Pharmacokinetics of Concurrent Administration ofFosamprenavir and Atazanavir without Ritonavir inHuman Immunodeficiency Virus–Negative Subjects

MaryPeace McRae, Pharm.D., Ph.D., Patrick G. Clay, Pharm.D., FCCP,Peter L. Anderson, Pharm.D., and Alan G. Glaros, Ph.D.

Study Objective. To quantify the pharmacokinetics of amprenavir andatazanavir (administered as the prodrug fosamprenavir) alone and incombination in human immunodeficiency virus (HIV)–negative subjects.

Design. Randomized, open-label, three-way crossover study.Setting. Research facility.Participants. Eleven men and 10 women who were seronegative for HIV.Intervention. Subjects were randomized to 14-day treatment periods of

fosamprenavir 1400 mg once/day, atazanavir 400 mg once/day, or fosamprenavir1400 mg plus atazanavir 400 mg once/day; after a washout period of at least 21days between each treatment, they received the other two treatments.

Measurements and Main Results. Subjects underwent 24-hour pharmacokineticsampling at baseline and on day 14 of each treatment period. Primary outcomemeasures were area under the plasma concentration–time curve (AUC) andmaximum concentration (Cmax) for amprenavir and atazanavir. Atazanavirsignificantly enhanced the exposure of amprenavir. When fosamprenavir wasgiven alone, the geometric mean of amprenavir’s AUC was 20.2 µg•hour/ml(95% confidence interval [CI] 19.1–21.2 µg•hr/ml). When given incombination with atazanavir, amprenavir had an AUC of 39.8 µg•hour/ml(95% CI 38.7–40.9 µg•hr/ml). Similarly, the Cmax for amprenavir increasedfrom 4193 ng/ml (95% CI 3927–4459 ng/ml) to 6621 ng/ml (95% CI6427–6814 ng/ml) when given in combination with atazanavir. In contrast,AUC and Cmax for atazanavir significantly decreased when atazanavir wascoadministered with fosamprenavir; AUC decreased from 17.6 µg•hour/ml(95% CI 16.6–18.7 µg•hr/ml) to 11.8 µg•hour/ml (95% CI 11.3–12.3µg•hr/ml), and Cmax decreased from 2507 ng/ml (95% CI 2379–2635 ng/ml) to1832 ng/ml (95% CI 1752–1911 ng/ml). Adverse events were assessed at eachstudy visit and 1 month after the subjects completed the three treatments. Bothdrugs were well tolerated. One serious adverse event (grade 3 acutepancreatitis) occurred and resolved without further incident.

Conclusion. Atazanavir 400 mg/day plus fosamprenavir 1400 mg/daysignificantly decreased concentrations of atazanavir compared withstandard dosing regimens of each drug alone. This dosing scheme is not arecommended combination of dual, fully active protease inhibitors.

Key Words: human immunodeficiency virus, HIV, pharmacokinetics,fosamprenavir, atazanavir, amprenavir.(Pharmacotherapy 2009;29(8):937–942)

PHARMACOTHERAPY Volume 29, Number 8, 2009

Antiretroviral drug combinations that includeprotease inhibitors reduce the morbidity andmortality of persons infected with humanimmunodeficiency virus (HIV).1 Proteaseinhibitors, which are generally substrates ofcytochrome P450 (CYP) 3A, are less effectivewhen administered alone than when given inritonavir-boosted combinations.2 Ritonavirboosting refers to the use of ritonavir to inhibitCYP3A-mediated metabolism of a concomitantlyadministered protease inhibitor. A boosted regimensubstantially raises the plasma concentrations ofthe concomitant protease inhibitor and improvesthe antiviral response.3, 4

