March 30, 2006
It was about 10 years ago that the concept of highly active antiretroviral therapy (HAART) was born, in which at least three highly potent drugs are combined to suppress HIV. The initial HAART formula -- two nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) plus a third drug from another class -- for first-line HIV treatment has changed very little since its implementation. NRTIs remain the backbone of antiretroviral therapy, and it is very unlikely that this part of the HAART formula will change in the near future. New recommendations to consider when implementing HAART include the following:1
We have also learned that when initiating therapy, more is not necessarily better. That is, administering four drugs instead of the standard triple-drug cocktail may not necessarily result in a better outcome.2 Similarly, less treatment may also be risky, as several studies with monotherapy,3,4 and NRTI-sparing regimens have demonstrated.5,6
Currently, we have over 30 different formulations, reformulations or coformulations of antiretroviral drugs in pharmacologic circulation. Given the abundance of antiretroviral agents, it is hard to believe that new drugs are still needed for HIV treatment. However, there remain many barriers to overcome -- most importantly, the "chase-the-tail" dilemma in which the use of new antiretroviral agents results in the emergence of new drug resistance mutations that other future new agents or drug classes will have to overcome.
So, what new drugs are lined up for development over the next few years? This review details the most important agents that have been recently discussed at international conferences such as the 45th International Conference on Antimicrobial Agents and Chemotherapy (ICAAC 2005) in Washington, D.C., in December 2005 and the 13th Conference on Retroviruses and Opportunistic Infections (CROI 2006) in Denver, Colo., in February 2006.
At the 3rd International AIDS Conference (IAS 2005) in Brazil in July 2005, Cal Cohen, from Community Research Initiative of New England, presented the results of the placebo-controlled phase 2b study in which different doses of DFC were added to a failing regimen in 200 patients.9 At two weeks, patients who received 200 mg of DFC (the highest dose administered) demonstrated a 0.7-log reduction in viral load; patients not taking lamivudine or emtricitabine experienced greater viral load suppression with a 1.1-log reduction. At 16 weeks, the viral load reduction in patients not taking lamivudine or emtricitabine was 1.4 logs. This response was independent of the presence of the M184V mutation.
A comprehensive analysis looking at the impact of baseline resistance on the treatment responses observed in this study showed that the mutation Q151M was the only mutation associated with baseline phenotypic resistance or poor virologic outcome at week 16.10 Also, no specific resistance to DFC emerged during the study period. In fact, no novel mutations have been identified with the use of this compound.
It is unclear what role, if any, DFC may play in HIV treatment. Although DFC is generally well tolerated, it is known to cause asymptomatic hyperlipasemia when used in combination with didanosine. If this drug proves to be as potent in treatment-experienced patients who harbor NRTI-resistance mutations as it is in treatment-naive patients -- and despite the potential drug interactions with lamivudine, emtricitabine and didanosine -- it may be an agent worth developing. Phase 2b studies of DFC are currently in progress comparing 200 mg DFC without lamivudine or emtricitabine to 300 mg lamivudine on top of an optimized background therapy.
[Editor's Note: Following the publication of this article, the development of DFC was terminated, after researchers conducting a phase 2b study in treatment-experienced patients found that an "unacceptably high" number of patients -- quantified as "well above ... 10% to 15%" -- developed severe hyperlipasemia when switching to DFC from lamivudine or emtricitabine. More information is available in a separate article.]
At ICAAC 2005, data were presented on the already well-known TMC114 (darunavir) and the relatively unknown brecanavir (GW640385). These new PIs are novel in that they belong to a non-peptidic PI subclass. These non-peptidic compounds tend to bind tightly to the HIV protease enzyme, even in the presence of highly drug-resistant virus, thereby increasing their potency. In addition, all of the new agents appear to be fairly well tolerated despite requiring boosting with ritonavir.
At CROI 2005, the results of a phase 2 dose-finding study of TMC114 were presented.16 At a dose of 600 mg twice daily (boosted with ritonavir 100 mg), the mean change in viral load from baseline was -1.85 log10 copies/mL, with 72% of patients in this group achieving a viral load decrease of 1 log or greater. This dose has been selected for further development.
Two ongoing, multinational, phase 2b clinical trials with similar study designs are assessing the efficacy and safety of boosted TMC114 in heavily treatment-experienced patients. Both the POWER 1 and POWER 2 studies (Performance Of TMC114/r When Evaluated in triple-class-experienced patients with PI Resistance) randomized triple-class-experienced patients with at least one major PI mutation to take one of four TMC114 + ritonavir doses (400/100 mg once daily, 800/100 mg once daily, 400/100 mg twice daily or 600/100 mg twice daily) or the best-feasible PI-based control regimen. After the initial 24-week dose-finding phase, all patients were continued on the same dose according to the study design.
