Editorial

Mark Bennett Feinberg, MD, PhD, Emory Vaccine Research Center, Atlanta, Georgia

The advent of potent drugs to treat HIV infection and the development of increasingly effective treatment strategies have resulted in dramatic improvements in the prognosis and quality of life for HIV-infected persons. However, two major challenges now exist in developing even more effective treatments. First, treatment regimens typically require the use of three or more antiviral drugs, resulting in complex medication schedules that require patients to take multiple pills at multiple times each day. As the relative complexity and inconvenience of a specific treatment regimen is likely to impact a patient's adherence and their ultimate chance for long-term therapeutic success, the development of effective, but more convenient, antiviral drugs is an essential goal. Second, there remains a great need to develop alternative antiretroviral treatment regimens for patients who develop resistance to the antiviral drugs included in their initial (or subsequent) antiretroviral therapy combinations. In particular, there is substantial "cross-resistance" seen among different antiretroviral drugs of the same class, including the potent inhibitors of the HIV protease. As a result of the extensive cross-resistance between protease inhibitors (PIs), none of these such drugs currently available have been shown to durably suppress the replication of HIV variants that arise in individuals who had been previously and unsuccessfully treated with other PIs. The development of PIs that select for distinct mutations in the HIV protease gene than do previously available PIs would provide one useful path to successful treatment of patients who develop drug resistant HIV variants as a result of prior unsuccessful PI therapy. In addition, growing evidence indicates that strategies to alter the metabolism of PIs can result in substantially increased concentrations of the drug in the blood and appreciably prolong the half-life of the drug. Because the antiviral effect of any antiretroviral drug is influenced by its associated resistance pattern and by the level of the drug attained during treatment of HIV-infected persons, maintenance of high concentrations of a potent PI may, in certain instances, be able to successfully inhibit the replication of even those HIV variants that already harbor a number of mutations associated with PI resistance.

For these reasons, the development of the combined formulation of the potent PIs lopinavir (LPV) and ritonavir (RTV), known as Kaletra® (formerly called ABT-378/ritonavir), represents a promising addition to the available antiretroviral treatments for HIV infection. Lopinavir/ritonavir (LPV/RTV) acts as a potent inhibitor of the HIV protease, and its pharmacologic properties yield high levels of the drug in-vivo, thereby permitting a simple, twice a day treatment schedule. Furthermore, this combination drug therapy has proven itself to be an important component in the formulation of successful treatment regimens for patients who have failed prior therapy. While recent data on LPV/RTV indicate that it likely represents a valuable advance in HIV therapy, important questions remain concerning how and when to best exploit the drug's favorable properties. Should, for example, the drug become a routine component of initial combination therapy regimens, or alternatively, should it be reserved for those individuals who fail their initial antiretroviral treatments? What will be the long-term consequences of the lipid abnormalities seen in many LPV/RTV-treated patients, and how can these best be managed to minimize their clinical significance? Nevertheless, many patients who have appeared to have run out of options for successful antiretroviral therapy may now be able to obtain substantial benefit through the carefully considered formulation of salvage therapy regimens that include LPV/RTV as an active constituent drug.

Introduction

Based on the results of recent clinical trials, LPV/RTV was granted accelerated approval by the United States Food and Drug Administration on September 15, 2000 for the treatment of HIV infection in adults and children six months and older, in combination with other antiretroviral medications. Findings from a number of recent studies evaluating the pharmacology, clinical effectiveness, side effects, and drug interactions of LPV/RTV were presented during the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC). Collectively, these data indicate that LPV/RTV represents a promising new antiretroviral agent-its favorable pharmacologic properties and pattern of resistance mutations may enable PI-experienced patients whose resident HIV populations harbor multiple baseline mutations-which is associated with PI resistance-to achieve durable suppression of HIV replication.

