How much ritonavir is needed to boost protease inhibitors?
Systematic review of 17 dose-ranging pharmacokinetic trials
13 November 2009
Hill, Andrew; van der Lugt, Jasper; Sawyer, Will; Boffito, Marta
aPharmacology Research Laboratories, Liverpool University, Liverpool, UK
bHIVNAT, Thai Red Cross, AIDS Research Centre, Bangkok, Thailand
cMetaVirology Ltd, UK
dSt Stephens Centre, Chelsea and Westminster Hospital, London, UK.
Background: Ritonavir has been evaluated at boosting doses of 50-800 mg daily with seven protease inhibitors: amprenavir, atazanavir, darunavir, indinavir, lopinavir, saquinavir and tipranavir. Minimizing the boosting dose of ritonavir could improve tolerability and lower costs.
Methods: A MEDLINE search identified 17 phamacokinetic trials using different ritonavir doses with protease inhibitors. The dose of ritonavir used was correlated with plasma levels of each boosted protease inhibitor. For the five pharmacokinetic trials of lopinavir/ritonavir, a meta-analysis was used to estimate the effects of lopinavir dose versus ritonavir dose on lopinavir pharmacokinetics.
Results: Saquinavir, fosamprenavir and darunavir were boosted equally well by lower (50-100 mg) versus higher doses of ritonavir. Indinavir, tipranavir and lopinavir were boosted more by higher ritonavir doses. Data on atazanavir were inconclusive. The ritonavir dose-dependence of boosting effects did not correlate with their bioavailability or their effects on ritonavir plasma levels. Atazanavir and indinavir raised plasma ritonavir levels by 69-72%, whereas saquinavir had no effects on ritonavir. Darunavir, lopinavir, tipranavir and fosamprenavir all lowered ritonavir plasma levels. For the meta-analysis of lopinavir/ritonavir trials, the 200/150 mg twice daily (b.i.d.) dose of lopinavir/ritonavir (one Meltrex 200/50 mg tablet and one ritonavir 100 mg b.i.d.) showed lopinavir area under the curve and minimum concentration similar to the standard 400/100 mg b.i.d. dose.
Conclusion: It may be possible to use three protease inhibitors (saquinavir, amprenavir and darunavir) with lower doses of ritonavir. A 200/150 mg b.i.d. dose of lopinavir/ritonavir could lower costs while maintaining very similar lopinavir plasma levels to the standard dose. New pharmacoenhancer drugs may need to be used at different doses to boost different antiretrovirals.
The HIV protease inhibitor ritonavir was first developed as an active antiretroviral at a dose of 600 mg twice daily (b.i.d.). However, ritonavir showed dose-related rises in adverse events, which limited the use of the drug at the originally approved dose . In the first dose-ranging trial, diarrhoea was the most common adverse event, followed by nausea; in addition, there were dose-related rises in circumoral parasthaesia . In the second dose-ranging trial, which evaluated doses of 300-600 mg b.i.d., 10 of the 75 patients discontinued ritonavir for adverse events , which included elevations in cholesterol triglycerides and liver enzymes, gastrointestinal symptoms and circumoral parasthaesia. In the phase 3 clinical endpoint trial of ritonavir at the 600 mg b.i.d. dose, the most common treatment-limiting adverse events were nausea, vomiting, diarrhoea, weakness, altered taste and circumoral parasthaesia .
Ritonavir is a potent inhibitor of CYP3A4 (the primary enzyme involved in the metabolism of most protease inhibitors) and this forms the basis for its enhancement of concomitantly administered protease inhibitors. CYP3A4 is present in the intestinal tract and liver, where it plays a key role in first-pass metabolism of protease inhibitors. Ritonavir significantly increases the plasma concentrations of drugs primarily eliminated by CYP3A4 metabolism - these include clarythromycin, ketoconazole and HIV protease inhibitors . Ritonavir has been developed as a pharmacokinetic enhancer (booster) of seven protease inhibitors: atazanavir, darunavir, fosamprenavir, indinavir, lopinavir, saquinavir and tipranavir. Ritonavir can increase the plasma concentrations of these protease inhibitors either by improving bioavailability or increasing their elimination half-life in plasma. A wide range of doses of ritonavir have been used to boost protease inhibitors. First-line use of ritonavir-boosted protease inhibitors such as atazanavir/ritonavir, darunavir/ritonavir or lopinavir/ritonavir is recommended in international treatment guidelines .
