Clinical Aspects of Cytomegalovirus Antiviral Resistance in Solid Organ Transplant Recipients

Transcription

1 IMMUNOCOMPROMISED HOSTS INVITED ARTICLE David Snydman, Section Editor Clinical Aspects of Cytomegalovirus Antiviral Resistance in Solid Organ Transplant Recipients A. K. Le Page, 1,4 M. M. Jager, 1,2 J. M. Iwasenko, 1,4 G. M. Scott, 1,3,4 S. Alain, 6 and W. D. Rawlinson 1,3,4,5 1 Virology Division, Prince of Wales Hospital and University of New South Wales Research Laboratories, Randwick, Australia; 2 VU University Medical Center, Department of Medical Microbiology and Infectious Diseases, Amsterdam, the Netherlands; 3 School of Biotechnology and Biomolecular Sciences, Faculty of Science, and 4 School of Medical Sciences, Faculty of Medicine, University of New South Wales, Kensington, and 5 Virology Division, SEALS Microbiology Prince of Wales Hospital, Randwick, New South Wales, Australia; and 6 National Reference Center for Cytomegaloviruses, CHU de Limoges, Université de Limoges, Inserm UMR 1092, Faculté de médecine, Limoges, France Despite advances in the prophylaxis and acute treatment of cytomegalovirus (CMV), it remains an important pathogen affecting the short- and long-term clinical outcome of solid organ transplant. The emergence of CMV in a patient reduces the clinical efficacy of antiviral therapy, complicates therapeutic and clinical management decisions, and in some cases results in loss of the allograft and/or death of the patient. There is increasing use of antiviral prophylaxis after transplant with little expansion in the range of antiviral agents effective in treatment of CMV. Further understanding is needed of the risk factors for development of CMV antiviral and of therapeutic strategies for treating patients infected with resistant viruses. We review the current status of CMV in solid organ transplant recipients, and provide diagnostic and therapeutic suggestions for the clinician in managing antiviral. Keywords. CMV; cytomegalovirus; ; transplant. Human cytomegalovirus (CMV) is one of the most clinically significant viral pathogens causing infections after solid organ transplant (SOT), owing to the frequency of infection, the acute effects on patient and graft survival, and effects on long-term graft survival [1]. Developments in antiviral agents and preventive strategies over the past decade have significantly improved outcomes from CMV disease in SOT recipients. Kidney transplant recipients of high serologic risk, in which the donor is CMV seropositive (D + ) and the recipient seronegative (R ), on prolonged prophylaxis have a cumulative disease incidence 2 years after transplant of approximately 22% [2]. More frequent and prolonged prophylaxis with ganciclovir and Received 22 June 2012; accepted 26 November 2012; electronically published 12 December Correspondence: W. D. Rawlinson, Virology Division, SEALS Microbiology, Prince of Wales Hospital, Barker St, Randwick, NSW 2031, Australia (w.rawlinson@unsw.edu.au). Clinical Infectious Diseases 2013;56(7): The Author Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please journals.permissions@oup.com. DOI: /cid/cis1035 valganciclovir (valganciclovir [valgcv]) [2] has raised concern about antiviral (AVR), with the potential for cross- to currently available therapies [3, 4]. We here provide a brief update on CMV antiviral agents, with an in-depth focus on the clinical significance and risks of CMV AVR in SOT, and strategies to monitor, prevent, and treat AVR. ANTIVIRAL AGENTS The development of ganciclovir (GCV) in the mid- 1980s was a significant breakthrough in CMV therapy. It is a nucleoside analogue requiring phosphorylation by the viral protein kinase ( pul97) prior to inhibition of the viral DNA polymerase (UL54) (reviewed in [5]). The poor oral bioavailability of GCV led to development of oral valgcv, a prodrug with enhanced bioavailability and comparable efficacy to intravenous GCV for treatment of CMV disease in SOT recipients [6]. Both agents have been used in CMV-preventive regimens in prophylactic, preemptive, or combination algorithms largely based on donor/recipient serologic 1018 CID 2013:56 (1 April) IMMUNOCOMPROMISED HOSTS

2 Table 1. Alternative Antivirals in Development or Under Consideration for Use in Cytomegalovirus Resistant to Ganciclovir, Foscarnet, or Cidofovir Drug Name (Class) Dose Schedule (Phase of Development) Target Recent CMV Clinical Studies Antiviral Resistance + Cross- Reactivity With GCV/FOS/CDV Reference(s) IMMUNOCOMPROMISED HOSTS CID 2013:56 (1 April) 1019 AIC246 letermovir (3,4-dihydroquinazoline derivative) Artesunate (antimalarial) CMX-001 (Cidofovir prodrug, orally bioavailable lipid conjugate. CDV is nucleotide analogue of cytidine) Cyclopropavir CPV (Methylenecyclopropane nucleoside analogue) Fomivirsen (21-nucleotide [nt] antisense oligonucleotide with sequence complementary to CMV IE2 mrna) Leflunomide (Inhibitor of pyrimidine synthesis and protein kinase activity) mg/d dependent on clinical reason for use 200 mg bd 1 d then 100 mg/d for 28 d UL56/UL89 terminase complex Acts on cellular pathways altering host cell kinase signaling and NF-kappa B and Sp1 activation paths with secondary inhibition of synthesis of viral immediate early proteins Current (2012) phase II trial for prophylaxis in HSCT Recent single-case report of successful use in multidrug-resistant CMV in SOT Recent case series: used in preemptive CMV treatment in HSCT demonstrating reduced viral loads or plateau in viral load increase. Single case report of unsuccessful use in Unknown UL54 Current phase II III in treatment of CMV disease Unknown Current approval only for intraocular application for retinitis Off-label use for CMV treatment in SOT using a loading dose of 100 mg/d orally for 5 d, then 40 mg/d orally, with dose adjustment according to levels of the teriflunomide metabolite Similar to GCV as phosphorylated by pul97 IE2 mrna Current phase I trial of oral use in healthy volunteers Nil known in nonretinitis disease? Viral capsid assembly Low long-term suppressive response ( 50%) in 2010 studies of use in CMV treatment in SOT No with GCV, FOS, CDV Mutations in UL56 confer [59] [60] Unknown [13] [14] [15] CDV probably same UL54 exonuclease, A987G mutations as for CDV Cross- with GCV mutations UL97 M460I/T, H520Q,C603R with 8 20 increase in IC 50 No cross- with GCV UL97 mutation L595S Probable cross- with FOS [9] [53] [61] Unknown [62] In vitro activity against GCVresistant CMV [16] [63]

