The HIV/AIDS Drug Pipeline: A Status Report

Introduction

While most people in the world with HIV/AIDS lack access to treatment, many of those who take anti-HIV medications are quickly exhausting the benefit of available treatment options. The problems associated with the 16 approved anti-HIV drugs -- viral resistance, long-term complications and side effects, adherence issues, and high cost -- are well documented. Studies have shown that the virological failure rate of highly active antiretroviral therapy (HAART) continues to be greater than 50% per year in most U.S. clinics. Yet only two genuinely new anti-HIV compounds, Kaletra (lopinavir/ritonavir) and tenofovir DF (Viread), have been approved by the U.S. Food and Drug Administration (FDA) since the beginning of 2000. A number of promising drug candidates appear to be floundering or have been discontinued. Some people fear that the research pipeline is drying up at a time when ever more individuals with HIV infection require new medications.

Despite several setbacks, however, there are imminent signs of progress in drug development. A few new agents are expected to be approved later this year. Others are becoming available, albeit in tightly restricted quantities, through expanded access programs (EAPs). Importantly, the notion of drug failure is not always well understood by people with HIV and their providers. Failure in terms of surrogate markers, such as the inability to achieve an undetectable viral load, does not necessarily translate into clinical failure, or deterioration of an individual's health. Recent studies indicate that disease progression is slowed and an immune system benefit is maintained (though not indefinitely) in people who continue to take "failing" medications. By contrast, people not taking anti-HIV therapy are more likely to have faster rates of immune system decline.

The following survey of drugs in the research pipeline is not an exhaustive list of experimental agents, but rather an outline of the major trends in current anti-HIV research. Readers should keep in mind that while advances in medicine depend on human research, transposing clinical trial data to utility in the real world is often difficult; study subjects may be more motivated, more closely monitored, and less representative of the variety of people who are likely to use the therapy under investigation.

Approved Drug Classes

A handful of experimental therapies in the three approved classes -- protease inhibitors (PIs), non-nucleoside reverse transcriptase inhibitors (NNRTIs), and nucleoside reverse transcriptase inhibitors (NRTIs) -- may provide benefit to people who are in need of third-line therapy, that is, those whose dominant HIV isolates (strains) have become resistant to many or all of the currently available drugs. Even if these new therapies are approved, they will require continued assessment for long-term efficacy and for toxicities that emerge over time.

Protease Inhibitors

Atazanavir

Atazanavir (Zrivada, formerly known as BMS-232632) is the first PI drug likely to be taken once per day (see the relative merits of once-daily dosing). In addition, the pill burden (requirement) of two capsules per day should be manageable, and atazanavir does not need augmentation, or boosting, with ritonavir (Norvir). Another advantage of this new PI is its reportedly neutral effect on levels of blood lipids (fats, primarily cholesterol and triglycerides). Atazanavir's lipid profile appears to be unique, as raised blood fat levels, which increase the risk of future cardiovascular events, are a major problem of the PI drug class. For more information on this topic, see "Cardiovascular Disease in People With HIV" in this issue.

In a recent study of 85 PI-experienced subjects with viral loads between 1,000 and 100,000 copies/mL, lipid levels remained stable -- and some even decreased -- by week 48 in those taking atazanavir (400 mg or 600 mg) and saquinavir (Fortovase) plus two NRTIs, compared with those taking ritonavir, saquinavir, and two NRTIs. (Earlier studies have shown that atazanavir and saquinavir work synergistically, or with enhanced effects, though combining the two substantially raises the pill burden.) For example, fasting triglyceride levels fell by an average of 27% from baseline among those in the 600 mg atazanavir arm, while they increased by an average of 93% from baseline in the ritonavir arm. Three people in both atazanavir arms (9% and 11%, respectively) discontinued the study due to treatment-related adverse events, whereas seven people (30%) did so in the ritonavir arm.

Viral load was reduced by 1.66 log copies/mL in the ritonavir arm vs a 1.44 log decrease in the 400 mg atazanavir arm and a 1.19 log decrease in the 600 mg atazanavir arm; details about which subjects achieved an undetectable viral load were not reported. (Most experts seem to agree that PIs and NNRTIs should be able to reduce viral load levels by at least 1.5 log copies/mL, or about 96%, within the first few weeks, and NRTIs, by about 1.0 log copies/mL, or 90%. See sidebar.) Other studies suggest that atazanavir has antiviral activity comparable to that of nelfinavir (Viracept). In vitro (test-tube) studies indicate that atazanavir should benefit people with resistance to one or two approved PI drugs, but its efficacy may be limited in people with resistance to multiple PIs.