Ritonavir boosting is not without limitations,however. Important limitations are the potentialfor deleterious lipid and glucose toxicities,tolerability issues, and serious drug interactionswith ritonavir, even at boosting doses.2 Thecombination of two different protease inhibitorshas been attempted to improve antiviral responseswithout ritonavir boosting. This approach hasbeen reserved mostly for antiretroviral-experiencedpatients who may be infected with a drug-resistant virus. Unfortunately, concomitant useof protease inhibitors has sometimes led tounexpected and deleterious drug interactions. Anotable example is the marked reduction inconcentrations of lopinavir and amprenavir whenlopinavir-ritonavir and fosamprenavir arecoadministered.5

Two potentially favorable protease inhibitorsfor combination are atazanavir and fosamprenavir(which is metabolized to amprenavir on absorp-tion). Both are substrates of CYP3A.6, 7 Atazanaviris an inhibitor and amprenavir is an inhibitor-inducer of CYP3A. As single agents, both drugsare well tolerated, with low glucose and lipidtoxicities, and they have complementary dietaryrequirements. However, given the possibility ofdetrimental interactions, it is imperative to

formally test the pharmacokinetics of fosam-prenavir plus atazanavir before this combinationcan be considered a clinically viable and fullyactive dual protease inhibitor. The objective ofthis study was to quantify the pharmacokineticsof fosamprenavir and atazanavir alone and incombination in HIV-negative subjects.

Methods

Study Design

This was a randomized, open-label, three-way,crossover study. All subjects provided writteninformed consent, as approved by the institutionalreview board, before undergoing any studyprocedures.

Study Participants

Men and women aged 18–65 years who wereseronegative for HIV (confirmed by using OraQuick;OraSure Technologies, Inc., Bethlehem, PA) wereeligible for inclusion. Subjects had to weigh55–95 kg and have a body mass index of 19–30.4kg/m2. Subjects were excluded from participationif they had a history of sulfonamide allergy, anyclinically significant concomitant diagnoses, or asignificant clinical history of alcohol or illicitdrug abuse. Subjects were also excluded if theywere pregnant or if they had any baselinelaboratory values rated greater than grade 2 usingthe grading scale from the Division of AcquiredImmunodeficiency Syndrome (DAIDS), NationalInstitute of Allergy and Infectious Diseases.Hepatitis status was not evaluated.

Subjects had to agree to refrain from usingtobacco or drinking grapefruit juice for 2 weeksbefore and during the study. No concomitantdrug therapy other than occasional use ofacetaminophen and diphenhydramine wereallowed during the study.

Study Protocol

Using a randomization schedule, each subjectwas assigned to one of six sequences consistingof three 14-day treatment periods separated by awashout of at least 21 days (Table 1). Subjectsbegan with oral fosamprenavir 1400 mg once/day,atazanavir 400 mg once/day, or fosamprenavir1400 mg plus atazanavir 400 mg once/day, whichwas switched in subsequent periods so theyreceived all three treatments. Fosamprenavir 700mg (Lexiva; GlaxoSmithKline, Research TrianglePark, NC) was supplied by the manufacturer.

938

From the Departments of Pharmacology andMicrobiology (Dr. McRae) and Basic Medical Sciences (Dr.Glaros), and the Dybedal Clinical Research Center (Dr.Clay), Kansas City University of Medicine and Biosciences,Kansas City, Missouri; and the Department ofPharmaceutical Sciences, University of Colorado Denver,Denver, Colorado (Dr. Anderson).

Financial support provided by GlaxoSmithKline, ResearchTriangle Park, North Carolina.

Presented in part as a poster at 13th Conference onRetroviruses and Opportunistic Infections, Denver,Colorado, February 5–8, 2006.

Address reprint requests to Patrick G. Clay, Pharm.D., FCCP,Dybedal Clinical Research Center, Kansas City University ofMedicine and Biosciences, 1750 Independence Avenue, KansasCity, MO 64106; e-mail: pclay@kcumb.edu.

KINETICS OF FOSAMPRENAVIR PLUS ATAZANAVIR IN HEALTHY SUBJECTS McRae et al

Atazanavir (Reyataz; Bristol-Myers Squibb,Princeton, NJ) was acquired from a localpharmacy. Subjects were instructed to take allstudy drugs in the morning with a light meal.

