptive therapy and prophylaxis with ganciclovir in a previous meta-analysis (21), the patients receiving valganciclovir in this study had lower overall incidence of CMV disease both in the prophylactic (9.9%) and preemptive groups (2.6%). Furthermore, the incidence of rejection, graft loss and mortality in the patients in this study was also lower than those reported with the use of ganciclovir (21). Such reduced risk of CMV disease and CMV-associated indirect outcomes in this study might be attributable to the improvement in transplant patient care, advances in immunosuppressive therapy and wider availability of rapid and reliable diagnostic tools for CMV.

Several limitations of this review deserve to be acknowledged. Data regarding CMV disease and other outcomes are reported in a descriptive fashion only without comparative statistical analysis. Statistical comparisons of pooled data from various studies are not considered scientifically appropriate unless comparable groups or the same type of patient population are compared. For example, if some studies included more high-risk patients (R–/D+), the percentage of disease could be higher than in studies that included largely low-risk patients. Statistical analysis on summed data therefore could lead to erroneous conclusions. There were far fewer studies of preemptive therapy in the literature, and most studies were from single centers with small sizes. A meta-analysis by definition is a combination of multiple studies that address a research related hypothesis. In theory, combining multiple studies attempts to overcome the limitation of statistical power associated with small sample sizes in individual studies. However, given the paucity of randomized controlled studies for the prevention of CMV with valganciclovir, it is not possible to perform a meta-analysis for our study. In addition, several factors with potential to compromise the findings of this study exists, such as different CMV serostatus in study populations, lack of uniform diagnostic criteria for indirect effects of CMV and variability in valganciclovir dosage used. Nevertheless, valganciclovir is the most commonly prescribed antiviral agent for CMV prevention nowadays (18,19), and our study offers the most relevant and comprehensive information regarding its use for CMV prevention in SOT patients.

In summary, preemptive and prophylactic strategy with valganciclovir use has successfully decreased the incidence of posttransplant CMV disease in SOT patients. However, most patients, in particular those belonging to the R–/D+ group with CMV disease in the current era present with late-onset CMV disease. Prophylaxis is uniquely associated with late-onset disease while preemptive therapy approach is protective against late-disease. Future studies should focus on head to head comparison between preemptive therapy and prophylaxis for the prevention of CMV disease and clearly assess the efficacy of different preventive strategies for the indirect outcomes associated with CMV infection. Optimal approach is the one that has a protective effect on both CMV disease and indirect outcomes or tips the balance in favor of achieving best outcomes after transplantation.

Ancillary