Effect of efavirenz treatment on the pharmacokinetics of nelfinavir boosted by ritonavir in healthy volunteers

Abstract

Aims

A once-daily (q.d.) nucleoside-sparing regimen can prevent mitochondrial toxicity, overcome viral resistance and improve compliance. In the present study the effect of efavirenz on the pharmacokinetics and tolerability of once-daily nelfinavir/ritonavir was evaluated in healthy subjects.

Methods

This was a multiple-dose, open-label, single-group, two-period study in 24 healthy subjects. Each received from days 1–10 (period 1): 1875 mg nelfinavir plus 200 mg ritonavir q.d. with a 300-kcal snack. During days 11–20 (period 2) efavirenz 600 mg q.d. was added to the regimen. Blood samples were collected up to 24 h after dosing on days 10 (period 1) and 20 (period 2). High-performance liquid chromatography methods were used for the determination of the concentrations of all compounds. The main pharmacokinetic parameters were calculated using noncompartmental methods.

Results

All subjects completed the study. After the first period mean nelfinavir AUC_{0−24 h}, C_max and C₂₄ were 49.6 mg h⁻¹ l⁻¹, 5.0 mg l⁻¹ and 0.37 mg l⁻¹, and the sum of nelfinavir plus its active metabolite M8 C₂₄ was 0.83 mg l⁻¹. The relative bioavailability, expressed as a geometric mean ratio (90% confidence interval) for nelfinavir AUC_{0−24 h}, C_max and C₂₄ of period 2 compared with period 1 was: 1.30 (1.21, 1.40), 1.29 (1.19, 1.40) and 1.48 (1.32, 1.66). The sum of nelfinavir and M8 C₂₄ in period 2 was 0.99 mg l⁻¹, an increase of 19%. No serious adverse events occurred.

Conclusions

The studied regimens were well tolerated. Nelfinavir/ritonavir given together with efavirenz resulted in a 48% higher mean C₂₄ concentration for nelfinavir, and the sum of nelfinavir and M8 C₂₄ concentrations was 0.99 mg l⁻¹. Efavirenz exposure in this study was similar to that reported previously, and therefore can be used effectively in combination with ritonavir and nelfinavir.

Introduction

Since the introduction of protease inhibitors (PIs) in the mid 1990s for the treatment of HIV infection, life expectancy has increased substantially [1, 2]. Nevertheless, treatment with highly active antiretroviral therapy (HAART) can still be improved, as efficacy is not 100%. Major concerns in treatment with HAART are the occurrence of adverse events [3], failure of compliance [4] and development of viral resistance [3]. To improve the treatment of HIV-infected persons these problems need to be overcome. A nucleoside-sparing regimen can prevent nucleoside reverse transcriptase inhibitor (NRTI)-induced mitochondrial toxicity, prevent NRTI (cross-) resistance and/or lead to an increase in the susceptibility of HIV to non-nucleoside reverse transcriptase inhibitors (NNRTIs) [5]. To improve compliance, and thus virological outcome [4], simplification of antiretroviral dosing regimens seems to be effective [6, 7].

We have previously demonstrated that nelfinavir in combination with ritonavir can be given once daily, at optimal doses of 2000 mg and 200 mg, respectively [8]. As a potent inhibitor of the cytochrome P450 (CYP) enzyme CYP3A4 [9], ritonavir is used in these circumstances to boost plasma nelfinavir concentrations to achieve a durable therapeutic response. CYP3A4 is mainly responsible for the metabolism of nelfinavir [10]. However, the formation of the virologically active [11] metabolite nelfinavir-hydroxy-t-butylamide (designated M8) is primarily dependent on CYP2C19 [12]. M8 is subsequently metabolized by CYP3A4 into inactive metabolites [12]. Therefore, the combination of nelfinavir with ritonavir will result in both increased nelfinavir and M8 concentrations [13, 14].

In the current study nelfinavir was given in form of new 625-mg tablets instead of 250-mg tablets. The new formulation was choosen for two reasons. It is expected to be marketed within a relative short period and enables the dosage of 1875 mg q.d. to be taken as three tablets only, which is more convenient for the patients. Ritonavir was dosed at 200 mg q.d. and efavirenz at 600 mg q.d., once daily in the evening, to avoid potential neurological adverse events [15]. Nelfinavir should be given during the day and with food [16]. In the present study nelfinavir, ritonavir and efavirenz were all taken at bedtime (23.00 h) together with a snack of around 300 kcal.

