Dyslipidemia and coronary heart disease (CHD) are increasingly recognized in persons with HIV infection.1,2 Many antiretrovirals, including efavirenz (EFV), are associated with increases in serum lipids often requiring lipid-lowering therapy.3-5 EFV, a nonnucleoside reverse transcriptase inhibitor (NNRTI), is commonly used because of its proven durable efficacy and tolerability. EFV is also a mixed inducer/inhibitor of cytochrome P450 (CYP) 3A4, and many hydroxymethylglutaryl coenzyme A (HMG-CoA) reductase inhibitors are primarily metabolized via CYP3A4.6 Most drug interaction studies done in vivo have demonstrated that EFV is a potent inducer of CYP3A4 and CYP2B6.7
Protease inhibitors (PIs) that are known to influence CYP3A4 activity have varied effects on the metabolism of HMG-CoA reductase inhibitors, including the potential for clinically important adverse effects attributable to CYP3A4 inhibition of simvastatin (SIM) metabolism.8-14 There are no published data on drug-drug interactions between NNRTIs and HMG-CoA reductase inhibitors. Because it is likely that the patients using antiretroviral regimens containing EFV may require concomitant lipid-lowering therapy, it is important to establish the safe and effective use of statins with EFV. Therefore, we evaluated the effect of EFV on the pharmacokinetics and short-term lipid-lowering effects of 3 HMG-CoA reductase inhibitors in HIV-seronegative healthy volunteers.
AIDS Clinical Trials Group (ACTG) A5108 was a phase 1 open-label study examining the effect of EFV on the pharmacokinetics of SIM, atorvastatin (ATR), and pravastatin (PRA). The study was performed in healthy adult HIV-seronegative volunteers to avoid potential confounders from comorbidities and HIV management medications and to avoid the development of antiretroviral drug resistance. The primary objective was to examine the effect of EFV on the pharmacokinetics of statins. Secondary objectives were to investigate the effect of statins on the non-steady-state pharmacokinetics of EFV and non-steady-state changes in serum low-density lipoprotein (LDL) cholesterol secondary to statins.
Subjects were accrued into the 3 study arms between November 2001 and May 2002. Subjects had no underlying illnesses, were not taking chronic medications, and had baseline laboratory values within normal limits (fasting triglyceride levels <400 mg/dL). Women of reproductive potential were excluded. Subjects who did not complete the 3 pharmacokinetic evaluations were replaced to achieve the required sample size.
The study consisted of 3 treatment arms: arm A (40 mg of SIM daily), arm B (10 mg of ATR daily), and arm C (40 mg of PRA daily) (Fig. 1). Medications were taken in the evening as is recommended for EFV and statins. For the first 3 days, the subjects self-administered statins, and on the fourth day, the statin was administered under directly observed conditions. Blood samples were obtained before drug administration as well as 30 minutes and 1, 2, 3, 4, 6, 8, 12, and 24 hours after statin administration for pharmacokinetic assay. After the 24-hour blood draw, the subjects were administered 600 mg of EFV and were instructed to take 600 mg of EFV at bedtime for the next 9 days without the statin. The 11th dose of EFV was administered under directly observed conditions, and samples were collected over 24 hours as described previously. Subsequently, subjects resumed statin dosing and were asked to take EFV and statin together for the next 2 evenings. For the final day, both drugs were administered under directly observed conditions and samples were again collected over 24 hours. Adherence to study medications was measured using pill counts and questionnaires. Subjects were evaluated at each study visit for adverse events. The protocol required subjects experiencing any grade 2 or higher signs, symptoms, or laboratory abnormalities to terminate study drugs. The protocol was approved by the local institutional review boards of participating institutions. Informed consent was obtained from all subjects. The study was approved by the scientific committees of the Adult AIDS Clinical Trials Group (AACTG) and the National Institute of Allergy and Infectious Diseases (NIAID) Division of AIDS Clinical Science Review Committee.
SIM and SIM acid plasma concentrations were assayed at Merck and Co. (West Point, PA) using liquid chromatography/tandem mass spectrometry (LC/MS/MS).15 Quantification of plasma concentrations of HMG-CoA reductase inhibitors was also performed using an enzyme inhibition assay. This assay measures plasma levels of active and total inhibitors, which are generated by base hydrolysis.16
ATR and 2 active metabolites, 2-hydroxy ATR and 4-hydroxy ATR, were assayed in plasma by turbo ion spray LC/MS/MS in the positive ion mode at Advion BioSciences, Inc. (Ithaca, NY).17 Unchanged ATR and total active ATR (derived from adding concentrations of ATR and the 2 active metabolites) are presented.
