 |
|
|
| |
9th International Workshop on Clinical Pharmacology of HIV Therapy
|
| |
| |
Jennifer J. Kiser, Pharm.D.
Assistant Professor
School of Pharmacy
University of Colorado Denver
jennifer.kiser@uchsc.edu
Courtney V. Fletcher, Pharm.D.
Dean and Professor
College of Pharmacy
University of Nebraska Medical Center
cfletcher@unmc.edu
Abbreviations:
%CV, percent coefficient of variation
3TC, lamivudine
AUC, area under the concentration-time curve
ARV, antiretroviral drug
ATV, atazanavir
C12, concentration 12 hours after the dose
Cmin, minimum concentration
CYP, cytochrome P 450
DRV, darunavir
LPV, lopinavir
ETV, etravirine
EVG, elvitegravir
FTC, emtricitabine
IC95, concentration of drug required to inhibit viral replication by 95%
NNRTI, non-nucleoside reverse transcriptase inhibitor
NRTI, nucleoside reverse transcriptase inhibitor
PACTG, Pediatric AIDS Clinical Trials Group
PD, pharmacodynamics
PI, HIV Protease Inhibitor
PK, pharmacokinetics
RAL, raltegravir
RTV or r, ritonavir
SQV, saquinavir
T20, enfuvirtide
TDF, tenofovir disproxil fumarate
TPV, tipranavir
UGT, uridine glucuronosyl transferase
VL, HIV viral load
ZDV, zidovudine
The 9th International Workshop on Clinical Pharmacology of HIV Therapy was held April 7-9, 2008 in charming New Orleans, LA. This year's workshop featured 83 abstracts including 27 oral presentations and six invited lectures.
Fred Lopez, M.D., gave the opening presentation on the state of health care in New Orleans during and after hurricane Katrina. Dr. Lopez is Associate Professor and Vice Chair, Department of Medicine at Louisiana State University Health Sciences Center in New Orleans. His superb presentation as an insider at Charity Hospital was a sobering reminder of the state of health care and emergency preparedness, but also an uplifting presentation on the response to this crisis by the health care profession. The October 13, 2005 issue of the New England Journal of Medicine has a series of articles on these topics.
David Haas, M.D., followed Dr. Lopez with a presentation on HIV pharmacogenomics. Dr. Haas is Associate Professor of Medicine, Microbiology and Immunology, and Director of the AIDS Trials Program at Vanderbilt University in Nashville. Dr. Haas has played an instrumental role in moving this field forward domestically, and most recently internationally with his efforts to develop a DNA repository from participants in ACTG studies at international sites. David discussed recent work related to the methods for the collection of DNA, and instrumentation for the study of genetic variability. Dr. Haas reviewed recent findings in the field, and perhaps most importantly, he reminded us that genetic associations must be validated in separate data sets.
Dr. Nancy Kass, Professor of Bioethics and Public Health at Johns Hopkins University discussed the ethics of healthy volunteer studies. Healthy volunteer studies are increasingly common in the development of antiretroviral drugs to answer phase I questions, and drug-drug interactions. Dr. Kass discussed aspects of her research highly relevant to these topics, including participant motivation, informed consent, justice and international research ethics.
Deanna Kroetz, Ph.D., gave a state-of-the-art presentation on pharmacokinetic and pharmacodynamic implications of drug transporters. Dr. Kroetz is Professor of Biopharmaceutical Sciences and Pharmaceutical Chemistry at the School of Pharmacy at the University of California San Francisco. We have learned that genetic variability in drug metabolizing enzymes contributes to the variability in plasma concentrations seen among patients taking PIs or NNRTIs. It is now clear that genetic variability in membrane drug transporters contributes to this variability as well and extends to the NRTIs. Dr. Kroetz's presentation primarily focused on the role of the multidrug resistance membrane (MRP) transporters, which are efflux pumps that have been shown to export a number of antiretroviral drugs out of cells. It seems clear that an understanding of the degree to which antiretroviral agents are substrates, inducers or inhibitors of these transporters will be as important as this is for drug metabolizing enzymes.