Ritonavir is formulated in 100 mg capsules and the dose used for boosting protease inhibitors is normally 100 mg, either once daily or b.i.d. The most widely used doses of these seven protease inhibitors, combined with ritonavir are shown in Table 1. A new heat-stable 100 mg tablet formulation of ritonavir has been developed [6,7]. Ritonavir is known to elevate lipid levels even at doses of 100-200 mg daily [8,9], and the lipid elevations and gastrointestinal adverse events experienced during treatment with protease inhibitors may be partly caused by ritonavir [10,11]. In studies of healthy volunteers, higher ritonavir exposures were correlated with increases in lipids . Minimizing the dose of ritonavir used could improve tolerability while lowering pill counts and costs. It may be possible to formulate lower doses of ritonavir - for example, a 50 mg heat-stable tablet. Other pharmacokinetic enhancers are currently in development [12,13] and the experience from ritonavir may help us to understand how to use these new drugs to boost a range of HIV protease inhibitors.
The objective of this review is to describe what doses of ritonavir are best associated with its boosting effect on seven different protease inhibitors and to predict whether the doses of either ritonavir or the boosted protease inhibitors could be improved in the future.
We conducted a detailed MEDLINE search for public domain publications and conference abstracts to identify clinical trials of the pharmacokinetics of seven protease inhibitors, with different doses of ritonavir. We used a MEDLINE search with the search terms of each antiretroviral, followed by 'clinical trial' (e.g. 'lopinavir clinical trial'). In addition, we searched the US Food and Drug Administration (FDA) product labels for registration trials of each approved antiretroviral and searched for abstracts on clinical trials presented at the following conferences: Annual Conference on Retroviruses and Opportunistic Infections, International Clinical Pharmacology Workshop, International Conference on Antimicrobial Agents and Chemotherapy (ICAAC), European AIDS Clinical Society, International AIDS Conference (including IAS Pathogenesis Conference) and International Conference on Drug Therapy in HIV Infection. The search generated 2618 references, of which 17 were multiple-dose pharmacokinetic trials evaluating different doses of ritonavir to boost protease inhibitors. Ritonavir was normally dosed using 100 mg capsules, except for the trials of 50 mg ritonavir, in which ritonavir liquid was used. These trials are described in detail in Table 2 [14-29]. The effects of each protease inhibitor on ritonavir concentrations are shown in Table 3 [18,21,30-34]: these results are from cross-over trials, in which ritonavir was used with and without each protease inhibitor.
A meta-analysis of the five pharmacokinetic trials of lopinavir/ritonavir was conducted. For each trial, the mean protease inhibitor maximum concentration (Cmax), minimum concentration (Cmin) and area under the curve (AUC) were recorded, with corresponding estimates of variance (standard error or standard deviation). Within each trial, the geometric mean ratio (GMR) of Cmax, Cmin and AUC was calculated for each dose versus the standard 400/100 mg b.i.d. dose. This approach was used to standardize the estimates of relative bioavailability between the trials and to reduce between-study variation. These estimates of GMR from the individual trials were combined in a meta-analysis to compare the GMR and 95% confidence intervals (CIs) of each nonstandard dose versus the standard 400/100 mg b.i.d. dose of lopinavir/ritonavir. A bootstrapping method was used to calculate the 95%CI around each predicted dose. This method calculates only the ratio of predicted drug concentrations versus the standard dose: as the pharmacokinetic data are from different populations, the ratios versus standard dosing should, therefore, be more reliable than absolute concentrations of lopinavir.
The approved dose of saquinavir/ritonavir is 1000/100 mg b.i.d.; once daily doses (1500/100 and 2000/100 mg once daily) are also being evaluated. Originally, saquinavir was available as 200 mg capsules, but recently a 500 mg tablet has been introduced. Saquinavir had a very low bioavailability as the original hard-gelatin formulation (Invirase; Roche, Nutley, New Jersey, USA) and was the first protease inhibitor to be combined with ritonavir [30,35,36]. In one of the earliest studies, ritonavir increased the Cmax of saquinavir by 20-fold and inhibited clearance, which raised the AUC of saquinavir by 9.6-fold and the Cmin by over 20-fold . Saquinavir does not affect the pharmacokinetics of ritonavir [4,14,30].