3 Table 1 continued. Antiviral Resistance + Cross- Reactivity With GCV/FOS/CDV Reference(s) Recent CMV Clinical Studies Dose Schedule (Phase of Development) Target Drug Name (Class) [7] [11] [42] Resistance described in clinical practice with UL97 T409 and H411Y mutations No cross- reported with GCV MBV induced in vitro UL97 mutations V353A, L397R, T409M, H411L/N/Y in the kinase ATP-binding site not cross-reactive with GCV. UL27 mutations in vitro lead to (low level) Phase III HSCT prophylactic study completed no difference to placebo UL97 inhibition: viral encapsidation and nuclear egress of viral particles Prophylactic dose in phase III HSCT trials where no difference was found to placebo: 100 mg bd for 12 wk MBV (Benzimidazole L-riboside ATP competitive inhibitor) Abbreviations: ATP, adenosine triphosphate; bd, twice daily; CDV, cidofovir; CMV, cytomegalovirus; FOS, foscarnet; GCV, ganciclovir; HSCT, hematopoietic stem cell transplant; IC50, drug concentration that reduces viral growth by 50%; MBV, maribavir; mrna, messenger RNA; SOT, solid organ transplant; TID, tissue invasive disease. risk. Toxicity associated with prolonged use of GCV and valgcv, especially reversible bone marrow suppression, remains a limitation. Foscarnet (FOS) and cidofovir (CDV) are intravenous CMV antivirals that also target the viral polymerase [7]. They are more commonly used as second-line treatment or for AVR, often limited by frequent nephrotoxicity. GCV and CDV, once phosphorylated, (CDV via cellular enzymes) compete with deoxyguanosine triphosphate binding of UL54 with consequent inhibition, with FOS blocking the UL54 pyrophosphate binding site (reviewed in [8]). Alternative antivirals are few, although trials are being conducted for less toxic agents with different mechanisms of action (Table 1). They include an oral CDV prodrug (CMX- 001) with lower nephrotoxicity [9], new antivirals with alternative targets (maribavir [MBV], letermovir) and drugs currently used in other settings (leflunomide, artesunate). MBV is a UL97 inhibitor with good oral bioavailability and low toxicity and has theoretical benefits for inhibiting crossresistant viruses (reviewed in [7, 10]). However, in phase III clinical trials, as prophylaxis in hematopoietic stem cell transplant (HSCT) recipients, and more recently in D + /R liver recipients, MBV prophylaxis has been demonstrated as inadequate [11, 12]. Orally administered artesunate has low toxicity and in vitro inhibits CMV, including GCV-resistant viruses [13], although clinical efficacy is currently unclear [14, 15]. Oral leflunomide has anti-cmv activity, although low (approximately 50%) response rates suggest it may best be used in combination [16]. CMV intravenous immunoglobulin (IVIG) may have some role in CMV therapy and prevention, especially where recipients have low gammaglobulin levels or in heart/lung transplant [17]. CMV ANTIVIRAL RESISTANCE Definitions, Prevalence, and Outcome A practical definition of CMV AVR is a clinical profile of increasing or high plateaued viral load in the presence of adequate antiviral therapy administered for >2 weeks, with confirmed on genotypic and/or phenotypic testing. Most reports of AVR occur in patients with CMV disease on long-term (>6 weeks) GCV, although may occur in asymptomatic viremia [18, 19]. Clinical (clinical profile) and virologic (laboratory evidence) are not always linked, with virologic reported in up to 50% of cases of clinical [3, 19]. Accurate estimates of SOT AVR risk and prevalence are confounded by differing AVR definitions, testing methods, the diversity of SOT subjects, and selection bias. Table 2 reviews prevalence in publications reflecting the recent era 1020 CID 2013:56 (1 April) IMMUNOCOMPROMISED HOSTS

4 Table 2. Cytomegalovirus Antiviral Resistance in Recipients of Solid Organ Transplants: Key Studies IMMUNOCOMPROMISED HOSTS CID 2013:56 (1 April) 1021 (Reference) Method of Assessment (Limaye 2002 [20]) Review including cohort and case studies of, (Boivin 2009 [21]) Prospective cohort study of as part of RCT: the VICTOR study [6] Prospective monitoring at days 0, 21, 49 of therapy for CMV disease (Hantz 2010 [19]) Prospective cohort study Prospective monitoring after day 21 of therapy for infection or disease according to variable center practice Population Studied: SOT Type, (No. of Patients), Inclusion/ Exclusion Criteria Review of factors associated with AVR in mixed SOT studies from Includes heart, kidney, kidney pancreas, liver, and lung SOT: mixed. Numbers with CMV infection: heart (20), kidney (216), liver (23), lung (17) Adult recipients Life-threatening disease or GFR, 10 ml/min excluded Therapy Type, Duration, and Dose Where Stated Variable treatment in different centers, with IV or oral GCV Rx for CMV disease randomized to valgcv (900 mg/bd) or IV GCV (5 mg/kg/ bd), both GFR adjusted, for 21 d, followed by secondary prophylactic (maintenance) valgcv for 28 d Primary prophylaxis received by 44% 46% of patients SOT: mixed Rx according to local Numbers with CMV center practice: infection: valgcv, IV GCV, or heart (28), kidney FOS, doses not (224), liver (24), lung defined (7), multiorgan (4) Primary prophylaxis Cohort studied given at the included HSCT discretion of the recipients local center Resistance Definition/ Testing Methods Genotypic and phenotypic 51 to 510 d after methods for UL97 only. therapy Antiviral load measurements not in common practice prior to 2002 Genotypic testing of viremia of UL97 + UL54 detected at days 0, 21, 49 of Rx and during CMV clinical recurrence Confirmed defined as a previous report with marker transfer phenotypic Probable genotypic defined as new mutation at codon of previous identified Clinical (day 21 replication) with genotypic +/ phenotypic testing Onset of Resistance (Posttherapy, Infection or Transplant as Resistance Risk Factors Indicated) Where