New HIV Protease Inhibitors

New Options in the PI Class for Treatment-Naive Persons


The past decade has seen gigantic changes in the options for the treatment of HIV. The use of antiretroviral medications has dramatically changed the quality and quantity of life for persons living with HIV infection. The first potent antiretroviral medication combinations (termed highly active antiretroviral therapy, HAART) usually included protease inhibitors (PIs). PI-based HAART regimens were tough to take, with frequent dosing and lots of pills; frequent side effects and toxicities added more difficulty with adherence. Ultimately, these issues led to short-lived treatment benefit and drug resistance.

Because of these issues, the easier-to-take non-nucleoside reverse transcriptase inhibitor-based regimens became very popular, pushing protease inhibitors to later rounds of HAART treatment.

However, the story isn't over -- active research programs have continued to improve PIs, attempting to retain the potency strengths of PIs while addressing the previous limitations.

Benefits of Early Protease Inhibitors

Two thousand-three marked the sixteenth year of licensed therapies for the treatment of HIV-1 infection. Until 1995, HIV doctors only had nucleoside reverse transcriptase inhibitors (NRTIs, nukes). This mono- and dual-drug era was one of desperate times with high death rates and lots of complications.

In the winter of 1995-96, the introduction of the first HIV protease inhibitors (saquinavir, indinavir and ritonavir) revolutionized the medical care of persons living with HIV infection. The addition of a PI to dual nuke therapy resulted finally in the suppression circulating HIV to undetectable levels.

Several early clinical trials showed the ability of PI-containing HAART to suppress HIV and increase CD4 cell counts. Subsequently, several studies reported dramatic reduction in the death rate in HIV infected individuals.

Limitations of Early Protease Inhibitors

Not long after we celebrated the successes of PI-based HAART came the realization that not everything was great -- issues like high pill burden, need for frequent dosing or short-term side effects. As the life expectancy increased for HIV-infected persons, concerns of long-term toxicities emerged.

Early HIV-1 protease inhibitors treatments involved many pills (sometimes more than 10 a day) and frequent dosing schedules. Saquinavir, indinavir, ritonavir, nelfinavir and lopinavir/ritonavir all needed to be taken with a mind towards diet. Only amprenavir (APV) does not have a food requirement. Dosing of indinavir needed even more behavioral changes; the need to drink lots of water to avoid kidney stones. The recent reformulation of nelfinavir (NFV), from a 250 mg tablet to a 625 mg tablet is a recent effort to lower the pill burden of this PI -- from 10 a day to 4.

Short-term side effects of PI-based therapy are well known, with frequent gastrointestinal problems, like, nausea and diarrhea. These side effects often resulted in treatment discontinuation. Making matters worse, long-term toxicities (or fear of toxicity) emerged as we treated people longer with PI therapies. One main area of concern was body fat changes, now called, lipodystrophy. While it is increasingly understood that significant risk for lipodystrophy is often raised by non-drug factors, (such as severity of HIV disease or age) many early reports erroneously placed the sole blame for lipodystrophy on the PIs. Nevertheless, many PWAs would choose to avoid HIV therapy altogether, in order to avoid the stigmatizing effects of lipodystrophy.

Elevations in the blood chemicals cholesterol and triglyceride have been in persons taking all early PIs. Ritonavir-boosting generally makes these elevations even greater. The reason this is concerning is that abnormal cholesterol and triglycerides can raise one's risk of heart disease.

Ultimately, these limitations made adherence tough and negatively affected quality of life. It is clear that side effect aversion was a significant negative factor in taking pills -- a real problem in a world where long-term success requires near-perfect adherence. This issue is so significant that in one study, PWAs were willing to trade years of life and risk death in order to avoid medication side effects.

Challenges to Protease Inhibitors: Non-Nucleoside RT Inhibitors

The arrival of non-nucleoside reverse transcriptase inhibitors (NNRTIs, non-nukes) was met with skepticism. Simple and well tolerated drugs couldn't possibly match the potency of PI-based treatment.

Over time, multiple studies showed the superiority of NNRTI-based therapy, compared with PI-based therapy, even among the most ill patients, those with high viral loads or very low CD4 cell counts. Recent improvements in the formulation of the non-nuke efavirenz has yielded a one pill per day drug -- an industry benchmark.

The Rise of Boosted PIs

HIV PIs are metabolized by similar paths in the liver. The PI ritonavir has the unusual behavior of preventing the metabolism of most current PIs. When dosed together with other PIs, ritonavir increases the drug levels for all of the HIV protease inhibitors, except for nelfinavir. This "boost" decreases the dosing frequency and pill count for saquinavir, indinavir and amprenavir and eliminates the dietary restriction for indinavir.

Pharmacological boosting of lopinavir results in a clinically relevant product, the co-formulated lopinavir/ritonavir (LPV/RTV, Kaletra). In key studies, regimens that contained the boosted PI lopinavir/ritonavir were shown to be more potent than those that contained nelfinavir, and improved responses were seen in persons with very high viral loads. Because of these studies LPV/RTV is currently a component of one of the "preferred" initial antiretroviral regimens for initial therapy, a demotion of "unboosted" PIs to the alternate position.