Subjects underwent intensive 24-hourpharmacokinetic sampling at the start of eachtreatment period (baseline) and on day 14 ofeach treatment period. Subjects were instructednot to take any acetaminophen and/ordiphenhydramine during the 3 days before eachpharmacokinetic visit.

On the day of the pharmacokinetic visit,subjects were instructed to fast for 10 hoursbefore their appointment time. At the researchfacility, the morning dose of the drug wasadministered after a standardized meal was eaten,and the time was recorded. Blood and urine werecollected for laboratory safety analysis and formeasurement of trough plasma concentrations.Blood samples were taken at 1, 2, 3, 4, 6, 8, 12,14, and 24 hours after drug administration.Individual blood samples were collected in tubescontaining 6 ml of sodium citrate. The sampleswere immediately centrifuged, and plasma wasisolated within 1 hour of collection into two 1-mlcryovials. Samples were stored at -80ºC untilassayed.

Plasma concentrations of amprenavir andatazanavir were determined by using a validated,simultaneous procedure for high-performanceliquid chromatography. In brief, 0.2 ml of plasmawas extracted with tert-butylmethyl ether afterthe addition of 0.4 ml of 0.0125 M sodiumhydroxide and an internal standard (A86093;Abbott Laboratories, Abbott Park, IL).Chromatographic separation was accomplishedby using a 3-µm, 100 x 4.6-mm column (WatersCorp., Milford, MA) at 35ºC with a mobile phaseof acetonitrile and 20-mM acetate buffer. Thelimit of quantitation was 20 ng/ml for both

compounds. Within- and between-day coefficientsof variation for the quality controls duringvalidation were less than 15%. All samples froma given subject were assayed in the sameanalytical run.

All subjects were physically examined andmedically interviewed by a clinician at screeningand each visit. Laboratory parameters obtainedat baseline were reevaluated at each visit. Inaddition, adverse events were assessed byadministering a questionnaire during each visit.The DAIDS grading scale was used to documenttoxicity.8 Any toxicities of grade 3 or higher wereverified within 1 week; if they recurred, thesubject was withdrawn from the study. Acomprehensive safety visit was conducted 1month after the subjects completed all threestudy periods.

Adherence to the study drug regimen wasassessed by using questionnaires and pill countsat each study visit.

Pharmacokinetic and Statistical Analysis

Noncompartmental pharmacokinetic analysiswas performed by using WinNonLin software,version 5.2 (Pharsight Corp., Mountain View,CA). Primary outcome measures were area underthe plasma concentration–time curve (AUC) andmaximum concentration (Cmax) over the dosinginterval for amprenavir and atazanavir. In addition,minimum concentration (Cmin) was calculated.

The sample size was calculated to estimate thenumber of subjects needed to detect a meandifference of 30% between experimentalconditions. With � set at 0.05, a final sample of16 subjects provided sufficient power (≥ 0.80) todetect a significant effect for Cmax and AUC.

Geometric least squares means andcorresponding 95% confidence intervals (CIs)were calculated for each group. The Wilcoxonsigned rank test was used to compare variouspharmacokinetic parameters between groups.Subjects’ pharmacokinetic data were included foreach complete period. If a subject did notcomplete an entire period, only the safety data forthat period was used.

Results

Twenty-one healthy volunteers completed 61of 63 treatment periods. After participating inthe first two treatment arms of the study, onesubject withdrew participation because of jobrelocation. Another study participant completedonly two of three pharmacokinetic visits because

939

Table 1. Treatment SequencesPeriod 1 Period 2 Period 3

Sequence (14 days) (14 days) (14 days)1 A B C2 B C A3 C A B4 A C B5 B A C6 C B A

Each subject was randomized to one of these six sequences.Treatments A, B, and C were fosamprenavir 1400 mg once/day,atazanavir 400 mg once/day, and fosamprenavir 1400 mg +atazanavir 400 mg once/day, respectively. Each treatment wasseparated by a washout period of at least 21 days.

