Pharmacokinetic variability and strategy for therapeutic drug monitoring of saquinavir (SQV) in HIV-1 infected individuals

Abstract

Aims

To investigate the pharmacokinetic profile of the protease inhibitor saquinavir (SQV) after multiple doses in HIV-positive patients and to evaluate the possibility of predicting total body exposure of SQV from concentrations determined at single time points.

Methods

Twenty HIV-positive patients on steady-state treatment with SQV (Hard-Gel-Capsule, Invirase®) were enrolled in this study. Serial blood samples were obtained during a dosing interval. SQV plasma concentrations were determined by high performance liquid chromatography (h.p.l.c.) and pharmacokinetic parameters were determined by noncompartmental techniques.

Results

There was a marked interindividual variability in SQV pharmacokinetic parameters with a 11-fold variability in total systemic exposure to SQV, as expressed by AUC(0,8h) values (range: 268–3009 ng ml⁻¹h, CV: 69%). The oral clearance shows an interindividual CV of 75%. A strong correlation (r=0.94) was found between SQV plasma concentration at 3 h (C_{3 h}) and AUC(0,8h).

Conclusions

This study shows that C_{3 h} is a good predictor for total body exposure of SQV and might be useful to predict SQV disposition in HIV-positive patients on steady-state treatment.

Introduction

Saquinavir (SQV) is a peptide derivative anti-HIV protease inhibitor (PI).

The SQV formulation commonly available in most countries and formulated as a hard-gel capsule (HGC-SQV, Invirase®) has been utilized in many clinical studies and is prescribed in combination therapies for many thousands of HIV-infected individuals.

The current FDA-approved dose for HG-SQV is 600 mg three times daily when administered in combination with two nucleoside analogues. Due to poor bioavailability of SQV (≈4%) [1, 3] and marked interindividual variation in plasma levels, there is concern about development of resistance resulting from subinhibitory plasma levels.

When this study was planned and conducted, Invirase® was the only SQV formulation available at our Institution. A new formulation (SGC-SQV, soft-gel capsule, Fortovase) has recently been approved for use in the USA [4].

The aim of this investigation was to determine the pharmacokinetic interindividual variability of SQV in HIV-positive patients and to assess the predictability of individual SQV concentrations at different time points for the area under the curve (AUC), in order to predict SQV total body exposure.

Methods

Study population and sample collection

Twenty clinically stable subjects with HIV-infection (confirmed by ELISA and Western-blot analysis), aged between 18 and 65 years, with normal renal function (creatinine <1.5 times the upper limit of the normal range) on steady-state treatment with SQV were considered eligible for this kinetic study. Approval for the study was obtained from the local ethics committee. The patients were enrolled in the study only after being informed about the purposes of the investigation and after giving their informed consent to it. Patients were also receiving two anti-HIV nucleoside analogue reverse transcriptase inhibitors (NRTIs) in combination with SQV. No medications known to interfere with SQV metabolism were prescribed during the study period.

SQV was administered at the oral dosage of 600 mg three times daily, taken with a full meal and with two glasses of commercially available grapefruit juice, within 1 h before and/or after swallowing SQV capsules.

Blood samples were taken by venepuncture at the following times: 0 (trough level, just before SQV administration), 1, 2, 3, 6 and 8 h after drug administration.

Blood samples were collected in heparinized tubes, cooled in ice and centrifuged for plasma at 3000 g within 1 h from sampling. The separated plasma was heated to 56° C for 50 min to inactivate HIV and then stored at −20° C until analysis.

Analytical methods

SQV concentrations in plasma were analysed by a specific and validated reverse-phase high-performance liquid chromatography assay (h.p.l.c.), based on the recent technique developed by Merry et al. [5]. The limit of quantification was 10 ng ml⁻¹. Inter-assay and intra-assay precision was assessed with three different control samples containing nominal concentrations of 50, 100 and 1000 ng ml⁻¹. The coefficients of variation were <10% for all the three concentration points.

Pharmacokinetic and statistical analysis

From the plasma concentration-time data, pharmacokinetic parameters were determined by noncompartmental techniques (PNCA Computer Program, Creteil, France).

The highest measured plasma concentration and the corresponding time were defined as C_max and t_max, respectively. The minimum plasma concentration during a dosing interval (C_min) and the trough plasma level (C₀, just before SQV administration) were also observed. The area under the plasma concentration-time curve (AUC) during a steady-state dosing interval was estimated by the linear trapezoidal method from time zero to the end of the dosing interval following the test dose (AUC(0,8 h)).

Total drug oral clearance (CL/F, where F is the bioavailability) was calculated dividing the dose by AUC. Data are reported as mean±standard deviations (s.d.) or median and range.

The average steady-state concentration (C_av) was calculated as AUC/τ, where τ is the dosing interval. The percentage peak-trough fluctuation (fluctuation index, FI) was calculated as ([C_max−C_min]/C_av)×100. A linear regression analysis was used to determine the best fit between each time point SQV plasma concentrations or C_max or C_min, and the corresponding full 6-sample AUC (AUC(0,8 h)). The relationship between each time point SQV plasma concentration and AUC(0,8 h) was evaluated by the Pearson’s correlation coefficient.