Current boosted PIs are improvements from the early days, but limitations persist. Co-formulated lopinavir/ritonavir and the new 625 mg formulation of nelfinavir are good efforts in lowering the pill burden. Gastrointestinal side effects and elevations in cholesterol remain problems and for some persons, serious intolerance to ritonavir makes boosting an unacceptable option.

Therapeutic Challenges and New Protease Inhibitors

The treatment of HIV has continued to evolve since 1995, with ever improving prognosis and expanding therapy options. Because of effective HAART, HIV disease need not be fatal, and there inspection of the potency, long-term effectiveness and tolerability of treatments. Non-nuke-based therapies now permit very low pill burden and well tolerated options. Drug resistance is an essential consideration in treatment selection, even for the first treatment regimen in therapy-naive persons. Pill count, dosing frequency and dietary issues are central to adherence. Clinical trials of new drugs need to assess for potential toxicities, both short- and long-term, that were not appreciated during the discovery programs of earlier PIs.

New Protease Inhibitors

Three new HIV protease inhibitors are receiving attention because of recent clinical studies; atazanavir (ATV, Reyataz®) and fosamprenavir (APV, Lexiva) are already approved by the US Food and Drug Administration; tipranavir is now in phase III clinical trials. The improved potency, tolerability, dosing flexibility and resistance properties of the medications has sparked new interest in the role that PIs may play in the treatment of persons living with HIV.

Atazanavir (ATV, Reyataz)

Atazanavir was approved by the FDA in July 2003. ATV is the second HIV PI that is approved for once-daily dosing (along with boosted amprenavir). When ATV is dosed with ritonavir, ATV drug levels are increased (called boosted ATV). The usual dose is two 200 mg capsules, once-daily. ATV causes little disturbance in blood lipids and has a promising resistance profile in therapy-naive persons. ATV should be taken with food.

Studies in Treatment-Naive Persons

ATV has been given to treatment-naive persons in several studies. All involve the use of unboosted ATV; there has been no study of boosted ATV to date.

Early clinical trials (called studies AI424-007, -008) showed that the ATV was about as potent as NFV, with similar proportions of patients achieving undetectable viral loads and similar increases in CD4 cell rise. Elevations in blood levels of bilirubin were frequently observed, particularly among persons who received the highest dose (600 mg qd). In the latest clinical trial, AI424-045, therapy-naive persons received either unboosted ATV or efavirenz (Sustiva) with fixed dose ZDV/3TC (Combivir). This study showed that unboosted ATV seems to work as well as efavirenz. This is one of very few clinical trials that has compared a newer PI to efavirenz and suggests similarity between the two classes of drugs for first line therapy.

Studies in Treatment-Experienced Persons

The use of ATV in therapy-experienced individuals has been explored in two clinical trials. Study AI424-043 compared unboosted ATV against LPV/RTV and optimized dual nuke background in patients who were experiencing treatment failure. In this study, unboosted ATV was not as potent as lopinavir/ritonavir, and study subjects receiving ATV experienced less reduction in plasma HIV RNA and fewer ATV patients achieved undetectable viral loads.

Study AI424-045 is an ongoing study that compares boosted ATV/RTV (300 mg/100 mg qd) versus lopinavir/ritonavir versus ATV/SQV (saquinavir) in therapy-experienced persons. Patients received tenofovir as part of the randomized treatment (in conjunction with a NRTI). Preliminary 24-week data from this study was recently presented. The proportion of persons with viral load below LOQ (level of quantification or undetectable) was similar in the ATV/RTV and LPV/RTV groups, whereas the ATV/SQV groups tended to have poorer response. The 48-week data from this trial is anxiously awaited. It is tempting, though premature, to speculate that ATV/RTV will have comparable potency and tolerability in treatment-naive persons.

Side Effect Profile

ATV is generally very well tolerated, though it has slightly more gastrointestinal side effects than NFV. The drug's once daily, two pill count profile is the lowest pill burden of any currently approved HIV protease inhibitor. A distinguishing characteristic of the protease inhibitor is the lack of effect on blood cholesterol and triglycerides.

The principle and characteristic side effect associated with ATV therapy is elevations in the blood level of bilirubin, a condition called hyperbilirubinemia. Bilirubin is a byproduct of liver metabolism. Hyperbilirubinemia was the most common laboratory abnormality in ATV clinical trials. Clinical cases of jaundice yellowing of the skin have been reported in 11% of all study subjects; 9% reported yellowing of the eyes ("scleral icterus"), though these events rarely resulted in having to stop the medicine. These clinical symptoms are reversible upon treatment discontinuation or interruption.

Mild abnormalities in the electrical patterns in the heart occurred in some patients during clinical studies. These changes were asymptomatic but, because of this cautionary note, ATV should be used with caution in persons with preexisting cardiac conduction abnormality.

Drug-to-Drug Interactions

There are a number of significant drug-drug interactions with ATV. ATV should not be taken with commonly used antacids -- H2 receptor antagonists and proton pump inhibitors. If taken with tenofovir, ATV needs to be boosted, because tenofovir results in ~ 20% reductions in ATV levels.

