 |
|
|
| |
8th International Workshop on Clinical Pharmacology of HIV Therapy
|
| |
| |
April 2007, Budapest, Hungary
Jennifer J. Kiser, Courtney V. Fletcher
University of Colorado at Denver and Health Sciences Center
The 8th International Workshop on Clinical Pharmacology of HIV Therapy was held April 16-18, 2007 in the lovely Budapest, Hungary. 193 HIV clinical pharmacologists attended this year's Workshop and more than half were new attendees. This year's meeting included eighty posters, 30 platform presentations, six invited lectures, and a roundtable discussion on the optimal design of drug interaction studies.
The plenary lectures covered a variety of topics of relevance to clinical pharmacology and the therapeutics of treating HIV infection. Dr. Kevin Park reminded the audience of the high frequency of drug-induced adverse reactions and their consequences including hospital admissions and drug withdrawals from the market, and discussed mechanisms of drug-induced hepatotoxicity. Dr. Judith Currier gave a state of the art lecture on women's health topics in HIV focusing on virologic, pharmacologic and complication of treatment issues. Dr. Currier noted that to date most clinical trials are underpowered to detect differences with respect to safety and efficacy between men and women and the need for future studies to address this problem. Dr. Daria Hazuda discussed the development of integrase inhibitors and illustrated the promise these agents hold for treatment of HIV infection. Dr. Anton Pozniak's lecture focused on treatment and drug-drug interactions in the setting of HIV and tuberculosis co-infection. This lecture dealt with the sober statistics of infection with tuberculosis, noting one death from tuberculosis every 15-20 seconds, questions of when to start antiretroviral therapy in the co-infected person, and the management challenges of drug-drug interactions. There remain critical gaps in our understanding of the pharmacotherapy of HIV and tuberculosis coinfection that need urgent attention. Dr. Terry Blaschke discussed a variety of issues related to the co-formulation of antiretroviral agents noting the usefulness of co-formulated drugs and the variety of challenges that must be addressed in their development. The last lecture, given by Dr. Bruno Stieger discussed the role of membrane transporters in drug pharmacokinetics and pharmacodynamics. This is a topic of increasing importance for antiretroviral agents, with drug-drug interactions such as those between rosuvastatin and lopinavir/ritonavir, and pravastatin and darunavir/ritonavir (discussed later) likely being transporter mediated. Drs. Thomas Kakuda and Courtney Fletcher led a roundtable discussion on the role of phenotyping cocktails to identify drug interactions, the optimal design of drug-drug interaction studies, and the importance of determining which drug-drug interactions may be of clinical significance for patients.
It is not possible to cover all of the abstracts presented at the Workshop, so we have chosen to focus on studies we felt were well designed and most clinically relevant. This report is divided into four sections: Drug-Drug Interactions, Pharmacokinetic Data with Existing Antiretroviral Compounds, Antiretroviral Pharmacokinetics in Special Patient Populations, and the Pharmacology of Investigational Drugs.
Drug-Drug Interactions
Previous studies have shown significant hepatotoxicity in healthy volunteers receiving rifampin in combination with saquinavir/ritonavir (Schutz M. 6th International Workshop on Clinical Pharmacology of HIV Therapy, 2005, abstract 35) and lopinavir/ritonavir (La Porte CJL. AAC 2004;48(5):1553-1560). At this year's Workshop, a study detailing adverse events in healthy volunteers receiving the combination of rifampin and lopinavir/ritonavir was presented (abstract 51). Volunteers received 600 mg of rifampin daily for 5 days; then either 600/150 mg or 800/200 mg of lopinavir/ritonavir was added to the rifampin. By the second day of the combination, 10/11 subjects suffered nausea and vomiting and all 11 had liver enzyme elevations and thus the study was terminated. The liver enzyme elevations peaked 2-3 days after discontinuing the combination (n=2 with grade 2, n=9 with grade 4 elevations). Liver enzymes returned to normal in all subjects during follow up and there were no signs of clinical hepatotoxicity. The mechanism(s) for the development of severe hepatotoxicity with the combination of rifampin plus saquinavir/ritonavir or lopinavir/ritonavir in healthy volunteers is unclear. There remains a real unmet need for effective strategies for concomitant treatment of tuberculosis and HIV, and in particular whether rifampin can be given with any ritonavir-boosted protease inhibitor. This latter question needs to be addressed in healthy volunteer studies because of the risk of suboptimal protease inhibitor concentrations; however, it is clear that the combination of ritonavir-boosted protease inhibitors and rifampin may pose significant risks to these subjects and a high degree of caution is warranted for any future studies.
