Drug Interactions and Pharmacokinetics of Antiretrovirals (Part 1)
Stephen Piscitelli, Pharm.D.
Associate Director, Clinical Pharmacology
Adding NNRTI to boosted PI regimen: SQV/RTV, APV "908" prodrug, IDV/RTV
Allopurinol as a DDI booster
APV + Kaletra
Methadone & LAAM: delavirdine, nelfinavir, RTV/SQV once daily
Issues in clinical pharmacology continue to be a key aspect of antiretroviral therapy. Two poster sessions were devoted to studies that characterized drug interactions and pharmacokinetics. However, many of the studies at the 41st ICAAC were confirmations of previously reported data or studies that would not appear on the surface to have clinical relevance. Compared to the past five years, the number of important papers appeared to be lower than normal. This may be due to the large knowledge base we have already acquired, resulting in fewer important interactions that remain to be studied, or it may be due to the fact that there are fewer new drugs released within the past year for which pharmacokinetic data still need to be presented. Finally, some investigators may have chosen to skip ICAAC and focus their most clinically important work for presentation at the upcoming Retrovirus Conference.
Adding an NNRTI to a "Boosted" PI
The combination of an NNRTI and dual PIs is becoming increasingly common, especially in treatment-experienced patients. A number of abstracts evaluated the addition of efavirenz (EFV) to a RTV-boosted PI. Piliero and colleagues evaluated the effect of EFV on the pharmacokinetics of ritonavir (RTV) and saquinavir (SQV).1 Healthy volunteers received 10 days of SQV/RTV 400/400 mg and then added EFV 600 mg QHS for 14 days. There was an insignificant 10% decrease in the saquinavir mean trough after treatment with EFV. Of note, there were significant increases in cholesterol and triglycerides between baseline and day 24 of the study. Mean cholesterol increased by 56 mg/dl and triglycerides increased by 201 mg/dl. A similar study with comparable results was previously reported at the 7th CROI in a small number of HIV-infected patients.2 Both studies demonstrate that SQV/RTV 400/400 mg regimens are not significantly altered by the CYP3A4 induction effects of efavirenz.
New Amprenavir Prodrug "908" + Efavirenz
A similar study at this meeting evaluated GW433909, a prodrug of amprenavir
(APV) given once daily in combination with RTV (200 or 300 mg QD) in healthy
volunteers.3 After 14 days of dual PI therapy, EFV 600 mg QD was added for
an additional 14 days. As above, the addition of EFV did not produce any
significant changes in the APV mean trough concentration or AUC in the 300 mg
RTV group. A decrease in the trough of approximately 40% was noted in the 200
mg group. However, 25% of the subjects were discontinued due to adverse
effects (n=7) or lost to follow-up (n=1). This study does provide support
for once daily dosing of 908 when combined with RTV, and suggests that the
addition of EFV will not decrease APV levels. The toxicity in this study
raises some concerns, although healthy volunteers often do not tolerate
antiretroviral drugs as well as HIV-infected patients. Side effects included
diarrhea, oral numbness/tingling, mild CNS effects and rash. It is unclear
whether toxicity was related to high peak concentrations or some other
pharmacokinetic correlate. Follow-up studies in HIV-infected patients will
further define the tolerability of this once daily dosing regimen.
However, in twice daily 908+EFV regimen a dose of 700/100 is preliminarily
recommended with or without EFV. It appears an extra
100mg RTV is required only in the QD regimen. But, as I say studies are
ongoing and this is preliminary information. Preliminary studies including the one above suggest that when 908 plus Retonovir is used with EFV once daily for patients with PI resistance or resistance to APV, 300mg of Retonovir appears needed rather than 200mg. In PI naive patients, a once daily regimen of 1400mg of 908 plus 200mg RTV is being studied and preliminarily recommended.
A third abstract from Moreno and colleagues in Spain examined 27 patients receiving three different RTV/IDV regimens of 400/400 mg, 800/100 mg, or 800/200 mg.4 Sixteen of these patients were taking concomitant NNRTIs and 11 received only the dual PIs. The investigators examined trough concentrations and inhibitory quotients (Cmin divided by the protein corrected IC95) between the two groups. No significant differences were noted in the pharmacokinetics of IDV or RTV between the two groups. Although they stated that no differences in IDV levels or IQs were seen in any of the three dosing schedules, the sample size was likely too small to detect even a moderate change.
