| |
Pharmacokinetics, pharmacodynamics and safety of once-daily versus twice-daily dosing with enfuvirtide in HIV-infected subjects
|
| |
| |
AIDS: Volume 20(3) 14 February 2006 p 397-404
....Ctrough was 57% lower with QD, and there was trend towards less antiviral efficacy with QD....
Authors; Thompson, Melaniea; DeJesus, Edwinb; Richmond, Garyc; Wheeler, Davidd; Flaherty, Johne; Piliero, Peterf,*; True, Andreag; Chiu, Yu-Yuanh; Zhang, Yingg; McFalls, Emilyg; Miralles, G Diegog; Patel, Indravadan Hh
From the aAIDS Research Consortium of Atlanta (ARCA), Atlanta, Georgia
bIDC Research, Altamonte Springs, Florida
cNorth Broward Hospital District, Fort Lauderdale, Florida
dIDP Inc., Annandale, Virginia
eNorthwestern University, Chicago, Illinois
fAlbany Medical Center, Albany, New York
gTrimeris Inc., Morrisville, North Carolina
hRoche, Nutley, New Jersey, USA.
Abstract
Objective: To investigate the pharmacokinetics, safety/tolerability and antiviral activity of enfuvirtide administered once-daily (QD) versus. twice-daily (BID).
Design: An open-label, randomized, multiple dose, two-period crossover study comparing 180 mg enfuvirtide, two injections QD versus 90 mg enfuvirtide, two injections, BID.
Methods: Steady-state intensive pharmacokinetic samples were obtained on days 7 and 14.
Results:
Thirty-seven subjects received at least one dose of enfuvirtide. Thirty-three subjects completed both dosing periods.
The regimens were bioequivalent based on the ratio of geometric mean area under the curve (AUC)0-τ [112 ± 6.2 μg- h/ml QD; 115 ± 6.4 μg-h/ml 2 X BID; QD/BID 0.98; 90% confidence interval (CI) 0.89,1.07].
The maximum observed plasma concentration within a dosing interval (C max) was 49% higher for QD (9.5 ± 2.7 μg/ml) versus BID (6.3 ± 1.7 μg/ml) and the pre-dose plasma concentration (C trough) was 57% lower for QD (1.6 ± 1.1 μg/ml) versus BID (3.8 ± 1.3 μg/ml).
The LSM decrease in viral load from baseline to day 7 was 1.0 ± 0.14 log10 (n = 18) for QD and 1.4 ± 0.2 log10 (n = 17) for BID (LSM difference 0.385; P = 0.07).
Linear regression analysis suggested that decline in viral load up to day 7 was associated with C trough but not C max or AUC. There were no significant differences in adverse events between the two dosing regimens.
Conclusions: Administration of enfuvirtide 180 mg QD results in bioequivalence compared with 90 mg BID based on AUC with a similar short-term safety profile, but a trend towards a weaker antiretroviral effect. Larger and longer-term studies are needed to determine if 180 mg once daily is an effective dosing alternative for enfuvirtide.
Results
Study population
Thirty-seven patients (35 male, two female) were enrolled at six North American sites between April 2003 and August 2003. All patients were randomized (19 to group A, 18 to group B) and received at least one dose of enfuvirtide. Four subjects (10.8%) discontinued study medication early: two due to adverse events other than injection site reactions on days 2 and 6; one because of injection site reactions on day 9; and one due to a protocol violation on day 7. Therefore, the safety population comprised all 37 patients and the efficacy analysis was carried out on data from 36 patients (one patient did not have a post-baseline viral load measurement). Summaries for pharmacokinetic parameters included 33 patients for QD dosing and 34 patients for BID dosing. The analysis of bioequivalence for AUC0-τ was performed on data from the 33 patients who completed both dosing periods.
Patient demographics and baseline characteristics are summarized in Table 1. There were no major baseline differences between patients in group A and patients in group B. Nucleoside reverse transcriptase inhibitors, non-nucleoside reverse transcriptase inhibitors and protease inhibitors were used as part of the background regimen in 97.3, 16.2 and 86.5% of patients, respectively. The majority of patients (91.9%) had three to five agents in their background regimen and combinations of nucleoside reverse transcriptase inhibitors and protease inhibitors were most common (78.4% of patients). The majority of patients had fewer than two active agents in the background regimen based on genotypic testing (GSS < 2; 24/37, 65%). More subjects had a GSS of zero in group A (9/19, 47%) than in group B (6/18, 33%; Table 1).
