icon star paper   Articles  
Back grey_arrow_rt.gif
Coadministration of Milk Thistle and Indinavir in Healthy Subjects
  Robert DiCenzo, Pharm.D., Mark Shelton, Pharm.D., Kelly Jordan, Pharm.D., Christine Koval, M.D., Alan Forrest, Pharm.D., Richard Reichman, M.D., Gene Morse, Pharm.D.
Pharmacotherapy 23(7):866-870, 2003. \0xA9 2003 Pharmacotherapy Publications
Posted 08/06/2003
Abstract and Introduction
Study Objective: To determine if milk thistle (silymarin) alters the pharmacokinetics of indinavir. Design: Sequential crossover trial. Setting: General clinical research center. Subjects: Ten healthy subjects. Intervention: Indinavir 800 mg 3 times/day was given for four doses on days 1 and 2. Silymarin 160 mg 3 times/day was given on days 3-15. On day 16 and for one dose on day 17, both drugs were given at the same dosages.
Measurements and Main Results: Indinavir's pharmacokinetic parameters were evaluated at steady state both before and after administration of 14 days of silymarin. Blood samples were collected -0.25, 0.5, 1, 2, 3, 4, and 5 hours after indinavir dosing and assayed by high-performance liquid chromatography. The final pharmacokinetic model had first-order absorption after a lag time, and two compartments with first-order elimination from the central compartment. When given alone and combined with silymarin, respectively, the geometric mean (95% confidence interval [CI]) steady-state indinavir area under the plasma concentration-time curve was 20.7 hrmg/L (15.3-28.2 hrmg/L) and 19.4 hrmg/L (15.8-23.6 hrmg/L) and the trough plasma concentration was 0.340 mg/L (0.232-0.497 mg/L) and 0.232 mg/L (0.129-0.419 mg/L). Conclusion: Silymarin has no apparent effect on indinavir plasma concentrations.
As complementary alternative medicine (CAM) is practiced increasingly among adults infected with human immunodeficiency virus (HIV), it is ever more necessary to know if CAM will influence antiretroviral therapy. A recent survey of 118 HIV-infected patients receiving highly active antiretroviral therapy (HAART) showed that 38% of respondents were taking CAM. Of particular concern is that physicians were unaware of this in 67% of patients taking herbal products.[1]
Approximately one third of HIV infected patients are coinfected with hepatitis C virus (HCV). Hepatitis C infection can progress to liver failure, and this progression may be especially rapid in coinfected patients.[2] Milk thistle, an herbal supplement, is purported to decrease the risk of developing liver failure; therefore, a number of HIV-infected patients who are also infected with HCV are interested in taking the supplement.[3]
The active component of milk thistle is thought to be silymarin. Silymarin has the potential to influence drug metabolism by decreasing the metabolic activity of cytochrome P450 (CYP) 3A4, a ubiquitous enzyme responsible for hepatic and intestinal metabolism of many antiretroviral agents, including indinavir.[4, 5] Silymarin also may alter drug absorption, distribution, and elimination through inhibition of P-glycoprotein (P-gp).[6] P-glyco-protein is an active drug transporter, has a wide substrate range, and is abundant in the apical membrane of many pharmacologically important barriers such as intestinal epithelium, and the blood-brain, blood-nerve, blood-testis, and maternofetal barriers.[7] Flow cytometric analysis confirmed significant P-gp expression in both CD4+ and CD8+ cells; therefore, in addition to playing a role in limiting drug bioavailability, P-gp may limit brain, testis, fetal, and intracellular penetration of drugs.[8]
Indinavir belongs to the protease inhibitor class of antiretroviral drugs and is an effective component of HAART. It is also a substrate of both CYP450 and P-gp.[5, 7, 9] The purpose of this trial was to determine if silymarin influences the pharmacokinetics of indinavir.
This sequential, crossover trial was conducted in healthy men and women not of childbearing potential (bilateral tubal ligation, hysterectomy, menopause) between 18 and 55 years of age who had a negative HIV test, normal clinical laboratory blood tests, and normal physical examination. Exclusion criteria were smoking in the past year, ingestion of milk thistle within 30 days, use of agents known to influence CYP450, use of herbal supplements, and active or history of liver disease.
