Bilirubin - a biomarker of atazanavir exposure in HIV/AIDS patients
D. Rekić (1), O. Clewe (1), D. Röshammar (2), L. Flamholc (3), A. Sönnerborg (4), V. Ormaasen (5), M. Gisslén (6), A. Äbelö (1), M. Ashton (1)
(1) Unit for Pharmacokinetics and Drug Metabolism, Department of Pharmacology, Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden. (2) AstraZeneca R&D Mölndal, Sweden. (3) Department of Infectious Diseases, Malmö University Hospital, Malmö, Sweden. (4) Department of Infectious Diseases, Institution of Medicine, Karolinska University Hospital and Karolinska Institet, Stockholm, Sweden. (5) Department of Infectious Diseases, Ullevål University Hospital, Oslo, Norway. (6) Department of Infectious Diseases, Sahlgrenska University Hospital, University of Gothenburg, Sweden
Objectives: The protease inhibitor (PI) atazanavir is currently recommended for the first line treatment of HIV-1 infected patients. Although elevated bilirubin levels are commonly observed in patients on a atazanavir treatment it is an uncommon cause of treatment discontinuation. Several studies have shown a concentration dependent increase in bilirubin levels (1). This study aimed to quantify the relationship between bilirubin and atazanavir through the use of modelling and investigate the possible uses of bilirubin as a biomarker of atazanavir exposure.
Methods: The population pharmacokinetic/pharmacodynamic analysis was performed using NONMEM VI based on atazanavir (n=200) and bilirubin (n=361) plasma concentrations of the atazanavir arm (n=82) in the NORTHIV trial (2). A one compartment model with fixed first order absorption and lag-time was fitted to atazanavir concentrations. Individually predicted plasma concentrations were used to drive the bilirubin time course in each individual. An indirect response model adequately described the atazanavir-bilirubin relationship. Simulations of the typical individual were performed to assess the influence of non-adherence and suboptimal exposure on bilirubin levels.
Results: Oral clearance and volume of distribution were estimated at 6.47 L/h and 93.6 L, respectively. The between subject variability was estimated at 53% for clearance and at 43.8% for volume of distribution. Bilirubin baseline was estimated to 7.69 μmol/L while the fractional turnover rate (Kout) was estimated to 0.423 h-1. Atazanavir was estimated to inhibit bilirubin elimination with an Imax of 0.91 and an IC50 concentration of 0.303 μmol/L. The between subject variability of the baseline was estimated to 32%. The half-life of uninhibited bilirubin was estimated to 1.64 h, in fair agreement with a literature beta-phase value of 1.16 h (3). Atazanavir inhibition resulted in a seven-fold increase in bilirubin half-life.
Conclusions: The atazanavir-bilirubin relationship was well described by the proposed model. PK parameter where in agreement with previously reported values (4). Based on simulations, bilirubin was a good indicator of non-adherence and suboptimal exposure in patients.
 Karlström, O., Josephson, F. & Sönnerborg, A. Early virologic rebound in a pilot trial of ritonavir-boosted atazanavir as maintenance monotherapy. JAIDS Journal of Acquired Immune Deficiency Syndromes 44, 417 (2007).
 Josephson, F. et al. The relation between treatment outcome and efavirenz, atazanavir or lopinavir exposure in the NORTHIV trial of treatment-naïve HIV-1 infected patients. European journal of clinical pharmacology 66, 349-57 (2010).
 Berk, P.D., Howe, R.B., Bloomer, J.R. & Berlin, N.I. Studies of bilirubin kinetics in normal adults. Journal of Clinical Investigation 48, 2176 (1969).
 Dickinson, L. et al. Population pharmacokinetics of ritonavir-boosted atazanavir in HIV-infected patients and healthy volunteers. Journal of Antimicrobial Chemotherapy, (2009).