Notably, atazanavir can increase levels of bilirubin, a pigment produced by the breakdown of hemoglobin, which at high levels can cause jaundice (yellowing of the skin and whites of the eyes) in some people. Those genetically predisposed to increased bilirubin levels one day may be detected using a screening test and advised not to take atazanavir -- a potential innovation from the new field of pharmacogenomics, or individualizing drug use based on a person's genetic characteristics. Researchers have indicated that atazanavir also may cause prolongation of the QT interval, which is associated with heart arrhythmias and can lead to cardiac arrest and sudden loss of consciousness or death, particularly in children and young adults. (The QT, or QTc, interval refers to a phase in the heart's electrical impulse cycle as measured by electrocardiogram, or ECG. Women tend to have slightly longer QT intervals than men.) Another limitation of atazanavir is the lack of data on interactions with other anti-HIV agents.

Despite these drawbacks, Bristol-Myers Squibb (BMS, based in New York City), initiated an EAP in May 2002 that provides atazanavir free of charge to limited numbers of people from two categories: those who cannot control their HIV infection using approved standard-of-care agents, and those with severe HAART-associated hyperlipidemia (high lipid levels) despite using lipid-lowering therapy. Pending ongoing research on drug interactions in treatment-naive people, EAP enrollees may not take ritonavir or Kaletra, and should be cautious if taking indinavir (Crixivan) or any NNRTI drugs. For more information about the atazanavir EAP, call 877-7BMS-EAP (877-726-7327).

Tipranavir

Tipranavir is the first of a new class of nonpeptidic PIs, which are more easily synthesized and manufactured -- and therefore potentially less expensive -- than the current peptide-based PIs. (Peptides are short chains of amino acids.) Tipranavir's activity against HIV strains with decreased susceptibility to all other PIs, presumably a result of the drug's unique molecular structure, makes it highly attractive. Recent data showed that among 41 NNRTI-naive people (nine female) without adequate viral suppression on their second PI-based regimen, only one (2%) developed high-level resistance (a greater than ten-fold increase in mutant virus) to tipranavir. Six (14%) developed mild resistance to the drug, which was associated mostly with the V82T and L33I resistance mutations. After 48 weeks of follow-up, viral load reductions from baseline were nearly identical among study subjects who had up to five or more than five PI resistance-conferring mutations (decreases of 2.39 and 2.24 log copies/mL, respectively). The most common adverse events were diarrhea and nausea within the first month; most resolved with or without treatment.

Unlike atazanavir, tipranavir must be boosted with ritonavir to compensate for its relatively short half-life and poor bioavailability (the degree to which the drug is absorbed and circulated in the body). Ongoing Phase IIb studies will help determine the optimal dose of tipranavir/ritonavir, which remains problematic since the tipranavir hard capsule was reformulated into a lower-dose (250 mg), self-emulsifying drug delivery system (SEDDS) midway through the study described above. The tipranavir/ritonavir combination is being tested in doses of 500 mg/100 mg, 500 mg/200 mg, and 750 mg/200 mg BID (twice per day). If current studies warrant long-term use of tipranavir, an EAP is expected to begin in the third quarter of 2003. Tipranavir is under development by Boehringer Ingelheim Pharmaceuticals (Ridgefield, Connecticut).

908

908, also known as GW433908 or fos-amprenavir, is a prodrug of the approved PI amprenavir (Agenerase). A prodrug is an inactive precursor of a drug that is converted by metabolic processes in the body into the drug's active form. Amprenavir is known to have poor bioavailability; reformulating the drug into 908 should reduce pill burden and thwart resistance by increasing levels of amprenavir in the body. The current formulation of amprenavir must be taken as eight large capsules twice per day and is not boosted with ritonavir; researchers are evaluating 908 with and withoutb ritonavir boosting. Dosages under investigation include 1) two 908 tablets twice per day, 2) one 908 tablet with one ritonavir capsule twice per day, or 3) two 908 tablets with two ritonavir capsules once daily. Pending results of ongoing Phase III studies, GlaxoSmithKline (GSK; Research Triangle Park, North Carolina) anticipates submitting 908 for FDA approval in the fourth quarter of 2002.

Other PIs

TMC-114 is a second-generation PI drug highly active against HIV with PI resistance mutations in vitro, and against HIV isolates taken from people with up to 100-fold reduced sensitivity to at least one approved PI. (Second-generation antiretroviral drugs are designed to be more potent and easier to tolerate than currently available drugs.) A Phase IIa study began enrolling in Europe in May 2002 to evaluate a new formulation of TMC-114; polyethylene glycol in the original agent had caused excessive diarrhea at higher doses. If TMC-114 is boosted with low-dose ritonavir, it may be available as one pill taken once daily. This new PI is under development by Tibotec-Virco NV (Mechelen, Belgium), which was acquired by Johnson and Johnson in early 2002.