A phenotyping cocktail study was conducted to evaluate the in vivo effects of tipranavir on various cytochrome (CYP) P450 enzymes. Sixteen healthy volunteers were given single doses of caffeine (CYP1A2 probe), warfarin (CYP2C9 probe), and dextromethorphan (CYP2D6 probe) at baseline and 10 hours after a first dose and a steady-state dose of tipranavir/ritonavir 500/200 mg twice daily (abstract 52). At steady state, tipranavir/ritonavir was found to induce CYP1A2 and CYP2C9 because the ratios of caffeine and warfarin AUCs at steady state tipranavir/ritonavir vs. baseline were 0.57 and 0.88, respectively. The ratio of dextromethorphan AUC at steady state tipranavir/ritonavir vs. baseline was 6.24 suggesting tipranavir/ritonavir is a potent inhibitor of CYP2D6. Data presented at the 2007 CROI (Vourvahis M. 14th CROI, 2007, abstract 563), demonstrated that tipranavir/ritonavir is a potent intestinal CYP3A inhibitor and has modest effects on P-glycoprotein. Collectively, these data provide some insight into the complex mechanisms of previously identified interactions with tipranavir/ritonavir including the ability to inhibit and induce different drug elimination pathways simultaneously, and may assist in prediction of other interactions with this compound. Enzyme and transporter phenotyping studies are emerging as a new tool to aid in the prediction of drug-drug interactions. At this time, the greatest benefit of probe studies appears to be in identifying potential drug interactions that do not need further clinical evaluation.
The Department of Health and Human Services Guidelines for the Use of Antiretroviral Agents in HIV-Infected Adults and Adolescents recommends starting with a 5 mg dose of tadalafil (not to exceed 10 mg in a 72 hour period) for the treatment of erectile dysfunction in persons with HIV on a protease inhibitor. However, the results of a healthy volunteer study presented at the Workshop suggest that this lower 5 mg dose of tadalafil only needs to be used during the first few days of initiating a tipranavir/ritonavir-based regimen, but after 7-10 days on tipranavir/ritonavir, no tadalafil dose adjustment is necessary (abstract 61).
Pravastatin is one of the most widely used HMG-CoA reductase inhibitors in persons with HIV because of its low propensity for CYP-mediated drug interactions. However, Vanitha Sekar presented data on an unexpected interaction between darunavir/ritonavir and pravastatin (abstract 54) in 14 healthy volunteers. In this study, pravastatin AUC and Cmax were increased 81% and 63%, respectively (on average) when combined with darunavir/ritonavir 600/100 mg twice daily. Interestingly, there was substantial interindividual variability in the magnitude of this interaction, with not all volunteers having an increase in pravastatin exposure with the addition of darunavir/ritonavir and others having up to a 3 to 10 fold increase. The mechanism for this interaction is speculative, but is likely mediated by darunavir and/or ritonavir's inhibition of organic anion transporting polypeptide 1B1 (OATP1B1) located on the basolateral side of the hepatocyte. Pravastatin is a high affinity substrate for OATP1B1. A previously presented study identified a similar unexpected interaction between another OATP1B1 substrate, rosuvastatin, and lopinavir/ritonavir (Hoody DW. 14th CROI, 2007, abstract 564). A study is ongoing to elucidate the purported mechanism for the interaction between darunavir/ritonavir and pravastatin. Clinicians may wish to avoid the combination of darunavir/ritonavir and pravastatin until more data are available.
