Portal de Periódicos da CAPES

Optimal use of maraviroc in clinical practice

2008; Lippincott Williams & Wilkins; Volume: 22; Issue: 17 Linguagem: Inglês

10.1097/qad.0b013e3283136d95

1473-5571

Vincent Soriano, Anna María Geretti, Carlo Federico Perno, Gerd Fätkenheuer, Deenan Pillay, Jacques Reynes, Giuseppe Tambussi, Vincent Cálvez, José Alcamı́, Juergen Rockstroh,

HIV/AIDS Research and Interventions

Introduction Current recommended regimens to treat HIV infection consist of a combination of two nucleoside reverse transcriptase inhibitors (NRTIs) and either a nonnucleoside reverse transcriptase inhibitor (NNRTI) or a protease inhibitor. Cross-resistance and toxicities derived from the long-term use of these agents, however, often limit their sustained therapeutic success. For this reason, the search for new antiretroviral agents continues actively. New classes of antiretroviral drugs targeting different steps of the HIV replication cycle (i.e., viral entry or integration) have recently been approved, and others (i.e., maturation inhibitors) are in advance stages of clinical development. Viral entry is an attractive step for the development of new agents against HIV variants resistant to current antiretroviral drugs. HIV gains entry into CD4-expressing cells through a series of sequential interactions between the envelope glycoprotein gp120 and the CD4 receptor and one of two coreceptor molecules, chemokine (C-C motif) receptor 5 (CCR5) or chemokine (C-X-C motif) receptor 4 (CXCR4), which are expressed on the surface of target cells [1,2]. A conformational change in gp41 immediately thereafter permits the penetration of viral RNA molecules into the cell. The fusion inhibitor, enfuvirtide (Fuzeon; Roche, Basel, Switzerland) [3], and the small molecule CCR5 antagonist, maraviroc (Celsentri or Selzentry; Pfizer Ltd., Sandwich, UK) [4], are the first entry inhibitors approved for clinical use in HIV-infected persons. Maraviroc is the only approved CCR5 antagonist and the single oral HIV entry inhibitor in clinical use. Maraviroc specifically inhibits the replication of R5-tropic HIV variants by an allosteric mechanism after binding to the transmembrane CCR5 coreceptor cavity. The present article reviews the main features of maraviroc and discusses the optimal clinical scenarios in which the drug can be used. Mechanism of action of small molecule chemokine (C-C motif) receptor 5 antagonists These compounds represent the first anti-HIV agents targeting a host cell protein instead of a viral enzyme. In the HIV entry process, the CD4–gp120 interaction induces conformational changes in the viral envelope that expose a chemokine receptor-binding site and allows the CD4–gp120 complex to interact with a chemokine coreceptor, typically CCR5 or CXCR4. The CD4–gp120 complex binds to either coreceptor through interactions with the V3 region of gp120, although other HIV gp120 regions such as V1/V2, C4 and the bridging sheet may also be involved [5]. The gp120–coreceptor binding map suggests that for R5 viruses, the N-terminal domain and the second extracellular loop (ECL2) of the CCR5 coreceptor are essential for coreceptor recognition and therefore for inhibitory activity (Fig. 1).Fig. 1: Mechanism of action of maraviroc. CCR5, chemokine (C-C motif) receptor 5; ECL, extracellular loop; gp, glycoprotein; MVC, maraviroc.The use of CCR5 or CXCR4 coreceptor by HIV-1 is mainly, although not exclusively, determined by the amino acid sequence of the V3 region of gp120 [6,7]. Accordingly, HIV-1 isolates are classified as either R5 tropic, X4 tropic or dual/mixed tropic depending on their coreceptor use [8]. The term dual/mixed is used to describe a virus population that may contain either dual tropic viruses that can use both chemokine coreceptors or a mixture of viruses that exclusively use CCR5 or CXCR4. In the mid-1990s, specific chemokine receptors were identified as essential for HIV-1 entry into host cells. The natural ligands of CCR5 are the beta-chemokines, regulated upon activation, normal T-cell expressed and secreted (RANTES), macrophage inflammatory protein (MIP)1α and MIP1β, which were shown to be the HIV suppressive factors released by CD8+ T cells [9]. These observations, together with the recognition that genetic defects in CCR5 expression significantly reduced HIV acquisition and disease progression [10,11], provided the basis for the development of HIV coreceptor antagonists as antiretroviral drugs. In August 2007, more than a decade after the discovery of the critical role of chemokines and their coreceptors in HIV infection, maraviroc received approval for the treatment of HIV infection as the first coreceptor antagonist [4]. Maraviroc (formerly UK-427 857) is an orally bioavailable noncompetitive (allosteric) inhibitor of the CCR5 chemokine coreceptor with potent in-vitro activity (IC90 ∼ 2 nml/l) [12–14]. Maraviroc binds to the transmembrane coreceptor cavity within the 