All outpatient and inpatient procedures were performed in the General Clinical Research Center (GCRC) at the University of Rochester Medical Center. Subjects received indinavir (Crixivan; Merck & Co., Inc., Whitehouse Station, NJ) 800 mg at 8 A.M., 4 P.M., and midnight on day 1. The first dose was administered in the GCRC and subjects reported back to the center that evening for the first of two overnight stays. On day 2 subjects received indinavir 800 mg in the fasted state (no food within 8 hours) at 8 A.M.; plasma samples were drawn -0.25, 0.5, 1, 2, 3, 4, and 5 hours after this dose. Subjects received a standard breakfast consisting of 4 ounces of skim milk, 3/4 ounce of corn flakes, and 4 g of sugar 1 hour after the morning dose and were discharged after the last blood draw. They were instructed not to drink alcohol or citrus beverages on days 1, 2, 16, and 17. On day 3 subjects began taking silymarin 160 mg (standardized milk thistle; General Nutrition Corp., Pittsburgh, PA) 3 times/day. On day 16 they began taking indinavir with silymarin in the same dosages, and returned to the GCRC in the evening for their second overnight stay. On day 17 subjects took their last doses of silymarin 160 mg and indinavir 800 mg. Plasma sampling times and meals were identical to those for day 2. After the last blood draw on day 17 subjects were discharged from the GCRC. They returned to the center between days 42 and 56 for a physical examination, when blood was drawn for clinical laboratory tests and an HIV test.
Each blood sample was centrifuged for 10 minutes at 800 x g. Plasma was separated into three aliquots and stored at -70\0xA1C until shipped to the University at Buffalo Adult AIDS Clinical Trials Group Pharmacology Support Laboratory, Buffalo, NY. Indinavir plasma concentrations were quantified by a validated high-performance liquid chromatography (HPLC) assay with ultraviolet detection. Indinavir interassay coefficients of variation (CVs) were 3.2% at 75 ng/ml and 2.8% at 3500 ng/ml, and intraassay CVs were 1.7-8.5% at 75 ng/ml and 0.3-3.4% at 3500 ng/ml. Lower limit of quantification for the assay was 12.5 ng/ml.
Before subject enrollment, 10 milk thistle capsules were assayed for silymarin content by a validated HPLC method (Research Triangle Laboratories, Inc., Raleigh, NC). The amount of silymarin was 173 mg/capsule, and all milk thistle dispensed during this study was from the same lot number.
Indinavir plasma concentrations were first modeled in Adapt II using maximum likelihood.[10, 11] For all modeling methods, observed data were weighted by the inverse of the fitted variance. The variance model assumed a linear relationship between standard deviation and fitted concentration. Model discrimination was performed using Akaike's information criterion. Once the structural model was developed, final pharmacokinetic estimates were calculated by a MAP Bayesian approach using iterative two-stage analysis. Indinavir concentrations 8 hours after dose administration (Cmin) were computed as the fitted Cmin 8 hours after the dose. Maximum concentration during the dosing interval (Cmax) was determined by visual inspection. Ten subjects were required to have 80% power to detect a 50% difference in indinavir area under the plasma concentration-time curve (AUC0-8) using a two-sided paired t test with an a error of 0.05 assuming a 40% CV.[12] Confidence intervals (CIs) were constructed on results of analyses of variance of each pharmacokinetic parameter or its log transform. Bioequivalence was defined as the 90% CI of the geometric mean ratio of AUC0-8 or Cmax being within Food and Drug Administration (FDA) guidelines (0.80-1.25). All statistical analyses were conducted using SAS system, version 8 (SAS Institute, Cary, NC).