At IAS 2005, the 24-week results of the POWER 1 study conducted in Europe, Brazil and Australia, which enrolled 318 treatment-experienced patients, were revealed.17 TMC114 + ritonavir was found to be significantly more effective at reducing HIV viral loads when compared with other PIs (P < .001 for any dose of TMC114 + ritonavir versus control) based on an intent-to-treat analysis. Of the patients receiving the most effective TMC114 + ritonavir regimen at 600/100 mg twice daily, 77% achieved at least a 1-log decrease in HIV-1 RNA versus 25% of the control-treated patients. Furthermore, TMC114 + ritonavir was effective in patients with three or more primary PI mutations.
The 24-week preliminary results for the counterpart POWER 2 study conducted in the United States with 278 treatment-experienced patients were discussed at ICAAC 2005.18 Again, patients were randomized to receive an optimized background regimen (OBR) plus one of four doses of TMC114 + ritonavir, or an optimized background regimen plus an investigator-selected control PI(s).
The baseline demographic characteristics among the various groups were very well matched.18 Patients in the control arm received fosamprenavir (FPV, 908, Lexiva, Telzir; 42%), saquinavir (30%), lopinavir (LPV; 28%), atazanavir (ATV, Reyataz; 11%), indinavir (IDV, Crixivan; 4%) or nelfinavir (NFV, Viracept; 2%). All patients received PI boosting with ritonavir.
At 24 weeks, the percentage of patients achieving undetectable HIV-1 RNA levels (less than 50 copies/mL) was 39% in the highest TMC114 + ritonavir dose group compared with 7% in the control arm.18 The results also showed that 62% of TMC114 + ritonavir patients achieved a viral load reduction of at least 1 log or greater at a dose of 600/100 mg twice daily.
Regarding safety data, the incidence of adverse events (of any grade) was comparable between TMC114 + ritonavir and control patients.19 The most common adverse events reported included headaches and nausea, which each occurred in 17% of individuals in all TMC114 + ritonavir arms compared with 17% and 9% of patients, respectively, in the control arm. At least one serious adverse event was reported in 15% of patients in the TMC114 + ritonavir groups in comparison with 8% of control patients. Overall, TMC114 + ritonavir was generally well tolerated and had a low rate of treatment discontinuation due to adverse events. This safety profile is similar to that previously observed in the POWER 1 study.
It is interesting and informative to directly compare and also combine the data for both POWER studies. The preliminary efficacy and safety results from the POWER 2 study validate the findings observed by the European researchers in the POWER 1 study and support the usefulness of this compound in the treatment-experienced population. The increased virologic response rate observed in POWER 1 over POWER 2 might be a reflection of the more advanced population that was treated in POWER 2.
Another very important observation made in the POWER 2 study involved the difference in TMC114 + ritonavir efficacy when patients were stratified by enfuvirtide (T-20, Fuzeon) experience. Sixty-four percent of patients who received enfuvirtide for the first time with TMC114 + ritonavir 600/100 mg twice daily achieved undetectable levels of HIV-1 at 24 weeks (less than 50 copies/mL) compared with only 30% of patients who received TMC114 + ritonavir without enfuvirtide.
These data support the findings of other similar studies, such as the RESIST-1 and RESIST-2 studies (Randomized Evaluation of Strategic Intervention in Multi-Drug Resistant Patients with Tipranavir) on tipranavir, regarding the benefits of adding enfuvirtide to an optimized background regimen that includes at least one new drug. These results are also consistent with the recently updated recommendations in the Antiretroviral Treatment Guidelines from the U.S. Department of Health and Human Services.1 Moreover, these data come at a critical time given the recent initiation of an international expanded-access program for TMC114 + ritonavir. Treatment-experienced patients who require new therapeutic options may benefit from the simultaneous use of enfuvirtide with this investigational agent.