Pharmacokinetic Data with Existing Antiretroviral Compounds
Marta Boffito presented data on the plasma and intracellular concentrations of abacavir when given as either 600 mg once daily or 300 mg twice daily (abstract 13). 27 subjects (9 females) completed the study. Plasma exposures of abacavir were similar between patients on 300 mg twice daily and 600 mg once daily, while Cmax was 109% higher and Ctrough 63% lower, as expected, following 600mg once daily vs. 300 mg twice daily. However intracellular carbovir triphosphate AUC and Cmax were 32% and 99% higher in those on 600 mg once daily. Plasma concentrations of abacavir were 38% higher in females even after adjusting for weight, and intracellular carbovir triphosphate AUCs were 2-fold higher in females. The finding of a gender effect with regard to intracellular triphosphate concentrations of nucleosides was echoed in a study (abstract 56) of tenofovir. This investigation found no difference in plasma tenofovir concentrations between men and women but did show that women had approximately 50% higher intracellular tenofovir concentrations than did men. The clinical significance of higher intracellular carbovir-triphosphate and tenofovir-diphosphate in women is unknown. Of most concern would be whether the higher concentrations of these pharmacologically-active moieties predispose women to greater risk of adverse reactions. These studies, in conjunction with previous reports showing that women had higher intracellular triphosphate concentrations of zidovudine and lamivudine (Anderson PL. AIDS 2003: 17:2159-68) provide a pharmacokinetic basis to warrant further studies of gender-based differences in nucleoside phosphorylation that incorporate virologic, immunologic and safety evaluations.
Stavudine is part of generic, fixed dose antiretroviral combination products available in developing countries. However, its use has been associated with toxicities. Thus the pharmacokinetics, efficacy and safety of lower doses of stavudine are being explored. Gilles Peytavin presented data from 57 patients (median weight 72 kg), who had been on a stavudine-containing regimen for a median of 6 years, who decreased their stavudine dose from 40 mg twice daily to 30 mg twice daily (abstract 33). Eleven of the 57 subjects underwent intensive pharmacokinetic studies while on 40 mg and 30 mg twice daily. Despite stavudine AUC and Cmax being reduced by 31% and 44%, respectively, with the dose reduction, 98% and 93% of subjects had viral loads of less than 400 copies/mL at 24 and 48 weeks after stavudine dose reduction. Additionally, 6 of 17 patients who reported symptoms of neuropathy on 40 mg reported an improvement at week 48. These data are encouraging for the potential for stavudine dose reduction in countries without access to other less-toxic nucleoside analogs.
Ritonavir is frequently used to "boost" the concentrations of other protease inhibitors. However these concomitant protease inhibitors have differing effects on ritonavir concentrations. Higher ritonavir concentrations may lead to rises in lipids and gastrointestinal adverse effects; thus, there may be differences in tolerability between ritonavir-boosted protease inhibitor regimens due to differences in ritonavir concentrations. Marta Boffito presented data from 16 studies on the effects of eight protease inhibitors on ritonavir concentrations when used as a pharmacokinetic booster (abstract 50). Overall, atazanavir and indinavir were found to increase ritonavir concentrations by 62 and 72%, saquinavir showed no significant effect, while amprenavir, nelfinavir, darunavir and lopinavir lower ritonavir levels. Tipranavir showed the greatest reduction in ritonavir concentrations by 90% (hence the reason for using a 200 mg dose of ritonavir to boost this agent). The question posed by these investigators of the differential effects of PIs on ritonavir is interesting and clinically relevant. This initial effort to present a comprehensive examination of the pharmacokinetic effects of protease inhibitors on ritonavir concentrations warrants additional study of these issues.
Antacids decrease tipranavir exposure by about 25%, thus the potential for an interaction between omeprazole and tipranavir requires investigation. Charles laPorte and colleagues reported the results of a drug-drug interaction study in 15 healthy volunteers between tipranavir/ritonavir and the proton-pump inhibitor omeprazole (abstract 59). This study was designed to evaluate the pharmacokinetics of single doses of tipranavir/ritonavir (500/200 mg) given with food alone and after 5 days of omeprazole 40 mg once daily. The geometric mean ratios (and 90% confidence interval) for the tipranavir AUC and Cmax were 1 (0.89, 1.12) and 1.05 (0.94, 1.17), respectively. These data indicate that omeprazole had no adverse affect on tipranavir bioavailability and concomitant therapy with food should be acceptable. In addition, these authors investigated the effect of food on the pharmacokinetics of tipranavir/ritonavir, 500/200 mg twice daily in healthy volunteers. 32 of 35 participants completed this study. The geometric mean ratios (and 90% confidence interval) for the tipranavir pharmacokinetic characteristics given fasted versus fed were: AUC, 0.99 (0.88, 1.11); Cmax, 1.02 (0.91, 1.14); and Cmin, 1.02 (0.84, 1.23). These data would suggest that tipranavir can be given either with or without food. However, these data are in contrast to data in the manufacturer's product information that described an enhanced bioavailability of tipranavir when given with high-fat meals (868 kcal, 53% derived from fat, 31% derived from carbohydrates) with a 31% increase in AUC (1.23-1.39). The most prudent recommendation until the complete data from this food effect study are available (and perhaps a regulatory agency review) is to continue with the recommendation to administer tipranavir/ritonavir with food.