Ten subjects (seven men) completed the trial. Mean \0xB1 SD age and weight were 32.5 \0xB1 10.0 years and 83.6 \0xB1 20.1 kg, respectively. All subjects were serum negative for HIV-1 and HIV-2 antibodies, and all women had negative serum pregnancy tests both before and after receiving study agents. Indinavir and milk thistle were well tolerated. One subject had elevated fasting blood glucose levels on both pharmacokinetic sampling days; however, this subject's fasting blood glucose concentrations were elevated before and after receiving study drug. One subject had an elevated alanine aminotransferase level before receiving study drug that decreased to normal by the end of the study. No other laboratory tests showed clinically significant deviation from normal. One subject reported very mild gastric reflux and another reported an episode of malaise while receiving indinavir. An episode of mild nausea and mild headache was reported by one subject receiving both indinavir and milk thistle. No adverse events were reported while receiving milk thistle alone. One subject took hydrochlorothiazide, one took estradiol and medroxyprogesterone, and three took multivitamins throughout the study.
The final pharmacokinetic model had two compartments, first-order absorption following a lag time, and first-order elimination from the central compartment. Individual pharmacokinetic parameters are shown in Table 1. Indinavir geometric mean (95% CI) steady-state AUC0-8 when given alone and with silymarin was 20.7 hrmg/L (15.3-28.2 hrmg/L) and 19.4 hrmg/L (15.8-23.6 hrmg/L), respectively (Figure 1). The difference [95% CI] between the geometric mean AUC0-8 for indinavir given alone and with silymarin (1.38 hrmg/L [0.79-1.48 hrmg/L]) was not significant (p=0.64), and changes in AUC0-8 did not follow an apparent trend (Figure 2). The geometric mean [95% CI] difference in Cmax (1.0 mg/L [0.88-1.4 mg/L]) also was not significant (p=0.30).
Neither the geometric mean ratio [90% CI] of indinavir AUC0-8 (0.93 hrmg/L [0.72-1.2 hrmg/L]) nor the geometric mean ratio [90% CI] of indinavir Cmax (0.89 mg/L [0.72-1.1 mg/L]) met the FDA definition of bioequivalence. This was most likely due to small sample size and intrasubject and intersubject variability in exposure, as this study was not powered to determine bioequivalence.
Milk thistle did not significantly influence indinavir trough concentrations. Geometric mean (95% CI) indinavir Cmins when given alone or with silymarin were 0.340 mg/L (0.232-0.497 mg/L) and 0.232 mg/L (0.129-0.419 mg/L), respectively. When comparing indinavir administered alone versus with silymarin, indinavir geometric mean [95% CI] apparent oral volume of distribution (54.1 mg/L [27.9-104.6 mg/L] vs 58.6 mg/L [40.4-85.6 mg/L]) and half-life (3.1 hrs [1.2-8.5 hrs] vs 3.1 hrs [1.3-7.2 hrs]) were not significantly different.
Milk thistle extract is a common herbal supplement. Milk thistle is a biennial herb that consists of a mixture of flavonolignans that are present in the fruit, seeds, and leaves of the plant. Flavonolignans are produced in plants through coupling of a flavonoid and a phenylpropanoid. Silymarin, a mixture of flavonolignans (silybin, silydianin, silychristin), is primarily composed of silybin, and typical extracts of milk thistle contain 70-80% silymarin.[3] Milk thistle is considered a hepatoprotectant, and purported to be active against acute or chronic viral hepatitis, toxin- and drug-induced hepatitis and cirrhosis, and alcoholic liver disease. Silymarin is reported to inhibit nitric oxide production, increase levels of glutathione in liver and intestines, scavenge free radicals, prevent lipid peroxidation, and stimulate ribosomal RNA polymerase and subsequent protein synthesis.[3, 13-15]
Preliminary data suggest that silymarin may influence the metabolic capacity of CYP3A4, a CYP isoenzyme responsible for hepatic and intestinal metabolism of many important classes of drugs.[4, 5] Most protease inhibitors, including indinavir, are primarily metabolized by CYP3A4. In vitro it is both an inhibitor and substrate for P-gp, an active drug transporter of the adenosine triphosphate-binding cassette transporter family. Both infection of T cells with HIV-1 and administration of zidovudine to HIV-infected T cells resulted in elevated expression of P-gp.[16, 17] Since P-gp limits the bioavailability and the brain, testis, intracellular, and fetal penetration of drugs, effective inhibition of P-gp may increase the concentration of indinavir in blood and possibly in viral sanctuary sites. The importance of investigating potential herbal-antiretroviral drug interactions is further highlighted by trials in which St. John's wort significantly decreased indinavir plasma concentrations, and garlic significantly decreased plasma concentrations of the protease inhibitor saquinavir in healthy adults.[18, 19]