Pharmacokinetic results
Total AUC0-τ from the QD regimen (120 ± 41.1 μg/ml) was comparable to that achieved for the BID regimen after multiplying the AUC0-τ value by 2 to account for the evening dose (121 ± 34.5 μg/ml; Fig. 2 and Table 2). The ratio of the geometric least squares mean AUC0-τ from the ANOVA analysis for QD (112 μg-h/ml; n = 33) to that for BID (BID X 2 = 115 μg-h/ml; n = 34) was 0.98 with a 90% CI of 0.89-1.07, which fell within the range of 0.80-1.25. Thus, the two dosing regimens were considered bioequivalent based on the log-transformed values of AUC0-τ.
Secondary pharmacokinetic parameters are summarized in Table 2. C max was 49% higher with the QD regimen than with the BID regimen. The mean C trough was approximately 57% lower than the value of 3.77 μg/ml achieved with the BID regimen. During the QD dosing period, 13 of 33 patients had C trough values below 1 μg/ml, a trough level considered acceptable/targeted based on earlier pre-clinical and clinical studies. CL/F was similar for the two dosing regimens.
Safety results
The mean (± standard deviation) enfuvirtide exposure for all randomized subjects was 13.1 (± 2.7) days. Table 3 lists all adverse events (excluding ISRs), irrespective of relationship to treatment that occurred in greater than 5% of subjects in either dosing regimen by preferred term. Excluding ISRs, the number of subjects reporting at least one adverse event by body-system class was similar during the QD and BID dosing periods (18/36, 50.0% QD and 15/35, 42.9% BID). For both dosing periods, gastrointestinal disorders (including nausea and vomiting) were the most commonly reported adverse events (10/36, 27.8% QD and 8/35, 22.9% BID). Respiratory, thoracic and mediastinal disorders (cough, pharyngolaryngeal pain and respiratory failure) were more common during the QD dosing period (4/36, 11.1%) than during the BID period (0/35) whereas nervous system disorders (including agitation and headache) were somewhat more common during the BID dosing period (5/35, 14.3%) than the QD period (3/36, 8.3%). Infections occurred in three of 36 (8.3%) and two of 35 (5.7%) subjects during the QD and BID periods, respectively. With the exception of general disorders and administration site conditions during the BID dosing period (3/35, 8.6%), disorders in all other body class systems occurred in less than 6% of subjects.
Excluding ISRs, three of 36 (8.3%) subjects experienced enfuvirtide-related adverse events during the QD dosing period versus five of 35 (14.3%) subjects during the BID dosing period. The most common adverse events considered related to enfuvirtide were gastrointestinal disorders (three patients during each dosing period). One subject discontinued enfuvirtide on day 6 during the BID dosing period due to worsening of an erythematous rash over the trunk/arms, which was considered related to enfuvirtide by the investigator. Only one patient experienced serious adverse events. This patient discontinued enfuvirtide and withdrew from the study on day 2, during QD dosing, due to pancreatitis. Four subjects (10.8%) had at least one treatment-emergent grade 3 laboratory abnormality and none experienced grade 4 abnormalities.
One subject discontinued because of injection site reaction on day 7 during the BID dosing period. This patient reported grade 3 erythema and grade 4 induration.
Efficacy results
Because of the limited sample size and short exposure times, only an exploratory analysis of efficacy between 180 mg QD and 90 mg BID up to day 7 was undertaken. By day 3, the median decline in viral load from baseline was -0.19 log10 copies/ml for QD and -0.39 log10 copies/ml for BID, and a further decline was seen in both groups up to day 7 (-0.97 log10 copies/ml QD and -1.43 log10 copies/ml BID). At day 7, the least squares mean (± standard error) was -1.03 (± 0.14) for QD and -1.41 (± 0.15) for BID. The difference (QD - BID) of the least squares means was 0.38, which was not considered statistically significant (95% CI, -0.04 to 0.81; P = 0.074).
Although confounded by the crossover on day 8, an exploratory analysis of change in viral load from baseline was also conducted for day 14. There was a significant difference in the least squares mean reduction in viral load between subjects starting with QD dosing [group A, -1.20 (± 0.15)] and subjects starting with BID dosing [group B, -1.91 (± 0.15); group A - group B = 0.71; 95% CI, 0.27-1.15; P = 0.002). By day 14, two patients in group A and none in group B had experienced a rebound in viral load to within 0.5 log10 copies/ml of baseline. Neither of these patients had any active agents in their background regimen (GSS = 0).