2, 3, 6 and 7 helix [13,14]. Binding triggers conformational changes, especially in the ECL2 region of the CCR5 coreceptor, which prevents interaction with the V3 crown of gp120 (Fig. 1). Many other compounds are currently being tested as coreceptor antagonists, and some are in advance stages of clinical development (Table 1) [2]. Vicriviroc (formerly known as Schering D) is on phase III clinical trials. It is an orally bioavailable inhibitor of R5 viruses and, similar to maraviroc, interacts directly with the CCR5 transmembrane cavity, inhibiting gp120 binding by an allosteric mechanism [15]. The clinical development of a third CCR5 inhibitor, aplaviroc, was discontinued because of increased risk of hepatotoxicity [16], despite significant antiviral potency [17].Table 1: Entry inhibitors already approved for clinical use or in advanced clinical development.Efficacy and safety of maraviroc in registrational trials The in-vivo antiviral activity of maraviroc was initially described in a monotherapy dose-ranging study that compared once daily and twice daily doses of maraviroc (25, 50, 100, 150, 300 mg) with placebo over 10 days in HIV-infected drug-naive individuals with R5 virus phenotype. A reduction in plasma HIV RNA of at least 1 log copies/ml was observed with doses of 100 mg or more [18]. Subsequently, in the MOTIVATE 1 (conducted in North America) and MOTIVATE 2 (conducted in Europe, Australia and the United States) trials, triple class-resistant patients harbouring R5 viruses at baseline were randomized to receive either maraviroc (150 or 300 mg) once daily or twice daily or placebo, each combined with an optimized background regimen (OBR). In these double-blind, placebo-controlled studies, a combination of maraviroc and OBR demonstrated significantly greater virological and immunological efficacy and similar safety profile than OBR alone. At 24 weeks, the proportion of patients with plasma HIV RNA load of less than 50 copies/ml was doubled in the maraviroc arm than in the placebo arm (45 vs. 23%, respectively; P < 0.0001) [19,20]. Importantly, responses were maintained at 48 weeks (45 vs. 17%, respectively; P < 0.0001) (Fig. 2) [21,22].Fig. 2: Efficacy of maraviroc in the MOTIVATE and MERIT trials: 48-week results. D/M, dual/mixed; EFV, efavirenz; ENF, enfuvirtide; MVC, maraviroc; OBR, optimized background regimen; ZDV, zidovudine; 3TC, lamivudine. *Noninferiority margin in −10%.As expected, maraviroc demonstrated the greatest virological efficacy in treatment-experienced individuals when combined with at least one active agent in the OBR. In particular, a beneficial effect of combining the two entry inhibitors, maraviroc and enfuvirtide, was observed. The proportion of patients who achieved a viral load of less than 50 copies/ml was more elevated when enfuvirtide was used for the first time (Fig. 2) [23]. Similar enhanced responses were obtained in patients who received maraviroc along with a protease inhibitor in the absence of major protease resistance mutations. These data illustrate the importance of including additional fully active agents when employing a novel drug in HIV treatment-experienced patients. The A4001029 study compared the activity of maraviroc with that of placebo in treatment-experienced individuals with dual/mixed, X4 viruses or undetermined tropism who also received OBR [24]. There were no significant differences in plasma HIV RNA responses between the two arms, indicating a lack of recognizable antiviral activity of maraviroc in the presence of dominant quasispecies of X4 or dual tropic viruses. On the contrary, there was no evidence at week 48 of any adverse effect on viral load or CD4 cell count in the maraviroc group when compared with the control arm. Overall, the study confirmed that maraviroc should only be used in the setting of pure R5 viruses. Because R5 tropic viruses predominate in early stages of HIV disease, the chances of success using CCR5 antagonists may be highest in early infection. The MERIT trial evaluated the safety and efficacy of maraviroc (300 mg) twice daily as compared with that of efavirenz (600 mg) once daily, both with zidovudine and lamivudine, as initial therapy in drug-naive HIV-infected patients. In an intent-to-treat evaluation, the trial failed to demonstrate noninferiority of either the once daily or twice daily maraviroc arms relative to efavirenz, using the attainment of plasma HIV RNA load of less than 50 copies/ml at 48 weeks as the primary end-point [25]. Interestingly, although more patients discontinued maraviroc because of lack of efficacy than efavirenz (12 vs. 4%), fewer patients discontinued maraviroc than efavirenz because of adverse events (4.2 vs. 13.6%). The 96-week data of the MERIT study are awaited. Meanwhile, a detailed analysis of the virological correlates of treatment failure at 48 weeks has recently been presented [26]. In 13 patients (3.8%) receiving maraviroc, a switch in viral tropism