Saquinavir/ritonavir was evaluated in pharmacokinetic trials at doses of saquinavir from 400 to 800 mg b.i.d., combined with ritonavir doses of 200-400 mg b.i.d. . Later, once daily doses of 1200/100 and 1200/200 mg were compared . The study of b.i.d. treatment showed slightly higher saquinavir drug levels when used with higher doses of ritonavir ; by contrast, the study of once daily treatment showed slightly lower saquinavir drug levels when given with the higher dose of ritonavir . A meta-analysis of these two trials showed no significant correlation between the dose of ritonavir used and saquinavir Cmax or Cmin . More recently, a Thai cross-over pharmacokinetic study compared doses of 1500/50 mg with 1500/100 mg once daily, in 18 HIV-1-infected patients. The pharmacokinetic parameters of saquinavir 1500 mg once daily were equivalent when used with the 50 mg dose, compared with the 100 mg dose of ritonavir .
Another strategy evaluated was to use ritonavir once daily to boost saquinavir b.i.d: the AUC of saquinavir 1000 mg b.i.d. was significantly lower when given with 100 mg ritonavir once daily versus 100 mg b.i.d. . These results reinforce the need to give the boosted protease inhibitor with ritonavir at the same time to maximize the boosting effect.
Fosamprenavir was originally formulated as 150 mg amprenavir capsules and was then reformulated to 700 mg fosamprenavir tablets. In an early pharmacokinetic trial, ritonavir increased the AUC of amprenavir by 3.3-4-fold and the Cmin by 10.8-14.3-fold . The main effect of ritonavir is on inhibiting clearance of amprenavir - there is a minimal effect of ritonavir on the Cmax of amprenavir [17,18]. Coadministration of fosamprenavir has been shown to lower the AUC of ritonavir by 28-53% in different studies [18,19].
Twice-daily doses of ritonavir of 50 and 100 mg were found to have equal boosting effects on amprenavir 600 mg b.i.d. in a German study of healthy volunteers . The same study found that 100 and 200 mg once daily doses of ritonavir had equal boosting effects on amprenavir 1200 mg once daily . A review of other pharmacokinetic trials of amprenavir with ritonavir also showed similar boosting effects of different ritonavir doses, ranging from 100 to 300 mg b.i.d., on amprenavir .
The fosamprenavir formulation was evaluated in several studies using different doses of ritonavir. The COL10053 trial, conducted on healthy volunteers, showed bioequivalent amprenavir AUC and Cmax and slightly lower Cmin, when used at the 1400/100 mg once daily dose, compared with 1400/200 mg once daily . Two studies used fosamprenavir 700 mg b.i.d. with either 100 or 200 mg ritonavir and showed no effect of the higher ritonavir dose on fosamprenavir exposure [20,21].
The approved doses of darunavir/ritonavir are 800/100 mg once daily and 600/100 mg b.i.d. Darunavir is formulated as 300, 400 and 600 mg tablets. The main effect of ritonavir is on inhibiting clearance of darunavir, increasing the elimination half-life in plasma to 15 h  and increasing the plasma darunavir AUC 11-fold . There is relatively little effect of ritonavir on the Cmax of darunavir. In a cross-over pharmacokinetic study, darunavir lowered the plasma AUC of ritonavir by 14% .
One early clinical pharmacology trial evaluated darunavir with doses of 600/200 mg and 1200/200 mg once daily. The plasma exposure for darunavir for these doses was similar to once daily darunavir/ritonavir doses containing 100 mg of ritonavir once daily . Since then, darunavir has only been evaluated with ritonavir at the 100 mg b.i.d. or once daily doses.
The approved dose of atazanavir/ritonavir is 300/100 mg once daily, but the 400 mg unboosted dose is also approved in the United States. Atazanavir is formulated in 150, 200 and 300 mg tablets. Coadministration of ritonavir increases the atazanavir AUC 3.4-fold. The Cmax of atazanavir is increased 1.9-fold, but the Cmin is raised 11.9-fold: the main effect of ritonavir on atazanavir pharmacokinetics is inhibition of atazanavir clearance . In addition, atazanavir increases the AUC of ritonavir by 39-69% and also increases saquinavir AUC levels by 60% .
