- Top of page
- Materials and methods
Although saquinavir (SQV) is a potent and specific HIV protease inhibitor (PI) [1,2], its usage relative to other PIs was initially limited by the low bioavailability (< 4%) of the original hard gel capsule (HGC) formulation (Invirase®) . However, administering SQV in a soft gelatine capsule (SGC) formulation (Fortovase®) substantially increases plasma levels  and significantly improves antiretroviral activity . The improved pharmacokinetic profile apparently allows twice-daily administration of SQV-SCG, with a triple therapy regimen incorporating 1600 mg twice daily providing equivalent efficacy at 48 weeks to a triple therapy regimen incorporating 1200 mg three times daily .
The efficacy of SQV is closely related to its plasma drug concentration [7,8]. However, although the introduction of the SGC formulation has increased the absorption of SQV, further increases are still limited by extensive first-pass metabolism, particularly by the cytochrome P450 enzyme CYP3A4 present in the gut wall and liver [9,10]. The PI ritonavir (RTV) has potent inhibitory effects on the CYP3A enzyme system ; coadministration of SQV and RTV leads to substantial increases in the bioavailability of SQV (illustrated by significant increases in Cmax), irrespective of whether the SGC or HGC formulation is used [11–13]. This interaction is presumably mediated through inhibition of CYP3A4 by RTV, leading to inhibition of SQV first-pass metabolism by this enzyme. Pharmacokinetic interactions may also involve RTV-induced reductions in P-glycoprotein-mediated transport of SQV in the gut [14,15]. Co-administration of SQV and RTV appears to have limited effects on the systemic metabolism of SQV, as there are relatively small alterations in the clearance rate of SQV in vivo.
The ‘boosted’ PI combination of SQV/RTV is now frequently employed at a dose of 400 mg RTV plus 400 mg SQV twice daily, and provides potent and sustained clinical activity [13,17]. However, particularly from the perspective of tolerability and patient compliance, it may be beneficial to maximize the plasma exposure to SQV, a relatively well tolerated drug , while minimizing the recognized dose-related toxicity of RTV . Strategies have therefore investigated the combination of ‘mini-doses’ of RTV with SQV in an attempt to obtain the enhancement of SQV exposure with lower doses of RTV [16,19,20]. However, the optimal RTV dose to achieve this pharmacokinetic enhancement is unknown and a variety of dosages are used in clinical practice.
In two dose-ranging clinical pharmacology studies, combined administration of RTV and SQV (given twice daily in one study and once daily in the second) led to several-fold increases in SQV exposure [16,21]. These increases were also associated with reductions in intersubject variability in the pharmacokinetics of SQV compared with those seen following the administration of SQV alone [16,22]. Although trough SQV levels were substantially higher than the EC50, the combination regimens were well tolerated, with those employing lower doses of RTV being associated with fewer adverse events [16,22].
This paper reports a combined analysis of the pharmacokinetic data from the two dose-ranging studies discussed above [16,22]. The aims were to further explore the impact of RTV on the pharmacokinetics of SQV, utilizing the larger combined sample size in order to determine the correlation between RTV dosage and the degree of enhancement of SQV levels. This should help provide information regarding the optimal combined twice-daily dosing regimen for use in future clinical trials. However, conclusions may be generally applicable to other dosage regimens, such as once-daily administration of SQV.
- Top of page
- Materials and methods
These retrospective analyses of pharmacokinetic data from 97 healthy volunteers further characterize the interaction between RTV and SQV. These results show for the first time that the pharmacokinetic enhancement of SQV exposure provided by RTV is not significantly affected by the dose of RTV when dosages ranging from 100 to 400 mg twice daily are employed. Recently, further data from 271 patients treated either with SQV-SGC or combinations of SQV (SGC or HGC) plus RTV have confirmed that SQV exposure is related to the dose of SQV and not the dose of RTV (100–400 mg twice daily) used .