PRA plasma concentrations were analyzed by turbo ion LC/MS/MS in the positive ion mode at Advion BioSciences, Inc.18 EFV plasma assays were performed at the University of Alabama Pharmacology Support Laboratory of the ACTG.19
Serum total cholesterol, triglycerides, and high-density lipoprotein (HDL) concentrations were measured at local site laboratories using standard enzymatic assays. Calculated LDL was determined using the Friedewald equation.
Calculation of Area Under the Curve
Systemic exposure to the statins and EFV was quantified by calculating the area under the curve (AUC) of the drugs before dosing to the end of the dosing interval (0-24 hours). Concentrations below the lower limit of quantification (LLQ) were assigned a value of LLQ/2. Actual sample times were used. If the 0-hour concentration was unavailable and the 24-hour concentration was available, an imputed 0-hour concentration taking the value of the concentration at 24 hours (C24h) was used, and vice versa. AUCs were estimated according to the linear trapezoidal rule, without extrapolation beyond the dosing interval.
The pharmacokinetic parameters tested were AUC, maximum concentration (Cmax), and minimum concentration (Cmin). Means, standard deviations (SDs), coefficients of variation (CVs), and medians and their 95% confidence intervals (CIs)20 were reported. Median within-subject differences (eg, day 18 − day 3) as a percent of the earlier value (eg, day 3) were also calculated, as were 90% CIs around geometric means of within-subject ratios (GMRs), for comparison with no-effect boundaries of 80% to 125% as recommended by the US Food and Drug Administration (FDA) when evaluating bioequivalence.21,22
To test the null hypothesis of no difference in statin exposure before versus after initiation of EFV (and in EFV exposure with and without statin coadministration), the Wilcoxon signed rank test was applied to within-subject differences in AUC and Cmax values. Reported P values are 2-sided and not adjusted for multiple comparisons.
Fixing type I and II error rates at 5% and 10%, respectively, assuming within-subject CVs of 37% for SIM and ATR AUCs and assuming the use of the 2-sided paired t test for log AUCs, 12 subjects per arm provided 90% power to detect decreases (increases) of 38% (60%) or larger attributable to EFV in SIM and ATR AUCs. Assuming a within-subject CV for PRA AUCs of 31%, a sample size of 12 provided 90% power to detect a decrease (increase) in PRA AUCs of 33% (50%) or larger attributable to EFV. To adjust for the lower efficiency of the Wilcoxon test relative to the paired t test and to account for missing results, 2 additional subjects per arm were added, bringing accrual targets to 14 per arm.
Serum lipid levels were obtained under fasting conditions (≥8 hours from last ingestion) before dosing on days 0, 3, 15, and 19 (before statin administration and after 4 doses of statins and before and after EFV initiation). The Kruskal-Wallis (KW) test was used to compare baseline LDL across arms. The Wilcoxon signed rank test was used (1) to compare statin-associated LDL changes without EFV (day 4 − day 0) and with EFV (day 19 − day 15) each with the hypothesized value of 0, (2) to compare statin-associated LDL changes without EFV (day 4 − day 0) and with EFV (day 19 − day 15) with each other, and (3) to compare the first (day 0) and second (day 15) LDL baseline values (compare the difference from 0). No adjustments were made for multiple comparisons.
A total of 52 subjects were enrolled. Three subjects in the SIM arm exhibited study-defined toxicities: 1 experienced a grade 2 platelet count elevation, another had grade 2 back pain, and the third had grade 2 diarrhea and insomnia. One subject on the ATR arm developed grade 2 creatine kinase elevation. No subjects on the PRA arm exhibited study-defined toxicities. No subjects developed alanine aminotransferase (ALT) or aspartate aminotransferase (AST) elevations during the study.
Of the 52 subjects who enrolled, 42 (81%) were eligible for statistical analysis of pharmacokinetic data. Of the 10 subjects ineligible for pharmacokinetic analysis, 4 were ineligible because of improper specimen collection, 2 discontinued early because of on-study toxicities, 2 discontinued early because of baseline toxicities not available until after the first dose of the study drug, and 2 were noncompliant with dosing and/or visits. Each arm accrued 14 subjects who were eligible for the pharmacokinetic analysis; however, samples for 1 subject were unavailable, leaving 13 subjects on the PRA arm. Most subjects reported that they took 100% of their prescribed doses; high adherence was confirmed by pill counts.
