The Body Covers: The 10th Conference on Retroviruses and Opportunistic Infections
CROI 2003: Emerging HIV Therapies
February 14, 2003
Research presented regarding several drugs most likely to be soon available to patients included atazanavir (ATV, Zrivada), T-1249, T-20 (enfuvirtide, Fuzeon), tipranavir (TPV), GW433908 ("908") and extended release d4T (stavudine, Zerit).
Atazanavir (ATV, Zrivada), manufactured by Bristol-Myers Squibb (BMS), is administered as two pills once daily. Unlike other protease inhibitors (PIs), in clinical trials, ATV has not been observed to significantly increase cholesterol or triglycerides or induce glucose intolerance. Rob Murphy1 presented the results of a rollover study that enrolled antiretroviral naive subjects originally randomized to d4T (stavudine, Zerit) + 3TC (lamivudine, Epivir) and either ATV 400 mg/day (once daily) or ATV 600 mg/day or nelfinavir (NFV, Viracept) 1250 mg/day. Those who at the end of the 48-week study had HIV RNA levels less than 10,000 copies/mL, were either continued on ATV at their originally assigned dose (if in the ATV arms) or if they were in the NFV arm, switched to ATV 400 mg once daily.
Of the 369 subjects completing the parent study, 346 joined the rollover study. Overall, 75 percent of the rollover participants had a viral load below 400 copies/mL (77 percent in ATV 400 mg, 76 percent in ATV 600 mg and 71 percent in NFV arm) at entry into the rollover. CD4+ cell counts at entry into the rollover was around 500 cells/µL. The cumulative virologic responses, after 48 and 108 weeks on study drugs, using an intent-to-treat analysis in each of the arms were as follows:
CD4+ cell counts increased during the first 24 weeks of the rollover by approximately 35 cells/µL, with each arm experiencing similar median count elevations.
Lipids were measured and, not unexpectedly, switching from NFV to ATV led to significant declines in total cholesterol (-16 percent), fasting LDL cholesterol (-20 percent) and fasting triglycerides (-25 percent) at 24 weeks following enrollment into the rollover study. As seen previously, subjects treated with ATV didn't experience major increases in lipids. The major toxicity associated with ATV is indirect hyperbilirubinemia, which was observed in the vast majority of subjects treated in this study with ATV. Abnormal bilirubin (any grade) was reported in 76 percent of those who switched from NFV to ATV, 83 percent receiving ATV 400 mg once daily and 88 percent of those treated with 600 mg once daily; grade 3 or higher hyperbilirubinemia was seen in 13 percent, 26 percent and 44 percent of subjects in these arms, respectively. Surprisingly, given the rate of high level hyperbilirubinemia, jaundice was only reported in 22 percent of the ATV 600-mg subjects and 13 percent of all the subjects taking ATV 400 mg during the rollover study.
This study provides longer-term evidence of decent rates of virologic suppression in patients receiving two nucleoside analogues and ATV. The equal efficacy of the two ATV doses and the increased toxicity associated with the 600 mg strength of the drug supports the continued study of ATV 400 mg once daily. The lipid friendly nature of this PI will be its major selling point. However, the real world experience and reaction to the hyperbilirubinemia ATV elicits remains to be seen.
Another issue swirling around ATV is the potential for cross-resistance with other PIs. The patterns of drug resistance that develop during ATV treatment were also presented at the conference. A Bristol-Myers Squibb-sponsored evaluation of isolates was obtained from treatment naive and experienced subjects enrolled in clinical studies of ATV.2 Two patterns of mutation evolution appeared to emerge dependent on prior PI experience. The I50L mutation, a unique mutation associated with ATV, was seen in over two dozen isolates. This interesting mutation appears to sensitize the virus to other PIs. Isolates without the I50L mutation were recovered from subjects with other PI experience and in contrast to those with the I50L mutation had reduced susceptibility to other PIs. This study adds to the available evidence suggesting ATV may be most useful as part of an initial antiretroviral regimen rather than a component of a salvage therapy.