The rationale for the combination of LPV with RTV in Kaletra® is found in the ability of RTV to potently inhibit the cytochrome system and alter its own metabolism, as well as those of other protease inhibitors. RTV has previously been proved effective in favorably altering the pharmacokinetics of other PIs, such as saquinavir (Fortovase®, Invirase®) and indinavir (Crixivan®). As a result of the RTV-mediated increases in plasma drug levels and drug half-lives, the dosing interval of the second PI can be lengthened. By formulating LPV with RTV, substantial increases in the oral bioavailability of the drug and in the levels of plasma drug concentrations achieved are realized. As a result, a simplified dosing schedule is possible, and drug concentrations can be reached in treated patients that are in excess of the concentrations needed to inhibit the replication of many PI-resistant HIV variants. It is recommended that LPV/RTV be given with a patient's choice of food, twice daily for adults, using either three capsules or five milliliters of a liquid formulation (total of 400 mg lopinavir/100 mg ritonavir for each of the two daily doses). The dose for children six months to 12 years is based on weight.

In his presentation on lopinavir/ritonavir, Dr. Kempf, a researcher from Abbott Laboratories (the manufacturer of LPV/RTV), noted that the company had specifically set out to identify a compound that would be enhanced by ritonavir (Norvir®) to a greater extent than any other currently available PI. In vitro experiments indicate that the concentration of ritonavir needed to inhibit the metabolism of lopinavir is approximately ten times less than that needed to inhibit the metabolism of indinavir sulfate (Crixivan®), for example.

"Within this regimen, lopinavir really provides the antiviral activity, and ritonavir is only there to boost its pharmacokinetics," Dr. Kempf said. "But because lopinavir is very sensitive to ritonavir, we can achieve very high blood levels with good tolerability."

Clinical Trial Results

Updates on three clinical trials of LPV/RTV were presented at the ICAAC meeting. One presentation focused on the comparative activity of LPV/RTV versus nelfinavir mesylate (Viracept®)-the PI component of the initial combination therapy used for antiretroviral-naive patients. A second presentation provided information about the long-term effectiveness (96 week follow-up) and tolerability of LPV/RTV in combination with stavudine (Zerit®) and lamivudine (Combivir®, Epivir®). The third presentation described results of a study evaluating the utility of LPV/RTV as a component of a "salvage therapy" regimen for treatment of individuals who had failed prior PI therapy due to the development of drug resistance.

Additional presentations described the virologic and genotypic data obtained in the course of this study that correlate drug resistance patterns and clinical outcomes.

A large, randomized, double-blind, Phase III study compared lopinavir/ritonavir, as a component of the initial antiretroviral treatment regimen in antiretroviral-naive patients, with nelfinavir, both used in combination with stavudine and lamivudine. In this study, the 653 participants enrolled had a mean plasma HIV RNA level of 4.9 log₁₀ copies/mL and a mean CD4+ T-cell count of 260 cells/mm³. Data from this study were evaluated using both "intent-to-treat" (ITT) and "on-treatment" (OT) analyses. ITT analysis incorporated data on all study participants, including those who left the study early for any reason and those who were considered treatment failures. OT analysis included data for patients who remained on treatment and for whom results were available at a particular timepoint. As the goal of any anti-HIV treatment regimen is to suppress levels of ongoing virus replication to below the limits of detection of sensitive assays for plasma HIV ribonucleic acid (RNA), this virologic assay provided an important end-point for the study. After 40 weeks of therapy, plasma HIV RNA levels were less than 400 copies/mL in 79 versus 64 percent (by ITT analysis) and 94 versus 83 percent (by OT analysis) with LPV/RTV and nelfinavir, respectively. When a more sensitive assay with a detection limit of 50 copies of HIV RNA/mL of plasma was used, suppression of HIV replication was also better in the LPV/RTV group. Tolerability of the two regimens was similarly good in both treatment groups. However, the rate of moderate to severe elevations in non-fasting triglyceride levels was higher in LPV/RTV than nelfinavir-treated participants.

Dr. Constance Benson (University of Colorado) presented an update on a cohort of previously untreated patients who had received LPV/RTV in combination with stavudine and lamivudine for 96 weeks. In contrast to the study described above, this study did not include comparisons with alternative potent combinations. Nevertheless, the LPV/RTV plus stavudine and lamivudine combination used in this long-term study performed very well clinically and was very well tolerated. A total of 86 of the initial 100 patients treated remained on therapy at 96 weeks, and only two patients discontinued treatment as a result of drug toxicity (diarrhea in one, and abnormal liver function tests in another). Fourteen percent of patients had cholesterol levels in excess of 300 mg/dL, while 12 percent had triglyceride levels greater than 750 mg/dL. Virologic analyses showed that 78 percent of the treated group maintained plasma HIV RNA levels of less than 50 copies/mL at 96 weeks when analyzed by ITT methods.