Results

The mean (±s.d.) and median (range) values for the main pharmacokinetic parameters of SQV, with the corresponding coefficients of variation (CV%), are summarized in Table 1.

Table 1.

Pharmacokinetic parameters of SQV in 20 HIV infected patients.

Measured peak plasma concentration (C_max) values were 230±148 ng ml⁻¹ (median: 182 ng ml⁻¹; range: 84–585 ng ml⁻¹). Time to reach peak concentration (t_max) was a median of 3 h (range: 2–6 h). The minimum SQV plasma concentrations measured during a dosing interval (C_min) was 38±30 ng ml⁻¹ (median 29 ng ml⁻¹; range: 12–144 ng ml⁻¹). In 10 out of 20 patients, the plasma levels 1 h after drug administration were lower than the corresponding trough plasma concentrations (C₀).

The mean area under the plasma concentration-time curve in the dosing interval (AUC(0,8 h)) was 983±679 ng ml⁻¹ h (median: 759 ng ml⁻¹h; range: 268–3009 ng ml⁻¹h).

During the dosing interval, an average steady-state concentration (C_av) of 122±85 ng ml⁻¹(median: 95 ng ml⁻¹; range: 34–376 ng ml⁻¹) was observed.

The mean apparent oral clearance (CL/F) was 14±10.5 l h⁻¹kg⁻¹ (median: 12.1 l h⁻¹kg⁻¹; range: 2.2–44.8 l h⁻¹kg⁻¹). The mean percentage peak-trough fluctuation (FI) during the 8 h dosing interval, was 163±34% (median: 165%, ranging from 90 to 221%).

In contrast with SQV trough (C₀) plasma concentrations, the drug C_min and C_max values were significantly related to the AUC, by the equations: AUC (ng ml⁻¹ h)=24.1 C_min (ng ml⁻¹) (P<0.01, r=0.88) and AUC (ng ml⁻¹ h)=4.28 C_max (ng ml⁻¹) (P<0.01, r=0.93), respectively.

Single point determination of SQV during the 8 h dosing interval were highly correlated to the systemic drug exposure at the 3 h time point SQV concentration (C_{3 h}). The data suggest that C_{3 h} provides a reliable and consistent estimate of the full AUC profile during a dosing interval (r=0.94) (Figure 1).

Figure 1.

Correlation between SQV AUC values and SQV plasma concentrations at 3 h (r=0.94).

The difference between estimated AUC from the concentration at C_{3 h} and the calculated AUC(0,8 h) was not significantly different (983±634 ng ml^{−1 h}vs 983±679 ng ml⁻¹h, P>0.05).

Discussion

The pharmacokinetic parameters of SQV obtained in this study agree with the values obtained in previously reported data from HIV-infected patients receiving multiple oral doses of 600 mg three times daily [1, 5, 8].

A marked interindividual variability in SQV pharmacokinetic parameters was observed, as shown by their high coefficients of variation (CV%).The oral clearance (CL/F) of SQV shows an interindividual coefficient of variation of 75%. Such variability is the result of both the variability of the drug reaching the systemic circulation (F) and the variability in systemic clearance (CL) of the drug. In our cohort of patients, there is 11-fold variability in SQV total systemic exposure, as expressed by AUC(0,8 h) variability, similar to that (12-fold) demonstrated by Merry et al. [5] in seven HIV-infected patients treated with SQV. Barry et al. [7] recently also demonstrated that plasma SQV concentrations are extremely variable among HIV subjects and that many patients may have SQV levels below the IC₉₅ value of 25 ng ml⁻¹. Thus, the authors suggested the utility of therapeutic drug monitoring in patients receiving SQV, since the response is more reliably correlated with drug exposure than dose. Despite limited data on SQV pharmacokinetics and efficacy being available, relationships have been found between SQV disposition and efficacy in terms of a decrease in HIV-RNA load and an increase in CD4+lymphocytes [1, 6]. Measurement of AUC provides a reliable method to estimate SQV exposure; however, this procedure is not widely acceptable in a clinical setting due to the need of multiple sampling.

Morning trough plasma concentrations (C₀) provide poor predictability for SQV body systemic exposure. In some individuals 50% absorption of SQV after oral dosing does not start immediately, as suggested by SQV plasma concentration values at 1 h after drug administration, being lower than the value just before drug administration (C₀). This delay in absorption of SQV is the basis of the unsatisfactory relationship between trough plasma levels (C₀) and AUC(0,8 h). Alternatively, SQV concentrations at 3 h have an excellent correlation with AUC and can predict 94% of drug exposure with a single concentration. On the basis of our results, a 3 h SQV concentration of about 150 ng ml⁻¹appears to be associated with an AUC(0,8 h) of about 700 ng ml⁻¹ h (average value obtained with the current SQV-HGC recommended dosage of 600 mg×3/day), considered predictive of the antiviral efficacy of the drug [1, 6]. Consequently, C_{3 h} may be useful as clinical indicator of total body systemic exposure of SQV in an outpatient setting.

Moreover, further studies are necessary to clarify the relationship between 3 h SQV concentration and clinical outcome. These data provide a contribution to the optimization of SQV therapy, when given as hard-gel capsule.