We could not find any cross-over pharmacokinetic trials evaluating atazanavir with different doses of ritonavir. However, one Canadian cohort study found that switching from 300/100 to 300/200 mg once daily did not raise atazanavir Cmin significantly in 28 patients .
Indinavir is manufactured as 400 mg tablets, but there is no approved dose in combination with ritonavir. A pharmacokinetic trial on 73 healthy volunteers evaluated indinavir/ritonavir at doses of 800/100, 800/200, 800/400 and 400/400 mg b.i.d., with a control dose of 800 mg thrice daily (unboosted). Ritonavir increased the Cmax of indinavir by 49-77%, but also inhibited clearance, increasing the Cmin over 10-fold. These boosting effects led to a three-fold rise in the AUC . Indinavir increases the AUC of ritonavir by 72% . The clearance of indinavir is partly related to the dose of ritonavir used. For example, the mean 12 h indinavir Cmin was 1.4, 3.1 and 3.1 mg/l for the 800/100, 800/200 and 800/400 mg b.i.d. doses, respectively: this may suggest a saturation of the ritonavir boosting effect at 200 mg ritonavir b.i.d. However, the 800/100 mg b.i.d. dose of indinavir/ritonavir was better tolerated, compared with the doses of 800/200, 800/400 or 400/400 mg b.i.d. .
Tipranavir is formulated in 250 mg capsules and the approved dose of tipranavir/ritonavir is 500/200 mg b.i.d. A clinical pharmacology trial evaluated tipranavir doses of 250-1000 mg b.i.d. in combination with ritonavir at doses of 100 or 200 mg b.i.d. . Tipranavir/ritonavir at the 500/200 mg b.i.d. dose led to higher tipranavir drug concentrations than the 500/100 mg b.i.d. dose. Tipranavir is a powerful inducer of CYP3A4 metabolism and the drug transporter P-glycoprotein and lowers ritonavir concentrations by 90% .
The approved dose of lopinavir/ritonavir is 400/100 mg b.i.d. Originally, the two protease inhibitors were coformulated in a soft-gelatin capsule with 133 mg of lopinavir and 33 mg of ritonavir (three capsules b.i.d.). The new heat-stable formulation (200/50 mg - two tablets b.i.d.) showed 18% higher plasma AUC lopinavir levels and 24% higher Cmax levels than the soft-gelatin formulation , but also less variable plasma drug concentrations, no food dependence and no need for refrigeration . Coformulated lopinavir/ritonavir tablets have also been used with additional ritonavir capsules to further increase lopinavir exposure .
Ritonavir increases the Cmax of lopinavir over 10-fold and inhibits clearance: in the first single-dose pharmacokinetic trials, a ritonavir dose of 50 mg led to a 77-fold rise in the AUC of lopinavir 400 mg . Lopinavir lowers ritonavir plasma drug levels: at the standard 400/100 mg b.i.d. dose, the plasma AUC of ritonavir for lopinavir/ritonavir is 50% lower than that for saquinavir/ritonavir 1000/100 mg b.i.d. . Ritonavir concentrations are also further reduced when higher doses of lopinavir are used with a fixed dose of ritonavir .
Five clinical trials have evaluated the clinical pharmacology of lopinavir/ritonavir using different doses of ritonavir.
Phase 1 pharmacokinetic trial on healthy volunteers
Phase 1 pharmacokinetic trial, a multiple-dose study, evaluated lopinavir/ritonavir doses ranging from 200/50 to 600/50 mg b.i.d., 200/100-400/100 mg b.i.d. and a 600/50 mg once daily dose. In this trial, the pharmacokinetics of lopinavir was dependent on ritonavir dosing. Higher doses of lopinavir used did not lead to dose-proportional absorption . Summary results are shown in Fig. 1: there was a large rise in lopinavir Cmin when the ritonavir dose was raised from 50 to 100 mg b.i.d. The 600/200 mg once daily dose led to a lopinavir Cmax over three times higher than the 600/50 mg dose . However, raising the lopinavir dose from 200/100 to 400/100 mg b.i.d. raised the lopinavir AUC by only 35%.