The primary objective of this study was to characterize the pharmacokinetics of nelfinavir boosted by ritonavir when administered once daily with efavirenz in healthy subjects. The secondary objective was to determine the influence of nelfinavir when combined with ritonavir on the pharmacokinetics of efavirenz. Tolerability of the once-daily regimen was also studied, as well as the influence of giving nelfinavir in the evening as opposed to the morning.

Methods

Study design

This was a multiple-dose, open-label, single-group, two-period study in 24 healthy subjects (12 males). Subjects received the following treatments: period 1 (days 1–10), 1875 mg nelfinavir plus 200 mg ritonavir to be taken once daily at 23.00 h with a light snack of around 300 kcal; period 2 (days 11–20), 1875 mg nelfinavir plus 200 mg ritonavir and 600 mg efavirenz to be taken once daily at 23.00 h with a light snack (around 300 kcal). Blood samples were collected throughout a 24-h period on days 10 and 20 following a light meal of 315 kcal. This consisted of one slice of bread with butter and cheese, or peanut butter, or two slices of sausage, together with a glass of semiskimmed milk (3.75% fat). All other meals eaten on the pharmacokinetic evaluation days were standardized (breakfast 485 kcal, lunch 656 kcal and dinner 1231 kcal). Medication was swallowed with 200 ml of noncarbonated water.

Subject selection

All subjects had to be in generally good health, appropriate for their age as established by medical history, physical examination, electrocardiography, blood pressure, heart rate, and the results of biochemistry, haematology and urinalysis performed within 3 weeks before the first dose. Subjects had to be between 18 and 65 years old. Body mass index had to be in the range of 18–30 kg m⁻². Subjects were not allowed to smoke more than 10 cigarettes, two cigars or two pipes per day, for at least 3 months prior to the study. The protocol was explained comprehensively to all subjects, and written informed consent was obtained prior to screening.

Exclusion criteria included a febrile illness within 3 days before the first dose, exposure to any drug, except for paracetamol, hormonal contraceptives and loperamid, participation in another drug trial and/or donation of blood within 60 days before the study, and hypersensitivity to nelfinavir, ritonavir or efavirenz. In addition, counselling and confirmation of using adequate contraception were required for all females of child-bearing age. Pregnant or breast-feeding subjects were excluded. The study was approved by the Regional Ethics Review Board, located in Nijmegen, the Netherlands

Blood sampling

Blood samples of 5 ml were collected in heparinized hard plastic tubes at the following time points: predose and at 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 8, 10, 12, 16, 20, and 24 h after drug intake on days 10 and 20, and at 9 h postdose on days 1, 4, 7, 14 and 17. Blood samples were centrifuged at 2500 × g for 10 min at 4 °C. Plasma was stored at −18 °C within 2 h after collection.

Clinical assessment

Blood samples after fasting for serum biochemistry [including glucose, total bilirubin, direct bilirubin (in case total bilirubin was above normal range), aspartate amino transferase (ASAT), alanine amino transperase (ALAT), gamma glutamyl transferase (GGT), alkaline phosphatase, creatinine kinase, amylase, total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), triglycerides, creatinine and potassium], haematology and a urine sample for urinalysis were taken at days 1, 4, 7, 11, 14, 17 and 21. In females of childbearing age an instant β-hCG urine pregnancy test was performed at each visit. A urine drug screen for tetrahydrocannabinol (THC), morphine, cocaine and amphetamines was performed on day 1 and at the start of study days 10 and 20, using the Instacheck^TM Multi-Drug Screen panel (Forefront Diagnostics, San Diego, CA, USA).

Blood pressure and heart rate were monitored throughout the study.

Subjects were monitored for adverse events by the medical and nursing staff at the trial centre. Subjects also reported any adverse events in response to general questioning. For each adverse event the following information was recorded: onset and resolution date and time, intensity, relationship to trial medication, action taken, and outcome. All adverse events occurring between the first intake of the trial medication and 30 days after the end of the trial were reported. Adverse events were classified as unrelated, doubtful, possible and probable, and were assessed for intensity according to AIDS Clinical Trial Group (ACTG) classifications: mild (symptoms do not interfere with daily activities), moderate (symptoms interfere with daily activities), severe (symptoms markedly interrupt daily activities) or serious. The latter were defined as those that at any dose resulted in death, were life threatening, required in-patient hospitalization or prolongation of existing hospitalization, resulted in a persistent or significant disability/incapacity or were congenital anomalies/birth defects.