Encouraging results were also presented from a smaller, open-label, Phase II study of the activity of LPV/RTV in patients who had previously been treated with other PIs, but who had evidence of persistent virus replication-plasma HIV RNA levels exceeded 1,000 copies/mL while on continuing therapy. Fifty-seven patients who were non-nucleoside reverse transcriptase inhibitor (NNRTI) naive, but who had been previously treated with an average of seven antiretroviral drugs including an average of three PIs, were studied. Participants were treated with one of two doses of LPV/RTV (440 mg/100 mg or 533 mg/133 mg twice daily) plus efavirenz (Sustiva®) and nucleoside analog RT inhibitors (NRTIs) of the investigator's choice. At baseline, 68 percent of participants had four-fold or greater decreases in susceptibility to three or more PIs. Forty-three percent also had more than a ten-fold decrease in susceptibility to LPV/RTV, compared to the wild-type-fully drug-sensitive-virus. Despite the presence of substantial baseline phenotypic resistance to PIs (and to LPV/RTV itself), however, higher rates of suppression of HIV replication were achieved than have been seen in other salvage therapy studies. After 24 weeks of therapy, 75 percent of participants had plasma HIV RNA levels of less than 400 copies/mL when analyzed by ITT methods, and 86 percent had levels below 400 copies/mL when analyzed by OT methods. As would be expected, participants with higher baseline levels of phenotypic resistance to PIs were less likely to achieve or maintain viral suppression. However, even among those patients whose baseline HIV isolates manifested 20-40 fold-increased concentration of the drug needed to inhibit viral replication by 50 percent in in vitro cultures-also known as IC₅₀, a measure of the phenotypic resistance of the virus to a given drug-to LPV, 67 percent reached levels of plasma HIV RNA of less than 400 copies/mL. A higher percentage of participants in the higher dose LPV/RTV group (533 mg/133 mg) achieved a durable virologic suppression compared with the lower dose group. In the course of the study, it was noted that efavirenz lowered the LPV/RTV trough levels by approximately 33 percent, and this may account for the higher success rates observed in the higher dose LPV/RTV group. These study results are significant, given the known difficulty in maintaining durable suppression of HIV replication in patients who have failed prior PI therapy. While efavirenz clearly contributed to the therapeutic success observed in patients in this study, the effects of this NNRTI would have likely been transient at best without the substantial activity of LPV/RTV against the pre-existing drug resistant viral variants.

Baseline Genotype and Risk of Virologic Failure

Dr. Dale Kempf and colleagues have analyzed the genotype of viral variants present at baseline in the participants in the "salvage therapy" study described above, and they have assessed how the nature and number of protease resistance mutations affects treatment outcomes. Investigators determined the extent to which the presence of a range of baseline resistance to 11 mutations leading to changes in the amino acids, which are actually encoded for synthesis of the protease protein, known to be associated with a reduced phenotypic susceptibility to LPV/RTV affected response rates at 24 weeks. Baseline phenotypic and genotypic tests were done on all 57 patients enrolled in the study described above. "The median number of prior PI use in the study was three," Dr. Kempf noted. Results from baseline phenotypic and genotypic testing indicated that there was a broad range of susceptibility to LPV/RTV, from drug sensitivity levels similar to that of wild-type virus to a virus strain that was 96-fold more resistant than wild-type.

Most of the patients had four or more mutations-a reflection of their previous exposure to PIs, noted Dr. Kempf: "As such, they would not be expected to respond to a traditional PI." At 24 weeks, responses among 52 out of 57 evaluable patients were analyzed with respect to the number of baseline mutations. Ninety-six percent of patients with five or fewer mutations at study entry achieved HIV RNA levels of less than 400 copies/mL, and 88 percent of them achieved HIV RNA levels of less than 50 copies/mL. An intermediate response was seen in patients who had six or more mutations at baseline-76 percent of patients achieved HIV RNA levels of less than 400 copies/mL, and 57 percent of them achieved less than 50 copies/mL. "When we looked at patients with the highest number of mutations, response rates were statistically significantly lower than the other two groups," Dr. Kempf indicated. Specifically, only 33 percent of patients with eight to ten baseline mutations reached less than 400 copies/mL by week 24, while only 17 percent of patients achieved HIV RNA levels of less than 50 copies/mL.