Reported Confirmed or probable: 5/13 demonstrated from day 0 of therapy Median time 21 d Clinical suspected (and subsequently confirmed with virologic testing) at median of 46 d after infection SNR: D + R Persistent VL day 21 a - D + /R a Lung recipients b Baseline viral load at start of therapy b NS: antiviral (valgcv vs IV GCV), time from last transplant D + /R a (virologic ) NS: induction antilymphocyte antibody therapy Resistance Rate, Overall and/or According to Specific Treatment Regimen or Organ Type Heart 0.28%, kidney 0.54% 1%, kidneypancreas 13%, lung 3.3 9%, liver 0% 2.3% (valgcv) of infected, 3.6% (GCV) of infected, 17.6% in lung cohort, 3.7% in kidney cohort Overall = 5.2% of those infected Lung 14.3%, kidney 6.2% Outcomes for Resistant States Mutations Symptomatic UL97 only: disease in 27/30 patients Crude mortality rate 19% predominantly L595S, A594V, M460V (12/23) Failure to eradicate 10/13 UL97, (VL >600 copies/ 2/13 UL 54, ml at days 21 1/13 UL97/UL54, and 49) a MDR-nil One-third recurrent CMV disease, 1 death attributed to recurrent CMV 56.25%: persistent viremia (undefined) or severe disease Of whole cohort including HSCT group: 17/18 patients UL97, 5/18 UL54, 3/18 UL97/UL54, 4 patients MDR

5 1022 CID 2013:56 (1 April) IMMUNOCOMPROMISED HOSTS Table 2 continued. (Reference) Method of Assessment (Myhre 2011 [24]) cohort study review of cases already identified or subsequently identified as resistant (Martin 2010 [18]) Prospective cohort study Prospective monitoring at day 100 of prophylaxis (end of prophylaxis), during suspicion of CMV disease, or at time of organ biopsy up to 6 mo after transplant (Van der Beek 2010 [58]) cohort study review with tested during occurrence of any viremia up to 1 y after transplant Population Studied: SOT Type, (No. of Patients), Inclusion/ Exclusion Criteria SOT: kidney (1244) SOT: mixed Total patients studied: heart (12), kidney (33), liver (17), multiorgan (1) Pediatric recipients only 38% D + /R Total patients studied: kidney (60), kidneypancreas (11) Only D + /R Therapy Type, Duration, and Dose Where Stated Rx for disease: valgcv 900 mg/bd or IV GCV 5 mg/kg/bd (GFR adj) until CMV DNAemia below detection limit for at least 2 wk Generally 4/52 secondary prophylactic valgcv thereafter All patients had prevention: either preemptive: weekly screening and Rx with valgcv 900 mg daily (GFR adj) if viremia identified or prophylaxis (valgcv dose/ duration not defined) ValGCV prophylaxis daily for 100 d according to BSA equation [57] Rx: valgcv 900 mg/bd for 14 d or IV GCV 5 mg/kg/d for 14 d (if symptomatic) All patients had a prevention strategy: either preemptive: valgcv 900 mg/bd for 14 d or prophylaxis 900 mg/ d for 90 d followed by preemptive regimen All adj for GFR Resistance Definition/ Testing Methods Genotypic testing of clinically suspected cases. Confirmed and probable definitions as per Boivin 2009 used as reporting of Genotypic testing of UL97 and UL54, and recombinant phenotyping for new mutations Genotypic testing (UL97 +/ UL54) according to antivirals used Onset of Resistance (Posttherapy, Infection or Transplant as Resistance Risk Factors Indicated) Where Reported Range of d after infection (av 115 d) Range of d after treatment (av 108 d) Day 100 of prophylaxis in 1/1 infected patients During viremia after 100 d of prophylaxis = d Not able to be accurately inferred D + /R a Preemptive therapy a Resistance Rate, Overall and/or According to Specific Treatment Regimen or Organ Type Outcomes for Resistant States 14.6% of infected No patients with attributed previous preemptive deaths, no Rx, 6.3% of infected persistent patients previously viremia on prophylaxis following diagnosis of and subsequent Rx 2.2% of all patients at 100 d of prophylaxis (1/1 of viremic patients) 3/3 patients with viremia after prophylaxis, during first 6 mo after transplant All asymptomatic at detection of and remained asymptomatic thereafter 19% (4/21) of patients Viremia and with treatment symptoms failure (defined as resolved on CMV load of 1000 valgcv or GCV copies/ml after 2 Rx wk of Rx) 0% of patients at initial detection of viremia Mutations All (27) UL97 mutations All UL97 mutations All UL97 mutations

6 Table 2 continued. IMMUNOCOMPROMISED HOSTS CID 2013:56 (1 April) 1023 (Reference) Method of Assessment Population Studied: SOT Type, (No. of Patients), Inclusion/ Exclusion Criteria (Couzi 2012 [30]) Kidney (112) Only D + /R cohort study review with tested if increasing viremia during therapy up to 1 y after transplant (Eid 2008 [33]) cohort study review of identified cases of CMV associated disease (Boivin 2012 [27]) Prospective cohort study of as part of RCT: the IMPACT study [2] Monitoring from the IMPACT study during and postprophylaxis SOT: mixed Total patients studied: kidney (138), kidneypancreas (4), liver (58), and heart (25) Only D + /R Kidney (318) Only D + /R (Iwasenko 2011 [3]) SOT: mixed Case series Total patients studied: Review of AVR in kidney (10), heart (8), specimens sent to lung (9), liver (2) an Australian testing facility (included HSCT recipients) Therapy Type, Duration, and Dose Where Stated Resistance Definition/ Testing Methods ValGCV prophylaxis: Genotypic testing of both 900 mg/d for 3 mo UL97 and UL54: ValGCV preemptive requested with (initiated at 2000 increasing viremia copies/ml) or during therapy therapeutic; dose 900 mg/bd GCV preemptive or therapeutic: 5 mg/kg bd, duration determined by viremia. GFR adj Variable/individualized Genotypic testing of UL97 Rx of disease. and UL54 All received valgcv prophylaxis (900 mg/ dgfradj d) ValGCV prophylaxis (randomized in IMPACT to d): 900 mg/d GFR adj. Therapy for disease not defined by study Variable: GCV, valgcv +/ FOS, CDV Genotypic testing of both UL97 and UL54 ( +/ recombinant phenotyping) if (1) identified viremia (>600 copies/ml) at 100 d or end of prophylaxis (+ post hoc including any viremia during prophylaxis) or 2) with CMV disease up to month 12 after transplant Genotypic testing of both UL97 and UL54 Onset of Resistance (Posttherapy, Infection or Transplant as Resistance Risk Factors Indicated) Where Reported Mean of 134 ± 61 d Range of d N/A duration of (val)gcv