In the present investigation, for regression analyses generated from pharmacokinetic data, the r2 values for both the Cmax and Cmin models were high, indicating that the variables assessed (SQV dosage, any RTV and RTV dosage) account for most of the variability in the data. While once-daily regimens are also under investigation, it remains to be verified that this is a reliable dosing strategy in HIV-infected patients. A potential limitation of our approach to evaluating potential twice-daily dose regimens is that some of the data were derived from a study involving once-daily dosing. However, this factor would be expected to underestimate rather than overestimate the potential exposure to SQV seen with twice-daily dosing. In addition, estimation of the residuals of multivariate regression plots revealed no systematic difference in measurements between the data obtained from once- and twice-daily studies.
The results suggest that a dose of 100 mg RTV is sufficient to achieve enhancement of SQV levels. At this dose, RTV is not expected to have a significant antiviral effect (600 mg twice daily being the licensed dose). However, as RTV shows a dose-related toxicity [19,26], it is likely that the lowest dose of RTV will have the most favourable toxicity profile. Indeed, in one of the two studies from which the data used in this analysis were derived, although all the combinations assessed were generally well tolerated, the least number of adverse events was recorded for the SQV/RTV 800 mg/200 mg regimen (200 mg RTV was the lowest dose of RTV used in that study) .
In certain patients it may be appropriate to consider adding RTV at a therapeutic dose (400 mg twice daily or greater) . The advantage of this approach would be to put additional selective pressure on the virus and potentially decrease the likelihood of the selection of RTV- and/or SQV-related mutations. Another approach that has been increasingly studied is to use the mini-dose of RTV to boost two PIs simultaneously such as SQV and lopinavir (unpublished data). This may have additional advantages of better tolerability compared to regimens including high-dose RTV.
The pharmacokinetic profiles seen with PIs when coadministered with RTV as part of pharmacokinetic enhancement strategies can loosely be divided into two categories: (1) bioavailability increases, but the rate of clearance remains similar; and (2) bioavailability is relatively unaffected, but the rate of clearance is reduced. The pharmacokinetic profile of SQV boosted with RTV falls into the first of these categories. Large increases in the bioavailability of SQV are likely to reflect the combination of a reduction in first-pass metabolism (due to inhibition of CYP3A4 in the intestine and liver by RTV) [9,10], and increased absorption of SQV (due to inhibition by RTV of P-glycoprotein transport) [14,15]. Lopinavir (ABT-378), when boosted with RTV, also falls into the first of these categories, with a large increase in bioavailability seen with doses of RTV as low as 50 mg . By contrast, indinavir and amprenavir fall into the second of these categories, with the rate of clearance being reduced, leading to significant increases in the plasma half-life [29–31].
This pharmacokinetic analysis can be used to assist in the selection of an appropriate twice-daily dosage of SQV/RTV to take forward for further clinical study. The analysis shows that 100 mg twice-daily RTV is sufficient to achieve pharmacokinetic enhancement while limiting the toxicity associated with this agent. Irrespective of the dosage of RTV over the range 100–400 mg twice daily, there was a consistent rise in the Cmin and Cmax of SQV for each increase in SQV dosage.
Clear exposure–response relationships based on Cmin and area under curve (AUC) have been established for SQV-SGC administered as monotherapy in HIV PI-naive adults, and an in vivo EC50 of 50 ng/mL for SQV-SGC monotherapy has been calculated based on Cmin. The EC50 for SQV in combination therapy may be lower due to synergies between antiretroviral agents . Based on these data, a target Cmin of 100 ng/mL has been suggested to ensure an optimal response among PI-naive patients . However, few data are currently available to suggest target pharmacokinetic parameters in patients who have previously experienced virological failure while receiving PI-containing combinations, and who are likely to harbour a diverse population of viruses with a range of susceptibilities to the various antiretroviral agents in the new regimen.
In selecting an appropriate SQV dosage, several considerations need to be balanced. Maximizing SQV exposure and Cmin provides the greatest theoretical probability of completely inhibiting the replication of virus strains exhibiting a range of decreased susceptibilities to SQV. However, a high level of adherence to medications has been shown to be essential for complete and durable viral suppression , and compliance is affected by both the dosing regimen and tolerability.