Fosamprenavir (908, APV, Lexiva)

Fosamprenavir, also known as GW-433908 (908) was approved by the FDA in October, 2003. 908 is a chemical relative of the PI, amprenavir. The metabolism of 908 is inhibited by ritonavir, permitting pharmacologic boosting with twice-daily, un-boosted or boosted and once-daily, boosted dosing. Once-daily dosing requires ritonavir-boosting. The unboosted dose is two 700 mg tablets, twice-daily; boosted dosing requires one 700 mg tablet with one 100 mg ritonavir capsules twice-daily or two 700 mg tablets with two 100 mg ritonavir capsules once-daily. A characteristic adverse event is rash. 908 has a favorable lipid profile and promising resistance profile in therapy-naive persons. There are no dietary restrictions in taking 908.

Studies in Treatment-Naive Individuals

Two phase III clinical trials have evaluated 908 in therapy-naive individuals. The NEAT study compared unboosted 908, dosed twice daily to nelfinavir; the SOLO trial compared boosted, once-daily 908/RTV to nelfinavir.

In the NEAT study, subjects either received 908 twice daily or nelfinavir. While CD4 cell count rises were similar in both groups, a greater percentage of 908 subjects achieved undetectable viral loads especially those with high baseline viral loads, suggesting superior potency of 908.

The SOLO trial evaluated boosted, once-daily 908/RTV versus nelfinavir. In this study, similar percentages of subjects achieved undetectable viral loads at 48 weeks. In an exploratory analysis, a greater percentage of persons with very high baseline HIV viral loads (>500,000 copies/mL) or very low CD4 counts (<50 cells/mm3) achieved undetectable viral loads when they received 908 than NFV. More NFV subjects discontinued therapy because of virologic failure; more 908 subjects discontinued because of non-virologic reasons.

Studies in Treatment-Experienced Persons

The efficacy of boosted 908 was assessed in the CONTEXT clinical trial. PI-therapy-experienced patients were randomized to receive once- or twice-daily 908/RTV versus LPV/RTV. Preliminary 48-week data has been released (though not publicly presented). It reports under-performance of the once-daily arm. Very similar proportions of subjects achieved undetectable plasma HIV RNA levels in the twice-daily 908/RTV and LPV/RTV and there were similar increases in CD4 cell counts.

Side Effect Profile

Fosamprenavir is generally very well tolerated. In studies that compared 908 to nelfinavir, there were decreased rates of gastrointestinal side effects. Diarrhea occurred less frequently in persons taking 908 compared to NFV, this was true, even when 908 was boosted with ritonavir.

Like amprenavir, 908 causes rash in a small percentage of patients, ranging from 2% to 7% in recent studies. Small increases in cholesterol were seen in therapy-naive persons who received 908.

Drug-to-Drug Interactions

Fosamprenavir, like many other PIs is metabolized by cytochrome P450 -- to this extent the potential for drug-drug interactions is similar to most other PIs. A recent report showed that 908 should not be taken with LPV/RTV (Kaletra), because of lower drug levels.


Tipranavir (TPV) is currently in phase III clinical trials in HIV-infected individuals. TPV is the first of a new sub-class, the non-peptide inhibitors of HIV-1 protease. The pharmacokinetic properties of TPV require ritonavir boosting. The dose currently under investigation is two, 500 mg pills with 2, 100 mg ritonavir capsules, twice-daily. TPV has typical PI-associated gastrointestinal side effects, but a very encouraging ability to inhibit PI resistant virus.


The potency and tolerability of atazanavir and 908 add important new options in the care of persons living with HIV. Together with other protease inhibitors in development and rising rates of initial drug resistance, the protease inhibitor class has seen a renewed level of interest for the care of treatment naive persons.

Lower pill burden and improved tolerability compared with previous unboosted PIs is desirable -- the two pill per day atazanavir and the four pill per day 908 have achieved pill counts similar to the original, highly successful formulation of efavirenz. Pill counts also are lower than the industry standard lopinavir/ritonavir. Compared to early unboosted PIs, gastrointestinal tolerability is much improved.

Is PI drug discovery complete with atazanavir and 908? Low though current pill count might be, there still is room for improvement. Limitations of our existing data set remain -- evaluation for long-term safety or complications requires long-term monitoring of study subjects and an improved understanding of how HIV disease occurs. Drug-drug interactions remain problematic for most PI medications, particularly for atazanavir. The optimal choice of which medication class or medication combination to use continues to be the subject of debate.

The newer PIs will change the care for persons living with HIV. The improved tolerability and low pill count of new PIs will challenge the place that non-nukes now hold as first-line therapies. For persons who have inherited NNRTI resistance, PIs must be components of first-line combinations. It has been suggested that persons with very advanced disease might be more appropriately treated with PIs, because of their higher genetic barrier to resistance and the general low likelihood of getting high level PI drug resistance. For persons with high risk of heart disease, atazanavir's lipid profile makes this drug an attractive option. Lastly, for the many persons who are currently successfully taking first-line PIs, the newer drugs offer options for treatment simplification with fewer pills and improved tolerability.

Back to the Resolute! Spring 2004 contents page.