exposure at suspicion of 5/6 occurred during 100 or 200 d of prophylaxis, 1/6 after prophylaxis (within 12 mo after transplant Not able to be accurately inferred Peak viral load a (multiv) (5.25 log 10 copies/ml) Preemptive therapy a (univ) Viral eradication >8 wk a (univ) NS: induction with ATG 5/6 with had persistent viral replication prior to detection of N/A Resistance Rate, Overall and/or According to Specific Treatment Regimen or Organ Type Outcomes for Resistant States 13% of infected Lower egfr at 1 y patients, 3% for No difference in those with graft or patient prophylaxis, 16% for survival PET (but where considered as % of only those infected and Rx, prophylaxis = 13% and PET = 27% NS) 6.2% (4) of patients with CMV disease Overall: 2.2% (of all patients): 1.8% (3) with 100 d of prophylaxis, 2.6% (4) with 200 d prophylaxis 50% of tested patients Drug-associated nephrotoxicity with allograft loss in 2/4 patients with NR Mutations 3/4 UL97, ¼ UL54 3/6 were 5/6 UL97 asymptomatic 1/6 UL54 and cleared virus without Rx, 3/6 CMVS with resolution (including viremia) with Rx Not able to be accurately inferred 9/32 Ul97, 7/32 L97 + UL54 Studies include valganciclovir therapy and/or prevention and examine antiviral incidence, risk factors, and outcomes. Abbreviations: adj, adjusted; ATG, antithymocyte globulin; av, average; AVR, antiviral ; bd, twice daily; BSA, body surface area; CMV, cytomegalovirus; CMVS, CMV syndrome; egfr, estimated glomerular filtration rate; FOS, foscarnet; GCV, ganciclovir; GFR, glomerular filtration rate; HSCT, hematopoietic stem cell transplant; IV, intravenous; MDR, multidrug ; multiv, multivariate; N/A, not applicable; NR, not reported; NS, not significant; PET, preemptive therapy; RCT, randomized controlled trial; Rx, therapy; SNR, statistical significance not reported; SOT, solid organ transplant; TID, tissue invasive disease; univ, univariate; VL, viral load; valgcv, valganciclovir. a Statistical significance at P.05. b Approaching significance P =

7 of valgcv prophylaxis. A pre-valgcv review summarizes retrospective highly selected populations [20]. Of prospective SOT studies examining across organ types using both clinical and virologic definitions, is noted in up to 17.6% of viremic lung transplant recipients [21] and up to 6.2% of viremic kidney recipients [19]. Clinical outcome in -associated disease is generally poor, although confounded by selection bias. Higher rates of progressive viremia, CMV-associated mortality, and recurrent CMV are reported. This is most marked for lung recipients [21]. Crude mortality rates of up to 19% 22% have been reported in SOT recipients with CMV AVR [20]. Risk Factors A cohort study of oral GCV prophylaxis demonstrated the interdependence of host-, virus- and antiviral therapy associated risk factors for CMV disease, therapeutic failure, and AVR [22]. Among the SOT recipients, was more common in D + /R kidney-pancreas recipients who had received potent immunosuppression (OKT3). The pathway to commenced with D + /R transplant, leading to higher viral loads, with the additional factors of relative high immunosuppression, low drug concentration, and prolonged subclinical viremia, proposed to have led to selective pressure growth advantage for resistant strains [23]. Host-associated risk factors for AVR include type of transplant, with highest risk consistently demonstrated for lung and kidney-pancreas recipients [21, 22]. This may be associated with specific immunosuppression protocol risk, local organ CMV load, or anatomical factors. Degree of immunosuppression has been shown to influence AVR [22], although in studies over the past 2 decades (Table 2), use of specific antilymphocyte agent induction has not been demonstrated as a significant risk. Absence of preexisting CMV specific immunity (D + /R )isthe most consistently identified AVR risk factor [19, 21, 22, 24]. Antiviral therapy factors associated with AVR risk include delay in commencement of prophylaxis [25] and increased duration of antiviral exposure, which is shown using sequential GCV susceptibility assays [26]. Although has been described both during prophylaxis and at commencement of therapy [18, 21], prolonged duration of valgcv prophylaxis has not been shown to increase AVR [27]. Recent studies have demonstrated benefits of such prolonged prophylaxis with reduced CMV disease [2] and infection [28], especially for high-risk recipients, and this has been incorporated into guidelines [29]. Preemptive antiviral prevention has been advocated to reduce AVR, although surprisingly, AVR rates have been observed to be higher in preemptive therapy [30]. Most AVR has been reported in association with GCV or valgcv long-term use (usually >6 weeks), although exposure to other antivirals may contribute to. Prior use of UL97-activated antivirals (aciclovir, valaciclovir) with higher IC 50 (see Laboratory Monitoring below) for CMV and hence lower CMV inhibition may be a risk for AVR, although this has been uncommonly reported [31, 32]. Suboptimal drug concentration, especially with use of oral GCV, has been associated with AVR through in vitro modeling studies, with these results also reflected in case series [22, 23]. With use of intravenous GCV and valgcv, adequate GCV levels have been demonstrated in most patients with AVR [19]. There are, however, circumstances in which suboptimal exposure is more common with consequent greater potential AVR risk, such as in scenarios of malabsorption, poor compliance, intolerance, or in recovery of glomerular filtration rate (GFR) after transplant [33]. ValGCV dose alteration with GFR change or the occurrence of neutropenia has been identified more frequently in AVR [19]. Significant post-sot AVR risk factors in the era of valgcv and preventive therapy are primary infection with D + /R serostatus, day 21 persistent viremia, high peak viral load (>10 5 copies/ml in peripheral blood), and lung transplant (Table 2). Laboratory Monitoring The phenotypic assay is the reference standard for assessing antiviral susceptibility, although it is rarely used in clinical practice [34]. This method defines IC 50 (the drug concentration that