These abstracts confirm a number of previous studies that demonstrate how dual PI regimens containing RTV can prevent NNRTI-induced drug interactions. Instead of a significant decrease in PI levels when EFV or nevirapine is added to a sole PI, the presence of even low doses of RTV can avoid a clinically significant interaction. However, the available data clearly show that at least 300-400 mg per day of RTV is necessary.5,6 Studies of dual PI regimens that include only 200 mg per day of RTV have consistently shown that NNRTIs still decrease the PI concentrations.7,8 Although the study by Moreno did not show this, it was not designed properly to do so nor was the sample size large enough to show this. The message for clinicians is that EFV and NVP can be used with dual PI regimens, but 200 mg BID of RTV is required to avoid a decrease in the second protease inhibitor.
DDI and Tenofovir
An interaction between didanosine and tenofovir was reported at the IAS meeting earlier this year in Buenos Aires.9 Tenofovir increased DDI exposure in healthy volunteers by approximately 40%. Trough levels, peak concentrations and half-life were not significantly changed. There was much debate as to the clinical relevance of this data. Plasma levels of nucleoside RT inhibitors generally do not correlate with efficacy or toxicity since they are prodrugs that must be phosphorylated intracellularly to their active form. In an extension of this study, data from two, 24-week placebo-controlled tenofovir trials, which included 197 subjects receiving DDI, were examined.10 Comparisons in toxicity between groups receiving DDI+TDF and DDI+placebo showed no difference in the incidence of pancreatitis, neuropathy, or increase in amylase. While this analysis does not rule out the possibility of increased DDI toxicity from higher levels with TDF, it does provide some reassurance that there is not an obvious safety problem from this combination. Additional data from Phase IV studies will add additional insight into this interaction and its clinical relevance. Clinicians should not decrease the dose of DDI when using tenofovir unless signs and symptoms of DDI toxicity are present.
Allopurinol as a DDI booster?
Liang and colleagues presented a study with seven HIV-infected patients that demonstrated an increase in DDI absorption by the administration of allopurinol, which inhibits a step in DDI metabolism.11 These investigators had presented a similar study at a previous meeting in healthy volunteers and were now confirming their results in patients.12 After a single 400 mg DDI dose, patients received 7 days of allopurinol (300 mg) and then a second 400 mg dose of DDI. There was a 2.3-fold increase in both the Cmax and the AUC of DDI. The percent urinary recovery of DDI increased from 7.7% to 18.3%, demonstrating that allopurinol inhibits presystemic metabolism of DDI and enhances its oral systemic bioavailability. While these data are intriguing, they again raise the question of what do plasma levels of NRTIs really mean and would such an intervention actually lead to clinical benefit or just increased DDI toxicity? In addition, allopurinol can cause rash, nausea and other side effects that may overlap with antiretroviral toxicities causing a complicated clinical picture for the patient and physician. At this point, this remains a strategy that is scientifically interesting but not likely to be useful in the clinic.
APV + Kaletra: Is there an interaction here?
There has been much recent debate regarding the potential interaction between Kaletra (lopinavir/ritonavir) and APV. Data from the recent HIV Clinical Pharmacology meeting in Noordwijk, the Netherlands, suggested that there may be a detrimental interaction between these three protease inhibitors.13 Investigators from France examined LPV and APV levels in four groups of patients. Kaletra alone (n=11), APV/RTV 600/100 mg (n=36), Kaletra + APV 600 mg (n=10), and Kaletra + APV 750 mg (n=5).14 Median trough concentrations of APV 600 mg were decreased from 1755 ng/ml when given with RTV alone to 778 ng/ml when given with Kaletra. Also, lopinavir troughs were decreased from 3326 ng/ml alone to 2226 ng/ml with 600 mg APV and 1716 ng/ml with 750 mg APV. These data are suggestive of a bi-directional drug interaction lowering both PIs, however this study has limitations in both its sample size and its design. A comparison of concentrations across different patient groups is limited by a number of factors including immunologic status, concomitant meds, diet, and demographics of the groups being compared. The previous study also compared different patient groups or compared the data from the combined Kaletra+ APV group to historical controls. These two studies raise the question that there may be some reduction in PI levels perhaps by induction of metabolism or other mechanisms. Two studies with cross-over designs are currently underway and may help to finally sort out this interesting interaction. From a clinical standpoint, the concentrations of both APV and LPV in this study were still adequate and well above the IC50 values, even for many resistant HIV strains.