Another feature of ATV is that it doesn't require boosting with low dose ritonavir (RTV, Norvir) -- a pairing that could be disastrous for lipids. However, the concomitant dosing of ATV and another popular once-a-day antiretroviral, efavirenz (EFV, Sustiva), has been observed to reduce ATV levels by over 70 percent. RTV boosting at a dose of 100 mg once daily appears to preserve ATV exposure when administered with ATV 400 mg once daily and EFV 600 mg once daily in a small number of HIV-uninfected volunteers3. Another drug interaction study performed by the French ANRS suggests tenofovir (TDF, Viread) also can reduce ATV levels through mechanisms that remain unclear. The potential need for RTV boosting when ATV is combined with EFV and perhaps TDF threatens the niche this PI is carving out for itself. Further study of the effects of RTV boosting of ATV on lipids and the interaction between ATV and another potential once-daily agent, nevirapine (NVP, Viramune), as well as explication of the TDF-ATV interaction will be important if this agent is to gain a foothold as a once a day alternative to current PIs.
The entry inhibitor T-20 (enfuvirtide for lingual contortionists) is not yet out on the market but a new and improved descendent is quickly being developed. T-1249 is an injected compound that binds to a slightly different segment of gp41 than T-20 and is active against HIV-1 and HIV-2. To determine the safety and the antiviral effect of T-1249 among patients failing T-20, the manufacturer of both drugs administered T-1249 to 24 patients who were virologically failing T-20, as demonstrated by the consecutive detection of plasma HIV RNA levels between 5,000-500,000 copies/mL4.
All subjects discontinued their T-20, maintained their other antiretrovirals and initiated T-1249 for 10 days. Median time on T-20 was 70 weeks and the median time on T-20 after detection of virologic failure was 60 weeks. Viral load at the time of the commencement of T-1249 was 5.0 log10 copies/mL. Genotypic evidence of T-20 resistance was found in all subjects. Each participant completed the 10-day T-1249 course. The median viral load decline at the end of the 10-day treatment was 1.12 log10 copies/mL (95 percent CI, -1.5 to 0.83) and 63 percent of subjects had at least a 1 log10 drop in HIV RNA level. A longer duration of previous T-20 therapy was associated with a decreased likelihood of responding to T-1249. All seven subjects who took T-20 for less than 48 weeks after virologic failure achieved at least a 1 log10 drop for a median decline of -1.6 log10 copies/mL compared to -0.94 log10 copies/mL in those who had maintained T-20 for more than 48 weeks after failure. T-1249 seemed well tolerated during this short ongoing study.
These preliminary data indicate that T-1249 has activity in vivo against T-20 resistant virus. That there are limits to the ability of T-1249 to salvage patients failing T-20 due to the eventual development of cross-resistance, is demonstrated by the reduced effectiveness of T-1249 in patients who continued T-20 while experiencing long periods of high-level viral replication. T-1249 offers hope to those who may consider T-20 their last chance. The rest of us can only hope that the "son-of T-20" inherits a name that someone other than the staff at Trimeris Pharmaceuticals can pronounce.
5 described the histopathologic changes seen in the skin and soft tissue of patients developing injection site reactions to T-20. Injection site reactions are almost universal among persons using T-20. They are characterized as solely erythema to palpable nodules that can be quite mild and hardly noticed by the patient or at the other extreme be inflamed, indurated, erythematous and painful. Biopsies of injection sites in seven patients were performed. Four had palpable nodules while the others had erythema and mild induration, induration alone or no reaction. Pathologic examination of the tissue revealed an inflammatory infiltrate consistent with a localized hypersensitivity reaction with a concentration of eosinophils and histiocytes. In some cases, the inflammatory cells included multinuclear histiocytes aggregated around areas of altered collagen similar to the appearance of granuloma annulare and a rare reaction to systemically administered medications called granulomatous drug reaction. Not unexpectedly, altered immune function is posited to be the etiological force underlying the development of these reactions to T-20.
