To the Editors:
Atripla is the first fixed-dose combination (FDC) available in the United States to combine 2 different classes of antiretroviral drugs in a single pill, which can be taken once daily. This tablet combines the active ingredients of efavirenz, a nonnucleoside reverse transcriptase inhibitor, with emtricitabine, a nucleoside reverse transcriptase inhibitor and tenofovir disoproxil fumarate, a nucleotide reverse transcriptase inhibitor.1 The tablet may be taken alone or in combination with other antiretroviral products for the treatment of HIV-1 infection in adults.2 A major advantage of this FDC product is its ability to reduce pill burden, which may increase patients' adherence to an antiretroviral regimen.3 However, administration of Atripla is limited to patients who can swallow a tablet. A liquid formulation may benefit those unable to swallow tablets, yet no such formulation is commercially available. The objective of this study was to determine the relative bioequivalence of tenofovir, emtricitabine, and efavirenz in a compounded oral liquid formulation compared with the commercially available tablet formulation.
The study was conducted at the University of Alabama at Birmingham in Birmingham, Alabama. The institutional review board approved the study, and written informed consent was obtained before participation. HIV-seronegative men and women between 19 and 65 years of age who were in good health based upon medical history, physical examination and laboratory screening were eligible to participate. Exclusion criteria were pregnancy, breastfeeding, any concomitant medications, serum creatinine greater than 1.2 mg/dL, and liver function tests greater than 3 times the upper limit of normal.
Subjects were randomized to a single-dose, open-label, 2-period, cross-over bioequivalence study. The reference formulation was a FDC Atripla (Bristol-Myers Squibb, Princeton, NJ and Gilead, Foster City, CA) tablet consisting of 600-mg efavirenz, 200-mg emtricitabine, and 300 mg of tenofovir disoproxil fumarate, (FDN078A, lot number, July, 2010, expiration date). The test formulation was a compounded oral liquid formulation of Atripla tablet crushed dissolved in 5 mL of water and diluted to 20 mL with Ora-Sweet oral vehicle (Paddock Laboratories, Minneapolis, MN). The dilution process consisted of rinsing the pestle with 15 mL of the Ora-Sweet to include any residue from fillers in the crushed tablet. The 20-mL solution was drawn up into an oral syringe. Based upon prior internal stability data, the solution was prepared within 24 hours of administration to ensure drug stability in solution. Subjects arrived to the research center on day 1 of the study after an overnight fast beginning at midnight and lasting until 4 hours after taking the assigned antiretroviral formulation(s). A saline lock was placed in an arm, and blood samples were collected predose (time = 0), 0.25, 0.5, 1, 2, 4, 6, 8, and 12 hours after the dose. Subjects returned to the research clinic for additional blood samples at 24 and 48 hours after the dose. Subjects returned to the research center after a 14-day washout period to receive the alternative antiretroviral formulation followed by the same study procedures and blood sampling as described on day 1.
The sample size of 16 was based upon a 10% dropout rate (ie, due to lost to follow-up, treatment discontinuation, etc). Because we were expecting 2 subjects not to complete the study, we expected 14 evaluable subjects. If the discontinuation rate was greater than 10%, we would continue to enroll until 14 evaluable subjects were attained. The target sample size for the study was based upon the Food and Drug Administration (FDA) guidance on bioequivalence testing.4 Any participant who dropped out of the study was not included in the pharmacokinetic and relative bioequivalence analyses.
Efavirenz plasma concentrations were quantitated using an internally validated and externally cross-validated high-performance liquid chromatography with ultraviolet detection method. The assay is linear over a concentration range of 50-20,000 ng/mL. Plasma concentrations of tenofovir and emtricitabine were determined by a validated liquid chromqtogrqphy-tandem mass spectrometry assay. The assay is linear in the range of 10-5000 ng/mL.
Noncompartmental methods (WinNonLin Professional, Version 5.2; Pharsight Corporation, Mountain View, CA) were used to determine each subject's area under the concentration time curve (AUC) from 0 to last measured time point (AUCt), AUC from 0 to infinity (AUC∞), half-life (t ½), and apparent oral clearance (CL/F) for the test and reference formulations of tenofovir, emtricitabine, and efavrienz. Maximum plasma concentration (C max) and time to C max (T max) were determined from each subject's concentration-time curve data. Average bioequivalence was determined for each antiretroviral drug using the pharmacokinetic parameters C max and AUC∞. The WinNonLin Professional Bioequivalence Program for Two-Period Crossover Studies, Version 5.2 was used to determine the ratio of the test to reference formulation mean C max and AUC∞. Individual pharmacokinetic data were natural log-transformed according to FDA recommendations and the means and standard deviations calculated (Microsoft Excel 2007; Microsoft Corporation, Redmond, WA). The ratio of the test to reference formulation geometric mean C max and AUC∞ for each drug and the 90% confidence intervals around each geometric mean ratio were determined. Relative bioequivalence was met if the 90% confidence intervals around the C max and AUC∞ geometric mean ratios for each drug all fell within the FDA's predefined limits of 0.80 to 1.25.