At CROI 2006, investigators reported on the resistance profiles for the POWER studies, including the baseline and on-treatment, emerging mutations that were associated with a reduced virologic response. The activity of TMC114 was maintained even with up to nine baseline PI mutations, which was better than the activity observed with the competitive study PI. The mutations V32I, L33F, I47V, I84V and L89V were associated with a diminished virologic response when present at baseline or when developed upon virologic failure. An interesting observation that was reported was the fact that tipranavir appears effective for patients who had a viral load rebound while on TMC114.20
With the recent approval of tipranavir and the availability of expanded access for TMC114, many researchers are questioning whether one of these two new PIs is better than the other. Both drugs were developed with the same initial indication in highly treatment-experienced patients. The registrational studies for both drugs are also extremely similar in design (RESIST-1 and RESIST-2 for tipranavir, and POWER 1 and POWER 2 for TMC114), thereby inviting a cross-study comparison, which will need to be performed with a great deal of caution. It is clear that the 400 mg of ritonavir needed once a day to boost tipranavir does not help improve the efficacy of tipranavir over that of TMC114, which requires only 200 mg/day of ritonavir for boosting, although the extra 200 mg of ritonavir that is needed to boost tipranavir over TMC114 has not been associated with an increased incidence of adverse effects. The response rate and the durability of the antiviral response when either agent is coadministered with enfuvirtide are also important considerations. Based on preliminary data from the POWER 1 and 2 studies, the use of TMC114 over tipranavir is favored in light of these issues.
Little was known about brecanavir until the 2005 ICAAC meeting.22 The data presented there were from a planned, 24-week analysis of an ongoing open-label study (HPR10006) evaluating the safety, tolerability, antiviral activity and pharmacokinetics of ritonavir-boosted brecanavir. In this single-arm study, 31 HIV-1 infected adults (treatment naive and experienced) with CD4+ cell counts greater than 200 cells/mm3 and HIV-1 RNA levels above 1,000 copies/mL received brecanavir + ritonavir 300/100 mg twice daily in combination with two NRTIs based on patient medical history and baseline resistance tests.
The proportion of patients with HIV-1 RNA levels below 400 and 50 copies/mL were 81% and 77%, respectively, at week 24 based on an intent-to-treat, missing-or-discontinuation-equals-failure analysis.22 Patients entering the study with PI-sensitive (n = 25) and highly PI-resistant (n = 6) virus had similar response rates. The median increase in CD4+ cell count was 84 cells/mm3.
A safety assessment showed that brecanavir was well tolerated and caused few grade 2 through 4 drug-related adverse events and no serious adverse events.22 Most adverse events reported were mild in nature and did not require treatment modification or discontinuation. Clinically relevant changes in laboratory parameters were also infrequent.
Interestingly, tenofovir was not permitted as part of the patients' NRTI backbone in this study, because at the time of the study design, there was a lack of pharmacokinetic interaction data between tenofovir and brecanavir. However, at the same ICAAC meeting, another presentation addressed this very issue by describing the pharmacokinetic interactions between brecanavir and tenofovir in healthy volunteers.23 When tenofovir 300 mg was coadministered with brecanavir + ritonavir 300/100 mg twice daily, the regimen appeared to be well tolerated and resulted in 24% to 32% increases in tenofovir plasma concentrations, 15% to 20% increases in brecanavir concentrations, and 31% to 39% increases in ritonavir concentrations for both area under the curve and maximum plasma concentrations. Such differences in drug exposure are not thought to require changes in dosing, although monitoring for tenofovir-associated side effects or toxicity may be necessary.
In sum, ICAAC 2005 was a good coming-out party for brecanavir. Despite its molecular resemblance to the amprenavir (APV, Agenerase) molecule, brecanavir appears to demonstrate an even more potent antiviral effect.
The fact that brecanavir's anti-HIV activity can be generated with a IC50 in the femptomolar range, which is tens to hundreds of times lower than that of other conventional PIs, certainly supports the further development of this compound. It is hoped that the potency observed in vitro will translate into high potency in real people. The 24-week preliminary data demonstrating potent antiviral activity for brecanavir in both PI-sensitive and PI-resistant HIV-infected adults is certainly a step in the right direction. Another interesting attribute that will make this drug attractive to practitioners is the fact that brecanavir does not appear to inhibit or induce the cytochrome P450 CYP3A4 protein, which might potentially translate into fewer or less pronounced drug-drug interactions. Further studies exploring other brecanavir + ritonavir doses aside from the 300/100 mg used in the HPR10006 study are either planned or currently in progress.
There are a multitude of other PIs currently in preclinical development, including SPI-256, a highly potent compound with subnanomolar potency and broad in vitro activity against multidrug resistant strains.24 In a study presented at CROI 2006, this compound was reported to have a four to 50 fold greater potency than currently approved PIs, as measured by IC50 values using a PhenoSense assay. In addition, in vitro resistance to this compound requires the accumulation of multiple primary PI mutations. If this compound ever makes it to full development, it could potentially represent another agent useful for salvage therapy.