Administration of silymarin failed to influence either indinavir AUC0-8 or Cmax; therefore, at the dosage given in this study it does not appear to have a significant influence on CYP3A4-mediated hepatic or intestinal metabolism of indinavir. With the same indinavir dosage, others also failed to detect an influence of silymarin on the pharmacokinetics of indinavir in healthy adults.[20] Although neither the geometric mean ratio [90% CI] of AUC0-8 (0.93 mg/L [0.72-1.2 mg/L]) nor Cmax (0.89 mg/L [0.72-1.1 mg/L]) met the definition of bioequivalence, this study was not powered to test for bioequivalence. We also saw no apparent trend in the intersubject variability of AUC0-8 regardless of silymarin coadministration; therefore, the most likely reasons for failing to claim bioequivalence appear to be small sample size and intrasubject and intersubject variability in indinavir exposure.
The Cmin or trough value of a protease inhibitor such as indinavir may be an important marker for clinical outcome. Interest in this relationship is highlighted by the mounting body of evidence suggesting that Cmin or ratio of Cmin to viral genotypic, phenotypic, or virtual phenotypic results may be an important predictor of clinical outcome.[21-23] Milk thistle administration failed to influence indinavir Cmin.
Limitations of this trial include small sample size, no evaluation of the effect of higher dosages of silymarin on indinavir pharmacokinetics, omission of assaying silymarin blood concentrations, and the fact that indinavir is now commonly administered with drugs that inhibit its metabolism so as to provide a pharmacokinetic boosting effect. Although the trial had only 10 subjects, pharmacokinetic parameters were almost identical between the agents, and it is doubtful that even a large trial would find a clinically relevant difference.
The dosage we administered approximates the recommended hepatoprotectant dosage of silymarin at the time this trial was designed (480 mg/day) and falls within the range of common milk thistle extract dosages (300-600 mg/day).[3, 24] However, we cannot conclude that silymarin would not have influenced the pharmacokinetics of indinavir at higher dosages. Milk thistle is poorly absorbed after oral administration (bioavailability 23-47%); therefore, silymarin blood concentrations may have been too low to achieve an appreciable effect on either CYP3A4 or P-gp.[25] Although the silymarin content in the capsules was determined before administration, in the absence of silymarin blood samples we cannot rule out the possibility that inadequate silymarin absorption led to the observed absence of effect. Furthermore, protease inhibitors are commonly administered with drugs that alter CYP3A4-dependent metabolism; therefore, we cannot rule out the possibility that milk thistle may alter the pharmacokinetics of indinavir indirectly by influencing the pharmacokinetics of concomitant drugs.
Finally, although this trial failed to show an influence of silymarin on the pharmacokinetics of indinavir, it was not designed to examine P-gp-dependent alterations in the distribution of indinavir to potential viral sanctuary sites. More study is necessary before concluding that an herbal supplement with P-gp-inhibitory properties, such as milk thistle, will not increase intracellular or potential viral sanctuary site concentrations of indinavir.
In summary, silymarin 160 mg 3 times/day failed to influence the pharmacokinetics of indinavir in healthy subjects. Concomitant administration of milk thistle at this dosage would not be expected to alter indinavir AUC, Cmax, or Cmin to clinically significant extents.

Data are geometric mean (95% CI). AUC = area under the plasma concentration-time curve; Cmax = maximum plasma concentration after dose administration; Cmin, minimum plasma concentration after dose administration.
  icon paper stack View Older Articles   Back to Top   www.natap.org