Antiretroviral Pharmacokinetics in Special Patient Populations
Hepatic Impairment
There are limited data on the appropriate dosing of antiretroviral drugs in patients with varying degrees of hepatic impairment. There are no data on the pharmacokinetics of fosamprenavir when combined with ritonavir in patients with any degree of hepatic impairment. Josep Mallolas presented a study evaluating fosamprenavir dosing and pharmacokinetics in HIV-infected subjects with mild and moderate hepatic impairment (abstract 1). Thirteen subjects with mild hepatic impairment (Child Pugh score 5-6) received fosamprenavir 700 mg twice daily plus ritonavir 100 mg once daily (Group A). Ten subjects with moderate hepatic impairment (Child Pugh score 7-9) received fosamprenavir 300 mg twice daily (as the oral suspension) plus 100 mg of ritonavir once daily (Group B). Eight subjects with moderate hepatic impairment received fosamprenavir/ritonavir 700/100 mg once daily (Group C). Ten patients with normal hepatic function received fosamprenavir/ritonavir 700/100 mg twice daily (Group D/controls). All subjects underwent intensive pharmacokinetic studies (including measurement of unbound amprenavir concentrations at two time points) two weeks after initiating fosamprenavir/ritonavir. Patients in Group A had total amprenavir plasma AUCs 22% higher and Cmins that were similar to controls, but amprenavir unbound Cmin was 2-fold higher in patients in Group A vs. controls. Subjects in Group B had amprenavir AUC and Cmin 27% and 57% lower, respectively than controls, but the amprenavir unbound Cmin was similar for subjects in Groups B and D. Subjects in Group C had amprenavir AUC and Cmin 24% and 65% lower than controls, and unbound amprenavir Cmin that were 40% lower on average than controls. The investigators concluded that the reduced ritonavir dose regimen of fosamprenavir 700 mg twice daily plus ritonavir 100 mg once daily, is the appropriate dose for subjects with mild hepatic impairment. However, neither dosing strategy appeared adequate for subjects with moderate hepatic impairment. Thus, they speculate that fosamprenavir 450 mg twice daily plus ritonavir 100mg once daily would provide adequate exposures for patients with moderate hepatic impairment, though there are no pharmacokinetic or safety data with this dosing strategy.
The pharmacokinetics of maraviroc following a single 300 mg dose in subjects with mild and moderate hepatic impairment were compared to the pharmacokinetics in subjects with no hepatic impairment (abstract 8). Maraviroc AUC and Cmax were increased 25% and 11%, respectively in subjects with mild hepatic impairment relative to those with no hepatic impairment. Maraviroc AUC and Cmax were increased 46% and 32%, respectively in subjects with moderate hepatic impairment relative to those without hepatic impairment. Further studies are necessary to determine if dose adjustments may be necessary for subjects with moderate hepatic impairment.
Renal Impairment
Sangeeta Agarwala presented a study evaluating the pharmacokinetics of unboosted atazanavir in persons with severe renal impairment including those on hemodialysis (abstract 2). This was an open-label, parallel design study with 3 groups (controls, severe renal impairment not on hemodialysis, and hemodialysis) of HIV negative subjects (n=10 per group). Subjects with severe renal impairment not receiving dialysis had atazanavir AUCs 19% higher than age, weight, and gender matched controls with normal renal function. Subjects on hemodialysis had atazanavir AUCs 42% lower on dialysis days and 28% lower on non-dialysis days compared to controls. Though the mechanism for the reduction in atazanavir exposures in those on hemodialysis is not known, the investigators speculate that there may be decreased gastric acid production in hemodialysis patients. These investigators (from Bristol Myers Squibb) commented that for patients receiving hemodialysis clinicians could consider using atazanavir/ritonavir to compensate for the reduction observed in this study in atazanavir concentrations. While the basis for this recommendation is understandable it is important to stress that no pharmacokinetic or safety data are available at this time to support this recommendation.