Abbott 720 trial
In Abbott 720 trial, 100 treatment-naive patients were given 48 weeks of treatment with three different doses of lopinavir/ritonavir: 200/100, 400/100 and 400/200 mg b.i.d. . Mean 12 h AUC lopinavir plasma concentrations (evaluated in 45 patients at week 24) were higher for the 400/200 mg b.i.d. arm (mean 105.6 µgh/ml) compared with the 400/100 mg b.i.d. arm (mean 76.5 µgh/ml), suggesting that higher ritonavir dose increases lopinavir exposure . As with the previous study, doubling the lopinavir dose from 200/100 to 400/100 mg b.i.d. led to only a 50% rise in 12 h lopinavir AUC (from 50 to 76.5 µgh/ml).
All 16 patients treated with the 200/100 mg b.i.d. dose of lopinavir/ritonavir showed HIV RNA reductions below 50 copies/ml after 48 weeks of treatment, compared with 76% patients (37/49) given lopinavir/ritonavir 400/100 mg b.i.d. However, this apparent benefit in efficacy for the lower dose of lopinavir/ritonavir was driven more by adverse events than antiviral efficacy .
French kaledose trial
In a cohort study, 28 patients in France with HIV RNA levels below 50 copies/ml on lopinavir/ritonavir 400/100 mg b.i.d. had their dose reduced to 266/66 mg b.i.d., using the old soft-gelatin capsule formulation . The mean Cmin fell from 7.4 µg/ml, during treatment with the 400/100 mg b.i.d. dose, to 4.5 µg/ml, during treatment with 266/66 mg b.i.d. There were no data on AUC or Cmax from this study and it was not included in the meta-analysis.
Thai HIV Netherlands Australia Thailand 019 trial
Forty-eight treatment-naive patients were randomized to lopinavir/saquinavir/ritonavir in four different dosing schedules, in the absence of nucleoside reverse transcriptase inhibitors (NRTIs). Pharmacokinetics and efficacy were evaluated to week 24. For the 23 patients receiving a lopinavir/ritonavir dose of 400/100 mg b.i.d., the mean 12 h AUC was 125.7 µgh/ml, compared with 68.6 µgh/ml for the 24 patients receiving the 266/66 mg b.i.d. dose of lopinavir/ritonavir. The efficacy rates were similar in the treatment arms using lower doses of lopinavir/ritonavir, compared with those using higher doses . The changes in lipid profiles and body composition were more pronounced in the highest dose arm.
Abbott 049 trial
In Abbott 049 trial, a pharmacokinetic trial, 36 multiple protease inhibitor-experienced patients were randomized to receive lopinavir/ritonavir at either the 400/300 or 667/167 mg b.i.d. doses . Pharmacokinetic samples were obtained from 33 participants. Ritonavir concentrations were measured over 12 h, after 3 weeks of dosing. The results were compared with historical control data obtained from patients receiving lopinavir/ritonavir 400/100 mg b.i.d. with two NRTIs. The mean 12 h lopinavir AUC was similar, when comparing the 400/300 and 667/167 mg b.i.d. doses (144.9 and 164.5 µgh/ml, respectively). The 400/300 mg b.i.d. dose produced a 12 h AUC 59% higher than the 400/100 mg b.i.d. dose .
Meta-analysis of pharmacokinetic trials of lopinavir/ritonavir
In the meta-analysis of the clinical pharmacology trials described above, lopinavir plasma concentrations correlated strongly with both lopinavir and ritonavir doses used: a larger dose of ritonavir could compensate for a smaller dose of lopinavir. Figure 2 shows the predicted plasma drug levels of lopinavir for three new doses, relative to the standard 400/100 mg b.i.d. dose. The three doses selected were 200/50 mg b.i.d. (one lopinavir/ritonavir 200/50 mg tablet per dose), 200/150 mg b.i.d. (one lopinavir/ritonavir 200/50 mg and one 100 mg ritonavir tablet per dose) and 400/200 mg b.i.d. (two lopinavir/ritonavir 200/50 mg and one 100 mg ritonavir tablet per dose). GMRs are shown, with associated 95%CIs: a GMR of 1 (or 100%) would mean that the lopinavir plasma concentration for the new dose of lopinavir/ritonavir is the same as the standard dose.
