1. Home>
2. Specialties>
3. HIV/AIDS Clinical Care>
4. Feature

Update on HIV Protease Inhibitors

For the past few years, therapeutic strategies for treatment of HIV infection have been based principally on drugs that target HIV reverse transcriptase, such as nucleoside analogues. The clinical usefulness of these agents has been proved in a number of controlled clinical trials. However, the immunological improvement and viral replication inhibition obtained by these drugs are incomplete and the duration of clinical benefit is always transient. A number of factors may contribute to this phenomenon, including the low therapeutic index of the available drugs, the impossibility of a complete immunological reconstitution, the emergence of strains with reduced sensitivity, and the fact that these drugs may not be active on chronically infected cells such as monocytes/macrophages. The combination of two reverse transcriptase inhibitors has been shown to increase therapeutic activity of this class of drugs, as shown by the recently released results of ACTG 175 and the European Delta trial. (See back page.)

Recent insights into HIV pathogenesis (see ACC, Mar 95, p 24, and Aug 95, p 63) give rise to hope for an effective therapeutic intervention based on even more potent combinations of antiretroviral agents, possibly acting at different sites of HIV replication. HIV protease inhibitors are a new class of antiretroviral drugs aimed at a different target than reverse transcriptase. Protease inhibitors have a high antiviral potency and a favorable toxicity profile, so they may be good candidates for such a multicombination therapeutic approach.

General Characteristics of HIV Protease and Protease Inhibitors

The protease of retroviruses is an enzyme consisting of two identical protein chains that combine to form a single active site. During its replication HIV generates large precursor proteins -- mainly from the gag and pol genes -- that undergo processing when the viral particles bud out from the infected cells. The processing of the gag-pol polyprotein is driven by the HIV protease, which is able to autocatalytically cleave itself out of the precursor protein, then process the remainder into other structural and functional proteins, including reverse transcriptase, RNase, and integrase. If this process is inhibited, the viral particles are structurally disorganized, incapable of functioning, and, therefore, non-infectious.

Protease inhibitors began to be developed after the essential role of HIV protease in the biology of the virus had been recognized in 1988 and its primary structure described. Inhibitor design is based on constructing either peptide-based competitive mimetics of the protease substrate (peptidic inhibitors) or molecules that complement the active site of the enzyme (C2 symmetric inhibitors, which are only partially peptidic). A third chemical class of inhibitors is the nonpeptidic inhibitors; these may be simpler and cheaper to manufacture.

Overall, more than 20 laboratories have designed more than 300 HIV protease inhibitors. However, only a few have passed all the in vitro screening phases, particularly in terms of specificity (the ability to inhibit HIV protease without inhibiting human proteases). The high specificity of some protease inhibitors is expected to translate into lower toxicity, in particular, lower than nucleoside analogues that may interfere with cellular enzymes involved in the synthesis of nucleic acids.

It is worth noting that the design of HIV protease inhibitors clearly benefited from the past two decades' extensive research on specific inhibitors of human proteases, some of which are now in wide clinical use (e.g., angiotensin-converting enzyme inhibitors for treatment of hypertension).

At least ten HIV protease inhibitors are currently in clinical trial or ready to enter clinical development. They are all active in vitro against HIV-1 (and, with minor differences, against HIV-2) at nanomolar concentrations (whereas reverse transcriptase inhibitors are generally active at micromolar concentrations). In contrast to nucleoside analogues, which must be phosphorylated to become active within the cells, protease inhibitors do not need further intracellular processing to exert their action. Moreover, because they intervene at a post-translational phase of HIV replication, they are effective both in acutely and chronically infected cells. Because of this they also inhibit HIV replication in populations of cells such as macrophages, which are considered a major reservoir of HIV but do not efficiently metabolize nucleoside analogues.

Pharmacokinetic properties are crucial to an effective delivery of drugs to the target cell. Antiretroviral agents should have a high oral bioavailability and a low rate of excretion to permit oral dosing on a reasonable schedule. However, because of their chemical characteristics, peptidic protease inhibitors, in particular, may have suboptimal oral bioavailability. Another pharmacokinetic problem observed with some protease inhibitors is the binding to a specific plasma protein, alpha-acid glycoprotein (AAG), that may inhibit intracellular uptake. The development of one protease inhibitor (SC 52151) was stopped a few months ago because of ACTG 282 results that showed no activity in vivo, despite very high plasma levels, because of high-affinity binding to alpha-acid glycoprotein.

A large number of protease inhibitors are in early clinical development. A partial list includes the following compounds: VX-478 (Vertex/Glaxo Wellcome); AG1343 (Agouron); U-96988, U-103017, and U-104904 (Upjohn); BMS 182 and 193 (Bristol-Myers Squibb); DM 323 and DM 450 (DuPont Merck); KNI 227 and KNI 272 (Nikko Kyodo); SB 204144 and SB 206343 (Smith Kline); S338 (Monsanto Searle).