Among the 42 eligible subjects, median ages were 38, 32, and 25 years, respectively, for the SIM, ATR, and PRA arms. Most subjects were male (86%, 93%, and 100%, respectively); 64% were white, and 21% were black.
Median AUC decreases of the statins with EFV coadministration are shown in Figure 2; summary and inferential statistics are presented in Table 1. Administration of EFV resulted in a statistically significant median decrease of 58.0% in SIM acid AUCs (36.48 ng·h/mL before and 14.46 ng · h/mL after EFV; Wilcoxon signed rank test, P = 0.003). In addition, the active HMG-CoA reductase plasma AUC in the SIM arm significantly decreased by a median of 60.2% (P < 0.001). Administration of EFV resulted in a significant median decrease of 42.7% in the ATR AUC (11.2 ng · h/mL before and 6.56 ng · h/mL after EFV; P < 0.001). In addition, the total active ATR AUC significantly decreased by a median of 34.5% (P = 0.005). Finally, administration of EFV resulted in a significant median decrease of 40.4% in the PRA AUC (96.32 ng · h/mL before and 42.65 ng · h/mL after EFV; P = 0.005). None of the statins affected the non-steady-state pharmacokinetics of EFV (data not shown).
Baseline LDL cholesterol did not differ across arms (KW test, P = 0.224); means/medians were 117.0, 105.0, and 106.5 mg/dL on arms A, B, and C, respectively. Changes in LDL cholesterol are summarized in Table 2. After short-term administration (4 doses, 3.5 days) of each statin alone, median LDL changes were −37.0, −29.0, and −23.5 mg/dL on SIM, ATR, and PRA, respectively. Median LDL decreases after 4 doses of statins administered with EFV were attenuated, although still statistically significant. The median LDL decrease secondary to SIM was 11 mg/dL smaller when the subject also took EFV (P = 0.048). The difference in LDL changes secondary to ATR with versus without EFV coadministration was not statistically significant. The median LDL decrease secondary to PRA was 6.5 mg/dL smaller when EFV was coadministered; however this difference exhibited only a trend toward statistical significance. Median decreases in LDL cholesterol attributable to the 3 statins with and without EFV are shown in Figure 3. In all 3 arms, differences between the true baseline LDL (day 0) and the second “baseline” LDL (day 15) were not statistically significant (not shown in Table 2).
This study demonstrates that EFV induces the metabolism of 3 of the most commonly used statins. Coadministration of EFV resulted in a median decrease of 58% in SIM acid concentration, suggesting a potential for a clinically important interaction. In fact, decreases in LDL cholesterol were significantly attenuated by coadministration of EFV with SIM. Similarly, EFV reduced the total active ATV concentrations by a median of 34% and the PRA concentrations by a median of 40%. Importantly, there were no significant short-term adverse effects associated with the concomitant use of these medications. The combined findings demonstrate that EFV is a strong inducer of metabolism of all 3 statins, 2 of which are substrates for CYP3A4 (SIM and ATR) and 1 of which has more a complex metabolism, with CYP3A4 playing only a minor role (PRA). There is no evidence of alteration in the metabolism of EFV by any of these statins.
The changes in serum LDL cholesterol did not reach steady state after only 4 doses of statins; thus, these data are only suggestive of a pharmacodynamic consequence of EFV induction of statin metabolism. The hierachical attenuation of an LDL cholesterol-lowering effect by EFV corresponded to the decrease in exposure to the specific statins (SIM > PRA > ATR), adding credence to the hypothesis that induction of statin metabolism has physiologic consequence. Studies of longer duration are needed to clarify the pharmacodynamic consequences of these drug-drug interactions at steady-state concentrations of all drugs, however.