inhibits 50% of viral plaque number). The proposed cutoffs for antiviral susceptibilities are 6 μm for GCV, 2 μm for CDV, and 400 μm for FOS, although standardization of assays is difficult. Interlaboratory correlation may be improved by using a sensitivity index (SI 50 ), which is the IC 50 of the isolate divided by the IC 50 of the reference strain (usually strain Ad169). A >3-fold increase in SI 50 is defined as [19]. The process of phenotypic testing, however, is lengthy (approximately 4 weeks), labor intensive, and subjective. Currently it is mostly performed to assess newly described genotypes, with recombinant virus phenotyping (see below) increasingly used owing to the difficulties of culturing viral isolates. In the era of molecular diagnostics, CMV AVR testing is usually performed clinically using polymerase chain reaction amplification and sequencing of loci direct from clinical specimens (genotypic testing), with a potential turnaround time of 2 3 days. Loci are compared to known resistant or wild-type strains. The recent development of Web-based mutation databases has facilitated this, by enabling laboratories to easily access and contribute to reports of polymorphisms or mutations and their significance [35]. Where an unknown mutation is found at a codon hot spot in the setting of clinical, the virus is considered as a probable resistant strain. Validation of the clinical effect of a given mutation or deletion is now usually performed through marker transfer 1024 CID 2013:56 (1 April) IMMUNOCOMPROMISED HOSTS

8 to a wild-type strain [36]. This involves transfer of the mutation sequence to a reference CMV strain or a bacterial artificial chromosome virus clone (methods reviewed in [7]), and subsequent (recombinant) phenotyping. Using this technique, new mutations and natural polymorphisms may be identified. It also allows categorization of mutation sensitivity to antivirals. Where the IC 50 ratio is >5, (IC 50 ) is classed as major, with a change in antiviral therapy necessary [7]. Genotypic testing allows detection of mixed wild-type and mutant populations, although minorities of <10% 20% may not be identified using conventional DNA sequencing or restriction-length fragment polymorphism analyses [37]. These minority populations may be clinically relevant, with ultradeep pyrosequencing able to detect resistant minorities in clinical [37]. Because risk of may be influenced by suboptimal antiviral exposure, therapeutic drug monitoring (TDM) of plasma GCV may be a useful tool to monitor such risk. There is also potential to examine toxicity risk, with implications for reducing unnecessary antiviral withdrawal or dose reductions that may contribute to AVR. GCV TDM it is not currently in common use, although specific populations may benefit from its application [38]. This includes pediatric kidney transplant populations where low plasma drug levels can be associated with the relatively high creatinine clearance of the transplanted adult kidney [38, 39]. Trough level parameters associated with clinical efficacy, however, have not been clearly defined. Original efficacy data derived from adult HIV-infected populations refer to optimal trough levels of >0.6 mg/l, although outcome correlation is not consistent at this level [40, 41]. Of the few studies that examined GCV levels and, suboptimal levels were infrequently found [19, 39, 40]. Mutations, Mechanisms, and Multiple Strains The genetic changes associated with in UL97 and UL54 genes of CMV have been reviewed in detail [42]. The most common (>95%) AVR strains contain mutations in the UL97 gene, conferring to GCV and valgcv. This gene encodes the UL97 protein kinase responsible for GCV phosphorylation and activation. In the virus it may be involved with encapsidation, nuclear egress, and replication [7]. UL97 mutations involved in, however, have not clearly been shown to have significant negative effects on replication, and clinical isolates may demonstrate full pathogenicity [7, 23, 43], despite amino acid substitutions or short deletions [7, 24]. The most frequently identified UL97 mutations occur at codons 460, 520, and [43]. It is proposed that segments confer by modifying substrate recognition, and 460 and 520 by modifying ATP binding [43]. Multiple codon deletions are generally associated with higher levels of [7]. Identified point mutations such as N597D and C592G demonstrate lower levels of that can be overcome by optimized GCV dosing [23, 44]. The CMV UL54 gene encodes the DNA polymerase (pol) responsible for viral replication. UL54 mutations are proposed to cause by 3 mechanisms reduced antiviral affinity (the only proposed mechanism of FOS ), reduced DNA chain incorporation, or increased antiviral excision out of the DNA chain and this may be predicted by UL54 mutation location [45]. Typically UL54 mutations evolve following UL97 mutations, in the presence of continued GCV exposure, with isolated UL54 gene mutations uncommon [19, 46]. This may reflect reduced fitness of pol mutants, possibly linked to the key replicative function of UL54. Mutations in UL54 can demonstrate cross- between GCV and CDV (multidrug [MDR]), and less commonly FOS. CDV thus may occur even in the absence of specific exposure, although FOS will usually only be seen with exposure [47]. High degrees of GCV (>30-fold) are common with dual UL97/UL54 mutations [36]. Overall, UL54 mutations may not be infrequent, as demonstrated by recent cohort studies where up to 28% of cases had UL54 mutations, mostly in combination with UL97 (Table 2) [19, 21, 27]. Sequencing of clinical isolates has demonstrated the mutated codons that may be used to predict specific patterns [42]. Mutations at codons in genes more critical to replication, such as UL54, tend to be less variable. For the UL97 gene, codons are most commonly found mutated in SOT recipients, and are associated with a large number of heterogeneous mutations [43]. UL54 mutations resulting in are more widespread in the gene with more extensive sequencing required [42]. The evolution of MDR may also occur as a result of the accumulation of different mutated strains under antiviral selective pressure. Multiple strains with different AVR profiles may be present, complicating diagnosis and prediction of therapeutic response [44]. Multiple strains have also been demonstrated to compartmentalize, with AVR variability across organs [48]. CLINICAL STRATEGIES FOR AVR Testing CMV AVR testing should be considered based on both clinical scenario and viral load monitoring during antiviral therapy. At present there is no clear benefit in testing for AVR during prophylaxis or early therapy [21]. During therapy, most guidelines recommend weekly quantitative viral load monitoring [29]. Because up to two-thirds of patients may exhibit increased quantitative CMV DNA within the first 3 weeks of IMMUNOCOMPROMISED HOSTS CID 2013:56 (1 April) 1025