There are two related agents currently at different phases of development that promise activity against most viruses resistant to current NNRTIs: TMC125 (etravirine) and TMC278.
At the 2005 ICAAC, the 24-week primary analysis results of the TMC125-C223 study were presented,27 which were the same results presented at EAC 2005.28 For this study, patients with baseline HIV-1 RNA levels above 1,000 copies/mL, with documented NNRTI resistance and with three or more primary PI mutations were randomized 1:2:2 to receive an active control regimen (i.e., the best available regimen from licensed agents), TMC125 400 mg twice daily or TMC125 800 mg twice daily, respectively. The baseline demographic characteristics were very well balanced between both TMC125 groups and the active control group. At 24 weeks, the median viral load change between both TMC125 arms and the active control arm was statistically significant, while no differences were seen between the TMC125 400-mg and 800-mg groups.
More patients in the TMC125 groups achieved viral loads less than 50 copies/mL than in the active control arm, although these differences were not statistically significant. On the other hand, the proportion of patients who achieved a viral load less than 400 copies/mL in both TMC125 groups was significantly superior when compared with the active control group.
When the virologic response rate was evaluated based on the number of active drugs in the background regimen, the activity of TMC125 was found to be more potent when the background regimen included an additional one or two active drugs. For example, if there were two or more active drugs in the background regimen in addition to TMC125, this resulted in a 1.5-log viral load reduction; in comparison, only a 0.5-log reduction was observed in active control patients who received two or more active drugs. In addition, enfuvirtide-naive TMC125 patients who received enfuvirtide in their background regimen demonstrated a significantly superior virologic response (2.0-log reduction) versus those patients in the active control group who received enfuvirtide (1.0-log reduction).27
In terms of safety, there was a higher percentage of adverse events observed in the TMC125 groups versus the active control group, including diarrhea (20% versus 15%), rash (20% versus 8%), fever (17% versus 10%), headaches (13% versus 5%) and insomnia (13% versus 8%). There was also a higher number of patients with pancreatic amylase elevations, but few cases of pancreatitis. No central nervous system syndromes or hepatotoxicity were reported.
The analysis of the resistance data collected on this study was reported at CROI 2006.29 All patients had prior PI exposure (median of four PIs); 94% had NNRTI experience and 30% had prior enfuvirtide treatment. All patients had documented NNRTI resistance, although only 79% had NNRTI resistance present at baseline. The median baseline fold change in EC50 for efavirenz, nevirapine and TMC125 was 41, 61 and 1.7, respectively. Patients with zero, one or two NNRTI mutations at baseline achieved a mean decrease in HIV RNA of 1.82, 1.65 and 1.00 logs, respectively. In addition, mutations associated with an increased baseline EC50 for TMC125 were uncommon (prevalence of less than 3%). This analysis demonstrates that TMC125 retains activity in the presence of multiple NNRTI mutations, where current NNRTIs are not expected to be effective.
The results of this TMC125 study with highly treatment-experienced patients are not only important but immensely exciting. For once, we have a promising NNRTI with activity against NNRTI-resistant virus, and the agent appears to have a fairly safe toxicity profile. Even better news, it was recently announced that a new TMC125 reformulation is going to make it possible to dose the agent at 200 mg twice daily with a compact formulation (i.e., two tablets twice daily) that provides comparable exposure to the twice-daily, 800-mg dose achieved with the old formulation.
In contrast, it was recently announced that the TMC125-C227 study, in which patients who failed an initial NNRTI-containing regimen were randomized to receive TMC125 or a PI in combination with active NRTIs, was prematurely halted due to a suboptimal virologic response in the TMC125 arm in comparison with the PI arm. What do these results mean? Is TMC125 effective for NNRTI-experienced patients? The data from the TMC125-C227 study have not been released, so we don't yet have a clear understanding regarding the reasons for the study discontinuation. Despite this, it appears that TMC125 may be a good option for heavily NNRTI-experienced patients, but it may not necessarily represent the best option for NNRTI-resistant patients with other viable PI options.