Drug analysis

A validated high-performance liquid chromatography (HPLC) method with ultraviolet detection was used for the determination of nelfinavir, M8 and ritonavir concentrations in plasma as published previously [17]. The lower limit of quantification was 0.04 mg l⁻¹ for nelfinavir and ritonavir, and 0.10 mg l⁻¹ for M8. For nelfinavir, accuracy ranged from 96% to 100% depending on the concentration. Intraday and interday precision ranged from 2.1% to 7.5% and from 0.4% to 3.5%, respectively. For ritonavir, accuracy ranged from 102% to 108%, and intraday and interday precision from 2.0% to 8.1% and 0% to 2.4%, respectively. For M8, accuracy ranged from 93% to 108%, and intraday and interday precision from 2.8% to 4.3% and 2.0% to 3.0%, respectively.

A HPLC method with ultraviolet detection was also used for the analysis of efavirenz [18]. The lower limit of quantification was 0.20 mg l⁻¹. Accuracy ranged from 99% to 101%, and intraday and interday precision from 1.8% to 2.6% and from 1.1% to 2.8%, respectively.

Pharmacokinetic analysis

The following pharmacokinetic parameters were calculated using noncompartmental analysis (Excel version 2000, Microsoft Corporation 1985–1999): AUC_{0−24 h} (mg h⁻¹ l⁻¹), the area under the plasma concentration–time curve; C_max (mg l⁻¹), the maximum plasma drug concentration; C₂₄ (mg l⁻¹), the plasma drug concentration at 24 h after drug dosing; T_max (h), the time to reach maximum plasma drug concentration; T_1/2 (h), apparent elimination half-life; CL/F (l h⁻¹), apparent clearance; Vd/F (l), apparent volume of distribution. CL/F and Vd/F were not calculated for M8. If ritonavir plasma concentrations at the end of the 24-h dosing interval were lower than the quantification limit of 0.04 mg l⁻¹, they were estimated from the last quantifiable plasma concentration and the elimination constant.

Statistical analysis

For AUC_{0−24 h}, C_max and C₂₄ results were presented for treatment period 1 and treatment period 2 together with the ratios period 2 : period 1. The geometric mean and min–max range corresponding to mean and min–max range in the logarithmically transformed domain were given for both periods. The parametric point estimate (ratio estimate) and 90% confidence intervals were calculated for the ratio period 2 : period 1. Treatments were considered bioequivalent if the respective 90% classical confidence intervals for the AUC_{0−24 h}, C₂₄ and C_max ratios were included within the range of 0.80–1.25 [19]. T_max, T_1/2, CL/F and Vd/F were considered to be secondary characteristics, the analysis of which was explorative. T_max data were not log-transformed and were compared between treatment periods using a Wilcoxon signed ranks test.

Pharmacokinetic parameters for efavirenz were compared with historical controls derived from a previously published study in healthy subjects [18].

Data on adverse events were classified according to system organ class using MedDRA V 4.1 coding (MedDRA MSSO, Reston, VA, USA). Frequencies and percentages of occurred adverse events were tabulated. Laboratory values were graded according to toxicity scales. For ALAT, ASAT and GGT Grade 1 was defined as 1.25–2.5 times upper limit of normal (ULN), Grade 2 was 2.6–5.0 times ULN. For amylase Grade 1 was 1.1–1.3 times ULN and Grade 2 was 1.4–2.0 times ULN. For total cholesterol Grade 2 was 6.19–7.77 mmol l⁻¹ and Grade 3 was 7.77–10.35 mmol l⁻¹. Grade 1 cholesterol was not used as it overlapped with the laboratory normal ranges. For triglycerides Grade 2 was 4.52–8.47 mmol l⁻¹ and Grade 3 was 8.48–13.55 mmol l⁻¹. For creatinine and total bilirubin Grade 1 was 1.1–1.5 times ULN. Grade 1 hyperglycaemia was 6.42–8.91 mmol l⁻¹. For creatine phosphokinase Grade 1 was 1.1–2.0 times ULN, Grade 2 was 2.1–4.0 times ULN and Grade 3 was 4.1–6.0 times ULN. For HDL- and LDL-cholesterol no toxicity grades were defined. In addition, for each laboratory parameter the median change from the baseline value to the highest observed value was calculated as well as the median change from the baseline value to the last observed value in the study.