Linear regression analysis suggested that the decline in viral load up to day 7 was more closely related to C trough (r 2 = 0.229) than to C max (r 2 = 0.0029) or AUC0-τ (r 2 = 0.028) over the limited data range observed in this study. An exploratory subgroup analysis showed that the reduction in viral load from baseline was greater for BID dosing than for QD dosing at all time points, regardless of baseline GSS.
Discussion
This study investigated the steady-state pharmacokinetic parameters of a QD dosing schedule for enfuvirtide administered as two 90 mg subcutaneous injections given QD in comparison with enfuvirtide administered as a single 90 mg subcutaneous injection given twice daily. The ratio of the least squares means for QD/BID was 0.98 with a 90% CI of 0.89-1.07 and therefore the two dosing regimens were considered bioequivalent based on AUC.
C max was 49% higher and C trough was approximately 57% lower with the QD regimen than with the BID regimen. Presently there are insufficient data to clearly define what the optimal or minimum trough concentration for enfuvirtide should be. However, it was recently reported that there was a weak relationship between trough concentrations of enfuvirtide and antiviral activity in the TORO studies [26]. Furthermore, it was suggested that a C trough of a low as 1 μg/ml may be sufficient for patients who are able to combine enfuvirtide with one other active drug (phenotypic sensitivity score or PSS ≥ 1). The data reported by Mould et al. was confined to patients receiving 90 mg BID therefore these data may be confounded by an insufficient spread of trough concentrations to accurately define any exposure-effect relationship. However, it does leave the possibility that trough concentrations somewhat below those achieved with the BID administration of enfuvirtide may under defined circumstances be effective.
In the present study an exploratory pharmacokinetic-pharmacodynamic analysis suggested a correlation between viral load reduction during treatment and C trough (r 2 = 0.229), but not AUC or C max. The lower C trough observed with 180 mg QD dosing with enfuvirtide should be investigated in larger, longer-term trials in order to assess the potential for sustainable efficacy relative to 90 mg BID.
Exploratory analysis of efficacy between the two dosing regimens focused on decline in viral load over the first 6 full days of treatment. The median reduction in viral load was greater in subjects who received enfuvirtide BID (group B) than in subjects who received enfuvirtide QD, and although the difference was not statistically significant (P = 0.07) it may signify a trend towards a weaker antiviral effect. The larger proportion of patients with no active drugs in the background regimen in group A may have contributed to the smaller reduction in viral load in this group. However, an exploratory sub-group analysis suggested that this is not the case, since the sample size limited the possibility to identify any influence of the baseline differences in GSS on decline in viral load between the two groups.
The number of subjects reporting treatment-emergent adverse events was similar for both dosing regimens. Gastrointestinal disorders were most common for both dosing regimens but respiratory, thoracic and mediastinal disorders were more common during QD dosing, whereas nervous system disorders were somewhat more common during BID dosing. A total of four of 37 randomized subjects discontinued the study prematurely. Two patients discontinued due to adverse events (excluding ISR) - one during the QD and one during the BID dosing regimen. One subject discontinued study due to ISR during the BID regimen and one subject was withdrawn by the sponsor due to a major protocol violation. Thus, overall, it appears that the safety and tolerability between the two dosing regimens was comparable. However, detailed ISR data was not collected for patients who did not report a serious ISR or ISR leading to discontinuation, since the dosing intervals were short (7 days) and ISR tend to stabilize following 2 weeks of enfuvirtide treatment [27].
Based on these results, additional investigation of 180 mg QD dosing for enfuvirtide is required to determine whether this dosing regimen provides long-term efficacy and safety profiles comparable to the 90 mg BID dosing regimen in a larger and more diverse group of subjects.
Introduction
Enfuvirtide (ENF, FUZEON; Roche, Nutley, New Jersey, USA and Trimeris, Inc., Morrisville, North Carolina, USA) is the first of a new class of antiretrovirals known as HIV-1 fusion inhibitors. With the availability of enfuvirtide there are now four classes of antiretrovirals for the treatment of HIV. Enfuvirtide is a synthetic 36-amino acid peptide, the sequence of which was derived from the gp41 transmembrane glycoprotein of HIV-1 [1,2]. Enfuvirtide disrupts the conformational changes that occur following binding of the HIV-1 envelope to the CD4 cell-surface receptor preventing viral entry [1,2]. In adult and pediatric clinical trials enfuvirtide has proved to be well tolerated with potent antiviral activity [3-9]. Results from the 96-week Phase III trials TORO 1 and 2 showed that enfuvirtide added to an optimized background (OB) of oral antiretrovirals significantly improved virologic response, increased the time until virologic failure and increased CD4 cell count compared with treatment with OB alone [6,9].