In another interesting study, this one conducted at Chelsea and Westminster Hospital in London, the pharmacokinetics, safety and efficacy of TMC125 200 mg twice daily (new formulation) and TMC114 + ritonavir (600/100) twice daily plus NRTIs (with or without enfuvirtide) was evaluated in 10, three-class-resistant patients with a mean viral load of 4.6 logs. No significant pharmacokinetic interaction between TMC125 and TMC114 was observed. At week 6 all patients have achieved at least a 2-log decrease in HIV RNA. This combination was well tolerated and it has promising efficacy against three-class-resistant patients.30
These findings were not reproduced in another study that combined TMC125 and boosted tipranavir.31 In that study, 24 healthy patients were given TMC125 800 mg twice daily (old formulation) and tipranavir + ritonavir (500/200) twice daily, both alone, and then in combination for eight days. Nine patients discontinued the study before completion. Of the remaining 15 trial participants, seven experienced grade 3 adverse events (lab abnormalities). The steady-state exposure (AUC12h) of TMC125 was significantly reduced by 76% when both drugs were coadministered compared to when they were given alone. TMC125 increased the steady-state exposure of tipranavir by 18% and of ritonavir by 23%. Therefore, due to the significant and clinically relevant decrease of TMC125 exposure, the coadministration of these two drugs is not recommended.
In this study, TMC278 was found to be safe and well tolerated. Headaches represented the most common adverse effect reported but occurred with equal frequency in the TMC278 and placebo arms. No genotypic changes associated with antiretroviral resistance were observed during the course of the study.
TMC278 is metabolized through CYP3A. Pharmacokinetic drug interaction studies with tenofovir have shown that TMC278 exposure is not affected by coadministration with tenofovir, but tenofovir exposure increases by approximately 24%. Another significant interaction occurs with lopinavir/ritonavir.33 The exposure of TMC278 was substantially increased without concomitant changes in lopinavir exposure, thereby suggesting that TMC278 has a minimal effect on the induction or inhibition of CYP3A. It is also likely that a dose adjustment for TMC278 will be required when coadministered with lopinavir/ritonavir.
A follow-up study in treatment-experienced patients using a once-daily dose of TMC278 is currently in progress. In addition, a phase 2b dose-finding study is currently ongoing in treatment-naive individuals in which various doses of TMC278 are being compared with efavirenz, all of which are being given in combination with two NRTIs. So, TMC278 is another interesting drug worthy of close attention.
The current enfuvirtide formulation has been confounded by the logistics of complex subcutaneous injections twice daily. At CROI 2006, investigators reported that they have identified two potential next-generation fusion inhibitor candidates with substantial improvements over enfuvirtide in potency, durability and pharmacokinetics.34
Entry inhibitors represent an attractive area for drug development given the multiple steps in viral attachment and entry that may be therapeutically targeted. Several compounds currently under development are able to successfully inhibit the CCR5 chemoreceptor molecule. Unfortunately the development of this class of compounds has been compromised by recent events including the discontinuation of the CCR5 inhibitor aplaviroc (APL, GSK873140) following reports of severe hepatotoxicity in several patients.35
At the 2005 ICAAC meeting, the results of several studies were presented describing the pharmacokinetic interaction between aplaviroc and several other compounds, including tenofovir,36 efavirenz,37 atazanavir + ritonavir38 and fosamprenavir + ritonavir + efavirenz.39 Sadly, these well-collected data cannot be extrapolated to the maraviroc (UK-427,857) compound, and, if anything, these data will just have a historical pharmacokinetic value.
The phase 2 trial of vicriviroc was an international study in which 92 treatment-naive, R5-tropic patients, with a CD4+ cell count greater than 150 and an HIV-RNA viral load greater than or equal to 5,000 copies/mL, were randomized to three doses of vicriviroc (25, 50 and 75 mg) versus placebo monotherapy for 14 days. Patients were then rolled over to a continuation phase using those three doses of vicriviroc versus efavirenz, both in combination with zidovudine/lamivudine (AZT/3TC, Combivir). During the initial two weeks there was a dose response to treatment as measured by HIV-RNA decline, with the largest drop (1.34 logs) observed with the 75-mg dose. The study was stopped because the proportion of patients who experienced a virologic breakthrough on that dose was 17% versus 4% in the control arm.
At the doses used in this study, vicriviroc + zidovudine/lamivudine did not provide the same sustained response observed with efavirenz + zidovudine/lamivudine. It is unclear if there is an issue with the in vivo potency of this compound, or if the doses evaluated in this study were all just too low. It is also possible that the two-week monotherapy lead-in period may have played a role in causing these failures. Overall, vicriviroc was safe and well tolerated, with no hepatocellular injury observed during the course of the study. In addition, tropism shift was infrequent and seen in all groups, including placebo; and, in most cases, it occurred during the first two weeks.
Intentions were announced during the 2006 CROI meeting to continue the development of this drug by exploring higher doses, given that the safety analysis at the 75-mg dose was clean. So, the ultimate dose and role of this compound remains to be defined. In addition, phase 2 studies in treatment-experienced patients are still ongoing under the sponsorship of the Adult AIDS Clinical Trials Group (ACTG).