However, three HIV protease inhibitors are in more advanced phases of clinical development: saquinavir (formerly Ro-31 8959, developed by Hoffman-La Roche), indinavir (MK-639, developed by Merck), and ritonavir (ABT-538, developed by Abbott). Some in vivo activity data on these drugs are impressive, although based only on changes in virological (in particular, plasma HIV RNA) and immunological (mainly, CD4 lymphocyte) markers. Clinical data are expected to emerge from ongoing phase III trials and it is hoped they will clarify the usefulness of this new class of compounds in the management of HIV disease.

Saquinavir

This compound has been extensively tested in various phase I/II trials in both previously treated and untreated patients.

In antiretroviral-naive patients, saquinavir as monotherapy induced a median decrease of plasma HIV RNA of 0.6 log, comparable to the decrease previously seen with nucleoside monotherapy. In a study of saquinavir (1800 mg per day) in combination with AZT (600 mg per day) in patients with no prior antiretroviral therapy, a peak median decrease of plasma RNA of 1.5 log was seen at 2 weeks, and a median 99-cell increase in CD4 lymphocytes was still present at 20 weeks.

Another study (ACTG 229) evaluated a triple combination of saquinavir plus AZT plus ddC in 300 patients with advanced HIV infection who had already received AZT treatment. The triple combination was associated with significantly greater increases in CD4 cell count and reductions in viral load than either of the two control combinations (saquinavir plus AZT and AZT plus ddC).

In ACTG 229, CD4 and RNA responses in the triple combination arm were still different from baseline values after 48 weeks, but in both trials viral load and CD4 responses tended to return to pretreatment values over time. Quite consistently in all trials, drug-resistant strains of HIV have been isolated in 50% to 60% of the treated patients after one year of treatment.

Overall, phase I/II trials of saquinavir, either as monotherapy or combined with AZT and ddC, have proved that, despite its low bioavailability (4% in its present formulation), the drug has a definite antiviral activity in vivo. Indeed, the results of a pilot trial testing higher doses of saquinavir showed that, at 7200 mg per day, saquinavir monotherapy was able to induce a reduction in viral load of nearly 2 logs, suggesting that the current dosage of 1800 mg per day is indeed suboptimal. However, a new, enhanced, oral formulation of saquinavir, which is expected to have five times the bioavailability of the parent compound, may shortly enter clinical trials.

In all trials saquinavir was found to be well tolerated either alone or in combination; the majority of adverse events were mild in intensity and considered unrelated to saquinavir therapy.

Clinical studies of this compound in progress include a large, multinational, phase III trial of triple combination (saquinavir plus AZT plus ddC) versus double combinations (saquinavir plus AZT and AZT plus ddC) versus AZT monotherapy, in 3000 antiretroviral-naive patients with CD4 cell counts in the 50 to 350 range.

Saquinavir recently entered accelerated approval procedures in the U.S.

Indinavir

Indinavir is an orally bioavailable protease inhibitor that has shown, in phase I/II studies, a relevant antiviral effect in vivo, with 1- to 2-log reductions in plasma HIV-1 RNA copies and CD4 cell increases of up to 80 to 140 after a few weeks of treatment at a dosage of 400 to 600 mg four times daily. RNA levels returned to baseline after 12 to 24 weeks, in association with the appearance of viral strains with genotypic changes in the protease gene. However, new data gathered from patients receiving higher doses of indinavir may suggest a sustained effect of the compound at 52 weeks.

Indinavir was well tolerated in all trials; principal side effects were a dose-related hyperbilirubinemia (15% of the patients) and nephrolithiasis, possibly due to crystallization of the compound.

Ongoing clinical trials are now evaluating indinavir (at a dose of 800 mg three times a day) as monotherapy and in combination with stavudine or AZT plus 3TC in patients previously treated or untreated with AZT.

Ritonavir

Ritonavir (formerly ABT-538) is an orally bioavailable protease inhibitor. It is currently in a syrup formulation that must be kept refrigerated. The antiviral activity and safety of this C2 symmetric protease inhibitor were tested in phase I/II clinical trials. In one trial a decrease in HIV RNA plasma copies of 1 to 2 logs was seen after 2 weeks of treatment. After 32 weeks a median increase of 230 CD4 cells and a mean 0.8-log decrease in RNA level were still maintained in some patients receiving a dose regimen of 600 mg every 12 hours.

However, over time, and as a function of ritonavir dose, these measures of viral replication tended to return to baseline values. This phenomenon was temporally associated with the emergence of viral variants with progressively diminishing in vitro susceptibility.

In another trial, an average 180-cell increase in CD4 count was observed. Some patients on the high doses had peak increases in CD4 cell counts of 300 to 400 and viral load reductions of 2.7 log.

It is also worth noting that, although the trials were not designed to detect clinical efficacy, definite signs of clinical improvement were seen in some of the treated patients.

Overall, ritonavir appears to be well tolerated, with side effects including only nausea, diarrhea, perioral paresthesia, increased transaminases and raised triglycerides.

Phase II/III clinical trials with ritonavir will evaluate this protease inhibitor either alone or in combination with AZT, in patients with and without previous antiretroviral experience.