Dr. Shiew Mei Huang from the Office of Clinical Pharmacology at the FDA gave an enlightening presentation on drug interactions and package insert labeling issues. Dr. Huang started her comments by reminding the audience of drugs that have recently been withdrawn from the market for serious adverse reactions. She then discussed issues of study design and decision tree approaches for whether a drug is a substrate, inhibitor or inducer of CYP metabolism, and similar approaches for drugs as substrates and inhibitors of membrane transporters. She ended with comments regarding the labeling of drug interaction information. For more reading on this topic, see Huang SM et al. Clin Pharmacol Ther 2007; 81:298-304.
The final plenary was given by Scott Letendre, M.D., from the Neurobehavioral Research Center at the University of California, San Diego. Dr. Letendre is a leader in the field of HIV-associated neurocognitive disorders and the importance of penetration of antiretroviral drugs into the central nervous system (CNS). This was an outstanding presentation on PK focusing on penetration of antiretroviral agents into the cerebrospinal fluid (CSF), and importantly on PD, or the effect of that concentration on viral load in the CSF. Dr. Letendre discussed some of his latest work using a CSF penetration score for an antiretroviral regimen and association of that score with viral loads in the CSF (see Letendre S et al. Validation of the CNS Penetration-Effectiveness Rank for Quantifying Antiretroviral Penetration Into the Central Nervous System. Arch Neurol. 2008;65(1):65-70.)
The International Workshops on Clinical Pharmacology of HIV Therapy have always provided a venue for interactions of scientists from academia, industry and the FDA. A FDA-Industry roundtable discussion at this workshop focused on the interpretation and communication of drug interaction data. In this roundtable, Drs. Kimberly Struble and Richard Bertz discussed the recent labeling changes in the package insert for atazanavir concerning the use of acid reducing agents. This discussion highlighted issues of the synthesis of drug-drug interaction studies and the distillation of those studies into a simplified presentation/recommendation in the FDA-approved package insert. Drs. Kellie Reynolds and Thomas Kakuda discussed issues regarding the interpretation of drug-drug interaction studies for the recently approved NNRTI, etravirine. The theme of their discussion was clinical significance, and what information on exposure-response relationships is needed to determine whether a drug-drug interaction is likely to result in a clinically significant effect. This meeting is one of the very few venues where this kind of interchange takes place, and I think this was one of the highlights of the meeting.
It is not possible to cover all of the abstracts presented at the meeting, 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, the Pharmacology of Investigational Drugs, and Antiretroviral Pharmacokinetics in Special Patient Populations.
Drug-Drug Interactions
Abstract O19 revealed an unexpected 30% reduction in RAL C12 in the presence of LPV/r. Despite the reduction, all but one of the subjects maintained a C12 above the estimated IC95 of 15 ng/mL. RAL is primarily metabolized by UGT1A1 with minor metabolism via UGT1A3 and UGT1A9, thus this interaction is likely the result of LPV or RTV's induction of RAL's glucuronidation, but no glucuronide metabolites were quantified.
ATV is a CYP2C8 inhibitor in vitro and RTV's in vitro effects on CYP2C8 are contradictory, thus rosiglitazone was used as a CYP2C8 probe to evaluate the effects of ATV and ATV/r on CYP2C8 activity (abstract O5). With unboosted ATV, rosiglitazone single dose AUC was increased 35%, suggesting ATV is a weak CYP2C8 inhibitor, however rosiglitazone AUC was slightly reduced (17%) when given with ATV/r, suggesting that RTV offsets the ATV CYP2C8 inhibition. Thus, CYP2C8 substrates may have increased exposures in the presence of unboosted ATV, but may be slightly reduced with ATV/r. CYP2C8 is polymorphic, but there were too few patients with the various CYP2C8 polymorphisms to definitively determine the impact of genotype on rosiglitazone PK or the magnitude of these drug interactions.
PK of Existing Drugs
Concentration-effect relationships are needed for RAL, but difficult to discern due to the substantial variability in PK. PK relationships with week 16 viral load were evaluated using geometric mean C12 in ~850 patients and with an average of 4 samples for each patient taken throughout different times of the dosing interval (called GM All) in 457 patients. No PK-PD relationship were identified for RAL C12, but those patients with a VL greater than or equal to 50 copies/mL had a RAL GM All of 309 nM vs. 446 nM for those with VL less than 50, but GM All appeared to be a less important predictor of response than other factors like baseline viral load and having other active ARV drugs in the regimen. This analysis is a good start, and a useful metric for RAL exposure may have been identified; but, additional studies are still needed to determine the PK-PD relationships for this drug.