Results

Twenty-four subjects (12 males and 12 females) completed the study. All subjects were caucasian except for one Black male. The mean age of the subjects was 45 years, mean weight 72 kg, and mean height 1.74 m.

Pharmacokinetic results for nelfinavir, its active metabolite M8, ritonavir and efavirenz are presented in Table 1. Plasma concentration vs. time plots are presented in Figure 1 for nelfinavir and in Figure 2 for the sum of nelfinavir and M8. An increase in AUC₀₋₂₄ (+ 30%), C_max (+ 29%) and C₂₄ (+ 48%) for nelfinavir was seen when combined with efavirenz in period 2. As a result, these parameters were not bioequivalent over periods 1 and 2. Nelfinavir C₂₄ increased in 21 subjects, and decreased in three (by −38%, −12% and −5%).

Table 1.

Pharmacokinetic data for nelfinavir, M8, ritonavir and efavirenz

	Geometric mean (range)		Relative bioavailability*
N = 24	Period 1†	Period 2‡	GMR (90% CI)	P-value§
Nelfinavir
AUC0−24	49.6 (22.6–94.6)	64.3 (30.0–106.6)	1.30 (1.21, 1.40)	<0.01
Cmax	5.0 (2.4–8.2)	6.4 (3.5–10.7)	1.29 (1.19, 1.40)	<0.01
C24	0.37 (0.12–1.33)	0.55 (0.14–1.26)	1.48 (1.32, 1.66)	<0.01
Tmax¶	4.0 (2.5–6.0)	4.0 (2.0–6.0)		0.64
T1/2	5.4 (4.0–8.4)	5.7 (4.1–8.0)		0.02
CL/F	38.1 (20.9–83.5)	29.6 (17.5–66.0)		<0.01
Vd/F	296.6 (181.4–590.2)	242.9 (164.3–461.9)		<0.01
M8
AUC0−24	39.0 (26.0–68.2)	36.1 (20.3–75.2)	0.93 (0.87, 1.00)	0.09
Cmax	3.4 (2.0–5.8)	3.0 (1.7–5.3)	0.87 (0.81, 0.92)	<0.01
C24	0.46 (0.22–1.16)	0.44 (0.17–1.31)	0.96 (0.85,1.09)	0.59
Tmax¶	5.0 (4.0–6.0)	6.0 (4.0–8.1)		0.01
T1/2	6.4 (4.5–8.9)	6.3 (4.7–9.1)		0.39
Ritonavir
AUC0−24	17.5 (7.8–31.8)	14.0 (4.7–28.4)	0.80 (0.72, 0.88)	<0.01
Cmax	2.4 (1.2–5.1)	1.8 (0.59–3.2)	0.76 (0.67, 0.85)	<0.01
C24	0.04 (0.00**−0.14)	0.04 (0.01**−0.10)	0.88 (0.73, 1.06)	0.24
Tmax¶	4.0 (1.5–8.0)	3.5 (1.5–8.0)		0.96
T1/2	3.3 (2.2–4.1)	3.5 (2.4–4.7)		0.02
CL/F	11.4 (6.3–25.5)	14.3 (7.1–41.9)		<0.01
Vd/F	54.3 (29.2–100.8)	72.1 (31.0–231.5)		<0.01
Efavirenz
AUC0−24		64.0 (37.9–156.9)
Cmax		4.7 (2.9–8.7)
C24		2.0 (1.1–5.5)
Tmax		2.5 (1.0–10.0)
T1/2		30.4 (14.5–73.8)
CL/F		8.3 (2.9–14.0)
Vd/F		365.3 (157.9–864.3)

GMR, Geometric mean ratio; CI, confidence interval; AUC₀₋₂₄ (mg h⁻¹ l⁻¹), area under the concentration–time curve; C_max (mg l⁻¹), highest observed plasma concentration; C₂₄ (mg l⁻¹), trough concentration at 24 h; T_max (h), sampling time for C_max; T_1/2 (h), elimination half-life; CL/F (l h⁻¹), apparent clearance; Vd/F (l), volume of distribution.