There are three fully-enrolled, ongoing maraviroc studies, two of them with experienced patients (one exclusively in patients with R5-tropic virus and the other in patients with X4/R5 dual tropism). The third study contains naive, R5-tropic patients. It is evaluating three different dosages (once daily and twice daily) of maraviroc + zidovudine/lamivudine, and comparing them with a control arm of efavirenz + zidovudine/lamivudine. On Jan. 24, 2006, in a press release, we learned that the Data Safety Monitoring Board recommended the discontinuation of the once-daily dose arm in this naive study. An analysis of the first 205 patients given 300 mg of maraviroc once daily in combination with zidovudine/lamivudine found that, after 16 weeks of treatment, the combination did not prove itself to be "non-inferior" to efavirenz + zidovudine/lamivudine. The preliminary efficacy and safety results of the these three pivotal maraviroc clinical trials in treatment-naive and experienced patients, which should be presented sometime later this year, are eagerly awaited.
A Canadian company is also working on a CXCR4 antagonist (AMD11070) and the results of their phase 1 clinical trial are expected to be presented at a future conference in 2006. The development of these agents are important when we consider that at late stages of HIV infection, CXCR4 viral phenotypes have been linked with rapid progression to AIDS. It is postulated that a CXCR4 antagonist, in combination with standard antiretroviral therapy, may be advantageous in preventing AIDS.
Most of the entry inhibitors that are currently in development interact exclusively with host factors. These agents are likely to have activity against drug-resistant viral strains. The patterns for the development of resistance are also expected to be completely different than those that have been seen with currently approved drugs, so there should be no worries about cross-resistance with existing antiretrovirals. In general, because entry inhibitors function outside the cell, they are also unlikely to interact with intracellular components, thus possibly decreasing the risk of metabolic complications.
Given the numerous entry inhibitor molecules that are currently in the early stages of development, it is tempting to speculate about the possibility of combining several of these agents once their efficacy and safety have been established as the slide below shows.
In a plenary session at the 2005 ICAAC meeting, Dr. Robert Doms from the University of Pennsylvania discussed the possibility for synergism between entry inhibitors.45 He explained the different mechanisms of HIV viral entry into host CD4+ cells and the different targets for possible therapeutic intervention. But in reality, although there are currently many entry inhibitor candidates, only a few will likely achieve full development.
There are several integrase inhibitors that have made it to clinical development. One compound with promising activity, S1360, which was designed through a collaboration between two companies, was discontinued due to extremely poor bioavailability, despite its excellent antiviral activity.46 At the 2005 CROI meeting, data on the integrase inhibitor compound L-000870810 was presented.47 Despite excellent potency, this drug was also subsequently discontinued due to toxicity observed in dogs during long-term dosing. But there are two other compounds that are gradually advancing along the pipeline of antiretroviral development.
MK-0518 is metabolized mainly via glucuronidation (UGT1A1), and it is not a potent inhibitor or inducer of CYP3A4, therefore it does not require, nor should it be boosted by ritonavir. Drug interaction studies support the coadministration of MK-5018 with other antiretrovirals without dose adjustment.
At CROI 2006, data was presented from the randomized, double-blind, dose-ranging study in which triple-class, antiretroviral-experienced patients failing their current regimen were randomized to receive 200, 400, or 600 mg of MK-5018 twice daily versus placebo, in combination with an optimized background therapy (OBT).49
Patients at baseline had an HIV RNA greater than 5,000 copies/mL and a CD4+ cell count greater than 50 cells/mm3. Because this compound is potentially a substrate for the UGT1A enzyme, which could be inhibited by atazanavir, patients in this study were also stratified between non-atazanavir-containing optimized background therapy and atazanavir-containing optimized background therapy. They were also stratified between non-enfuvirtide-containing optimized background therapy versus enfuvirtide-containing optimized background therapy. The results of that stratified data were not reported in this presentation.
The 16-week interim analysis showed that 65% to 85% of patients in the three MK-0518 dosing arms (~40 patients per arm) had HIV RNA less than 400 copies/mL, compared to 20% in the placebo (optimized background therapy) group. Similarly, 56% to 72% of patients in the three MK-0518 dosing groups had HIV RNA less than 50 copies/mL. The drug was very well tolerated with only two patients not completing the study (one due to a lack of efficacy and the other due to unrelated death).