"So far, the best predictor of response is to simply count the mutations. If there are five or fewer mutations at baseline, there is a high likelihood the drug will be highly active in patients," Dr. Kempf said. Studies indicate that plasma levels achieved by LPV/RTV are approximately 75-fold greater than those needed to suppress replication of wild-type virus.

Hence, in patients with up to five mutations, "you would still expect good viral suppression, and that is what we saw," Dr. Kempf said. He also noted that it did not seem to matter which mutations were present at baseline in terms of achieving a good response to LPV/RTV. Indeed, many of the viruses tested at baseline carried common mutations for PIs, including 82, 84, and 90.

Other investigators collaborating with Abbott also presented data concerning the impact of baseline genotype on the risk of subsequent virologic rebound following initial viral suppression while on continued LPV/RTV treatment. As discussed by Dr. S. Brun, 23 out of 227 patients enrolled in three Phase II clinical trials evaluating the new PI either did not achieve HIV RNA levels below 400 copies/mL on lopinavir/ritonavir, or they had a sustained viral load rebound in excess of 400 copies/mL while on therapy.

"We found three distinct patterns of drug resistance during rebound from lopinavir/ritonavir, and these patterns could be differentiated on the basis of the baseline ABT-378/ritonavir mutation scores," Dr. Brun reported. Four antiretroviral-naive patients and six PI-experienced patients had mutation scores of zero to two. "At rebound, no additional PI mutations occurred in these patients, and consequently, no phenotypic resistance to protease inhibitors developed," Dr. Brun noted.

At the other end of the spectrum, there were eight multiple PI-experienced patients who entered the trials with over six mutations. On initial rebound, investigators again noted no evolution in protease resistance in these patients, although reverse transcriptase resistance occurred in both patient groups, he added. The more interesting group included five patients who were either single or multiple PI-experienced prior to being treated with LPV/RTV. These patients had four to five mutations associated with reduced PI susceptibility at baseline.

"This was the only group in which we saw an accumulation of further mutations in the protease enzyme on rebound and the subsequent development of phenotypic resistance to lopinavir/ritonavir," Dr. Brun said. The reduction in phenotypic susceptibility to the new PI was also fairly striking, ranging anywhere from nine to 99-fold relative to wild-type virus, he added. At the point of rebound, these patients had also accumulated an additional one to four protease mutations.

Patients who developed reduced susceptibility to the new PI during rebound remained or became resistant to indinavir sulfate, ritonavir, and nelfinavir mesylate, he added. In contrast, they remained susceptible in vitro to saquinavir, amprenavir (Agenerase®), and tipranavir (a new nonpeptidomimetic protease inhibitor now being developed by Boehringer-Ingelheim).

Thus, investigators suggested that together with ritonavir enhancement, "saquinavir, amprenavir, and tipranavir may be useful for salvage when ABT-378 resistance is present." However, any such predictions based on laboratory analyses will need to be rigorously evaluated in clinical trials designed to identify effective treatment options for patients who develop resistance to LPV/RTV.

Lopinavir/Ritonavir and Relevant Non-PI CYP3A Interactions

As a potent inhibitor of the P450 3A isoenzyme (CYP3A), LPV/RTV has the potential to interact with other drugs that are dependent on the same enzyme for clearance, including those that may be used to counteract the metabolic side-effects of its use, such as the observed elevations of cholesterol and triglyceride levels. Noting that the use of various statins is increasingly common among HIV patients for the treatment of dyslipidemias, Dr. Robert Carr, a senior pharmacokineticist at Abbott Laboratories, explored the potential for drug interactions with atorvastatin calcium (Lipitor®) and pravastatin sodium (Pravachol®). Twelve volunteers received four days of treatment with atorvastatin calcium, 20 mg/day, or pravastatin sodium, 20 mg/day, given together both in the presence and absence of LPV/RTV, in standard doses of 400/100 mg twice a day for 14 days.