^*Relative bioavailability of period 2 (test) over period 1(reference).

^†Combination of 1875 mg nelfinavir and 200 mg ritonavir.

^‡Combination of 1875 mg nelfinavir and 200 mg ritonavir and 600 mg efavirenz.

^§P-value based on a paired samples t-test with 95% confidence.

^¶Median and range; Wilcoxon-signed rank test.

^**If ritonavir plasma concentrations at the end of the 24-h dosing interval were lower than the quantification limit of 0.04 mg l⁻¹, the actual concentrations were calculated on the basis of the last quantifiable plasma concentration and the elimination constant.

Figure 1.

Mean (± SD) plasma nelfinavir concentration–time profiles (n = 24) for study days 10 (♦) (combination of 1875 mg nelfinavir and 200 mg ritonavir) and 20 (▪) (combination of 1875 mg nelfinavir and 200 mg ritonavir and 600 mg efavirenz). From this figure it becomes clear that plasma nelfinavir concentrations are increased after the addition of efavirenz to the regimen on day 20 in comparison with day 10

Figure 2.

Mean plasma nelfinavir plus M8 concentration–time profiles (n = 24) for study days 10 (♦) (combination of 1875 mg nelfinavir and 200 mg ritonavir) and 20 (▪) (combination of 1875 mg nelfinavir and 200 mg ritonavir and 600 mg efavirenz). In this figure the sum of plasma nelfinavir and M8 mean concentrations is plotted against time. It becomes clear that M8 substantially contributes to the trough concentrations, but also maximal concentrations on both study days 10 and 20

In contrast to the increased nelfinavir parameters, AUC₀₋₂₄, C_max and C₂₄ of the active metabolite M8 remained bioequivalent after the addition of efavirenz to the regimen.

Decreases were observed in ritonavir AUC₀₋₂₄ (− 20%), C_max (− 24%) and C₂₄ (− 12%) after the addition of efavirenz to the regimen, resulting in a lack of bioequivalence over periods 1 and 2.

All 24 subjects reported one or more adverse events out of a total of 225. The most frequently reported were diarrhoea (75.0%), dizziness (58.3%) and nausea (37.5%). Table 2 lists all adverse events that were suffered by two or more subjects and which were possibly or probably related to the study medication. Six of the 225 reported adverse events were unrelated, and were therefore not included in further analysis. One hundred and twenty-one adverse events were reported during treatment with nelfinavir/ritonavir (period 1), vs. 98 when nelfinavir/ritonavir was given together with efavirenz (period 2). No serious adverse events were reported during either treatment period. One adverse event of lower abdominal pain was classified as being severe, and it occurred during period 1. The majority (85%) of all adverse events were mild. The causal relationship to study medication was reported to be doubtful for 29%, possible for 41% and probable for 30% of adverse events. The occurrence of adverse events did not lead to temporary or permanent discontinuation of trial medication, nor to dose modifications. Five percent of the adverse events were treated with concomitant therapy. Although diarrhoea was reported frequently, the use of loperamide for its relief was indicated in only one subject during period 1 and in another subject during period 2.

Table 2.

Summary of adverse events occurring in two or more subjects throughout the study, considered possibly or probably related to study drug by the investigator