from R5 to dual/mixed was noticed between the screening and baseline samples despite an interval shorter than 6 weeks. In this subset of patients, the response to maraviroc was significantly reduced (54.6 vs. 7.1% achieved <50 HIV RNA copies/ml) as compared with efavirenz. In contrast, in patients harbouring R5 viruses at the time of beginning maraviroc therapy, similar responses were seen on comparing maraviroc and efavirenz groups (69.3 vs. 68%, respectively) (Fig. 2). These data indicate that the presence of X4 or dual tropic variants at baseline predicts maraviroc failure in drug-naive patients. Repeat tropism testing or the use of a more sensitive assay would have allowed the identification and exclusion of patients with X4 dual/mixed virus populations. The results of the MERIT trial illustrate the current challenges of conducting clinical trials in treatment-naive patients, as the current standard of care achieves a high degree of virological and immunological success that is difficult to overcome. In this context, the tolerability and safety profile of new drugs play a key role in determining the likely use in clinical practice. An additional consideration with regard to the convenience of once daily dosing for most initial current antiretroviral regimens and information on the potential use of maraviroc once daily is still limited. Maraviroc has demonstrated an excellent safety profile in clinical trials. Earlier data on CCR5 antagonists raised concerns about their potential for hepatotoxicity. The clinical development of aplaviroc was halted following the report of episodes of idiosyncratic hepatotoxicity, most likely intrinsic to the molecule rather than reflecting a class toxicity [16]. Although confounded by other factors, only one case of severe hepatotoxicity accompanied by allergic features was reported in the maraviroc registrational clinical trials [27]. Subsequently, an increase in liver enzymes was observed in treatment-experienced patients receiving maraviroc, although there was no overall increase in grade 3/4 liver function test abnormalities. In the MOTIVATE trials, no difference in the rate of adverse events between maraviroc and placebo was observed, and no excess hepatotoxicity was recorded in the maraviroc arm [20,21]. A further concern related to the use of CCR5 inhibitors was the potential for an increased risk of malignancies, as suggested by the ACTG 5211 trial conducted with vicriviroc [28]. In the MOTIVATE trials, a total of 11 malignancies were reported, including three Kaposi's sarcomas and three lymphomas in the maraviroc arms and three Kaposi's sarcomas and two lymphomas in the placebo arms. Because a greater number of patients were enrolled in the maraviroc arms, a lower incidence of malignancies was in fact observed using maraviroc [21,22]. In the MERIT trial, the rate of all serious adverse events and grade 3/4 adverse events was lower in the maraviroc arm than in the efavirenz arm. In particular, the incidence of hepatotoxicity or malignancies was very low in both groups. Interestingly, lipid elevations were more pronounced in the efavirenz arm than in the maraviroc arm [25]. CD4 recovery using maraviroc The 48-week results of the MOTIVATE 1 and 2 trials demonstrated significantly greater increases in CD4+ T-cell counts in both the once daily and twice daily maraviroc arms than in the placebo arm. The mean change in CD4 cell count (cell/μl) in the combined analysis was +116, +124 and +61, respectively [22]. In the A4001029 study, which assessed maraviroc activity in treatment-experienced individuals with dual/mixed or X4 viruses, despite lack of virological suppression, the mean gain in CD4+ T-cell count was greater in patients receiving maraviroc than in placebo (+65, +78 and +51, respectively) [24]. Similarly, in the MERIT trial, although the maraviroc twice daily arm failed to demonstrate virological noninferiority when compared with the efavirenz arm, it showed a significantly greater increase in CD4+ cell count (170 vs. 140 cell/μl) [25]. Taken together, these data suggest that maraviroc could provide an immunological benefit beyond that related to its direct antiviral activity. Interestingly, this phenomenon has also been reported with enfuvirtide, in which increasing CD4 cell count can be observed despite virological failure, occasionally with high levels of viraemia [29–31]. The mechanisms underlying this phenomenon are currently unknown. Several possible explanations have been proposed. The selection of specific resistance mutations to enfuvirtide could dramatically impair viral fitness [32,33] or modify the structural stability of the Rev-responsive element [31,34,35]. Alternatively, the entry inhibitor may specifically reduce immune activation. Reducing the sequestration of CD4+ T cells into the lymphoid tissue may allow their redistribution, thereby increasing the absolute number in the bloodstream [29]. Thus, HIV entry