Sixteen HIV-seronegative participants were enrolled between March and August 2009. One male dropped out of the study due to a scheduling conflict at the second study visit and 1 female dropped out of the study due to nausea and vomiting after the tablet administration during the first study visit. No other adverse side events were reported. Data from 6 men, all white and 8 women, 5 white and 3 African American were included in the final analysis. Mean ± standard deviation age and weight for 14 subjects completing the study were 33.3 ± 10.9 years and 85.7 ± 18.4 kg, respectively. Table 1 lists the geometric mean ratio and 90% CI C max and AUC 0-∞ for all antiretrovirals in both formulations.
Mean (± SD) plasma pharmacokinetics of efavirenz, emtricitabine, and tenofovir were similar between the test and reference formulations in 14 men and women. Study medications were well tolerated with the exception of gastrointestinal complains in one subject leading to inability to complete the study. The 90% CI for emtricitabine C max and AUC fell within the range of 0.8-1.25. Thus, bioequivalence was met. The 90% CI for efavirenz C max fell below the range of bioequivalence, whereas efavirenz AUC∞ fell slightly above the range. Tenofovir C max and AUC∞ fell above the range. Tenofovir C max and AUC∞ were approximately 40% and 20% higher, respectively, with the liquid formulation.
These data suggest that the liquid formulation is not bioequivalent to the tablet formulation for 2 of the 3 antiretroviral agents. The study results are clinically important for HIV-infected adults who cannot swallow tablets because a liquid formulation of the FDC product is not commercially available. Data are needed to determine if crushing the tablet and diluting with a palatable liquid alters drug exposure of each antiretroviral component. Ora-sweet was chosen as the liquid vehicle because it has a pleasing taste, is available at a low cost without a prescription, and is commonly used in compounding pediatric formulations. Furthermore, we chose a method for compounding the liquid formulation that would be reasonable for a patient or caregiver to conduct at home. However, there were several limitations to our study. First, efavirenz exposure was highly variable, approximately, 40%, and relative bioequivalence for efavirenz may have been more appropriately assessed with a larger sample size. The efavirenz half-life for healthy volunteers in this study is approximately 50 hours, and the sampling period of 48 hours accounted for only 50% of the total efavirenz AUC. Although sampling out to 76 hours would have been ideal, it was inconvenient to extend the sampling period for healthy volunteers participating in the study. Second, the clinical implications of these data are unknown and have not been tested in a clinical trial of HIV-infected patients. The magnitude of increase in tenofovir AUC with the liquid formulation is similar to that seen when the tenofovir tablet formulation is administered with many ritonavir-boosted protease inhibitors, which has demonstrated safety.5 The geometric mean ratio efavirenz AUC was slightly higher than the FDA acceptable range for bioequivalence but is unlikely to increase the incidence of efavirenz-associated toxicity. Because the current study was conducted in healthy volunteers after a single dose, the minimum plasma concentration (C min) after steady-state dosing could not be determined. Concentration-response relationships have been defined for many antiretrovirals, including efavirenz, based upon the steady-state C min.6 Minimum plasma concentrations may be adequate for this formulation when dosed to steady state in HIV-infected patients.
In conclusion, a FDC tablet of efavirenz, emtricitabine, and tenofovir crushed and mixed with Ora-sweet may be a viable option for HIV-infected patients unable to swallow tablets and are capable of swallowing and compounding an oral solution. However, possible risks and benefits for HIV-infected patients should be carefully considered before crushing Atripla tablets to construct a compounded oral solution.
Jennifer King, PharmD*
Matthew McCall, MD†
Anthony Cannella, MD†
Michael Anne Markiewicz, PharmD‡
Amanda James, BS*
Chantelle B. Hood, PhD*
Edward P. Acosta, PharmD*
*Department of Pharmacology and Toxicology, University of Alabama at Birmingham, Birmingham, AL, †Division of Infectious Diseases, Department of Medicine, University of Alabama at Birmingham, Birmingham, AL, ‡Department of Pharmacology, University of Alabama at Birmingham, Birmingham, AL