System organ class*	Description†	Whole study‡ frequency and percentage		Period 1§ frequency and percentage		Period 2¶ frequency and percentage
Gastrointestinal disorders	Abdominal discomfort	2	8.3%	2	8.3%	0	0%
	Diarrhoea	18	75.0%	10	41.7%	8	33.3%
	Dyspepsia	2	8.3%	0	0%	2	8.3%
	Flatulence	5	20.8%	4	16.7%	1	4.2%
	Loose stools	5	20.8%	4	16.7%	1	4.2%
	Nausea	9	37.5%	8	33.3%	1	4.2%
	Stomatitis	2	8.3%	1	4.2%	1	4.2%
General disorders and administration site conditions	Fatigue	7	29.2%	5	20.8%	2	8.3%
Metabolism and nutrition disorders	Anorexia	3	12.5%	1	4.2%	2	8.3%
	Hypercholesterolaemia	3	12.5%	1	4.2%	2	8.3%
Musculoskeletal and connective tissue disorders	Myalgia	5	20.8%	4	16.7%	1	4.2%
	Sensation of heaviness	2	8.3%	2	8.3%	0	0%
Nervous system disorders	Disturbance in attention	2	8.3%	0	0%	2	8.3%
	Dizziness	14	58.3%	4	16.7%	10	41.7%
	Headache	4	16.7%	3	12.5%	1	4.2%
	Paraesthesia	4	16.7%	2	8.3%	2	8.3%
	Paraesthesia circumoral	2	8.3%	1	4.2%	1	4.2%
	Somnolence	5	20.8%	2	8.3%	3	12.5%
Psychiatric disorders	Abnormal dreams	5	20.8%	1	4.2%	4	16.7%
	Agitation	2	8.3%	0	0%	2	8.3%
	Apathy	2	8.3%	1	4.2%	1	4.2%
	Insomnia	3	12.5%	1	4.2%	2	8.3%
Skin and subcutaneous tissue disorders	Sweating increased	3	12.5%	0	0%	3	12.5%

^*Adverse events were classified according to system organ class with MedDRA V4.1.

^†Description of adverse events according to MedDRA V4.1 preferred terms.

^‡Adverse events of both study periods 1 and 2 together tabulated as frequencies and percentages [calculated as frequency divided by number of subjects (24) times 100].

^§Adverse events of study period 1 tabulated as frequencies and percentages [calculated as frequency divided by number of subjects (24) times 100].

^¶Adverse events of study period 2 tabulated as frequencies and percentages [calculated as frequency divided by number of subjects (24) times 100].

Gastrointestinal disorders were observed more frequently during study period 1 than during period 2. Dizziness (41.7%) and psychiatric disorders such as abnormal dreaming (16.7%) and agitation (8.3%) were reported more frequently in period 2.

Laboratory abnormalities were recorded in all subjects. The results of laboratory safety analysis of biochemistry parameters are summarized in Table 3. Total cholesterol resulted in Grade 2 values in six subjects and Grade 3 values in four subjects. However, seven of these 10 subjects had Grade 2 toxicity values for cholesterol at screening. For triglycerides one subject showed Grade 2 toxicity and one subject Grade 3 toxicity. Both subjects had normal triglyceride values at screening. For haematology and urinalysis no clinically significant abnormalities were found.

Table 3.

Laboratory measurements

	Normal ranges males (females)	Number (Grade) of subjects with toxicity*	Change to maximum median (range)†	Change to final median (range)‡
Alanine amino transferase (U l−1)	0–50 (0–40)	3 (I)	2 (0–70)	−2.5 (−23 to 63)
Aspartate amino transferase (U l−1)	0–40	2 (I)	6.5 (0–28)	1 (−8 to 27)
Alkaline phosphatase (U l−1)	0–120	–	5 (0–34)	−6.5 (−26 to 34)
Gamma glutamyl transferase (U l−1)	0–50 (0–35)	1 (II)	5 (0–69)	3.5 (−14 to 52)
Amylase (U l−1)	0–53	1 (I), 1 (II)	5.5 (0–63)	−2 (−13 to 9)
Total cholesterol (mmol l−1)	3.9–6.5	6 (II), 4 (III)	0.95 (0.1–3.2)	0.65 (−0.5 to 2.1)
HDL-cholesterol (mmol l−1)	0.9–1.7 (1.2–2.3)	ND	0.1 (0–0.3)	−0.2 (−0.6 to 0.2)
LDL-cholesterol (mmol l−1)	3.5–4.5	ND	0.9 (0.1–3.2)	0.55 (−0.2 to 2.5)
Triglycerides (mmol l−1)	0.8–2.0	1 (II), 1 (III)	1.05 (0–6.7)	0.5 (−1.9 to 1.7)
Creatinine (µmol l−1)	60–120 (53–100)	2 (I)	4.5 (0–26)	−4 (−18 to 8)
Total bilirubin (µmol l−1)	3–17	1 (I)	6 (2–10)	1 (−4 to 5)
Glucose (mmol l−1)	4.0–6.0	1 (I)	0.1 (0–1.9)	−0.3 (−1.4 to 1.5)
Creatine phosphokinase (U l−1)	0–200 (0–170)	4 (I), 2 (II), 1 (III)	21 (0–614)	−23 (−157 to 84)

ND, Not determined.