inhibitors, because of their unique mechanisms of action, could provide an intrinsic beneficial effect on immune responses. Clearly, more research in this field is warranted in order to better understand this phenomenon and define its potential applications in clinical practice. HIV tropism assessment The antiretroviral activity of CCR5 antagonists is limited to R5 viruses. Therefore, treatment candidates are patients who lack detectable X4 viruses or dual tropic viruses. Several assays have been developed to determine HIV tropism in clinical samples (Table 2) [36]. The MT-2 assay was widely used during the late 1980s to test the cytopathic effect of HIV isolates and served to establish the classification of HIV strains into syncytium-inducing and nonsyncytium-inducing viruses. MT-2 cells express CXCR4 but not CCR5 receptors on their surface. Thus, any evidence of replication on culture of MT-2 cells indirectly reflects the presence of X4 viruses. The main disadvantage of the MT-2 assay is the requirement for viral isolates from stimulated peripheral blood mononuclear cells (PBMCs), which makes the assay very laborious and unaffordable for clinical use.Table 2: Tools for HIV tropism determination.Novel techniques to determine HIV tropism have been developed on the basis of recombinant virus phenotypic assays, making large-scale epidemiological and clinical studies possible. Parts or the whole env gene are amplified from plasma HIV RNA to generate recombinant virions, which are subsequently used to infect human cell lines expressing CD4 and either CXCR4 or CCR5 receptors. The Trofile assay [37] (Monogram Biosciences, South San Francisco, California, USA) has been extensively used in clinical trials and is currently in routine use for testing viral tropism in patients for whom treatment with maraviroc as part of their antiretroviral regimen may be considered. One of the limitations of recombinant virus assays is their limited sensitivity for the detection of dual tropic and X4 viruses when these are present as minority or low-frequency species, as their presence even as minority populations seems to significantly impair the antiviral activity. The recognition of X4 variants below the limit of detection of the Trofile assay at baseline in a small subset of HIV patients has been postulated to explain the subsequent virological failure in a subset of patients undergoing therapy with CCR5 antagonists. Monogram Biosciences has recently developed an enhanced version of the Trofile assay that is able to detect X4 variants representing only 0.3% of the viral population [38]. It is now critical to estimate a cut-off for the proportion of X4/dual tropic variants, above which virological failure using CCR5 antagonists would be predicted. Phenotypic assays using either HIV primary isolates or recombinant viruses are considered to be reliable. However, they are complex, expensive and require special facilities and expertise. Alternatively, HIV-1 coreceptor usage can be predicted using the amino acid sequence of the V3 region of gp120, as it is the main determinant of viral tropism [6,7]. Several genotypic algorithms have been developed and some are available on websites of free public access [36]. Geno2pheno: http://www.geno2pheno.org Wetcat: http://genomiac2.ucsd.edu:8080/wetcat/v3.html PSSM: http://ubik.microbiol.washington.edu/computing/pssm/ Comparative studies have examined the accuracy and concordance between genotypic and phenotypic tests to determine HIV tropism. Low et al.[39] evaluated the performance of several V3 algorithms in clinical samples collected from 977 antiretroviral-naive individuals who had been previously tested using the Trofile assay. Although the genotypic algorithms demonstrated high specificity (88–97%), the sensitivity to detect X4 variants was poor (22–45%). More recently, studies conducted in distinct populations and using other recombinant phenotypic assays as a reference have shown better sensitivity of genotypic tests for the detection of X4 variants, especially in antiretroviral-experienced patients, individuals infected with subtype B variants, and when using certain genotypic methods or combinations of them [40–47]. Although genotypic tools need further improvements before moving into clinical use, they may provide an easier and more rapid determination of tropism and, therefore, facilitate the assessment of whether treatment with CCR5 antagonists is appropriate. This may be particularly relevant for places where phenotypic assays are less accessible. Resistance to maraviroc The inhibitory activity of CCR5 antagonists may be evaded by two mechanisms: emergence of X4 or dual tropic viruses that preexist the introduction of therapy as a minority viral population that outgrows in the presence of the drug; selection of mutations in the HIV-1 gp120 molecule that allows the virus to bind to the drug-bound form of the CCR5 coreceptor [1,2]. This last mechanism