Perrine Courlet1, Monia Guidi1,2,3, Susana Alves Saldanha1, Felix Stader4,5, Anna-Katrin Traytel6,7, Matthias Cavassini8, Marcel Stoeckle4, Thierry Buclin1, Catia Marzolini4,5, Laurent A. Decosterd1, Chantal Csajka2,3,9 and the Swiss HIV Cohort Study.
1Service of Clinical Pharmacology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland 2Center for Research and Innovation in Clinical Pharmaceutical Sciences, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland 3Institute of Pharmaceutical Sciences of Western Switzerland, University of Geneva, University of Lausanne, Geneva & Lausanne, Switzerland 4Division of Infectious Diseases and Hospital Epidemiology, University Hospital of Basel, Switzerland 5University of Basel, Basel, Switzerland 6Division of Infectious Diseases and Hospital Epidemiology, University Hospital of Zurich, University of Zurich, Zurich, Switzerland. 7Institute of Medical Virology, University of Zurich, Zurich, Switzerland 8Service of Infectious Diseases, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland 9School of Pharmaceutical Sciences, University of Geneva, Geneva, Switzerland
Objectives: The prevalence of dyslipidemia in people living with HIV (PLWH) is increasing because of traditional risk factors, HIV infection and antiretroviral (ARV) therapy. Rosuvastatin is a widely prescribed lipid-lowering agent which is actively transported into the liver. Drug-drug interactions (DDIs) with ARV agents such as protease inhibitors (PIs) can occur via inhibition of rosuvastatin uptake in the liver, thereby potentially impairing the lipid response. The aims of this study were to characterize the pharmacokinetic (PK) profile of rosuvastatin, to quantify the magnitude of DDIs with ARV drugs, and to describe the relationship between rosuvastatin PK and non-HDL-cholesterol levels in PLWH.
Methods: Data were collected in PLWH from the Swiss HIV Cohort Study (SHCS), who were involved in rich (NCT03515772) and sparse sampling studies (SHCS #815). Total- and HDL-cholesterol (high density lipoprotein) before initiation of rosuvastatin treatment were retrieved from the SHCS database. Since PLWH have often higher than normal triglycerides values, low-density lipoprotein (LDL) values cannot be reliably derived using the Friedewald formula. Therefore, non-HDL-cholesterol levels were calculated by subtracting HDL-cholesterol from total cholesterol level, and were used to characterise the response to rosuvastatin treatment.
Analyses were performed in NONMEM with log-transformed rosuvastatin plasma concentrations, firstly using full PK profiles, and subsequently adding sparse data. The influence of age, body weight, creatinine clearance, aspartate (AST) and alanine (ALT) aminotransferases, sex, and boosted PIs on rosuvastatin PK was evaluated. The final PK model was combined with an indirect effect model to describe non-HDL cholesterol data. The non-HDL-cholesterol compartment was initialized with a baseline level, and an elimination rate (Kout) was defined as Kin/baseline, where Kin denotes the endogenous production rate which is inhibited by rosuvastatin. The influence of age, body weight, AST, ALT, diabetes, ARV treatment and presence of additional lipid-lowering agents was tested on baseline parameter of the PD model. Non-HDL-cholesterol levels were compared to the target value of 2.8 mmol/L [1].
Results: The PK analysis was performed first by using 65 rosuvastatin plasma concentrations collected in six PLWH enrolled in the study with rich sampling, before adding 89 plasma concentrations collected in 62 PLWH in the sparse sampling study. A two-compartment model with mixed first- and zero-order absorption best described rosuvastatin PK. In the complete PK analysis, the absorption rate constant and duration of zero-order absorption were fixed at 0.306 h-1 and 0.461 h, respectively, values estimated during the analysis of rich data. Typical values of rosuvastatin clearance, central and peripheral volumes of distribution, and inter-compartmental clearance were 122 L/h (between subject variability, BSV 50%) 153 L (97%), 1660 L and 67L/h, respectively. In univariate analyses, rosuvastatin central volume of distribution was significantly increased by 200% in women and decreased by 65% in PLWH receiving boosted protease inhibitors. However, no covariate was retained in the final PK model due to a lack of statistical significance during the multivariate analysis.
A total of 403 non-HDL-cholesterol values were included in the PK/PD model (253 and 150 values before and after rosuvastatin initiation, respectively). Baseline was estimated at 3.96 mmol/L (BSV 22%), Kin was estimated at 0.021 mmol/L/h, and IC50 (rosuvastatin concentration that produced a 50% inhibition of non-HDL-cholesterol production) at 9.5 ng/mL (83%). In the final model, baseline was increased by 12% in case of coadministration of ARV drugs with negative impact on lipids (i.e. PIs, efavirenz, cobicistat) and decreased by 14% in case of etravirine coadministration. The baseline value was surprisingly decreased by 43% between PLWH aged 40 to 80 years-old. Model-based simulations revealed that, under standard rosuvastatin doses of 5 mg and 20 mg once daily, 31% and 64% of PLWH might achieve non-HDL-cholesterol target, respectively.
Conclusions: High between-subject variability characterizes both rosuvastatin PK and PD profiles, and remains unexplained after inclusion of covariates. Demonstrating its limited potential for DDIs with ARV agents and its potent lipid-lowering effect, our results emphasize the safe and effective prescription of rosuvastatin in PLWH.
References:
[1] European AIDS Clinical Society. Guidelines version 10.0. 2019 [Available from: https://www.eacsociety.org/files/2019_guidelines-10.0_final.pdf].
Reference: PAGE () Abstr 9354 [www.page-meeting.org/?abstract=9354]
Poster: Drug/Disease Modelling - Infection