^*Number of subjects with toxicity graded I, Grade 1; II, Grade 2; III, Grade 3.

^†Median and range of the individual changes of 24 subjects from baseline (day 1) to maximum values. If day 1 was the maximum the result was 0.

^‡Median and range of the individual changes of 24 subjects from baseline (day 1) to last observed values. If day 1 was the maximum the result was a negative value.

Discussion

In the current study the combination of nelfinavir/ritonavir plus efavirenz dosed once daily was tested in comparison with the same combination without efavirenz. The results obtained in 24 healthy subjects showed an increased nelfinavir AUC₀₋₂₄ of 30%, C_max of 29% and a C₂₄ of 48% after addition of efavirenz. Exposure to M8, the active metabolite of nelfinavir, was unaffected. Ritonavir concentrations decreased after addition of efavirenz to the regimen, but its boosting effects were still present.

Efavirenz is known to inhibit CYP2C19 [15], which catalyses the formation of M8, the active metabolite of nelfinavir [10]. For this reason one would expect the coadministration of nelfinavir and efavirenz to result in lower M8 concentrations, combined with higher nelfinavir exposure. In accordance with this, it has been reported that the combination of nelfinavir 750 mg three times daily and efavirenz 600 mg o.d. leads to a 20% increase in nelfinavir AUC and a 37% decrease in M8 AUC [20]. Somewhat contrasting data come from a study in patients receiving a dual NRTI regimen with either efavirenz 600 mg o.d., or nelfinavir 1250 mg twice daily (b.i.d.), or the combination of efavirenz and nelfinavir [21]. In the latter study nelfinavir AUC₀₋₁₂, C_max and C_min were lowered by 37%, 21% and 65%, respectively, after the addition of efavirenz for 32 weeks. However, this study was an efficacy and not a bioequivalence study. After the initial 4 weeks of treatment in 40 patients no significant intra-individual differences in pharmacokinetics were seen, except for C_min, which was significantly lower after 32 weeks of treatment in 26 patients. No significant differences were noted in M8 exposure. A major difference with the current study, apart from nelfinavir dose, was the presence of ritonavir in our regimen. Ritonavir is a potent inhibitor of CYP3A4 [9], the enzyme responsible for further metabolism of M8 and of nelfinavir itself [10, 12]. The inhibition of M8 metabolism by ritonavir probably compensated for the decreased formation of M8 as a result of CYP2C19 inhibition by efavirenz. The alternative metabolism of nelfinavir, mediated by CYP3A4 [10], is also inhibited by ritonavir. This could explain the observed higher nelfinavir exposure combined with a bioequivalent M8 exposure over the different regimens in the current study.

For nelfinavir a minimal trough concentration of 0.80 mg l⁻¹ has been proposed [22]. In the current study nelfinavir given with ritonavir but without efavirenz resulted in a C₂₄ of 0.37 mg l⁻¹, increasing to 0.55 mg l⁻¹ when efavirenz was added. Both these values are too low to ensure viral suppression. If nelfinavir is dosed at 1250 mg b.i.d., plasma concentrations of M8 are 30% of those of nelfinavir itself [23]. Thus, the minimal trough concentration for nelfinavir and M8 together should be 1.0 mg l⁻¹. In the current study the sum of nelfinavir and M8 resulted in a C₂₄ of 0.83 mg l⁻¹ without efavirenz, which increased to 0.99 mg l⁻¹ when efavirenz was added.

The observed decrease in ritonavir AUC₀₋₂₄ of −20%, C_max of −24% and C₂₄ of −12% in the current study was unexpected, since a previous study found an increase in ritonavir exposure following efavirenz coadministration [24]. In the latter study ritonavir was dosed at 500 mg b.i.d., which, on a daily basis, is five times the dose used in the current study. Efavirenz is reported to have both inhibitory and inducing effects on CYP3A4 [25], and ritonavir depends mainly on CYP3A4 for its metabolism [9]. These data suggest that efavirenz coadministration may accelerate the metabolism of ritonavir administered at a low dose, which has also been reported previously [18]. Although concentrations of ritonavir decreased, it was still capable of boosting nelfinavir metabolism. It is unknown precisely what ritonavir exposure is necessary to benefit from its boosting effects.