Resistance to Protease Inhibitors

The emergence of resistance is considered to be one of the major reasons for drug failure. The issue of resistance to protease inhibitors is therefore of great interest.

Resistance will eventually arise with all effective antiretrovirals. Important issues of particular relevance for treatment, therefore, are the timing of the appearance of resistance and the possibility of delaying it.

In theory there may be mechanisms that apply to protease inhibitors that would make it possible for the problem of resistance to be overcome. The first would rely on the availability of compounds with a high therapeutic index that could be used at plasma concentrations high enough to exceed the susceptibility of the resistant virus. The second theoretical mechanism, which is an unproved research hypothesis, is based on the possibility of altering the replicative capacity of the resistant virus.

Available data seem to indicate that temporal patterns of resistance development are not identical for all protease inhibitors. Whether this is due to differentially selective pressures exerted by the drug or to the specificity of the drug/protease interaction is a question still under investigation.

Patterns of genotypic resistance also vary. Some inhibitors (i.e., indinavir and ritonavir) preferentially select for mutations at codons 82 and 84 in the protease gene, while others (saquinavir) seem to select for mutations at codons 48 and 90.

Recent observations have led to speculation that combination therapy with multiple protease inhibitors may yield multiple resistant variants; in fact, in one study, when viral isolates from patients harboring indinavir-resistant strains were tested against a number of other protease inhibitors, including ritonavir and saquinavir, some degree of resistance was found to all of these drugs. However, in another study, mutations at positions 10, 46, 63, 82, and 84, which are critical for broad cross-resistance, were not observed during prolonged saquinavir therapy. Thus, the fear that initial therapy with any one inhibitor may limit the benefit of subsequent treatment with the remaining compounds may not be a generalizable argument. However, it deserves further investigation.

Future Perspectives

The limited data available at this writing show that protease inhibitor treatment at appropriate doses may result in a dramatic decrease in circulating virus and a concomitant, sharp rise in CD4 count. These marker changes may indicate clinical benefit, but clearly this has to be confirmed by further trials.

Various organizations of people with HIV support wide availability of protease inhibitors, particularly for salvage therapy, and have exerted pressure toward this end. Others believe that more clinical information is needed before access is expanded. In the meantime, manufacturing difficulties and high dosing requirements combine to keep the protease inhibitors in extremely short supply. Hoffman La Roche started a compassionate program this past summer. Merck and Abbott are also planning compassionate-use protocols. However, because of the present low availability of the drugs, all programs plan to enroll only a limited number of patients worldwide.

The use of a potent but well-tolerated drug as monotherapy may still be a valuable option in some situations, such as advanced disease. However, protease inhibitors are most promising as agents to combine with other classes of antiretroviral drugs, such as nucleoside analogues, and with each other, e.g., a combination of two protease inhibitors with different resistance patterns. In fact, it is hoped that multiple drug therapy -- combined agents and sequentially administered combinations of agents -- may increase antiviral activity, may possibly delay or prevent the emergence of resistance, and, finally, have an impact on the natural history of HIV disease.

Protease inhibitors clearly represent a scientific advance in HIV therapeutic research. They are also are an example of rational, computer-driven drug design, an approach being applied to other potential HIV targets, such as HIV integrase. However, a number of important issues -- formulation, pharmacokinetics, dosing, long-term tolerability, durability of response, emergence of drug resistance, cost, and clinical effectiveness -- need to be investigated further and clarified before the role of protease inhibitors in the treatment of people with HIV can be accurately assessed.

— Stefano Vella, MD

Dr. Vella is an Associate Editor of ACC.

Published in Journal Watch HIV/AIDS Clinical Care October 1, 1995

Citation(s):

Condra JH et al. In vivo emergence of HIV-1 variants resistant to multiple protease inhibitors. Nature 1995 374 569-571.

• Medline abstract (Free)

Wei X et al. Viral dynamics in human immunodeficiency virus type 1 infection. Nature 1995 373 117-122.

• Medline abstract (Free)

Markowitz M et al. Evaluation of the antiviral activity of orally administered ABT-538, an inhibitor of HIV-1 protease. Second National Conference on Human Retroviruses and Related Infections, Washington DC, Jan 29-Feb 2 1995 Abstract 185 .

Mellors J et al. A randomized, double-blind study of the oral HIV protease inhibitor, L-735, 524 vs zidovudine in p24 antigenemic, HIV-1 infected patients with < 500 CD4 cells/mm³. Second National Conference on Human Retroviruses and Related Infections, Washington DC, Jan 29-Feb 2 1995 Abstract 183 .

Schapiro JM et al. First efficacy and safety results of the high dose saquinavir monotherapy trials. Second National Conference on Human Retroviruses and Related Infections, Washington DC, Jan 29-Feb 2 1995 Abstract LB2 .

Sommadossi JP et al. A human serum glycoprotein profoundly affects antiviral activity of the protease inhibitor SC-52151 by decreasing its cellular uptake. Second National Conference on Human Retroviruses and Related Infections, Washington DC, Jan 29-Feb 2 1995 Abstract LB4 .

Ho DD et al. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995 373 123-126.

• Medline abstract (Free)