The predicted lopinavir AUC and Cmin for the 200/50 mg b.i.d. dose of lopinavir were 47 and 54% lower, respectively, than the standard 400/100 mg b.i.d. dose, with 95%CIs not overlapping 1. However, the predicted lopinavir AUC and Cmin for the 200/150 mg b.i.d. dose (one 200/50 mg tablet, one ritonavir 100 mg tablet) would be close to the standard lopinavir/ritonavir dosing of 400/100 mg b.i.d., with 95%CIs either close to or overlapping 1. The predicted 12 h AUC for the 200/150 mg dose would be 15% lower than the standard dose, with 95%CIs from 36% lower to 13% higher than the 400/100 mg b.i.d. dose. The Cmin for the 200/150 mg b.i.d. dose would be 24% lower, with 95%CIs ranging from 52% lower to 20% higher than the standard 400/100 mg b.i.d. dose. A 400/200 mg b.i.d. dose of lopinavir/ritonavir led to a predicted 59% increase in AUC and a 69% increase in Cmin, compared with the standard 400/100 mg b.i.d. dose (Fig. 2).
The main efficacy trials of lopinavir/ritonavir were conducted using the soft-gelatin formulation, but the use of the new Meltrex formulation has led to lopinavir plasma AUC plasma levels 25-36% higher than the original soft-gel formulation in the Abbott 730 trial . Use of the Meltrex formulation for the 200/150 mg b.i.d. dose could raise the lopinavir drug levels, compensating for the slightly lower predicted levels at this dose.
Besides studying the optimal boosting dose of ritonavir, alternative pharmacoenhancers may improve future treatment for HIV-1-infected individuals. The ideal booster should have less toxicity and no anti-HIV activity. Antiviral activity is particularly undesirable when ritonavir is used in a nonprotease inhibitor-based setting: for example, to boost the integrase inhibitor elvitegravir. After unsuccessfully testing other CYP3A inhibitors, like ketoconazole , two new promising mechanism-based CYP 3A inhibitors were presented as an alternative for ritonavir.
The GS-9350 compound showed a similar pattern of enzyme inhibition specificity to ritonavir, with a many fold stronger inhibition for 2C8, 2C9 and 2D6 . When this pharmacoenhancer was administered with midazolam, a 95% reduction in oral clearance was seen for both the 100 and 200 mg dosage, which was comparable to the effect ritonavir had on the clearance. In a follow up study comparing tenofovir/emtricitabine/elvitegravir with either GS-9350 or ritonavir, elvitegravir boosted with 150 mg of GS-9350 showed similar concentrations as the elvitegravir boosted with 100 mg ritonavir; 100 mg of GS-930 showed inferior levels, especially for Cmin. No major safety concerns were raised in any stage of the drugs development, no effects on lipids in particular.
The other pharmacoenhancer was SPI-452 . To date, no short-term safety issues have been found. In vitro, this booster had a boosting effect comparable to ritonavir for a range of HIV protease inhibitors. In humans, the AUC of this drug was dose-proportional up to 100 mg and variable dose-proportional above 100 mg. Saquinavir concentrations were significantly increased when boosted with either 50 or 200 mg of SPI-452. Although 200 mg of SP-452 resulted in higher saquinavir concentrations, it did not seem proportional to the increased dose. The doses of 50 and 200 mg of SPI-452 showed similar boosting effects on darunavir, when measured as AUC, but the higher dose tended to raise the Cmin of darunavir more markedly. The highest dose of SPI-452 (200 mg) had the strongest boosting effects on atazanavir.
Despite these promising data for both pharmacoenhancers, long-term efficacy and safety data are needed before wide scale use can be approved. As SPI-452 shows a different interaction with darunavir compared with atazanavir, new boosters may need to be used at different doses to boost different antiretrovirals and other drugs.
The present systematic review of 17 clinical pharmacology trials shows that ritonavir boosts the protease inhibitors in different ways. For saquinavir, fosamprenavir and darunavir, there was no correlation between higher ritonavir doses and higher protease inhibitor concentrations. By contrast, for lopinavir, indinavir and tipranavir, use of higher doses of ritonavir led to higher concentrations of the boosted protease inhibitor. The data on atazanavir are from one cohort study and are inconclusive, but suggest that higher ritonavir doses may not further increase atazanavir exposure.