of escape from CCR5 antagonists indirectly proves that these drugs do not compete with the natural chemokine ligands targeting the same binding region but rather an allosteric inhibition may be occurring [1]. Changes in the V3 loop have been observed in patients experiencing maraviroc failure in the MOTIVATE trials. Mutations arose especially in the stem and tip of the V3 loop, but the pattern of amino acid changes seems to be quite heterogeneous on comparing distinct patients. Moreover, changes in other regions of gp120 outside the V3 loop were noticed [48]. Interestingly, shifts in the 50% inhibitory concentration (IC50), a common parameter used to quantify the level of resistance to other antiretroviral agents, could not be used as a marker of maraviroc resistance; in contrast, plateaus in the maximal percentage of inhibition were identified as a marker of maraviroc resistance, demonstrating the ability of the virus to use compound-occupied receptors [49]. Data from the MOTIVATE trials suggest that among treatment failures, the majority of patients in the maraviroc arms experienced a shift from R5 to X4 or dual/mixed tropic viruses, which was seen rarely in the placebo arm (64 vs. 5%, respectively) [22]. Subsequent clonal analysis in patients who failed maraviroc in the presence of X4 or dual/mixed variants suggested that in seven of 10 patients, these variants preexisted in the viral population at baseline below the level of detection of the standard tropism assay. Phylogenetic analyses showed that X4/dual/mixed variants detected at failure were genetically identical to X4/dual/mixed strains present at baseline as minor populations [50]. On the basis of these data, an outgrowth of X4 or dual tropic variants preexisting at baseline, rather than a shift in coreceptor usage at the individual virus level, seems to be responsible for virus escape from maraviroc in most cases. Interestingly, upon maraviroc discontinuation, R5 viruses tend to regain predominance in the virus population indicating that the X4 enrichment during therapy is not long lasting. Maraviroc drug interactions Maraviroc is a substrate for the P-glycoprotein and the cytochrome P450 (3A4), both of which are involved in the metabolism of multiple other agents, including several antiretroviral drugs. Therefore, drug interactions using maraviroc should be kept in mind. The main drug interactions of maraviroc, which require adjustments in drug dosing, are summarized in Table 3. Briefly, the standard 300 mg twice daily dose of maraviroc does not require changes when the drug is given concomitantly with methadone, statins, nevirapine, cotrimoxazole, etinilestradiol nor pegylated interferon and ribavirin. However, dose adjustments have to be made when maraviroc is used along with some potent inhibitors or inducers of CYP3A4 [51].Table 3: Drug interactions of maraviroc requiring dose adjustments.Maraviroc in special situations The unique features of maraviroc make this drug of special interest to be used in some particular circumstances and patient populations, such as preexposure and postexposure prophylaxis, tuberculosis treatment and viral hepatitis coinfection. Prevention of HIV infection CCR5 antagonists act very early in the virus life cycle, inhibiting the attachment of HIV particles to target cells, thereby blocking the R5 variants from entering into host cells and proceeding to viral replication. Because R5 viruses are generally responsible for the establishment of new HIV infections and predominate in early stages of disease [52], the use of CCR5 antagonists as preexposure and postexposure prophylaxis is currently one of the most active areas of research. Interestingly, maraviroc (300 mg twice daily) reaches similar concentrations in the plasma and the cervicovaginal fluid, achieving higher exposure to the drug in the genital tract than other antiretroviral agents [53]. In order to prevent or reduce the risk of sexual transmission of HIV, maraviroc could be given without prior testing of viral tropism. Preliminary evidence from anecdotal cases of occupational exposure to HIV has been reported, supporting the potential benefit of maraviroc for postexposure prophylaxis [54]. Clearly, studies are warranted to test the benefit of CCR5 antagonists to reduce sexual transmission of HIV. As with other antiretroviral drugs tested in this setting, high-risk populations such as prostitutes, partners of HIV-infected persons, etc. could be the study population. HIV patients with tuberculosis Maraviroc has demonstrated a favourable pharmacokinetic profile when given in combination with inducers of CYP3A4, such as rifampicin. Interactions between rifampicin and many antiretroviral agents often make the treatment of tuberculosis difficult in HIV-infected patients on HAART. In contrast, a simple doubling of the daily maraviroc dose can correct the decreased plasma exposure caused by the induction of CYP3A4 by rifampicin [51], facilitating the management of these patients. Hepatitis B or C coinfected patients The CCR5 coreceptor seems to play a role in chronic hepatitis B and C infections. The density of CCR5 molecules on the T-cell surface could determine the intensity of T-cell recruitment in the liver of patients with chronic viral hepatitis, influencing the extent of liver damage. Individuals heterozygous for the 32-bp deletion in the CCR5 coreceptor gene have substantially reduced levels of CCR5 on the cell surface. A recent report has associated the presence of the Δ32-ccr5 polymorphism with reduced risk of viral persistence following exposure to hepatitis B virus (HBV) [55]. Moreover, HBV fails to establish chronic infection in patients homozygous for the deletion. HBV-infected patients heterozygous for this allele show an amelioration of liver disease progression, suggesting that CCR5 contributes to viral persistence and replication in HBV infection. Similar observations have been reproduced in HIV/HBV-coinfected patients. Thus, CCR5 antagonists could be of therapeutic interest in patients with chronic hepatitis B alone or coinfected with HIV. For hepatitis C virus (HCV) infection, controversial results have been found. Although some authors have reported a relationship between low CCR5 levels and both increased viral clearance following HCV exposure and reduced liver inflammation after infection [56,57], others have not confirmed these findings [58]. Clearly, more studies addressing this issue are warranted, as current treatment for hepatitis C in HIV-coinfected patients has only limited efficacy. Patients with cardiovascular risk The recent recognition that many of the most commonly prescribed antiretroviral drugs (e.g., ritonavir-boosted protease inhibitors, efavirenz, abacavir and didanosine) may be associated with metabolic abnormalities and increased cardiovascular risk [59,60] has created a possible need for switching drugs in patients with undetectable viraemia [61]. In this specific situation, however, tropism testing is difficult because no virus can be obtained from plasma, preventing the use of maraviroc. The value of determining the tropism using either stored plasma specimens collected prior to undetectability or proviral DNA recovered from PBMCs is currently being tested, and preliminary data have provided promising results [62]. An alternative option could be the use of the drug without prior tropism testing with checking of viral load response soon thereafter (e.g., 4 weeks later). Patients with impaired CD4 recovery despite virological suppression on HAART Patients with impaired CD4 recovery despite virological suppression on HAART represent around 5–10% of patients with complete suppression of plasma HIV RNA under antiretroviral therapy. Because this subset of patients could be at increased risk for developing opportunistic complications, the search for alternative strategies to promote CD4 expansion, including the use of interleukin-2 [63] or intensification protocols [64], is being pursued. As discussed earlier, CD4 cell gains in the absence of complete viral suppression is a characteristic feature of entry inhibitors [29–35]. If definitively proven, this property could be used for managing the subset of patients unable to show an adequate CD4 cell recovery despite achieving sustained virological suppression with HAART. Because maraviroc is given orally, it would be preferred over enfuvirtide among HIV entry inhibitors. Summary Maraviroc belongs to a new class of anti-HIV drugs named CCR5 antagonists, which block HIV entry into cells. It has proven potent efficacy in treatment-experienced patients with multiple drug failure. Its favourable toxicity profile makes the drug attractive for consideration in other clinical scenarios, including patients with earlier stages of disease, cardiovascular risk and viral hepatitis coinfection. Because of its unique mechanism of action-blocking viral entry, studies in patients with primary HIV infection and in the context of preexposure and postexposure prophylaxis are particularly warranted. However, given its exclusive activity against CCR5 tropic strains, viral tropism testing is mandatory before using CCR5 antagonists in the clinic. Efforts to improve the reliability of viral tropism assessment using genotypic tests are underway, and it is likely that these tools will soon replace phenotypic assays, which are more expensive, difficult to perform and take a long time. Acknowledgements This work was funded by an unrestricted grant from Pfizer. We would like to thank Eva Poveda for excellent writing assistance and helpful comments. Specific tasks of authors: all participated in a 1-day meeting to discuss different aspects of maraviroc's clinical use. Each author addressed a different topic. V.S. wrote the draft, which was circulated among all coauthors. A final draft was prepared incorporating the suggestions from all authors.