Another noteworthy new PI that, unlike ATV, may have a role in the treatment of PI-experienced patients is tipranavir (TPV). As described in another conference coverage report on this site, TPV has been demonstrated in vitro to have activity against multi-PI resistant strains of HIV. Unfortunately, the drug is poorly bioavailable, requiring 12 pills three times a day. However, when taken with a lower dose RTV, dosing frequency and pill burden drop into the range of feasibility. Various concoctions of TPV and RTV have been studied in preliminary studies but the optimal dosing strategy has not been determined.
To obtain safety, pharmacological and efficacy data on candidate doses of TPV/RTV, the makers of this drug, Boehringer Ingelheim Pharmaceuticals, conducted the BI 1182.52 study6. This is a multi-centered, international, randomized and blinded study of three dosing strategies of TPV/RTV: 500 mg/100 mg twice a day, 500 mg/200 mg twice a day and 750 mg/200 mg twice a day. Eligible subjects had to have experience with agents of all three classes of approved antiretroviral therapy -- including two or more PIs, be failing their current PI-based regimen with an HIV RNA PCR level above 1,000 copies/mL and have at least one of a specified list of primary PI mutations (30N, 46I/L, 48V, 50V, 82A/F/L/T, 84V and 90M) but not have more than one of the following: 82L/T, 84V, 90M.
After entering the study, subjects discontinued their PI(s) and initiated one of the three study doses of TPV/RTV while maintaining their pre-study NRTIs. After 14 days, subjects were assessed for changes in viral load and at that point were permitted to have their regimens optimized by modifying their pre-entry antiretrovirals. For continued safety and efficacy data collection, subjects then continued on TPV/RTV and their optimized regimens. A total of 216 subjects were enrolled in the trial and randomized to the three dosage arms. Subjects were generally male (84 percent) and white (76 percent) and had a median entry viral load and CD4+ cell count of 4.53 log10 copies/mL and 153 cells/mm3, respectively. About a third of the subjects had prior exposure to lopinavir/ritonavir (LPV+RTV, Kaletra) and more than three-quarters had prior indinavir (IDV, Crixivan) therapy. There was significant evidence of significant PI resistance at entry, despite the inclusion criteria. Over 90 percent of the subjects had 10 or more mutations associated with resistance to all available antiretroviral drugs, and 41 of the isolates from the cohort had genotypic resistance patterns that indicated reduced susceptibility to all current PIs.
After 14 days there was a decline from a baseline HIV viral load of 0.9 log10 copies/mL in the 500/100 arm, a 1.0 log10 drop in the 500/200 arm and a 1.2 log10 fall in the 750/200 arm (p=NS). Response was truncated when multiple PI resistance mutations were present at baseline, but according to the presentation, it was only after 16 or more such mutations were detected that the efficacy of TPV/RTV waned significantly. TPV/RTV also continued to demonstrate an ability to reduce viral load, at least at 14 days after initiation, even when one to two of the so-called universal PI-associated mutations or UPAMs (l33I/V/F, V82A/F/L/T, I84V and L90M) were present baseline.
After 28 days, toxicity was assessed. Grade 3 or 4 increases in ALT, AST and total bilirubin were seen across all arms and were dose dependent as were adverse event related treatment discontinuations. TPV combined with RTV also clearly can increase lipid levels. Elevations of triglycerides to greater than grade 2 was seen in 9.5 percent of the 500/100 subjects, 15.5 percent of those receiving 500/200 and 11.3 percent of participants assigned to 750/200 after a month of therapy.