PHARMACOTHERAPY Volume 29, Number 8, 2009

of failure to appear for sampling on day 14.Mean adherence to the study regimen was above99%. Subject demographics are provided inTable 2. The participants were well balanced forsex and race.

Figures 1 and 2 graphically present thegeometric means for plasma concentration–timedata for amprenavir and atazanavir across thetreatment arms. For atazanavir alone, the AUCover the dosing interval (AUC0–�) was 17.6µg•hour/ml (95% CI 16.6–18.7 µg•hr/ml) anddeclined by 33% when the drug was given withfosamprenavir (11.8 µg•hr/ml, 95% CI 11.3–12.3, p<0.05; Table 3). In addition, the Cmax foratazanavir decreased from 2507 ng/ml (95% CI2379–2635 ng/ml) when taken alone to 1832ng/ml (95% CI 1752–1911 ng/ml) when combinewith fosamprenavir; this change represented a27% decrease in Cmax (p<0.05). In contrast,concentrations of amprenavir rose when fosam-prenavir and atazanavir were given together. TheAUC0–� and Cmax for amprenavir increased by97% and 58%, respectively (both p<0.05). The

AUC0–� increased from 20.2 µg•hour/ml (95% CI19.1–21.2 µg•hr/ml) to 39.8 µg•hour/ml (95%CI 38.7–40.9 µg•hr/ml), and Cmax increased from4193 ng/ml (95% CI 3927–4459 ng/ml) to 6621ng/ml (95% CI 6427–6814 ng/ml). The Cminvalues significantly differed for all comparisongroups (p<0.05).

Clinical adverse events such as pruritus, rash,gastrointestinal upset (diarrhea, abdominal pain,or nausea), headache, and oral paresthesias areshown in Table 4. They occurred at rates similaror lower than those observed when the agentswere given individually compared with historicalcontrols.6, 7

All clinical events (except serious adverseevents) were grade 2 or lower, required notreatment, and resolved without interruption ofstudy drugs. Only one serious adverse eventoccurred during the study. On day 3 of dosingfor fosamprenavir, a subject developed acutepancreatitis secondary to excessive alcoholingestion. This event required 24-hour observationof the subject at a local hospital, and the eventwas not deemed to be related to the study drug.The subject recovered and resumed dosing onday 5 without initially informing the study staffof the event; the subject had no recurrence of theepisode. Eleven subjects developed grade 3 or 4,transient, and self-limiting hyperbilirubinemiawhile they were receiving atazanavir, either aloneor in combination. All their laboratory valuesreturned to baseline by the start of the nexttreatment period and when they were evaluatedat the 1-month (mean ± SD 28 ± 3 days) safetyvisit after the study.

940

Table 2. Demographics of the 21 Study Subjects

Characteristic ValueAge, mean ± SD (yrs) 31.7 ± 9.5

No. (%) of SubjectsSex

Male 11 (52)Female 10 (48)

Race-ethnicityCaucasian 11 (52)African-American 9 (43)Hispanic 1 (5)

Figure 1. Amprenavir plasma concentration–time data(geometric means) across treatment arms.

Time (hour)

10 150 5 20 25

Am

pre

navi

rC

once

ntra

tion

(ng/

ml)

0

1000

2000

3000

4000

5000

Fosamprenavir alone

Fosamprenavir + atazanavir

Figure 2. Atazanavir plasma concentration–time data(geometric means) across treatment arms.