Interactions with Methadone or LAAM: delavirdine, nelfinavir, saquinavir/ritonavir once daily
The potential interactions between antiretrovirals and methadone are important due to the large population of HIV-infected subjects who are addicted to heroin. Many methadone interactions are large in magnitude and clinically relevant, such as the marked decrease in methadone levels that are seen with nevirapine and efavirenz. Such interactions can precipitate withdrawal symptoms and warrant close monitoring.
Investigators from Buffalo Institution? presented three separate abstracts that examined the effects of either methadone or the concomitant antiretroviral. The first abstract by Booker and colleagues demonstrated the pharmacokinetics of delavirdine, and its metabolite N-delavirdine, in 16 non HIV-infected subjects maintained on methadone and 15 non-opioid dependent controls.15 While large interpatient variability was observed -- as is common with most antiretrovirals -- no significant differences in DLV or its metabolite were observbed between the two groups. While DLV is not a commonly used antiretroviral, it is being considered in some cases as an alternative to ritonavir as a pharmacokinetic booster.
Smith et al evaluated the effect of methadone and LAAM, a long acting methadone product, on the pharmacokinetics of nelfinavir and its active metabolite, M8.16 A previous study demonstrated that nelfinavir did not significantly affect methadone thus, the other direction of this interaction was evaluated. Three groups of patients were studied; one maintained on methadone, one on LAAM, and a non-dependent controls.
Nelfinavir troughs and AUC were modestly higher in those patients on methadone compared to control, while M8 exposure was lower. M8 AUC was higher in the LAAM group but NFV AUC was similar.
Methadone interactions are very complicated as it is metabolized by various pathways and possibly has some effect on CYP3A4 and other isozymes. Nelfinavir and the formation of its metabolite also are complex making this study very difficult to understand from a mechanistic standpoint. Another complicated issue is how to interpret the M8 concentrations. It remains unclear as to the activity of the M8 metabolite and how it should be considered in pharmacokinetic data. Some studies simply add the NFV AUC and the M8 AUC while others only consider the parent drug and still other studies add only a portion of the M8 AUC. Overall, it appears that neither methadone nor LAAM would alter nelfinavir concentrations to any degree that would be clinically relevant. No dosage adjustment of nelfinavir is necessary based on these data.
The third abstract from Mark Shelton and colleagues evaluated SQV/RTV 1600/100 mg once daily in non HIV-infected volunteers maintained on methadone.17 Volunteers had pharmacokinetic sampling before and after 14 days of SQV/RTV. Unbound methadone (not protein bound) R and S isomers were evaluated as well as the pharmacokinetics of both PIs. No significant changes in methadone unbound isomers were observed. The authors noted that SQV concentrations were lower than those anticipated in patients, although most still had concentrations above 50 ng/ml, a value that is considered to be above the IC50 for wild type virus. Of note, healthy volunteers have been reported to demonstrate approximately half the blood levels of SQV compared to HIV-infected patients. Although the mechanism of this phenomenon is not well described, this may have been a reason for the lower-than-expected exposures. These data suggest that even boosting SQV still results in lower concentrations in healthy volunteers than in patients.
Studies that examine methadone interactions in HIV-infected patients are difficult to perform and these investigators should be commended. Such studies are also difficult to interpret since the methadone doses are usually quite variable (40-120 mg/day in these studies) and patients may be receiving other supportive medications that may complicate the findings. These three abstracts suggest that methadone and LAAM can be given safely to patients receiving delavirdine, nelfinavir, and SQV/RTV. Since interactions can be bi-directional, patients receiving methadone should always be monitored for signs of withdrawal and have doses adjusted as needed.