The Cmin values obtained with SQV dosages of at least 800 mg twice daily in association with mini-dose RTV were at least nine-fold above the in vivo monotherapy EC50, even at the lower extent of the 95% confidence interval. Nearly all patients receiving at least 800 mg SQV-SGC plus mini-dose RTV are therefore likely to maintain drug levels throughout the dosage interval that are adequate to suppress virus with moderately reduced susceptibility to SQV. However, higher dosages of SQV would increase further the confidence that all patients will attain adequate trough drug concentrations to suppress virus strains with substantially reduced SQV susceptibility.
The observation that drug exposure following the 1200 mg/200 mg SQV/RTV regimen was actually lower than the 1200 mg/100 mg SQV/RTV regimen was unexpected. While no definitive answer for this apparently counterintuitive result is currently available, there appears to be at least two theoretical answers. First, and perhaps most likely, is that this is a reflection of intrinsic pharmacokinetic variability. It might be argued that had the sample size been larger, this anomaly would not have been observed. Alternatively, this observation may uncover differences in the dose–response relationship for RTV-related enzyme induction and inhibition. It is theoretically possible that the 100 mg dose of RTV is predominantly associated with the cytochrome P450-related inhibition commonly associated with RTV, whereas at the slightly higher dose of 200 mg the P450-inducing effect of RTV results in a net reduction in inhibition relative to the 100 mg dose. At higher RTV doses there may be a net shift back towards greater P450-related inhibition.
Based on these considerations, a dosage of SQV-SGC/RTV 1000 mg/100 mg, which provides an estimated Cmin of 1387 ng/mL (95% CI 369–5208), has been selected for further clinical investigation. This dosage has been successfully employed in a study assessing the safety and efficacy of a five-drug regimen including efavirenz in patients who failed conventional highly active antiretroviral therapy (HAART) including indinavir or RTV [19,35]. At 24 weeks, 71% of patients achieved a plasma viral load < 500 copies/mL and 45% achieved a viral load < 50 copies/mL . At 48 weeks, these figures were 61% and 58%, respectively . Although efavirenz has been reported to reduce the exposure to SQV when administered as a sole PI , it had no marked effect on SQV plasma exposure when SQV was administered in combination with RTV at a dose of 400 mg/400 mg . Limited data suggest that a SQV/RTV 1000 mg/100 mg combination enables the coadministration of efavirenz ; however, this needs to be confirmed in clinical trials.
The large increases in Cmax seen with boosted SQV regimens may lead to questions regarding tolerability. At a 1000 mg/100 mg dose, the predicted Cmax for SQV is 4962 ng/mL. These exposures are similar to those levels which have been reported to be well tolerated in healthy volunteers , and presumably reflect the fact that SQV is generally well tolerated. By reducing the number of SQV capsules that need to be taken using a boosted regimen, it should be possible to achieve high exposure (e.g. Cmax and Cmin) while reducing the tolerability problems often associated with the administration of multiple capsules (so called capsule toxicity). The safety of the SQV/RTV 800–1200 mg/100 mg twice-daily combination in terms of SQV tolerability is supported by the fact that doses within this range were also found to be well tolerated at both 24 and 48 weeks in PI-experienced HIV-infected patients [19,35]. Although this study involved the HGC formulation of SQV, boosting with RTV produces a similar range of exposures, irrespective of the formulation [11–13].
In conclusion, increasing the exposure to SQV using RTV appears to be similar for doses between 100 and 400 mg. Given the potential advantages of using ‘mini-dose’ RTV to enhance SQV exposure (e.g. reduced toxicity, reduced pill burden and number of daily doses, leading to potential for improved compliance), we recommend further investigation of this strategy. A regimen of 800–1200 mg SQV plus 100 mg RTV is suggested based on both pill burden and a favourable exposure profile. A large clinical trial involving HIV-infected individuals is now ongoing to further evaluate the SQV/RTV 1000 mg/100 mg combination dose regimen.