Time (hour)

0 5 10 15 20 25

Ata

zana

vir

Con

cent

ratio

n(n

g/m

l)

0

500

1000

1500

2000

2500

Atazanavir alone

Atazanavir + fosamprenavir

KINETICS OF FOSAMPRENAVIR PLUS ATAZANAVIR IN HEALTHY SUBJECTS McRae et al

Discussion

Overall, the combination of fosamprenavir andatazanavir was well tolerated, with minimaladverse effects or toxicity. Because ofinteractions involving CYP enzyme activity, drugtransporter (P-glycoprotein) activity, and proteinbinding, drug interactions between proteaseinhibitors can be complex,9 as we observed inour study. Our data indicated that atazanavirmodestly but significantly boosted exposure toamprenavir. Increases in AUC, Cmax, and Cminfor amprenavir given with atazanavir were 97%,58%, and 297%, respectively. Fosamprenavir1400 mg once/day given with atazanavir resultedin an AUC for amprenavir of 39.8 µg•hour/ml(95% CI 38.7–40.9 µg•hr/ml) and a Cmax foramprenavir of 6.6 µg/ml (95% CI 6.4–6.8 µg/ml).These values were greater than those achieved inhistorical control subjects given unboostedfosamprenavir 1400 mg twice/day (33 µg•hr/ml,95% CI 27.6–39.2 µg•hr/ml) and 4.82 µg/ml(95% CI 4.06–5.72 µg/ml).7 However, the Cminfor amprenavir was 0.24 µg/ml (95% CI0.20–0.28 µg/ml) when atazanavir plusfosamprenavir was administered. These levels

were lower than historical control results forunboosted fosamprenavir given twice/day (0.35µg/ml, 95% CI 0.27–0.46 µg/ml).7

Of importance, concentrations of atazanavirdecreased when it was given in combination withfosamprenavir. For atazanavir, values of AUC,Cmax, and Cmin were reduced by 33%, 27%, and60%, respectively, when the drug was given incombination with fosamprenavir. The Cmin of 60ng/ml (95% CI 53–67 ng/ml) for atazanavir thatwe observed was well below the target Cmin

recommended for antiretroviral-naïve patients(150 ng/ml).2

These findings are consistent with those of afew small investigations. In one pharmacokineticstudy of the interaction between fosamprenavir(boosted with ritonavir and unboosted) andatazanavir, three ritonavir-intolerant subjectstook fosamprenavir 700 mg twice/day andatazanavir 400 mg once/day.10 Because of the lowatazanavir Cmin in the group, the investigatorsincreased the dosage of atazanavir from 400 mgonce/day to 400 mg twice/day. In an investi-gation of the pharmacokinetic interactionsbetween atazanavir plus fosamprenavir plus

941

Table 3. Steady-State Pharmacokinetic Parameters for Atazanavir and Amprenavir

Fosamprenavir +Parameter Atazanavir Alone Fosamprenavir Alone AtazanavirCmax (ng/ml)

Atazanavir 2507 (2379–2635) 1832 (1752–1911)Amprenavir 4193 (3927–4459) 6621 (6427–6814)

Cmin (ng/ml)Atazanavir 148 (130–167) 60 (52–67)Amprenavir 60 (53–67) 238 (200–275)

AUC (µg•hr/ml)Atazanavir 17.6 (16.6–18.7) 11.8 (11.3–12.3)Amprenavir 20.2 (19.1–21.2) 39.8 (38.7–40.9)

Data are geometric means (95% confidence intervals).Cmax = maximum concentration; Cmin = minimum concentration; AUC = area under the concentration-timecurve.

Table 4. Most Common Clinical Adverse Events

Abdominal OralVariable Pain Pruritus Rash Headache Paresthesia Diarrhea NauseaNo. (%) of all adverse events 12 (9.7) 12 (9.7) 10 (8.1) 9 (7.3) 13 (10.5) 15 (12.1) 9 (7.3)(n=124)

No. of Adverse EventsAtazanavir alone 3 1 1 3 1 4 1Fosamprenavir alone 5 7 7 3 4 6 4Fosamprenavir + atazanavir 2 2 1 2 5 3 2Before each treatment period 2 2 1 1 3 2 2

and at 1-mo follow-up

PHARMACOTHERAPY Volume 29, Number 8, 2009

ritonavir, AUC and Cmax values for atazanavirdecreased by 22% and 24%, respectively,compared with values observed in an atazanavir-plus-ritonavir arm.11