Nelfinavir was not expected to affect the pharmacokinetics of efavirenz [26]. Ritonavir at a dose of 500 mg b.i.d. has been shown to inhibit efavirenz metabolism, resulting in a 21% increase in AUC [24]. However, it is unclear whether the same effect would occur when ritonavir is used at the lower dose of 200 mg once daily. In a study of multiple doses of indinavir/ritonavir 800/100 mg b.i.d. in combination with efavirenz 600 mg, the AUC_{0−24 h}, C_min and T_1/2 of the latter were 56.2 h mg⁻¹ l⁻¹, 1.6 mg l⁻¹ and 35.1 h [18] after intake in the fasting. In the present study these data were 64.0 h mg⁻¹ l⁻¹, 2.0 mg l⁻¹ and 30.4 h. Although the comparison of efavirenz pharmacokinetics with historical controls did not reveal relevant differences, in the current study efavirenz exposure was somewhat higher. This could be the result of the intake of efavirenz with food in the present study, which has been reported to give a 17% increase in bioavailability [15].

In a previous multiple-dose study, where nelfinavir/ritonavir 2000/200 mg was given once daily to eight healthy subjects, either with a 610-kcal breakfast or with a 271-kcal breakfast [8], nelfinavir AUC, C_max and C_min were 57.6 h mg⁻¹ l⁻¹, 6.3 mg l⁻¹ and 0.59 mg l⁻¹, M8 C_min was 0.53 mg l⁻¹ following the 271-kcal breakfast. Nelfinavir exposure in this study was higher than that observed in our study after a 315-kcal snack in the evening. The lower nelfinavir concentrations in our study can partly be explained by a 6.25% lower dose. However, it is not possible to prove whether the remaining part of the observed difference results from the evening vs. the morning intake, and/or the difference in coadministered food. If the combination of nelfinavir/ritonavir is to be used with non-interacting NRTIs instead of efavirenz, it is preferable to administer the drugs in the morning with a 610-kcal breakfast, which has been reported to result in higher nelfinavir exposure [8].

In total, 225 adverse events were reported by the 24 healthy subjects in this study. However, the majority were mild in nature and none met the criteria for a serious adverse event. None led to withdrawal from the study, indicating that the medication was tolerated by the subjects. The most frequently reported adverse drug reaction was diarrhoea, which is a known side-effect of both nelfinavir and ritonavir, and occurred in 75% of the subjects. This seems to be less than the previously reported value of 100% after treatment with nelfinavir/ritonavir 2000/200 mg once daily [8]. The summary of product characteristics of nelfinavir states that 97.7% of mild to moderate severe diarrhoea occurs in patients treated with nelfinavir 750 mg thrice daily, without ritonavir [16]. It seems that the new formulation given at this dosage leads to less diarrhoea than the 250-mg tablets, notwithstanding the comedication with ritonavir. Treatment of diarrhoea with loperamid was only indicated in two subjects, which underlines the mild nature of this adverse event. More gastrointestinal disorders were reported during the first period of the study (nelfinavir plus ritonavir) than during the second period.

In the second period of the study, when efavirenz was introduced, dizziness, abnormal dreams and agitation occurred more frequently than in study period 1. These are known adverse events for efavirenz.

Increases in total cholesterol, leading to Grade 2 values in six subjects and Grade 3 values in four subjects, were observed. However, seven of these subjects suffered from Grade 2 values at screening, indicating that only three subjects showed increased values following drug treatment.

In conclusion, the studied regimens taken with a minimum amount of food at bedtime were tolerated well and thus considered safe to use in patients. Nelfinavir/ritonavir given together with efavirenz resulted in a 48% higher mean C₂₄ concentration for nelfinavir, and the sum of nelfinavir and M8 C₂₄ concentrations was 0.99 mg l⁻¹. From this study it became clear that not all subjects met the cut-off C_min of 1.0 mg l⁻¹ for nelfinavir and M8 together, which should be approached with great care. Efavirenz exposure in this study was similar to that reported previously, and it can therefore be used effectively in combination with ritonavir and nelfinavir.