Abstract O25 presented RAL Cmin in patients in the French RAL expanded access program who were on a variety of other ARV regimens. All patients were on RAL plus 2 NRTIs and/or T20 plus either DRV/r (group A, n=125), ATV/DRV/r (group B, n=18) or TPV/r (group C, n=32). 64% of the patients were also on ETV. 338 RAL troughs were obtained from 213 patients. RAL Cmin in Groups A, B, and C were 96 ng/mL, 182 ng/mL, and 52 ng/mL, respectively. Nine, 3, and 25% of patients in groups A, B, and C had RAL Cmin less than the IC95 of 15 ng/mL, respectively. There are multiple drug interactions at play here (including a potential increase in RAL levels due to TDF, and the percent of patients on TDF was not reported) and large interpatient variability in RAL Cmin, but these findings tend to confirm those of previous PK studies in healthy volunteers, that ATV increased RAL Cmin and TPV substantially reduced RAL levels. The BENCHMRK 1 and 2 studies showed similar virologic response rates at week 24 in those on TPV/r vs. other ARV with RAL, but the fact that 25% of patients in this study on TPV/r had RAL Cmin below 15 ng/mL raises questions as to how this interaction will manifest outside of the controlled clinical trial environment. We would suggest clinicians using this combination monitor their patients quite closely for virologic response.
Several PIs (SQV, LPV, DRV) and TMC278 were found to be substrates for the organic anion transporting polypeptide 1B1 (OATP1B1) transporter located on the basolateral side of the hepatocytes (abstract O4). At last year's workshop, an unexpected increase in pravastatin exposures in the setting of DRV/r was presented (Sekar V, 8th IWCPHT, #54). That study, along with two others with rosuvastatin (Kiser JAIDS 2008;47:570, Pham 15th CROI 2008 #767), identified an interaction that is most likely transporter-mediated and occurring at the level of the OATP1B1 transporter. Many statins are high affinity substrates for OATP1B1 and rely on it for uptake into the liver. The fact that the PIs are also substrates for this transporter suggests that the unexpected interactions observed between PIs and pravastatin and rosuvastatin may involve competition for transport into the liver via OATP1B1 with the PIs being higher affinity substrates for this transporter and preventing the transport of the statins. In further support of the role of OATP1B1 in PI transport, abstract O3 found that patients with the SLCO1B1 (the gene that encodes OATP1B1) 521T>C polymorphism had higher LPV exposures than those with wild-type genotype. The 521T>C polymorphism results in less LPV or RTV (or both) being taken up by the hepatocytes, thus these patients have higher plasma concentrations, but less drug penetrates the liver. This polymorphism has been associated with higher plasma levels with several of the statins as well. These abstracts highlight the increasing complexity of drug-drug interactions and the interplay between PK and pharmacogenetics.
TDF reduces unboosted ATV Cmin by 40%, however some patients still receive unboosted ATV in the setting of TDF due to RTV intolerance. Abstract O17 evaluated the PK of once vs. twice-daily unboosted ATV in the setting of TDF. Ten HIV-infected patients who had VL less than 50 copies/mL for at least 6 months on TDF, FTC, and unboosted ATV 400 mg once daily had intensive PK visits performed on ATV 400 mg once daily and 200 mg twice daily. The geometric mean ATV Cmin was 138 ng/mL once daily vs. 305 ng/mL twice daily. Six of the 10 patients had ATV Cmin below 150 ng/mL on once daily ATV vs. only 2 of 10 while on twice-daily ATV. Though giving unboosted ATV twice daily in the setting of TDF is probably better than giving it unboosted once daily, ATV should be given with RTV in the setting of TDF when at all possible because 20% of the patients still had ATV Cmin less than 150 ng/mL. Additionally, the patients selected for this study were undetectable on TDF, FTC, and once daily unboosted ATV and we know nothing about ATV PK in patients who may not achieve virologic suppression on this regimen.