"Pharmacokinetic assessments were obtained over a dosing interval for each parent drug, as well as each drug's metabolites," Dr. Carr noted. The statins were taken 30 minutes before breakfast, while LPV/RTV was taken 30 minutes after breakfast and dinner. Results indicate that there was a "very significant interaction" with atorvastatin calcium. Specifically, plasma levels of atorvastatin calcium increased approximately six-fold in the presence of LPV/RTV.

"Clinically, this means that if you give atorvastatin in doses of ten milligrams a day, the concentration of [the] parent drug in the blood would be equivalent to a dose of about 60 mg a day, and that does not take into account the effect ABT-378 had on the active metabolites," Dr. Carr commented. In contrast, plasma levels of pravastatin sodium increased by approximately 30 percent, which is not likely to be clinically significant, he added.

"Our advice is to stay with ten milligrams a day if you're using atorvastatin [with LPV/RTV] or use another statin [that LPV/RTV doesn't affect that much], like pravastatin," Dr. Carr concluded. Dr. Richard Betz, another pharmacokineticist at Abbott Laboratories, in turn, reported that the co-administration of lopinavir/ritonavir with efavirenz can reduce the concentrations of lopinavir, and, in particular, the trough concentrations by 40 to 45 percent. "Efavirenz concentrations were not affected by the co-administration of Kaletra®," Dr. Betz noted.

Given the high plasma concentration achieved with LPV/RTV, this interaction may not be that important, Dr. Betz suggested. But in patients who are highly antiretroviral experienced, with multiple genotypic or phenotypic resistance to lopinavir, "the dose [of lopinavir/ritonavir] should probably be increased," Dr. Betz said. Lopinavir/ritonavir is usually given in doses of 400/100 mg twice a day, which is three capsules, twice daily.

For heavily pre-treated HIV patients, Dr. Betz suggested physicians increase the dose to four capsules twice a day, or 533/133 mg twice a day. "It's up to clinicians' judgment to increase the dose or not, but we think that it is most important to do so in highly experienced patients," he confirmed. Finally, Dr. Betz and colleagues assessed bioequivalence between the liquid and soft gel capsule co-formulations, both of which are approved for HIV therapy in the United States.

"When we looked, in healthy volunteers, at blood concentrations of these two formulations in a cross-over fashion, we found that the liquid and the capsule produced very similar concentrations at the same dose, so they are interchangeable," Dr. Betz said. He also noted that the bioequivalence of the two formulations was assessed when volunteers took the drug with a meal containing modest amounts of fat.

If patients have a high fat meal-bacon and eggs at breakfast for example-"you do get a higher increase in the concentration of lopinavir from Kaletra®," Dr. Betz indicated, "but we don't believe that these increases are clinically important. As long as you take the drug with a meal, you should have adequate concentrations."

Unanswered Questions and Future Research Needs

In conjunction with results of other LPV/RTV Phase II studies, the data presented on LPV/RTV at the ICAAC meeting indicate that LPV/RTV can provide a potent, effective, relatively convenient, and well-tolerated protease inhibitor for use in combination therapy regimens for treatment-naive patients. However, a number of important questions about its optimal use in the care of HIV-infected persons will need to be resolved in future studies.

Importantly, it is not yet known which type of initial therapy regimen (e.g. PI-based, non-nucleoside reverse transcriptase inhibitor-based, or triple nucleoside-based) will be the most effective and least toxic in the long run. In the context of the PI component of combination antiretroviral therapy regimens, it is not yet known how the efficacy and tolerability of LPV/RTV will compare with other RTV-enhanced combinations of PIs (e.g., indinavir). Further, given that LPV/RTV has potent antiviral activity and activity against HIV variants that harbor some mutations that confer resistance to other PIs, physicians will need to decide whether it is best used as a first-line agent or reserved for patients who fail their initial treatment regimen. Toward this end, direct comparison studies will be necessary to determine which initial treatment selection most effectively preserves future treatment options for patients who develop resistance to the components of the initial regimen. While LPV/RTV appears to be very well tolerated, concerns exist about the potential long term consequences of the hyperlipidemia (observed in ten to 30 percent of LPV/RTV-treated individuals), as well as how to most efficiently manage this metabolic side-effect.