9 Figure 1. Therapeutic strategies for treatment of the immunosuppressed solid organ transplant recipient with suspected antiviral. Abbreviations: ART, artesunate; AVR, antiviral ; CDV, cidofovir; FOS, foscarnet; GCV, ganciclovir; IC, inhibitory concentration; IVIG, intravenous immunoglobulin; SI 50,IC 50 of the isolate divided by the IC 50 of the reference strain; TID, tissue invasive disease; VL, viral load. antiviral therapy, it is after this point, with persistent or increasing viral loads, that must be considered [19]. Where there has been adequate compliance and antiviral exposure, potentially assessed through GCV TDM, genotypic testing should be undertaken (Figure 1). A rapid result and appropriate change in antiviral therapy may improve outcome and reduce the chance of MDR. Continued monitoring of quantitative viral loads weekly during management of resistant infections is necessary. Information on antiviral use at each test helps laboratories target and expedite genotype testing. Where GCV has been used, the UL97 gene should first be analyzed, with codons initially sequenced, as these are the sites of canonical UL97 mutations. The finding of UL97 mutations should always direct sequencing of UL54 to assess for additional GCV and cross-. In the absence of UL97 mutations with a background of sole GCV use, GCV therapy should be optimized. If clinical suspicion of is high, UL54 testing should be performed, as lone UL54 mutations have been reported in this circumstance [46]. In the setting of FOS or CDV use, the UL54 gene should always be sequenced, with codons targeted. Even where GCV has not been used, UL97 should be sequenced, as this will provide most information in guiding alternative antivirals. Sequencing results can be compared with Web-based databases that report SI50 AVR levels for mutations vs the wildtype strain. While CMV AVR definitions vary, most authors use a SI 50 (or IC 50 ratio) of >3 as indicating, and an SI 50 of >5 as indicating high-level [7, 42]. The finding of unreported sequences in clinical should be considered to indicate -associated mutations and direct further assessment through phenotypic testing CID 2013:56 (1 April) IMMUNOCOMPROMISED HOSTS

10 Clinical specimens used for testing depend on the presence of viremia or tissue invasive disease. In lung transplant patients, CMV viral load is often more predictive when determined in lung specimens [3]. In gastrointestinal disease, blood results may be negative and colonoscopy specimens more accurate [3]. Persistent tissue invasive disease, despite lack of in blood isolates, may necessitate local sampling due to strain compartmentalization, although the risk-benefit ratio of invasive sampling needs consideration. Therapy Several strategies have been proposed for the treatment of CMV AVR in the SOT recipient, although evidence is limited to case reports and case series. In Figure 1 we present a management algorithm that extends that of current consensus guidelines [29]. All strategies must include careful consideration of immunosuppressant reduction where possible. This may include considering immunosuppressive conversion to sirolimus (mtor inhibitor), which has some limited clinical evidence of efficacy [49]. In the presence of significant, life- or sight-threatening CMV disease, an empiric change or addition to therapy should be made while awaiting testing. Where high-dose GCV has been already used [50], the addition or replacement with FOS is recommended. Multidrug therapies may be successful, with early studies of combination GCV and FOS showing promise [51]. Where UL97 mutations are found and can be categorized phenotypically at an SI 50 or IC 50 ratio of <5, GCV dose optimization is recommended. High-dose intravenous GCV (up to 10 mg/kg twice daily for normal renal function) may be used, although toxicity may be limiting or necessitate addition of granulocyte colony-stimulating factor [29, 50]. The presence of additional UL54 mutations conferring increased GCV would direct a change in therapy to FOS. In cases of lone high-level UL97 mutations, therapy should also be changed to FOS. GCV-CDV cross- from UL54 mutations is common and thus this agent would not be considered if UL54 mutations are present. If FOS occurs during therapy, cross- to GCV is uncommon and this agent could be used, provided there are no additional UL97 mutations. Where multi is confirmed and remaining options are limited, novel agents (Table 1) such as leflunomide or artesunate could be considered, although current evidence is case based. There is concern that leflunomide may not be adequate as a single agent and response may be slow; thus, early initiation is recommended [16]. The addition of CMV immunoglobulin is another measure that may be considered [52]. Future Strategies Newer antivirals for use in include a range of drugs with differing or undefined activity against CMV (Table 1). Many of these (such as cyclopropavir) are phosphorylated (as is GCV) by the UL97 kinase and act on the DNA pol [53], with cross- thus possible. Therapeutic immune manipulation through CMV specific adoptive T-cell therapy has reached phase II clinical trials in HSCT recipients [54], although the role is less defined in SOT [55]. Although currently experimental, these therapies may have future application in preventing and treating, providing long-term immunity, and reducing drug toxicity. CMV disease and risk of may also be prevented by active immunization. Of several vaccines, the glycoprotein subunit vaccines have progressed farthest through clinical trials, with demonstrated reduction in viremia duration in D + /R transplant recipients [56]. In conclusion, CMV AVR can be a challenging clinical problem in SOT. At present, there is little clinical evidence to guide decision making in managing these patients. We present an easy-to-follow algorithm for testing and managing (Figure 1). Publication of cohort studies where such strategies are used is a necessary validation step. Notes Acknowledgment. We thank Dr Sunwen Chou for his review of this article. Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed. References 1. Emery VC, Asher K, Sanjuan CD. Importance of the cytomegalovirus seropositive recipient as a contributor to disease burden after solid organ transplantation. J Clin Virol 2012; 54: Humar A, Limaye AP, Blumberg EA, et al. Extended valganciclovir prophylaxis in D + /R- kidney transplant recipients is associated with long-term reduction in cytomegalovirus disease: two-year results of the IMPACT study. Transplantation 2010; 90: Iwasenko JM, Scott GM, Naing Z, Glanville AR, Rawlinson WD. Diversity of antiviral-resistant human cytomegalovirus in heart and lung transplant recipients. Transpl Infect Dis 2011; 13: Scott GM, Isaacs MA, Zeng F, Kesson AM, Rawlinson WD. Cytomegalovirus antiviral associated with treatment induced UL97 (protein kinase) and UL54 (DNA polymerase) mutations. J Med Virol 2004; 74: Balfour H Jr. Management of cytomegalovirus disease with antiviral drugs. Rev Infect Dis 1999; 12(suppl 7):s Asberg A, Humar A, Rollag H, et al. Oral valganciclovir is noninferior to intravenous ganciclovir for the treatment of cytomegalovirus disease in solid organ transplant recipients. Am J Transplant 2007; 7: Chou S. Cytomegalovirus UL97 mutations in the era of ganciclovir and maribavir. Rev Med Virol 2008; 18: Gilbert C, Boivin G. Human cytomegalovirus to antiviral drugs. Antimicrob Chemother 2005; 49: Dropulic LK, Cohen JI. Update on new antivirals under development for the treatment of double-stranded DNA virus infections. Clin Pharmacol Ther 2010; 88: IMMUNOCOMPROMISED HOSTS CID 2013:56 (1 April) 1027

11 10. Avery RK, Marty FM, Strasfeld L, et al. Oral maribavir for treatment of refractory or resistant cytomegalovirus infections in transplant recipients. Transpl Infect Dis 2010; 12: Marty FM, Ljungman P, Papanicolaou GA, et al. Maribavir prophylaxis for prevention of cytomegalovirus disease in recipients of allogeneic stem-cell transplants: a phase 3, double-blind, placebo-controlled, randomised trial. Lancet Infect Dis 2011; 11: Winston DJ, Saliba F, Blumberg E, et al. Efficacy and safety of maribavir dosed at 100 mg orally twice daily for the prevention of cytomegalovirus disease in liver transplant recipients: a randomized, doubleblind, multicenter controlled trial. Am J Transplant 2012; 12: Efferth T, Romero MR, Wolf DG, Stamminger T, Marin JJ, Marschall M. The antiviral activities of artemisinin and artesunate. Clin Infect Dis 2008; 47: Wolf DG, Shimoni A, Resnick IB, et al. Human cytomegalovirus kinetics following institution of artesunate after hematopoietic stem cell transplantation. Antiviral Res 2011; 90: Lau PK, Woods ML, Ratanjee SK, John GT. Artesunate is ineffective in controlling valganciclovir-resistant cytomegalovirus infection. Clin Infect Dis 2011; 52: Avery RK, Mossad SB, Poggio E, et al. Utility of leflunomide in the treatment of complex cytomegalovirus syndromes. Transplantation 2010; 90: Snydman DR, Kistler KD, Ulsh P, Bergman GE, Vensak J, Morris J. CMV prevention on long-term recipient and graft survival in heart transplant recipients: analysis of the Scientific Registry of Transplant Recipients (SRTR) database. Clin Transplant 2011; 25:E Martin M, Goyette N, Ives J, Boivin G. Incidence and characterization of cytomegalovirus mutations among pediatric solid organ transplant patients who received valganciclovir prophylaxis. J Clin Virol 2010; 47: Hantz S, Garnier-Geoffroy F, Mazeron MC, et al. Drug-resistant cytomegalovirus in transplant recipients: a French cohort study. J Antimicrob Chemother 2010; 65: Limaye AP. Ganciclovir-resistant cytomegalovirus in organ transplant recipients. Clin Infect Dis 2002; 35: Boivin G, Goyette N, Rollag H, et al. Cytomegalovirus in solid organ transplant recipients treated with intravenous ganciclovir or oral valganciclovir. Antiviral Therapy 2009; 14: Limaye AP, Corey L, Koelle DM, Davis CL, Boeckh M. Emergence of ganciclovir-resistant cytomegalovirus disease among recipients of solid-organ transplants. Lancet 2000; 356: Emery VC, Griffiths PD. Prediction of cytomegalovirus load and patterns after antiviral chemotherapy. Proc Natl Acad Sci U S A 2000; 97: Myhre HA, Dorenberg DH, Kristiansen KI, et al. Incidence and outcomes of ganciclovir-resistant cytomegalovirus infections in 1244 kidney transplant recipients. Transplantation 2011; 92: Scott GM, Naing Z, Pavlovic J, et al. Viral factors influencing the outcome of human cytomegalovirus infection in liver transplant recipients. J Clin Virol 2011; 51: de Ona Navarro M, Melon S, Mendez S, et al. Assay of cytomegalovirus susceptibility to ganciclovir in renal and heart transplant recipients. Transpl Int 2002; 15: Boivin G, Goyette N, Farhan M, Ives J, Elston R. Incidence of cytomegalovirus UL97 and UL54 amino acid substitutions detected after 100 or 200 days of valganciclovir prophylaxis. J Clin Virol 2012; 53: Finlen Copeland CA, Davis W, Snyder LD, et al. Long-term efficacy and safety of 12 months of valganciclovir prophylaxis compared with 3 months after lung transplantation: a single-center, long-term followup analysis from a randomized, controlled cytomegalovirus prevention trial. J Heart Lung Transplant 2011; 30: Kotton CN, Kumar D, Caliendo AM, et al. International consensus guidelines on the management of cytomegalovirus in solid organ transplantation. Transplantation 2010; 89: Couzi L, Helou S, Bachelet T, et al. High incidence of anticytomegalovirus drug among D+R kidney transplant recipients receiving preemptive therapy. Am J Transplant 2012; 12: Michel D, Hohn S, Haller T, Jun D, Mertens