Based on these results and pharmacokinetic data presented at this conference7, the manufacturer has decided to pursue further clinical development of the 500/200 twice a day dosing strategy. The demonstration that this PI combination can provide durable viral load suppression beyond two weeks, particularly in PI-experienced patients, is the next hurdle that must be cleared. Further, more information needs to be collected on longer term lipid elevations and hepatic safety if this agent is to be considered as part of a first or second treatment regimen. Additionally, data presented earlier at this conference suggest that as many as 20 percent of subjects in a pharmacokinetics substudy receiving the 500/200 dose achieved a trough concentration of the drug below 20 µM -- a level 10 times the protein adjusted IC90 for PI-resistant HIV and a concentration target. An examination of the factors that may influence TPV levels needs to be included in the next phase of study.
In summary, these are encouraging early data regarding this important PI. Continued study of this drug will help determine where it may best fit as an addition to our growing armamentarium of antiretrovirals.
Given the large pill burden of amprenavir (APV, Agenerase), GlaxoSmithKline has developed GW433908 ("908"), a pro-drug taken as two pills twice a day. This drug has been studied in treatment naive and experienced patients and the results were presented in back-to-back oral presentations.8, 9 The NEAT Study is an international trial comparing 908 administered 1,400 mg twice a day versus NFV 1,250 mg twice a day when either were combined with abacavir (ABC, Ziagen) + 3TC as initial HIV therapy. A total of 251 subjects were enrolled and randomized 2:1 to 908 or NFV. All subjects had to have a viral load above 5,000 copies/mL. After 48 weeks, the proportion of subjects achieving an HIV RNA level below 400 and 50 copies/mL was higher in the 908 arm (66 percent and 58 percent) than in the NFV arm (51 percent and 42 percent). (Despite the certain presence of a biostatistician on the study team, p values were scarce in this and the subsequent presentation.) Differences in the rates of virologic failure, viral rebound and insufficient viral response also were seen and favored 908. Among those with a baseline viral load of 100,000 copies/mL, more were likely to achieve a drop to less than 400 copies/mL at week 48 of the study in the 908 arm than in the NFV arm (67 percent versus 35 percent). CD4+ cell counts increased by over 200 cells/µL in each group.
At a time when we are seeing rates approaching 80 percent or more for viral suppression to below 50 copies/mL during clinical trials of initial HIV therapy, these results are lackluster. Being superior to NFV proves only that you are better than NFV. Whether 908 can compete against other initial regimen heavy hitters such as the NNRTIs or even ATV needs to be better defined if this promising drug is to find a niche among the treatment naive.
Another 908 study examined the efficacy of the drug as a salvage agent -- the only real role papa APV has played. The CONTEXT study looked at the addition of 908 at 1,400 mg/RTV 200 mg once daily versus 908 at 700 mg/RTV 100 mg twice a day versus LPV+RTV three capsules twice a day to two NRTIs in PI-experienced subjects with an HIV RNA level of 1,000 copies/mL or greater on their current regimen. All subjects had to have a genotype indicating that NRTI therapeutic options remained available. Results after 24 weeks of study treatment were presented.
A total of 315 subjects were enrolled and received treatment in one of the three study arms. At baseline, the median CD4+ cell count of the cohort was 263 cells/µL and HIV RNA was 4.14 log10 copies/mL. More subjects in the 908 once-daily arm had prior antiretroviral experience than the other arms. As opposed to using the proportion of subjects who reached a set level of viral suppression (i.e., <400 copies/mL or <50 copies/mL), the primary objective was the time-averaged change in viral load from baseline. The drop in viral load at 24 weeks was -1.47 log10 for 908 once daily, -1.5 for 908 twice a day and -1.66 for LPV+RTV by intent-to-treat analysis. Statistically significant? As in the previous presentation on 908, no p values were provided.
Looking at the proportion of subjects who reached <50 copies/mL at six months: 41 percent of those on 908 once daily, 42 percent of those on 908 twice a day and 48 percent of those on LPV+RTV. Virologic failures seemed more common in the 908 once-daily arm (34 percent) compared to the 908 twice a day (27 percent) and LPV+RTV (21 percent) arms. Each arm saw an increase of CD4+ cell counts by 62-73 cells/µL.