We have previously demonstrated in AACTG A5047 that saquinavir plus ritonavir alters the pharmacokinetics of statins.8 These findings were confirmed with the PIs lopinavir/ritonavir and nelfinavir.14,23 Thus, we have extended our previous work to include NNRTIs. Similar interactions would be anticipated with nevirapine, an NNRTI also shown to induce CYP3A4.24 In contrast, delavirdine, an NNRTI that inhibits CYP3A4 metabolism, should have effects on statins similar to those of saquinavir/ritonavir.25
These findings were largely anticipated for SIM and ATR because of their known in vivo metabolism but were somewhat unexpected for PRA. SIM, a lactone prodrug of SIM acid, is the most lipophilic of the statins, and the drug is metabolized primarily by CYP3A4. Other inhibitors of CYP3A4 increase the formation of SIM acid with associated muscle toxicity.9,26 ATV is also a substrate of CYP3A4, but 2 of the metabolites formed are active and are partially responsible for the efficacy and toxicity of this drug.12 Thus, it was not surprising that EFV, an inducer of CYP3A4 metabolism, resulted in decreases in the levels of these 2 statins. PRA, in contrast, is hydrophilic, and systemic elimination does not use CYP3A4 oxidation to any great extent but rather uses multiple other oxidative and conjugative pathways.13 The mechanism of how EFV reduces concentrations of PRA is not known. EFV is not a known inducer of glucuronidation but may induce non-CYP3A4 oxidation.7 In addition, the pharmacokinetic disposition of PRA is dependent on organic anion transporters and is a substrate for OATP-C and MRP-2.27-29 It is unclear whether EFV induces these transporters to increase PRA hepatic elimination.
An even more complex question may be how to dose a statin when EFV is combined with a PI known to inhibit the metabolism of the statin. Three-way drug interactions are difficult to predict; thus, practitioners should err on the side of caution and start statins at a low dose and titrate upward until the goal serum LDL level is achieved without toxicity.
In conclusion, our data indicate that concomitant use of statins and EFV does not alter EFV exposure, and therefore presumably does not interfere with antiretroviral activity, but the significant induction of metabolism may require increased dosing of statins to achieve serum lipid concentration goals. Thus, clinicians need to be aware that the activity of these statins may be diminished when used in combination with EFV. Because the use of higher than recommended doses of statin has not been evaluated in clinical trials, this approach should only be used with increased surveillance for toxicity. Future studies are needed to determine the long-term effects of concomitant administration of EFV and statins as well as to demonstrate that increased doses of statins are safe and provide added benefit in persons with HIV infection.
The ACTG A5108 team thanks the volunteers for participating in this study and the study coordinators and sites for their hard work. Special thanks to W. J. Burning, M. J. Werder, E. Ferguson, M. Royal, R. DiFrancesco, S. Valle, J. Staggers, K. Ryan, R. Christensen, and M. Becker for their invaluable assistance. The team also thanks the University of Alabama Pharmacology Support Laboratory (E. Acosta) for running the EFV plasma concentrations. Participating ACTG A5108 sites and investigators are Stanford University (J. Norris and S. Valle), University of Minnesota (C. Fietzer and K. Fox), University of Cincinnati (T. Powell and D. Daria), Johns Hopkins University (I. Wiggins and D. Jones), Washington University (L. Kessels and M. Klebert), University of Colorado Health Sciences Center (J. Scott), University of Rochester Medical Center (J. Reid and R. Reichman), and University of Washington (J. Stekler and J. Conley).
1. Friis-Moller N, Sabin CA, Weber R, et al. Data Collection on Adverse Events of Anti-HIV Drugs (DAD) Study Group: combination antiretroviral therapy and the risk of myocardial infarction. N Engl J Med
2. Mauss S. HIV-associated and antiretroviral-induced hyperlipidaemia: an update. J HIV Ther
3. Fontas E, van Leth F, Sabin CA, et al. Lipid profiles in HIV-infected patients receiving combination antiretroviral therapy: are different antiretroviral drugs associated with different lipid profiles? J Infect Dis
4. Simon GL, Liappis AP, Granger SL, et al. Antiretroviral therapy-induced changes in lipoprotein subclass phenotype: comparison of protease inhibitor and non-nucleoside reverse transcriptase inhibitor regimens [abstract 14]. Antivir Ther