The other integrase inhibitor in development is GS 9137, formerly known as JTK-303. It made its debut at CROI 2006 with the presentation of three phase 1 to 2 studies. It is a dihydroquinoline carboxylic acid strand transfer inhibitor with a chemical structure resembling the quinolone compounds, but with no antibacterial activity or quinolone-associated toxicities.
Preclinical pharmacokinetic studies have demonstrated potent anti-HIV activity in vitro with a serum free IC50 of 0.2 nM and a protein binding-adjusted IC50 in peripheral blood mononuclear cells of 16 nM. It is metabolized via a combination of oxidative (CYP3A) and glucoronidation pathways. It is known to be a moderate inducer of CYP3A, but not an inhibitor.50
GS 9137 is fully active against nucleoside-, non-nucleoside- and PI-resistant isolates, and it has shown additive to synergistic activity with all other antiretrovirals. When given unboosted, single-dose pharmacokinetics in healthy volunteers have shown that the concentrations of GS 9137 are generally dose-proportional. Its oral bioavailability increases by threefold when given with food, and its half-life is approximately three hours.51 When GS 9137 is coadministered with ritonavir 100 mg, its pharmacokinetics are greatly improved. There is a twentyfold increase in oral bioavailability, no autoinduction of metabolism and its plasma half-life increases to approximately nine hours, allowing for once-daily dosing.
The proof-of-concept study presented was designed as a randomized, double-blind, placebo-controlled, 10-day monotherapy.52 Patients at baseline were either treatment naive or experienced, not receiving anti-HIV treatment or had an HIV RNA between 10,000 to 300,000. Participants were then randomized into five cohorts to receive placebo versus five different doses of GS 9137 (200 mg twice daily, 400 mg twice daily, 800 mg twice daily, 800 mg once daily or 50 mg once daily boosted with 100 mg of ritonavir).
The 24-hour, steady-state pharmacokinetics demonstrated that GS 9137 trough concentrations on the boosted 50-mg dose were well above what was seen with any of the twice-daily dosages. There was a dose-response relationship between the day 10 Cmin and antiviral effect.
All doses performed statistically better than placebo. The best responses were seen with the 400-mg twice-daily, 800-mg twice-daily and the boosted 50-mg once-daily arms, in which an approximate 2-log reduction on HIV RNA from baseline was observed. All patients on these three arms had a greater than 1-log reduction, and 50% of them had a greater than 2-log reduction in HIV-1 RNA.
GS 9137 was also well tolerated during this short trial, with no significant adverse events or laboratory abnormalities noted. After these results, plans are for phase 2b studies exploring other boosted dosages in treatment-experienced patients.
MK-0518 and GS 9137 appear to be extremely potent, as proven by these early results in which baseline viral HIV RNA dropped to ~2 logs. Of course, longer studies and follow-up will reveal the details regarding the long-term tolerability, safety and durability of antiviral responses. But with these exciting early results, we cannot help wonder about the possibilities for combining these agents with other new compounds such as entry inhibitors, and even the possibility that these two agents can be used in combination or sequentially. They would be a much needed addition for treatment-experienced patients who are running out of active agents.
In fact, it was recently announced that experienced participants in the ongoing phase 3 trial of MK-5018 would be allowed to add both enfuvirtide and TMC114 (under expanded access) to their optimized background regimen. This action could potentially offer highly experienced patients the opportunity to receive three drugs for which they may have full sensitivity. On the other hand, the use of these two drugs will be allowed in the ongoing phase 2B trial of GS 9137, but patients are not permitted by study design to receive TMC114 and GS 9137 concomitantly.
PA-457 has a 90% inhibitory concentration of 35 nM and demonstrates potent activity against wild-type and drug-resistant viral isolates. This activity is synergistic when combined with other approved drugs. Conversely, PA-457-resistant virus is sensitive to approved drugs.54 Pharmacokinetic studies have shown the half-life of PA-457 to be very long -- on the order of two to three days -- which results in significant drug accumulation and provides a low peak-to-trough ratio that allows once-daily dosing. PA-457 is metabolized by glucuronidation and does not inhibit hepatic P450 enzyme systems. In initial trials in healthy volunteers, the drug was found to be safe and well tolerated and demonstrated no end-organ toxicity.55
At CROI 2005, details of a PA-457 proof-of-concept trial were presented.56 This was a randomized, double-blind, single-dose study in which 24 HIV-infected patients received placebo or 75, 150 or 250 mg of PA-457. A maximal viral reduction of 0.73 log10 copies/mL occurred in patients treated with the higher doses of PA-457, and the antiviral effect persisted in some patients over several days.