Pediatrics
David Burger described the pharmacokinetics, efficacy, and tolerability of efavirenz tablets and capsules when used in children ages 2-16 years and dosed per the weight-based manufacturer's guidelines (abstract 11). 307 efavirenz plasma concentrations were obtained in 33 children. 8.8% of samples were less than 1000 ng/mL, while 14.7% of samples were greater than 4000 ng/mL. There was a trend towards a higher proportion of samples greater than 4000 ng/mL in the children who reported central nervous system adverse effects vs. those without these effects (p=0.23, 26 vs. 13%). All 27 children who remained on efavirenz achieved viral loads of less than 50 copies/mL, despite the occasional presence of subtherapeutic concentrations in 12 of these children most likely due to sporadic non-adherence. These data suggest that efavirenz tablets and capsules are effective in children who are able to tolerate the drug and provide some validation of the dosing guidelines for children contained in the manufacturer's dosing guidelines. This study raises the question as to whether those children who had CNS adverse events and were found to also have efavirenz concentrations greater than 4000 ng/mL might have benefited from therapeutic drug monitoring and dose adjustment.
The United States Food and Drug Administration requested that Roche evaluate the pharmacokinetics of ritonavir-boosted saquinavir following opening the Invirase capsules and dissolving them in various vehicles in an attempt to find an acceptable means of administering the drug to children 4 months to 5 years of age. Diane McKay (abstract 6) presented an open-label, randomized, four period, crossover relative bioavailability study of this strategy in 27 healthy adult volunteers. Volunteers received ritonavir 100 mg as the oral solution plus Invirase capsules either unopened (A), opened or suspended in simple syrup (B), baby formula (C), or jelly jam (D). The bioavailability of saquinavir was 10, 60 and 40% higher in simple syrup, baby formula, and jelly jam, respectively, relative to the unopened capsules. The next step is to evaluate this strategy in children. All three vehicles will be used in the studies with children.
Children appear to have faster apparent oral clearances of tenofovir and atazanavir than adults (Hazra R. AAC 2004;48(1):124-129, Kiser JJ. 12th CROI, 2005, abstract 767). Additionally, there is a bidirectional interaction between these drugs (Agarwala S. 6th International Workshop on Clinical Pharmacology of HIV Therapy 2005; abstract 16). Jennifer Kiser presented data evaluating the interaction between atazanavir/ritonavir and tenofovir in HIV-infected adolescents and young adults (abstract 12). Atazanavir pharmacokinetics appeared similar to values observed in older adults on the combination (Taburet AM, AAC 2004;48(6):2091-6). Though a higher tenofovir AUC may have been expected based on a previous study in healthy volunteers receiving the combination, this was not observed in this study. Apparent oral clearance increased as weight increased for atazanavir, ritonavir, and tenofovir. Estimated creatinine clearance was also associated with tenofovir apparent oral clearance. The geometric mean for intracellular tenofovir diphosphate concentrations in 22 subjects was 94 fmol/10^6 cells (63% CV) and were similar to that reported by Hawkins in a small number of patients (Hawkins T. JAIDS 2005;39(4):406-11).
