Virological response to atazanavir, ritonavir and tenofovir/emtricitabine: relation to individual pharmacokinetic parameters and adherence measured by medication events monitoring system (MEMS) in naïve HIV-infected patients (ANRS134 trial)

Background: Individual pharmacokinetic parameters and treatment adherence are key factors to reach fast and sustained virological response. Assessment of adherence by MEMS recording exact times of drug intakes allows quantification of drug exposure.
Methods: A prospective study was conducted in 35 naïve patients. Atazanavir (300 mg), ritonavir (100 mg), and tenofovir+emtricitabine were given once daily during 6 months. All drugs were supplied in bottles with a MEMS cap. Blood samples for pharmacokinetic analysis were drawn at week-4 (W4), then bimonthly. Atazanavir and ritonavir were assayed by HPLC (UV detection). Compartmental pharmacokinetic parameters were calculated with a one compartment model, using MEMS records for dosing-history. Clinical and biological data were collected at screening and monthly. Relationships between pharmacokinetic, adherence recorded with MEMS, pill counts and questionnaires, and immuno-virological response are analyzed.
Results: Median HIV viral load (VL) and CD4 count at inclusion were 23,200 copies/mL (range, 100-457,000) and 436 cells/µL (197-573). Median maximal (Cmax) and trough concentrations (Ctrough) of atazanavir at W4 were 3,774 ng/mL (1,500-10,956), and 552 ng/mL (40-695). Average persistence and adherence at W24 were 97% and 96% respectively. 88% of doses were taken on time (±3h). Median CD4 count at W24 was 480 cells/µL. 5/35 patients had a detectable VL (47-152 cp/mL) at W24: four had initial VL >100,000 cp/mL but correct adherence and the last patient was poorly adherent despite low initial VL. Among the 30 virological responders, 5/6 with initial VL >100,000 cp/mL had optimal therapeutic exposure (based on concentrations/adherence) versus 13/24 patients in whom VL was < 100,000 cp/mL. One patient experienced a severe hyperbilirubinemia and one a transient cytolytic hepatitis. Pill counts and questionnaires were less likely to detect adherence failure than MEMS.
Conclusions: Adherence records by MEMS allow quantification of individual drug exposure and may explain immunovirological response taking into account patient characteristics at treatment initiation.

C. Goujard1,2, A. Barrail-Tran3, X. Duval4, G. Nembot5, X. Panhard5, R. Savic5, D. Descamps6, B. Vrijens7, A.-M. Taburet3, F. Mentré5,8, ANRS 134 Study Group
1Bicetre Hospital, AP-HP, Internal Medicine Unit, Paris, France, 2INSERM 1018-CESP, University Paris 11, Le Kremlin Bicetre Cedex, France, 3Bicetre Hospital, AP-HP, Pharmacy, Le Kremlin Bicetre Cedex, France, 4Bichat Hospital, AP-HP, Paris, France, 5INSERM UMR 738, Bichat Hospital, Paris, France, 6Bichat Hospital, AP-HP, Virology, Paris, France, 7AARDEX Group, Vise, Belgium, 8Paris Diderot University, AP-HP, Paris, France