Currently, enfuvirtide is administered as a single subcutaneous injection, delivering 90 mg of active drug, administered twice daily (BID). The results of a survey carried out during the TORO trials indicated that most patients did not experience major difficulties with enfuvirtide reconstitution or self-injection [10], and adherence to enfuvirtide was greater than 85% in more than 90% of patients [6,9]. However, these patients received optimal training and support in a clinical trial setting and may have had a particularly high level of motivation. Highly active antiretroviral therapy (HAART) with oral agents has been less successful in unselected populations than in clinical trials - a difference that may, in part, be attributed to lower levels of adherence outside the controlled environment of a clinical trial [11,12]. Adherence to therapy has been shown to be one of the most important factors in determining the virologic response to HAART and even small degrees of non-compliance can have negative effects on clinical outcome [13-15]. Although many factors interact to determine patient adherence to antiretroviral regimens, convenience and tolerability appear to be the main drivers [16,17]. Data from other chronic disease areas suggest that fewer doses and lower numbers of pills improve adherence to treatment [18] and there is now a growing trend towards the development of antiretroviral regimens that can be taken once daily (QD) [19-24]. However, it should be acknowledged that in the treatment of HIV-infection once daily dosing regimens will initially be primarily targeted at early lines of therapy whereas in later lines of therapy, where use of enfuvirtide will be generally indicated, at least some of the antiretroviral therapy will still be administered twice a day.
Considering the additional complexity of reconstituting and administering a subcutaneous injection in comparison with an oral antiretroviral drug, reducing the frequency of enfuvirtide injections from twice daily to once a day dosing would be more convenient for some patients and may result in better compliance. To investigate the pharmacological feasibility of such a reduction in enfuvirtide dose frequency, we investigated the multiple-dose pharmacokinetics of 180 mg of enfuvirtide administered either as two 90 mg injections QD or one 90 mg injection BID. The study also included a comparison of the safety and tolerability of these two dosing schedules and an exploratory evaluation of short-term antiviral activity.
Methods
Study design
T20-104 was an open-label, randomized, two-period crossover trial to evaluate the multiple-dose pharmacokinetics of enfuvirtide given as a total daily dose of 180 mg via two 90 mg 1.0 ml QD injections or one 90 mg 1.0 ml BID injection. The primary objectives of the study were to determine and compare the multiple-dose pharmacokinetics of enfuvirtide for QD versus BID dosing. The secondary objectives were to determine and compare the safety and tolerability of the two dosing schedules. An exploratory evaluation of the differences in short-term antiviral activity between the initial QD and BID dosing periods was also conducted.
Patients were fusion inhibitor-naive, HIV-1-infected adults who had a plasma viral load ≥ 5000 copies/ml, and had previously been treated with nucleoside and non-nucleoside reverse transcriptase inhibitors and protease inhibitors. Patients received their current treatment regimen for at least 4 weeks prior to baseline.
Patients were randomized (1: 1) to receive enfuvirtide as two 1.0 ml, 90 mg subcutaneous injections QD (group A) or one 1.0 ml, 90 mg subcutaneous injections BID (group B) during study days 1-7. On day 8, patients in group A crossed over to BID dosing and patients in group B crossed over to QD dosing. The second dosing period continued until day 14 (Fig. 1). Randomization was stratified by screening viral load (< 50 000; 50 000-300 000; and > 300 000 copies/ml). All patients optimized their background antiretroviral regimen at day 1 based on the results of genotypic and phenotypic testing at screening and antiretroviral treatment history. The optimized regimen and other concomitant medications were to remain stable throughout the entire study. Changes to the background or concomitant medications were only permitted for toxicity purposes or if used to treat adverse events. All patients received training in sterile technique and the reconstitution and administration of enfuvirtide; they were instructed to rotate their injection sites and never to give the two daily injections at the same site. Institutional Review Board approval was obtained for all centres and written informed consent was obtained for each patient. The study was conducted in accordance with the ethical principles that have their origins in the Declaration of Helsinki.
