Plasma concentrations of maraviroc and raltegravir after dual therapy in patients with long-term suppressed viraemia: ROCnRAL ANRS 157 study

Sir, In a growing population of ageing and polymedicated HIV-infected patients, long-term tolerance remains a major public health issue. Current treatment guidelines even recommend the use of conventional tritherapy in the form of two NRTIs with a third agent from a different antiretroviral class.1 However, the long-term toxicity of these agents in ageing patients with comorbidities can be a limiting factor in treatment.2 The clinical evaluation of NRTI-, NNRTI- and PI-sparing strategies is needed. In light of the respective antiretroviral activities, pharmacokinetic profiles and metabolic characteristics of maraviroc and raltegravir,3,4 the trial ‘Raltegravir plus Maraviroc Therapy in Controlled HIV Patients Presenting with Lipohypertrophy’ (ROCnRAL ANRS 157; NCT01420523) proposed a dual regimen in patients with long-term virological suppression, clinical lipohypertrophy and R5-tropic B or CRF02_AG HIV-1 subtype infection (determined by Sanger DNA sequencing). The study protocol was reviewed and approved by the respective institutional ethics committees. The main objective of this non-comparative, Phase II, multicentre pilot trial was to assess the ability to maintain plasma HIV-1 RNA <50 copies/mL after 24 weeks of maraviroc plus raltegravir (300/400 mg twice daily). However, this study was prematurely discontinued because of the high treatment failure rate: 7 for the first 44 enrolled of 90 planned patients (5 virological failures and 2 serious adverse events).5 One of the mechanistic hypotheses for the unexplained virological failures was a deleterious drug interaction between maraviroc and raltegravir or lack of virological robustness of the dual regimen. This pharmacokinetic substudy aimed to (i) answer the question of drug interaction and (ii) assess free and total plasma concentrations of maraviroc and raltegravir during follow-up and at the time of virological failure. Eligibility criteria were as for the ROCnRAL trial.5 Comedications that might affect the respective pharmacokinetics were not allowed. Blood samples were collected at a mean of 12 ± 2 h after the last drug intake for assessing free and total trough plasma concentration (Cmin) at weeks 4, 24 and 48 and, if possible, at the time of virological failure. Plasma protein binding analysis involved an ultrafiltration assay with duplicate ultrafiltration (Centrifree, Millipore) (coefficient of variation <15%). The Cmin for maraviroc and raltegravir (and its inactive glucuronide metabolite) was determined by LC coupled with tandem MS (Waters Acquity UPLC-TQD; limits of quantification <1 ng/mL) as described previously,6 with some modification. The Cmin for maraviroc and raltegravir were interpreted by the antiviral thresholds of 50 ng/mL. We used a maraviroc threshold of 10 ng/mL that should saturate membrane CCR5 on all CD4 cells7 and a raltegravir threshold of 15 ng/mL (the IC95 in 50% human serum).8 Before the ROCnRAL trial was discontinued, 44 patients (86% male, median age 55 years) were enrolled. In pooled samples collected at weeks 4, 24 and 48, the median (IQR) total Cmin for maraviroc and raltegravir were 107 ng/mL (54–165) and 152 ng/mL (95–411), respectively. The free fractions were 42% and 20% for maraviroc and raltegravir, respectively. The Cmin was >10 ng/mL for most maraviroc free fraction samples (97%; 67/69) and ≥15 ng/mL for 87.5% of raltegravir free fraction samples (63/72) (Figure 1). The between- and within-patient variability in total Cmin was 105% and 35% for maraviroc and 117% and 61% for raltegravir, respectively. The metabolic ratio of glucuronide-raltegravir to raltegravir was ∼4.1 (2.3–6.4, n = 78). Box plots of free and total trough plasma concentration (Cmin) for pooled samples of maraviroc (MVC; white) and raltegravir (RAL; dark grey) at weeks 4, 24 and 48 after treatment. Horizontal lines in boxes are medians, top and bottom edges are IQRs and whiskers are ranges (10%–90%). Circles represent values outside the ranges. This pharmacokinetic study of the ROCnRAL trial confirmed the lack of a clinically relevant drug interaction between maraviroc and raltegravir in a population of mostly male HIV-infected patients with long-term suppressed viraemia and lipohypertrophy. Indeed, neither maraviroc, a CCR5 antagonist, nor raltegravir, an integrase inhibitor, demonstrated similar metabolic pathways. Maraviroc is a substrate of CYP3A4 whereas raltegravir is a substrate of UDP-glucuronyl transferase 1A1 (UGT1A1); both are substrates of P-glycoprotein.3,4 Our total Cmin values for maraviroc and raltegravir are consistent with those for healthy subjects9 and our free Cmin values were similar to those previously described (raltegravir, Isentress product information, US FDA).10 The Cmin for most maraviroc and raltegravir samples was considered effective considering the respective thresholds of 10 and 15 ng/mL. At each visit, patients were advised to take maraviroc plus raltegravir with food, which probably limited the interpatient variability in Cmin for both drugs. In this population of HIV-infected patients receiving the maraviroc plus raltegravir dual regimen, without any liver disease or drug interaction via UGT1A1, the Cmin was 4-fold higher for the glucuronide-raltegravir metabolite than raltegravir. In conclusion, this pharmacokinetic study of a maraviroc-plus-raltegravir dual regimen in HIV patients with long-term virological suppression confirms the lack of deleterious drug interaction between maraviroc and raltegravir. Despite an effective maraviroc and raltegravir Cmin, the lack of virological robustness should be investigated to detect some minor X4-tropic variants or integrase resistance mutations at baseline or other explanation for the treatment failure with maraviroc plus raltegravir. This work was supported by the Agence Nationale de Recherche sur le SIDA et les Hépatites Virales (ANRS), France, as well as ViiV Healthcare (France) and Merck Sharp & Dohme (France), who provided maraviroc and raltegravir, respectively, for this trial. M.-A.