1. Piliero PJ, Preston SL, Japour A, et al. Pharmacokinetics of the combination of ritonavir plus saquinavir with and without efavirenz in healthy volunteers. 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, IL, September 22-25, 2001, abstract A-495.
2. Hendrix CW, et al. Pharmacokinetics of the triple combination of saquinavir, ritonavir, and efavirenz in HIV positive patients. 7th Conference on Retroviruses and Opportunistic Infections, San Francisco, CA, Jan 30-Feb 2, 2000, abstract 79.
3. Wire MB, Preston SL, Ballow C, et al. An assessment of plasma amprenavir pharmacokinetics following administration of two GW433908 and ritonavir QD regimens in combination with efavirenz in healthy adult subjects. 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, IL, September 22-25, 2001, abstract I-1737.
4. Moreno A, Casado JL, Marti-Belda P, et al. Concomitant use of non nucleoside reverse transcriptase inhibitors does not decrease the inhibitory quotient of dual ritonavir/indinavir based therapy. 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, IL, September 22-25, 2001, abstract I-1728.
5. Degan O, Kurowski M, Van Lunzen J, et al. Amprenavir and ritonavir: intraindividual comparison of different doses and influence of concomitant NNRTI on steady-state pharmacokinetics in HIV-infected patients. 8th Conference on Retroviruses and Opportunistic Infections, Chicago, IL, Feb 4-8, 2001, abstract 739.
6. Piscitelli SC, Bechtel C, Sadler B, et al. The addition of a second protease inhibitor eliminates amprenavir-efavirenz drug interactions and increases plasma amprenavir concentrations. 7th Conference on Retroviruses and Opportunistic Infections, San Francisco, CA, Jan 30-Feb 2, 2000, abstract 78.
7. Kaletra product information, Abbott Laboratories, Abbott Park, IL, 2001.
8. Aarnoutse RE, Burger DM, Hugen PWH, et al. A pharmacokinetic study to investigate the influence of efavirenz on a BID indinavir / ritonavir regimen in healthy volunteers. 40th Interscience Conference on Antimicrobial Agents and Chemotherapy. Toronto, Canada, September 17-20, 2000, abstract 423.
9. Kearney B, Flaherty J, Sayre J, et al. A multiple dose randomized crossover drug interaction study between tenofovir DF and lamivudine or didanosine. 1st IAS Conference on HIV Pathogenesis and Treatment. Buenos Aires, Argentina, July 8-11, 2001, abstract 337.
10. Flaherty J, Kearney B, Wolf J, et al. Coadministration of tenofovir DF and didanosine: a pharmacokinetic and safety evaluation. 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, IL, September 22-25, 2001, abstract I-1729.
11. Liang D, Breaux K, Rodriguez-Barradas M, et al. Allopurinol increases didanosine absorption in HIV-infected patients. 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, IL, September 22-25, 2001, abstract A-498.
12. Liang. Allopuroinol in healthy volunteers
13. Peytavin G, Lamotte C, Duval X, et al. Amprenavir plasma concentrations are dramatically decreased by the association with ABT378/r in HIV-infected patients. 2nd International Workshop on Clinical Pharmacology of HIV Therapy. Noordwijk, the Netherlands, April 2-4, 2001, abstract 1.14.
14. Meynard JL, Poirer JM, Guiard-Schmid JB, et al. Impact of ABT 378/r on the amprenavir plasma concentrations in HIV-experienced patients treated by the association APV-ABT 378/r. 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, IL, September 22-25, 2001, abstract I-1736.
15. Booker B, Smith P, Forrest A, et al. Lack of effect of methadone on the pharmacokinetics of delavirdine & N-delavirdine. 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, IL, September 22-25, 2001, abstract A-490.
16. Smith PF, Booker BM, Difrancesco R, et al. Effect of methadone or LAAM on the pharmacokinetics of nelfinavir and M8. 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, IL, September 22-25, 2001, abstract A-491.
17. Shelton MJ, Cloen D, Berenson C, et al. Pharmacokinetics of once daily saquinavir ritonavir: effects on unbound methadone and alpha-1 acid glycoprotein. 41st Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago, IL, September 22-25, 2001, abstract A-492.