5. Tashima KT, Bausserman L, Alt EN, et al. Lipid changes in patients initiating efavirenz- and indinavir-based antiretroviral regimens. HIV Clin Trials
6. Lennernas H, Fager G. Pharmacodynamics and pharmacokinetics of the HMG-CoA reductase inhibitors. Clin Pharmacokinet
7. Adkins JC, Noble S. Efavirenz. Drugs
8. Fichtenbaum CJ, Gerber JG, Rosenkranz SL, et al. NIAID AIDS Clinical Trials Group. Pharmacokinetic interactions between protease inhibitors and statins in HIV seronegative volunteers: ACTG Study A5047. AIDS
9. Neuvonen PJ, Kantola T, Kivisto KT. Simvastatin but not pravastatin is very susceptible to interaction with the CYP3A4 inhibitor itraconazole. Clin Pharmacol Ther
10. Lees RS, Lees AM. Rhabdomyolysis from the coadministration of lovastatin and the antifungal agent itraconazole. N Engl J Med
11. Hare CB, Vu MP, Grunfeld C, et al. Simvastatin-nelfinavir interaction implicated in rhabdomyolysis and death. Clin Infect Dis
12. Malhotra HS, Goa KL. Atorvastatin. An updated review of its pharmacological properties and use in dyslipidaemia. Drugs
13. Everett DW, Chando TJ, Didonato GC, et al. Biotransformation of pravastatin sodium in humans. Drug Metab Dispos
14. Hsyu PH, Schultz-Smith MD, Lillibridge JH, et al. Pharmacokinetic interactions between nelfinavir and 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors atorvastatin and simvastatin. Antimicrob Agents Chemother
15. Zhao JJ, Xie IH, Yang AY, et al. Quantitation of simvastatin and its beta-hydroxy acid in human plasma by liquid-liquid cartridge extraction and liquid chromatography-tandem mass spectrometry. J Mass Spectrom
16. Rogers JD, Zhao J, Liu L, et al. Grapefruit juice has minimal effects on plasma concentrations of lovastatin-derived 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors. Clin Pharmacol Ther
17. Jemal M, Ouyang A, Chen B, et al. Quantitation of the acid and lactone forms of atorvastatin and its biotransformation products in human serum by high-performance liquid chromatography with electrospray tandem mass spectrometry. Rapid Commun Mass Spectrom
18. Mulvana D, Jemal M, Coates-Pulver S. Quantitative determination of pravastatin and its biotransformation products in human serum by turbo ion spray LC/MS/MS. J Pharm Biomed Anal
19. Turner ML, Reed-Walker K, King JR, et al. Simultaneous quantitation of protease inhibitors and non-nucleoside reverse transcriptase inhibitors in human plasma using a single high-performance liquid chromatography methodology. J Chromatogr B Biomed Appl
20. Hahn GJ, Meeker WQ. Statistical Intervals: A Guide for Practitioners
. New York: John Wiley and Sons; 1991.
21. FDA Guidance for Industry. In Vivo Drug Metabolism/Drug Interaction Studies-Study Design, Data Analysis and Recommendations for Dosing and Labeling
. US Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and Research (CBER); Rockville, MD. 1999.
22. FDA Guidance for Industry. Statistical Approaches to Establishing Bioequivalence
. US Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER); Rockville, MD. 2001.
23. Carr RA, Andre AK, Bertz RJ, et al. Concomitant administration of ABT-378/ritonavir (ABT-378/r) results in a clinically important pharmacokinetic (PK) interaction with atorvastatin (ATO) but not pravastatin (PRA) [abstract 1644]. Presented at: 40th Interscience Conference on Antimicrobial Agents and Chemotherapy; 2000; Toronto.
24. Back D, Gibbons S, Khoo S. Pharmacokinetic drug interactions with nevirapine. J Acquir Immune Defic Syndr
. 2003;34(Suppl 1):S8-S14.
25. Tran JQ, Gerber JG, Kerr BM. Delavirdine: clinical pharmacokinetics and drug interactions. Clin Pharmacokinet
26. Arnadottir M, Eriksson LO, Thysell H, et al. Plasma concentration profiles of simvastatin 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase inhibitor activity in kidney transplant recipients with and without ciclosporin. Nephron
27. Yamazaki M, Akiyama S, Ni'inuma K, et al. Biliary excretion of pravastatin in rats: contribution of the excretion pathway mediated by canalicular multispecific organic anion transporter (cMOAT). Drug Metab Dispos
28. Nakai D, Nakagomi R, Furuta Y, et al. Human liver-specific organic anion transporter, LST-1, mediates uptake of pravastatin by human hepatocytes. J Pharmacol Exp Ther
29. Mwinyi J, Johne A, Bauer S, et al. Evidence for inverse effects of OATP-C (SLC21A6) 5 and 1b haplotypes on pravastatin kinetics. Clin Pharmacol Ther
Keywords:© 2005 Lippincott Williams & Wilkins, Inc.
HIV infection; hydroxymethylglutaryl coenzyme A reductase inhibitors; statins; nonnucleoside reverse transcriptase inhibitors; drug interactions