Pharmacology of Investigational Drugs
Maraviroc is a CYP3A4 and P-glycoprotein substrate, thus previous interaction studies have shown that a lower dose of maraviroc, 150 mg twice daily, should be used in combination with protease inhibitors (excluding tipranavir/ritonavir). John Davis presented data on the interaction between maraviroc 150 mg twice daily and darunavir/ritonavir in 12 healthy volunteers (abstract 55). Maraviroc AUC and Cmax were increased 405% and 229%, respectively when combined with darunavir/ritonavir. This increase is consistent with other protease inhibitors and thus, the reduced 150 mg twice daily dose is recommended in combination with darunavir/ritonavir. In this study, as has been the case with all maraviroc drug interaction studies, maraviroc is administered in the fasting state, approximately 1.5 hour prior to the dosing of the protease inhibitor in combination with food. Food decreases maraviroc concentrations by about 50%, however there are no data on the magnitude of these drug interactions when administered with food. As more concentration-response data with this agent become available, it will be important to determine if the maraviroc concentrations achieved when taken in combination with other protease inhibitors and food fall within the therapeutic range for this drug. If so, that would eliminate the food incongruence between this agent and many other antiretroviral compounds.
Vicriviroc is an investigational CCR5 inhibitor. Charles Flexner presented data exploring the relationship between vicriviroc concentrations and antiretroviral effects (abstract 15). Two concentrations were obtained from 86 treatment experienced patients participating in a Phase II clinical trial of vicriviroc at doses of 5, 10, or 15 mg once daily in combination with ritonavir-boosted protease inhibitor regimens (Adult AIDS Clinical Trials Group study 5211). In this study, patients randomized to vicriviroc received the drug with their current failing antiretroviral regimens for two weeks, then patients continued on vicriviroc but their background regimen was optimized based on resistance test results. The concentration data from these 86 patients were combined with intensive pharmacokinetic data from 110 healthy volunteers from five Phase I vicriviroc studies to develop a pharmacokinetic/pharmacodynamic model. At week 2, a Cmin above 54 ng/mL (the EC90 for this drug is 56 ng/mL) and an AUC above 1460 ng*hr/mL were associated with greater viral load decreases. This relationship was no longer apparent at weeks 16 or 24, most likely because treatment response at that point also depended on having other active drugs in the regimen.
Data were presented on the combination of vicriviroc and tipranavir/ritonavir in 8 healthy volunteers (abstract 57). Volunteers received vicriviroc 15 mg once daily plus ritonavir 200 mg twice daily for two weeks. Tipranavir 500 mg twice daily was then added for an additional two weeks. Vicriviroc AUC and Cmax were reduced 6% and 12%, respectively when combined with tipranavir/ritonavir, thus consistent with other ritonavir-boosted protease inhibitors, no vicriviroc dose adjustment is necessary.
Tipranavir/ritonavir reduces the concentrations of many nucleoside analogs. However, the AUC and Cmax of apricitabine, an investigational nucleoside analog, are actually moderately increased by 40 and 25%, respectively by tipranavir/ritonavir (abstract 68). Adverse effects with the combination were mainly nausea, anorexia, headache, and elevated liver enzymes, which are consistent with the side effect profile of tipranavir/ritonavir. Additional studies of the safety of this combination and data on the intracellular concentrations of apricitabine (the active moiety of the nucleoside analogs) in combination with tipranavir/ritonavir are needed.
The pharmacophore of the HIV integrase inhibitors forms a complex with divalent cations (magnesium) at the active site of the integrase enzyme. This propensity for cation binding can result in an interaction with antacids at the level of drug absorption upon co-administration due to the high concentrations of di- and tri-valent cations in antacids. Thus the effect of simultaneous and staggered administration of antacids and elvitegravir (Gilead's investigational integrase inhibitor, also known as GS-9137) were presented (abstract 69). The effect of omeprazole 40 mg on elvitegravir absorption was also evaluated. The dose of elvitegravir used in this study was 50 mg with a 100 mg boosting dose of ritonavir administered once daily. Omeprazole did not alter elvitegravir absorption, thus this compound does not exhibit pH-dependent absorption. However, as expected, simultaneous administration of antacid reduced elvitegravir concentrations by about 50%. Separating the antacid by 2 hours only decreased elvitegravir concentrations by 10-20%, however elvitegravir concentrations were unchanged if separated by 4 hours.
Conclusions
The International Workshop on Clinical Pharmacology of HIV Therapy represents the largest scientific meeting focused on the clinical pharmacology of antiretroviral therapy. This meeting includes pharmacologists from academia, industry and regulatory authorities and continues to provide a forum to advance the science of the clinical pharmacology of antiretroviral agents, which translated to improved clinical outcomes of HIV therapy.
|
| |
|
|
|
|
|