The interaction between atazanavir andfosamprenavir is likely due to an induction ofhepatic CYP3A4 protein levels by fosamprenavir,which decreases plasma concentrations ofatazanavir. Another possibility is that protein-binding displacement interactions between thesetwo protease inhibitors could lower total drugconcentrations of atazanavir. Atazanavir andamprenavir, respectively, are more than 86% and90% bound to �1-acid glycoprotein andalbumin.6, 7 We did not measure unbound drugconcentrations, which are needed to determine ifprotein binding plays a role in the interaction.

Conclusion

Atazanavir 400 mg once/day plus fosam-prenavir 1400 mg/day significantly decreasedconcentrations of atazanavir compared withstandard dosing regimens of each drug alone.Therefore, this dosing scheme is not a recom-mended combination of dual, fully activeprotease inhibitors. Perhaps with different dosesor dosing schedules, the combination could beused for ritonavir-free boosting of amprenavirconcentrations. However, these dosing schemesshould not be used without additional study.

Acknowledgments

The authors would like to acknowledge and thankthe staff of the Kansas City Free Health Clinic, Patrick

F. Smith, Pharm.D., and A. David Lein, M.D., for theircontributions.

References1. Palella FJ Jr, Chmiel JS, Moorman AC, Holmberg SD.

Durability and predictors of success of highly activeantiretroviral therapy for ambulatory HIV-infected patients.AIDS 2002;16(12):1617–26.

2. Department of Health and Human Services Panel onAntiretroviral Guidelines for Adults and Adolescents.Guidelines for the use of antiretroviral agents in HIV-1-infectedadults and adolescents. Rockville, MD: U.S. Department ofHealth and Human Services; November 3, 2008. Available fromwww.aidsinfo.nih.gov/ContentFiles/Adultand AdolescentGL.pdf.

3. de Maat MM, Ekhart GC, Huitema AD, Koks CH, Mulder JW,Beijnen JH. Drug interactions between antiretroviral drugs andcomedicated agents. Clin Pharmacokinet 2003;42(3):223–82.

4. Walmsley S, Bernstein B, King M, et al. Lopinavir-ritonavirversus nelfinavir for the initial treatment of HIV infection. NEngl J Med 2002;346(26):2039–46.

5. Kashuba AD, Tierney C, Downey GF, et al. Combiningfosamprenavir with lopinavir/ritonavir substantially reducesamprenavir and lopinavir exposure: ACTG protocol A5143results. AIDS 2005;19(2):145–52.

6. Bristol-Myers Squibb Company. Reyataz (atazanavir sulfate)prescribing information. Princeton, NJ; 2008.

7. GlaxoSmithKline. Lexiva (fosamprenavir calcium) prescribinginformation. Research Triangle Park, NC; 2008.

8. Division of Acquired Immunodeficiency Syndrome. Table forgrading the severity of adult and pediatric adverse events.Rockville, MD: National Institute of Allergy and InfectiousDiseases; December 28, 2004. Available from www3.niaid.nih.gov/LabsAndResources/resources/DAIDSClinRsrch/PDF/DAIDSAEGradingTable.pdf.

9. Piscitelli SC, Gallicano KD. Interactions among drugs for HIVand opportunistic infections. N Engl J Med 2001;344(13):984–96.

10. Khanlou H, Bhatti L, Farthing C. Interaction betweenatazanavir and fosamprenavir in the treatment of HIV-infectedpatients. J Acquir Immune Defic Syndr 2006;41(1):124–5.

11. Wire MB, Shelton MJ, Lou Y, Agarwala S, Child M, Min SS.The pharmacokinetic interaction between fosamprenavir/ritonavir and atazanavir in healthy adult subjects (APV10018).Presented at the 10th European AIDS conference, Dublin,Ireland, November 17–20, 2005. Available from www.natap.org/2005/EACS/eacs_13.htm.

942