Potential gender and racial differences in DRV/r PK were evaluated in a PK substudy of 28 patients (predominantly women and minorities) in the Gender, Race, and Clinical Experience (GRACE) study. Despite a similar weight (mean 81 kg for both genders), DRV exposures were 20% higher in women vs. men, and interestingly, the RTV exposures were 70% higher in the women vs. men. The US RTV prescribing information reports no differences by gender in RTV PK, but this finding is consistent with higher RTV concentrations reported in women receiving LPV/RTV (Umeh O 14th CROI 2007, #786). The mechanism for this sex difference is unclear. Women did have, on average, 12% higher alpha-1 acid glycoprotein levels than men in this study. No differences were noted between black and Hispanic patients in DRV or RTV PK.
PK of Investigational Drugs
Elvitegravir (EVG) is metabolized by CYP3A, UGT1A1 and UGT1A3. Abstract O18 determined the PK of EVG in 15 healthy volunteers when given with RTV 100 mg once daily vs. ATV 400 mg once daily. EVG AUC was bioequivalent when given with ATV and RTV, and the mean (%CV) for the EVG trough was 89.9 (71.4, 113) with ATV vs. RTV. Thus, the good news is ATV may be an acceptable alternative booster to RTV for EVG, but the bad news is that the ATV mean trough with EVG was 74.5 ng/mL, which is lower than historical data and lower than the proposed target trough of 150 ng/mL. Additional studies will be required to determine if this combination is efficacious.
Abstract O14 evaluated the impact of various covariates on the PK of the once daily investigational NNRTI, TMC278 (rilpivirine), during the Phase IIb dose-finding trial. In this study, 268 naïve patients from several countries received 25, 50, or 100mg of TMC278 with either ZDV+3TC or TDF+FTC. Though region (Europe, US, and Russia), female gender, and TDF/FTC backbone were associated with lower TMC278 in univariate analysis, these relationships disappeared after adjustments for weight, but Asian race persisted as a predictor for slower TMC278 clearance even after controlling for weight. The mechanism(s) for slower TMC278 clearance in Asians is presently unclear, but it may possibly relate to reduced transport of TMC278 via OATP1B1 because of genetic polymorphisms in SLCO1B1 in Asians. Asians have slower statin clearances than other races. The 25mg once daily TMC278 dose has moved into Phase III trials.
The effects of various types of food on TMC278 were presented (abstract P32). TMC278 exposures are reduced 40-50% in the fasted state or when taken only with a nutritional drink. Thus, TMC278 should be taken with food, and preferably with the most substantial meal of the day.
PK in Special Patient Populations
PACTG 1066 is a pediatric PK, safety, and dose-finding RAL study. Abstract P8 presented preliminary PK and safety data for 15 children ages 6-19. In this study, RAL is added to a failing regimen, an intensive PK is performed in the fed state, and then ARV background regimens are optimized. If children fail to achieve a RAL AUC of >6.2 mg*hr/L (14.4 μM) and a C12 of >15 ng/mL (22 nM), then their RAL doses are increased and a repeat intensive PK performed. To date, a dose of 8 mg/kg twice daily has shown good tolerability and provided acceptable C12, but some children failed to achieve the AUC target at this dose. The PACTG 1066 study team feels the large variability of RAL concentrations in the fed state is hindering the determination of the optimal RAL dose, thus future enrollments will evaluate RAL PK at the 8 mg/kg twice daily dose, but in the fasted state.