Pharmacokinetic assessments
The primary pharmacokinetic parameter was the area under the plasma concentration-time curve from time 0 to τ, where τ is the length of the dosing interval (12 or 24 h, AUC0-τ). Secondary pharmacokinetic parameters were the maximum observed plasma concentration within a dosing interval (C max), the pre-dose plasma concentration on study days 6, 7, 13 and 14 (C trough), time to maximum plasma concentration (t max), and the apparent clearance after extravascular administration calculated from dose divided by the AUC0-τ (CL/F).
Pharmacokinetic assessments were carried out on days 6, 7, 13 and 14. On these days, the two morning injections (for QD) or morning injection and evening injection (for BID) were given in independent abdominal sites in the clinic by a clinician. For the QD dosing period, blood samples (5 ml each) were collected within 1 h prior to enfuvirtide administration on the penultimate day of the dosing period and at 0, 1, 2, 4, 6, 8, 10, 12, 18 and 24 h post-treatment administration on the last day of the dosing period. For the BID dosing period, blood samples were collected within 1 h prior to both the morning and the evening doses on the penultimate day of the dosing period and at 0, 1, 2, 4, 6, 8, 10 and 12 h post-dose on the last day of the dosing period.
Plasma concentrations of enfuvirtide were determined by a validated liquid chromatography-tandem mass spectrometry method [25]. The method utilized a 100 μl plasma sample and had a calibration range of 0.01-2.0 μg/ml. Intra- and inter-day precision for the assay ranged from 3.6-10.0% and 4.8-12.1%, respectively. Pharmacokinetic parameters were evaluated using the non-compartmental method with WinNonlin Professional Version 4.1 (Pharsight Corporation, Mountain View, California, USA).
Safety assessments
Safety assessments were conducted and recorded at each clinic visit. Safety parameters included clinical adverse events, serious adverse events and selected clinical laboratory test results. Adverse events were coded by body system and preferred term according to the Medical Dictionary of Drug Regulatory Affairs (MedDRA, Version 6.0). For the purpose of this study, only injection site reactions (ISRs) leading to discontinuation of enfuvirtide and grade 4 ISRs based on the overall grading (i.e. based on pain and discomfort and not on the grading of the individual signs/symptoms) were recorded on the study case report form.
Efficacy assessments
The exploratory efficacy parameter was change in log10 plasma HIV-1 RNA from baseline to day 7, after 6 complete days of dosing. Plasma HIV-1 RNA levels were measured at screening, baseline, and days 3, 7 and 14. Samples were tested at a central laboratory using Roche Amplicor HIV-1 Monitor Version 1.5 (Roche, Nutley, New Jersey, USA) with standard sample preparation (range: 400-750 000 copies/ml). Samples with < 400 copies/ml were re-tested using the Amplicor HIV-1 Monitor with ultrasensitive sample preparation (range: 50-75 000 copies/ml).
Data analysis and statistical methods
Pharmacokinetic parameters were calculated for all patients who completed at least one period of treatment and had sufficient pharmacokinetic data. The analysis of bioequivalence for AUC0-τ was based on patients who completed both treatment periods and had sufficient pharmacokinetic data. AUC0-τ data were log transformed and analysed using an analysis of variance (ANOVA) model including sequence, period and treatment as fixed effects, and subject within sequence as a random effect. The 90% confidence interval (CI) for the ratio of least squares means between test and reference treatment, derived from the ANOVA model, was used to assess bioequivalence. For QD versus BID comparisons of AUC0-τ, the QD regimen was compared with 2 X AUC0-12 for the BID regimen. The two treatments were considered bioequivalent if the 90% CIs for the ratio of log-transformed AUC0-τ between BID versus QD regimens were within the range of 0.8-1.25.
The safety analysis was carried out on all subjects who received at least one dose of enfuvirtide. The exploratory efficacy analysis was performed on all subjects who received at least one dose of enfuvirtide and had at least one post-baseline viral load measurement. Log10 plasma HIV-1 RNA change from baseline to day 7 was analyzed using a two-way ANOVA model including dosing regimen (BID versus QD), randomization stratum, and dosing regimen by stratum interaction. An exploratory pharmacodynamic analysis of HIV-1 RNA changes from baseline versus C max, C trough and AUC0-τ for the dosing periods up to day 7 was carried out using linear regression analysis. Efficacy results were also assessed for subgroups of patients with baseline genotypic sensitivity scores (GSSs) of ≦ 1 and ≥ 2. GSS was defined as the number of antiretroviral agents in the background regimen for which baseline genotypic testing classified the patient's virus as sensitive.
|
|
| |
| |
|
|
|