These are preliminary data. Further results following a year of treatment will be forthcoming. Information on baseline drug resistance and the patterns of resistance emerging during study treatment needs to be included in subsequent reports of these results.
In recognition of the once a day steamroller coming down the tracks, the makers of d4T have created an extended release formulation of d4T recently FDA approved. To determine if combining TDF with the new formulation of d4T has any significant impact on d4T levels, researchers from BMS performed a pharmacokinetics study in 18 healthy men10. Previous data have described an interaction between TDF and ddI (didanosine, Videx) in which blood levels of ddI are increased by concomitant administration of these agents.
The bottom line is that there was no significant effect of co-administration on d4T levels and the combo seemed to be well tolerated for the nine days the subjects took them.
In addition to these agents, which are at the daylight end of the drug pipeline, a host of compounds at earlier stages of development were discussed. Those that have been given to humans included FTC (emtricitabine, Coviracil), TMC-114 and TNX-355.
FTC, like its close relation, 3TC, is a cytosine analogue and can be administered once daily. The long term efficacy and safety of FTC was evaluated in an extension of a study in which patients on a 3TC-containing regimen that was their first antiretroviral combination, who had a viral load below 400 copies/mL, were randomized to either continue their 3TC-containing regimen or switch their 3TC to FTC 11. In the rollover study, subjects who maintained a viral load below 400 copies/mL until week 48 enrolled and those on FTC continued this agent. Subjects in the 3TC arm could now switch to FTC. A total of 294 subjects were originally assigned to FTC during the previous study and 215 entered the study extension. The poster focused only on the subjects who had been on FTC long term. Of the 294 subjects on FTC from the start, 152 (51 percent) maintained suppression of HIV-1 RNA = 400 copies/mL and 139 (47 percent) = 50 copies/mL through week 120 (2.3 years). The probability of virologic failure after four years of FTC-containing therapy was 11 percent. FTC was well tolerated.
FTC is clearly an agent about to make a splash. Another poster presentation demonstrated that FTC was superior to d4T in achieving viral suppression in treatment naive patients also taking ddI and EFV12. Almost three quarters of the 286 subjects assigned to FTC versus 58 percent of the d4T-receiving subjects had a viral load less than 50 copies/mL at 48 weeks (p=0.0001). Moreover, the FTC-treated subjects had a greater CD4+ cell count increase and were less likely to develop thymidine analogue associated mutations (TAMs) or ddI resistance. The recent acquisition of Triangle Pharmaceuticals -- the makers of FTC -- by Gilead, the manufacturers of TDF, means there will likely be a new lean, mean once-a-day nuke combo on the block.
This compound, billed as the "next generation PI," has a unique ringed chemical structure and in vitro activity against fiendishly drug resistant strains of HIV. A liquid formulation of TMC-114 was studied at three doses along with RTV -- 300 mg/RTV 100 mg twice a day, 600 mg/RTV 100 mg twice a day and 900 mg/RTV 100 mg once daily -- in 50 HIV-infected subjects who had multiple PI experience and were failing a PI-based regimen (viral load >2,000 copies/mL). Twelve of the subjects continued their failing regimen while the other subjects discontinued their PI in exchange for initiating one of the study assigned regimens while maintaining their pre-study NRTIs. After 14 days the TMC-114 regimen was stopped and optimization of the regimen was permitted.
Subjects were mostly male and white. Baseline viral load was 4.3 log10 copies/mL and the CD4+ cell count was 305 cells/µL. Over 50 percent were taking LPV+RTV as part of their pre-study regimen. The median number of PIs taken in all the arms was three to four and, not surprisingly, resistance to their pre-entry PI was plentiful by genotype and phenotype.