Abstracts O9 and O11 evaluated the PK of SQV 1000 mg, given as the 500 mg tablet formulation, boosted with RTV 100 mg, twice daily in pregnancy. 26 women from Europe and Thailand were included in the analyses reported in abstract O9: 10 were evaluated at week 20, 25 at week 33, and 5 at 4-6 weeks post-partum. No significant difference was found in the AUC of SQV in 9 women who were evaluated at week 20 and 33. None of the 26 women had C12 less than 0.1 mg/L, the minimum value for virologic response that has been defined for SQV. In a very similar study from Spain (abstract O11), the PK of SQV were evaluated in nine HIV-infected pregnant women at weeks 24 and 34, and 6 weeks post partum. 12-hour intensive PK curves were determined at each evaluation. Geometric mean AUC and C12 values were as follows: week 24, 26.7 mg*h/L and 0.61 mg/L; week 34, 14.5 mg*h/L and 0.54 mg/L; and 6 weeks post partum, 43.1 mg*h/L and 1.4 mg/L. Consistent with the findings in abstract O9, none of these 9 women had C12 values less than 0.1 mg/L at any time point evaluated. Both of these studies were consistent in their findings that none of the pregnant women receiving SQV/RTV, 1000/100 twice daily during pregnancy and post partum had C12 values less than 0.1 mg/L; abstract O11 differed from O9 with evidence for a decrease in the SQV AUC between weeks 24 and 34. Abstract O11 provided virologic and safety data showing that all women had viral loads less than 50 copies/mL at week 34 and post partum, and that no side effects related to SQV were observed. Collectively, the authors of these abstracts concluded that a regimen of SQV (tablets) and RTV at 1000/100 mg of each twice daily can be given to pregnant women in the third trimester of pregnancy without dose adjustment. These data certainly provide PK support for the use of the SQV/RTV 1000/100 twice-daily dose in HIV-infected pregnant women, as recommended by the Public Health Service Taskforce for Recommendations for Use of Antiretroviral Drugs in Pregnant HIV-Infected Women (Nov 2, 2007).
An open question for the dosing of PIs in adults is whether all should receive the same dose regardless of body weight, as presently recommended for all of the available PIs. Abstract P10 evaluated a low dose of ATV, given with RTV, in HIV-infected Thai adults. 22 persons on an ATV/RTV (300/100 once daily) based regimen with HIV-RNA less than 50 copies/mL were enrolled in this trial where their ATV dose was reduced to 200 mg with RTV 100 mg, once daily. The median (IQR) weight of the subjects was 59 kg (45.3-66.8). As expected, the concentrations of ATV declined after the dose reduction: median AUC, from 65.4 to 35.5 mg*h/L; and C24 from 1.0 to 0.5 mg/L. No subject had C24 values less than 0.15 mg/L, a value suggested as the minimum value for virologic response. At the start of this study, all subjects were experiencing hyperbilirubinemia. After dose reduction, unconjugated bilirubin decreased from 2.2 to 1.6 mg/dL. This study is suggestive that a lower dose of ATV may be suitable for low body weight persons. However, this study was only a 14-day PK evaluation, and virologic response can't be evaluated. Confirmation of this approach requires prospective evaluation.
The process of finding the right dose of antiretroviral drugs in HIV-infected children, domestically and internationally, remains a significant challenge. These challenges give rise, unfortunately, to differences in dosing recommendations. For example, the FDA approved dose of LPV/r for children 6 months to 12 years of age is approximately 230 mg/m2 (of LPV) twice daily. Some clinicians have chosen to use a higher dose in children of 300 mg/m2 based upon a concern for subtherapeutic concentrations in some children who receive the 230 mg/m2 dose. The World Health Organization (WHO) recommended dose of LPV/r is approximately 300 mg/m2 (of LPV) twice daily. Abstract P16 evaluated the WHO recommended LPV/RTV dose for children and a dose of approximately 220 mg/m2 twice daily. 23 children, 2-18 years of age, were randomized to receive either the WHO usual dose or reduced dose of LPV/RTV. PK evaluations were performed 4-6 weeks after treatment had started. Mean AUC and C12 values for the usual dose were 107 mg*h/L and 5.2 mg/L. For the reduced dose, AUC and C12 values were 84.6 mg*h/L and 3.8 mg/L. One child in the reduced dose group had a C12 value less than 1.0 mg/L, which is the suggested minimum value for LPV plasma concentrations. At week 24 of therapy, 74% of the children had HIV RNA values less than 400 copies/mL. The authors reported no difference between arms in virologic response, but actual values were not given. This abstract is consistent with known data for LPV in children: some children receiving doses of approximately 230 mg/m2 (of LPV) will have subtherapeutic concentrations; in this study that rate was 1 of 12. Unfortunately, this study sheds no new light on what dose of LPV/RTV clinicians should use for children: the FDA recommended dose (approximately 230 mg/m2) or the higher dose of approximately 300 mg/m2 as recommended by the WHO and discussed in the February 2008 version of the Department of Health and Human Services Guidelines for Use of Antiretroviral Agents in Pediatric HIV Infection.
|
| |
|
|
|
|
|