Data were presented from the 10-day monotherapy study of PA-457 at ICAAC 200557 and CROI 2006.58 This was a randomized, double-blind, placebo-controlled study conducted in treatment-naive and treatment-experienced patients currently off therapy and who had a baseline CD4+ cell count above 200 cells/mm3 and a viral load between 5,000 and 250,000 copies/mL. Patients received one of four oral doses of PA-457 (25, 50, 100 or 200 mg) once daily for 10 days.
As can be seen below, the viral load reductions achieved at different doses demonstrated a clear dose-response relationship. The 50-mg, 100-mg and 200-mg doses achieved viral load reductions that were statistically significantly different from that observed in the placebo arm, whereas the 25-mg response was no different from placebo. At day 11 of dosing, the median viral load reductions observed were approximately 0.5 logs for the 100-mg dose and 1.0 log for the 200-mg dose. These response rates did not differ in people with and without antiretroviral experience.
It is interesting to note that the viral decay curves for the 100-mg and 200-mg groups still appear to be declining at day 10. Viral load decay did not begin until approximately the second and third days of dosing, but the maximum viral load response was reached between one and three days after dosing was stopped.
Plasma trough concentrations were measured immediately prior to dosing on each morning.57 There was a clear proportionality between the dose and the trough concentration, with the 200-mg dose achieving trough concentrations approximately twice those of the 100-mg dose. The steady-state drug levels were reached at about day 7 of dosing. The shape of the viral response curves and the pharmacokinetic/pharmacodynamic relationship suggest that PA-457 200 mg may not attain the peak dose response possible; higher PA-457 doses may be needed.
Overall, the PA-457 pharmacokinetic level proved to be the best correlate of virologic response, with the exception of one person who never took antiretrovirals and who essentially had no response to PA-457 200 mg/day (only a -0.07-log change in viral load) for reasons that remain unclear. This person did attain adequate plasma levels of the maturation inhibitor, so this case certainly will require further investigation.
The safety profile of PA-457 over 10 days of dosing was similar to that previously reported. All adverse events were graded as mild or moderate in severity and did not differ between those patients receiving PA-457 and those receiving placebo. Similarly, there were no grade 3 or 4 laboratory abnormalities noted. One person with poorly controlled hypertension suffered a lacunar stroke during the trial, and the author noted that PA-457 could not be eliminated as a contributor.57
In summary, PA-457 demonstrated itself to be a potent inhibitor of HIV replication in this study. At a dose of 200 mg, the median viral load reduction was approximately 1 log, with reductions up to 1.7 logs noted in various individuals. This drug appeared to be safe and well tolerated. Moreover, there was no evidence for treatment-emergent mutations against PA-457 in a follow-up resistance analysis presented at CROI 2006.59 In that study, viral resistance to PA-457 was examined in vitro and in vivo. The in vitro resistant mutants were selected by serial passage at a suboptimal drug concentration. The in vivo samples were obtained from gag sequences from patients who participated on the 10-day, multidose monotherapy study. The in vitro selection identified five amino acid changes that independently confer PA-457 resistance, but none of those changes were observed in vivo.
Overall, these data validate maturation inhibition as a target for HIV therapeutics and support further development of these agents. The PA-457 dose of 200 mg/day appears to be the most effective of all the doses studied, but the question remains as to whether greater responses can be attained with doses above 200 mg. There are plans to move into phase 2b trials with PA-457 in 2006.
Much can be said for the continued development of new antiretroviral agents. Over the past few years, this research has led to significant improvements in how HIV is managed by health care providers given the availability of new formulations that reduce pill burden, more useful reformulations of older drugs and fixed-dose combination pills. The development of several highly effective new therapies, such non-peptidic PIs, second-generations NNRTIs, entry inhibitors, integrase inhibitors and maturation inhibitors, have the potential to offer significant benefits to patients with an extensive treatment history or with highly drug-resistant HIV.
The development of these new agents seems slow now in comparison with the drug development progress that we witnessed in the mid-1990s -- and for good reason. The agents recently approved by the FDA for HIV treatment have significantly raised the bar for what is to be expected from newer therapies, and we should not rush and compromise safety for convenience.
It is important that appropriate pharmacokinetic, pharmacodynamic and safety and efficacy studies are thoroughly conducted. We need to remember the painful lessons that HIV therapeutic history has taught us. Many mistakes do not bear repeating, such as those made with the dose-finding safety studies for stavudine (d4T, Zerit) or the gender dosing differences with the use of nevirapine, both of which were not discovered until years after the drugs were approved. Thankfully, we already have several good drugs to carry us through while we wait for the new generation of therapeutics.