Triangle Pharmaceuticals (Durham, North Carolina) announced in January 2002 that development of mozenavir (DMP-450), an investigational once-daily PI, was suspended because of disappointing study results, including reduced activity against virus resistant to indinavir (Crixivan) and ritonavir.

Non-Nucleoside Reverse Transcriptase Inhibitors

None of the experimental NNRTIs in the pipeline are beyond the early stages of development, and only a few show promising anti-HIV activity in people whose virus cannot be adequately suppressed using currently available NNRTIs. An improved resistance profile is essential for any second-generation NNRTI, as viral mutations that decrease susceptibility to one approved NNRTI are likely to confer resistance to all drugs in this class.

TMC-125

TMC-125 attracted attention at the 9th Conference on Retroviruses and Opportunistic Infections this past February when researchers reported on two Phase IIa studies showing that the drug induced rapid and impressive reductions in viral load levels in people with inadequate viral control while taking efavirenz (Sustiva) or nevirapine (Viramune) (median reduction of 0.9 log copies/mL), and in people who had never before taken anti-HIV therapy (median reduction of 1.92 log copies/mL). In people naive to therapy, TMC-125 reduced viral loads as potently as a five-drug regimen including agents from all three approved classes. TMC-125 also raised CD4 cell counts by 119 cells/mm³, compared with 60 cells/mm³ in a comparative analysis of people starting five drugs after no previous treatment. Significantly, these viral load reductions and CD4 cell count increases were measured within the first seven days of therapy in both TMC-125 studies. Recent reports indicate that response to anti-HIV therapy within the first week may predict long-term response to that treatment.

Headache and diarrhea are the most common adverse events associated with TMC-125. Further research on this drug, developed by Tibotec-Virco, no doubt will focus on reducing the onerous pill burden of 18 tablets (900 mg) taken twice per day. Phase IIb studies will enroll people who have used agents from all three approved drug classes; drug interaction studies are ongoing.

DPC-083

BMS's DPC-083 appears to benefit people whose HIV strains have decreased susceptibility to efavirenz and nevirapine. (The drug was originally developed by DuPont Pharmaceuticals, which was acquired by BMS in October 2001.) DPC-083 has a very long half-life -- approximately 100 hours -- which suggests once daily or even alternate-day dosing. (Taking a drug every other day, however, may have a negative impact on adherence.)

Fairly modest results have been achieved thus far. A recent Phase II comparison study in 134 antiretroviral-naive subjects (15% female, 17% non-Caucasian) showed similar rates of undetectable virus (fewer than 50 copies/mL) after 24 weeks among those taking one of three doses of DPC-083 (50, 100, or 200 mg once daily) plus standard efavirenz (600 mg) in an intent-to-treat analysis (all subjects were taken into account, including those who discontinued the study). Frequency of rash, a common side effect of efavirenz and other NNRTIs, in those taking DPC-083 was 15% (50 mg arm), 33% (100 mg), and 53% (200 mg), compared with 38% in the efavirenz arm. DPC-083 also may lead to central nervous system (CNS) side effects, although it appears to cause less dizziness than efavirenz.

Another recent Phase II study compared 100 mg and 200 mg once-daily doses of DPC-083 plus two NRTIs in people who previously had experienced viral rebound while taking nevirapine (61%) or efavirenz (39%). After eight weeks the strongest antiviral responses were found in those who not only switched their failing NNRTI to DPC-083, but also switched to at least one new NRTI. The mean decrease in viral load from baseline was 1.28 log copies/mL. These results are very preliminary and were not generated using a more useful intent-to-treat analysis, which would have taken into account the eight subjects (16% of total) who discontinued the study. Furthermore, data showing the response rate as number of subjects achieving fewer than 50 viral copies/mL, rather than 400 copies/mL as presented, would have been revealing.

Capravirine

In early studies capravirine (previously known as AG-1549 and S-1153) appeared to be ten times more potent than approved NNRTIs and to have activity against virus resistant to efavirenz (but not nevirapine). Phase II studies in people with NNRTI resistance and no PI experience, however, showed that those who took one of two doses of capravirine (1,400 or 2,100 mg) plus nelfinavir and two new NRTIs were no more likely to achieve viral loads of fewer than 400 copies/mL than those who took a placebo (an inactive substance). Further development of this investigational NNRTI by Agouron (a division of Pfizer, based in New York City) seemed doubtful following reports of vasculitis (blood vessel inflammation) in dogs given capravirine. However, no vasculitis has been found thus far in humans taking the drug, and studies of lower doses of capravirine (with vasculitis monitoring) are ongoing.