T. Aciclovir selects for ganciclovir-cross- of human cytomegalovirus in vitro that is only in part explained by known mutations in the UL97 protein. J Med Virol 2001; 65: Alain S, Hantz S, Scieux C, et al. Detection of ganciclovir after valacyclovir-prophylaxis in renal transplant recipients with active cytomegalovirus infection. J Med Virol 2004; 73: Eid AJ, Arthurs SK, Deziel PJ, Wilhelm MP, Razonable RR. Emergence of drug-resistant cytomegalovirus in the era of valganciclovir prophylaxis: therapeutic implications and outcomes. Clin Transplant 2008; 22: Drew WL, Miner R, Saleh E. Antiviral susceptibility testing of cytomegalovirus: criteria for detecting to antivirals. Clin Diagn Virol 1993; 1: Chevillotte M, von Einem J, Meier BM, Lin FM, Kestler HA, Mertens T. A new tool linking human cytomegalovirus drug mutations to phenotypes. Antiviral Research 2010; 85: Chou S, Van Wechel LC, Lichy HM, Marousek GI. Phenotyping of cytomegalovirus drug mutations by using recombinant viruses incorporating a reporter gene. Antimicrob Agents Chemother 2005; 49: Schindele B, Apelt L, Hofmann J, Nitsche A, Michel D, Voigt S. Improved detection of mutated human cytomegalovirus UL97 by pyrosequencing. Antimicrob Agents Chemother 2010; 54: Luck S, Lovering A, Griffiths P, Sharland M. Ganciclovir treatment in children: evidence of subtherapeutic levels. Int J Antimicrob Agents 2011; 37: Vethamuthu J, Feber J, Chretien A, Lampe D, Filler G. Unexpectedly high inter- and intrapatient variability of ganciclovir levels in children. Pediatr Transplant 2007; 11: Piketty C, Bardin C, Gilquin J, Gairard A, Kazatchkine MD, Chast F. Monitoring plasma levels of ganciclovir in AIDS patients receiving oral ganciclovir as maintenance therapy for CMV retinitis. Clin Microbiol Infect 2000; 6: Fishman JA, Doran MT, Volpicelli SA, Cosimi AB, Flood JG, Rubin RH. Dosing of intravenous ganciclovir for the prophylaxis and treatment of cytomegalovirus infection in solid organ transplant recipients. Transplantation 2000; 69: Lurain NS, Chou S. Antiviral drug of human cytomegalovirus. Clin Microbiol Rev 2010; 23: Chou S, Waldemer RH, Senters AE, et al. Cytomegalovirus UL97 phosphotransferase mutations that affect susceptibility to ganciclovir. J Infect Dis 2002; 185: Iwasenko JM, Scott GM, Rawlinson WD, Keogh A, Mitchell D, Chou S. Successful valganciclovir treatment of post-transplant cytomegalovirus infection in the presence of UL97 mutation N597D. J Med Virol 2009; 81: Cihlar T, Fuller MD, Cherrington JS. Characterization of drug -associated mutations in the human cytomegalovirus DNA polymerase gene by using recombinant mutant viruses generated from overlapping DNA fragments. J Virol 1998; 72: Boivin G, Goyette N, Gilbert C, Covington E. Analysis of cytomegalovirus DNA polymerase (UL54) mutations in solid organ transplant patients receiving valganciclovir or ganciclovir prophylaxis. J Med Virol 2005; 77: Smith IL, Cherrington JM, Jiles RE, Fuller MD, Freeman WR, Spector SA. High-level of cytomegalovirus to ganciclovir is associated with alterations in both the UL97 and DNA polymerase genes. J Infect Dis 1997; 176: Hamprecht K, Eckle T, Prix L, Faul C, Einsele H, Jahn G. Ganciclovirresistant cytomegalovirus disease after allogeneic stem cell transplantation: pitfalls of phenotypic diagnosis by in vitro selection of an UL97 mutant strain. J Infect Dis 2003; 187: CID 2013:56 (1 April) IMMUNOCOMPROMISED HOSTS

12 49. Ozaki KS, Camara NO, Nogueira E, et al. The use of sirolimus in ganciclovir-resistant cytomegalovirus infections in renal transplant recipients. Clin Transplant 2007; 21: West P, Schmiedeskamp M, Neeley H, Oberholzer J, Benedetti E, Kaplan B. Use of high-dose ganciclovir for a resistant cytomegalovirus infection due to UL97 mutation. Transpl Infect Dis 2008; 10: Mylonakis E, Kallas WM, Fishman JA. Combination antiviral therapy for ganciclovir-resistant cytomegalovirus infection in solid-organ transplant recipients. Clin Infect Dis 2002; 34: Sarmiento E, Fernandez-Yanez J, Munoz P, et al. Hypogammaglobulinemia after heart transplantation: use of intravenous immunoglobulin replacement therapy in relapsing CMV disease. Int Immunopharmacol 2005; 5: James SH, Hartline CB, Harden EA, et al. Cyclopropavir inhibits the normal function of the human cytomegalovirus UL97 kinase. Antimicrob Agents Chemother 2011; 55: Peggs KS, Verfuerth S, Pizzey A, Chow SC, Thomson K, Mackinnon S. Cytomegalovirus-specific T cell immunotherapy promotes restoration of durable functional antiviral immunity following allogeneic stem cell transplantation. Clin Infect Dis 2009; 49: Brestrich G, Zwinger S, Fischer A, et al. Adoptive T-cell therapy of a lung transplanted patient with severe CMV disease and to antiviral therapy. Am J Transpl 2009; 9: Pass RF, Zhang C, Evans A, et al. Vaccine prevention of maternal cytomegalovirus infection. N Engl J Med 2009; 360: Vaudry W, Ettenger R, Jara P, et al. Valganciclovir dosing according to body surface area and renal function in pediatric solid organ transplant recipients. Am J Transplant 2009; 9: van der Beek MT, Berger SP, Vossen AC, et al. Preemptive versus sequential prophylactic-preemptive treatment regimens for cytomegalovirus in renal transplantation: comparison of treatment failure and antiviral. Transplantation 2010; 89: Prichard MN, Kern ER. The search for new therapies for human cytomegalovirus infections. Virus Res 2011; 157: Kaul DR, Stoelben S, Cober E, et al. First report of successful treatment of multidrug-resistant cytomegalovirus disease with the novel anti-cmv compound AIC246. Am J Transplant 2011; 11: Chou S, Bowlin TL. Cytomegalovirus UL97 mutations affecting cyclopropavir and ganciclovir susceptibility. Antimicrob Agents Chemother 2011; 55: Azad RF, Driver VB, Tanaka K, Crooke RM, Anderson KP. Antiviral activity of a phosphorothioate oligonucleotide complementary to RNA of the human cytomegalovirus major immediate-early region. Antimicrob Agents Chemother 1993; 37: Waldman WJ, Knight DA, Blinder L, et al. Inhibition of cytomegalovirus in vitro and in vivo by the experimental immunosuppressive agent leflunomide. Intervirology 1999; 42: IMMUNOCOMPROMISED HOSTS CID 2013:56 (1 April) 1029