There was no clear correlation between the ritonavir dose needed to boost each protease inhibitor and the effect of each protease inhibitor on ritonavir levels. There was also no trend for protease inhibitors boosted in a particular way (e.g. mainly from reduced clearance) or those with low bioavailability as monotherapy to need higher doses of ritonavir. The mechanism of ritonavir boosting is not fully understood and may involve other important pathways in addition to CYP3A4 metabolism.
The minimum tablet strength for ritonavir is 100 mg, but two trials have evaluated 50 mg doses, using the liquid formulation: these suggest that 50 mg doses are sufficient to boost either saquinavir or fosamprenavir. It is unknown whether a 50 mg dose of ritonavir could also boost darunavir or atazanavir to the same extent as the approved 100 mg dose. A 50 mg dose of ritonavir might be easier to coformulate with protease inhibitors, which could improve adherence as well as lowering costs. A 50 mg ritonavir tablet would also be very useful for paediatric use. Further clinical trials evaluating saquinavir, atazanavir, fosamprenavir and darunavir with lower doses of ritonavir are warranted to define the minimum dose of ritonavir required for boosting: results from these trials could help to justify development of new heat-stable ritonavir tablets at lower doses.
The adverse event profile of boosted protease inhibitors is affected by both the dose of ritonavir used and the adverse events caused by the boosted protease inhibitor. For example, lopinavir/ritonavir 400/100 mg b.i.d. and saquinavir/ritonavir 1000/100 mg b.i.d. were compared in the GEMINI trial: both boosted protease inhibitors included 200 mg of ritonavir daily, and lopinavir is known to lower ritonavir drug concentrations. However, there were significantly more gastrointestinal adverse events in the lopinavir/ritonavir arm . Therefore, an optimal safety profile depends on both the choice of protease inhibitor and the lowest dose of ritonavir.
The meta-analysis of five clinical trials of lopinavir/ritonavir shows that the pharmacokinetics of lopinavir is highly dependent on the dose of ritonavir used, as well as the lopinavir dose. Consequently, it may be possible to use lower lopinavir dosing in exchange for a slightly higher dose of ritonavir. The potential from this result is a new dosage of lopinavir-ritonavir (200/150 mg b.i.d.), which would involve the same pill count (two b.i.d.) but with a lower treatment cost. The predicted lopinavir plasma exposure at 200/150 mg b.i.d. may be slightly lower than for standard 400/100 mg b.i.d. dosing, but in protease inhibitor-naive patients, strong efficacy has been shown for lopinavir/ritonavir at lower doses: either 200/100 mg b.i.d. in the Abbott 720 trial  or 266/66 mg b.i.d. in the HIV Netherlands Australia Thailand (HIVNAT) 019 trial . Therefore, a small reduction in lopinavir concentrations is unlikely to affect clinical efficacy, especially if the Meltrex formulation is used, with its higher bioavailability [43,46]. Lipid elevations on lopinavir have been shown to correlate with lopinavir exposure [49,50]. If the new dosage leads to similar lopinavir concentrations, the effects on lipids from the lopinavir component may also be similar. With the 200/150 mg b.i.d. dose, there would be a total daily ritonavir intake of 300 mg, compared with a 200 mg intake with the 400/100 mg b.i.d. dose schedule. However, as lopinavir is known to lower ritonavir concentrations by 50% , the actual difference in ritonavir exposures between the doses may be small. With lopinavir/ritonavir treatment, plasma concentrations of ritonavir are 10 times lower than those measured for lopinavir. Therefore, the ritonavir component may play a relatively small role in lipid elevations.
These results are from a meta-analysis of public domain data and need to be validated by a prospective cross-over pharmacokinetic trial. In such a trial, patients or volunteers could be dosed with either the 400/100 mg b.i.d. or 200/150 mg b.i.d. doses of lopinavir/ritonavir, and the lopinavir pharmacokinetics compared after intensive assessment at steady state on each dose. Safety parameters would need to be assessed as well.
In summary, this systematic review shows that ritonavir has dose-dependent boosting effects on the protease inhibitors lopinavir, indinavir, tipranavir, but lower and higher doses of ritonavir appear to have equal effects on saquinavir, fosamprenavir and darunavir. These results may help in the design of dose-ranging trials of other pharmacoenhancers currently in development.