Other NNRTIs

Other NNRTI drugs in development include two from Sweden's Medivir company: MIV-150, being codeveloped with Chiron (Emeryville, California) and soon to enter Phase II trials, and MV026048, which was licensed to Roche (Nutley, New Jersey) in April 2002 and is still in preclinical testing. Calanolide A is a naturally occurring NNRTI compound derived from a Malaysian rain forest plant and is under development by Sarawak MediChem Pharmaceuticals (Lemont, Illinois). It appears to be moving very slowly through the research pipeline. On the reformulation front, researchers at Boehringer Ingelheim are evaluating a once-daily version of nevirapine.

A number of initially promising compounds have been shelved by their developers -- a common occurrence in the pharmaceutical industry, since efficacy in humans rarely matches in vitro potency, and drug toxicities often sink a potential candidate. Emivirine (Coactinon) recently was discontinued by Triangle due to inadequate potency. BMS suspended development of DPC-961 after several study volunteers reported suicidal ideation, and DPC-963 appears to be on hold.

Nucleoside Reverse Transcriptase Inhibitors

NRTI drugs were the first class of antiretrovirals to be approved, beginning with AZT (zidovudine, Retrovir) in 1987. As with the other two approved classes, next-generation NRTIs will require improved potency with fewer side effects and more robust resistance profiles.

Emtricitabine

Emtricitabine (Coviracil, formerly known as FTC) is a new NRTI under investigation by Triangle. Early studies have shown that a once-daily 200 mg dose is optimal and able to reduce viral load by approximately 1.7 log copies/mL in people with no previous treatment experience. Emtricitabine appears to have potency equivalent to that of 3TC (lamivudine, Epivir) in treatment-experienced people; for several years the value of emtricitabine over 3TC has been questioned. Significantly, the two drugs' similar resistance profiles means that viral strains with reduced sensitivity to 3TC also evade the antiviral effects of emtricitabine. In addition, compared with viral resistance to 3TC, resistance to emtricitabine seems to be less associated with the M184V mutation, which confers a protective effect against AZT resistance.

If approved, emtricitabine may be desirable mainly for people who prefer once-daily dosing. Some subjects using this drug in trials have reported mild to moderate CNS symptoms, diarrhea, rashes, and biochemical abnormalities, including very high triglyceride levels. This drug also is being studied to treat hepatitis B virus (HBV) infection.

Triangle plans to submit a new drug application (NDA) with the FDA in the autumn of 2002 based on 24-week data from a third Phase III study (FTC-301). This study has enrolled approximately 560 people to compare emtricitabine with d4T (stavudine, Zerit), taken with ddI (didanosine, Videx) and efavirenz; the primary endpoint is undetectable viral load (fewer than 50 copies/mL).

Amdoxovir

Amdoxovir (formerly known as DAPD) is another NRTI being developed by Triangle. The body metabolizes the comparatively weak pill form of this drug (dioxolane purine) into a highly active form known as DXG, or D-dioxolane guanosine. Amdoxovir's unique chemical structure is believed to account for its potency against viral isolates resistant to AZT, 3TC, and abacavir (Ziagen), and some isolates with wide cross-resistance to NRTIs. Amdoxovir also appears to work against some viral strains with reduced susceptibility to certain NNRTIs (efavirenz and nevirapine). In fact, a number of resistance mutations associated with NNRTI drugs (for example, 103N, 106A, 108I, 181C, and 190A) appear to increase the efficacy of amdoxovir. Viruses with multiple mutations including K65R, F116Y, and Q151M, however, are less susceptible to amdoxovir.

This new NRTI may be especially suitable as a component of "salvage therapy," that is, for use in people whose predominant viral isolates no longer respond to a variety of approved anti-HIV drugs -- in this case, AZT and 3TC in particular. In preliminary studies, six people who added amdoxovir (500 mg twice daily) to their current, failing antiretroviral regimens experienced an average viral load reduction of 1.9 log copies/mL -- significantly more impressive than the average reduction of 1.0 log copies/mL seen in those taking amdoxovir alone after a "wash-out" period of no drugs.

Amdoxovir is currently in Phase II studies. The drug's activity, particularly against multi-NRTI-resistant viral isolates, appears to be augmented by coadministration with mycophenolate (mycophenolic acid, CellCept), an immunosuppressive drug approved for use in organ transplantation. The two drugs are expec