After two weeks, there was no change in viral load among those who continued their entire pre-study regimen but the median drop in HIV RNA in log10 copies/mL was 1.24 for the 300/100 twice a day, 1.50 for 600/100 twice a day and 1.13 for 900/100 once daily; a greater than 1 log10 reduction in viral load from baseline was seen in 69 percent, 92 percent and 69 percent of subjects, respectively. Gastrointestinal adverse effects were the most commonly reported with 32 percent of patients reporting diarrhea. Grade 3 (severe) or more ALT, AST or GGT elevations were seen in five subjects in the TMC-114 arms and one in the control group. Another subject in the active treatment arm developed a grade 4 rash.
The need for concomitant RTV with this new PI is a bit of a downer but may be worth it if the drug proves to have durable potency against multi-drug resistant HIV. The toxicity of the agent coupled with low dose RTV was not negligible in this pilot trial. Further study will clarify how well tolerated TMC-114 is in the current liquid formulation and if made into pill form.
One of the most interesting new compounds discussed at the conference was a humanized mouse antibody (TNX-355) designed to prevent post attachment HIV entry into the CD4+ cell by binding to the CD4+ receptor itself. Binding of this anti-CD4+ antibody is purported to interfere with gp120 and chemokine co-receptor interaction. In the lab the antibody inhibits R5 and X4 virus as well as all clades of HIV-1.
Sounds great, the down side? In macaques, antibodies against TNX-355 developed quickly, however in chimpanzees this has not been observed.
In this phase I open label study, 30 subjects with a CD4+ cell count above 100 cells/µL, a stable viral load above 5,000 copies/mL and on stable or no HIV therapy were assigned to one of five doses of the agent administered once only13. At baseline, 19 subjects were failing a regimen and all subjects were antiretroviral experienced. CD4+ cell count and HIV RNA level at entry were 341 cells/µL and 4.78 log10 copies/mL, respectively, at study entry.
A dose-dependent decrease in viral load was seen 14 days after a single infusion of the antibody with the two highest doses having significant effect. Overall, the drop in viral load was about 1.4 log10 copies/mL. Strangely, the CD4+ cell counts increased within a couple of days of the infusion, before the viral load started to drop, suggesting a change in the trafficking of these cells elicited by this antibody. An assay to measure CD4+ receptor coating by the antibody predicted viral response. No anti-TNX-355 antibodies were detected (humans are more like chimps than macaques). The agent was well tolerated.
The effect of the compound after a single dose on viral load was striking and suggests this is an agent that can be administered infrequently. Further study of this interesting and, thus far, impressive entry inhibitor is certainly something we will be looking for.
A potpourri of other agents in development were reported, most, however, have not been given to humans. One exception is DAPD (amdoxovir), a nucleoside analogue that performed well in a small study of treatment experienced patients14 but led to opacification of the lens of the eye in five of the 18 subjects -- which is a bizarre and concerning toxicity.
The remainder of drugs mentioned at Retrovirus were investigated in the lab. For instance, a new PI from Roche, RO033-4649, was evaluated against wild type and 10 PI-resistant mutant strains of HIV in the lab and demonstrated potent antiviral activity.15 Likewise, in vitro data from the study of three benzophone nonucleoside reverse transcriptase inhibitors (NNRTIs) from GlaxoSmithKline were presented and demonstrated activity against virus resistant to currently available NNRTIs.16 Another new and improved NNRTI is being looked at by Tibotec in the form of the dianilinopyrimidine compound TMC125 which, according to one report,17 may be able to assume various modes of binding, potentially increasing its activity and resistance threshold. Compounds that work at novel targets include new entry inhibitors such as AK602 -- a spirodiketopiperazine derivative that binds to CCR5. UK-427,857 is another CCR5 receptor antagonist administered to healthy volunteers in order to characterize its pharmacokinetics and preliminary safety. PRO 542 (CD4-IgG2) is an HIV-1 attachment inhibitor that, like TNX-355, is an antibody directed against the CD4 receptor, and PA-457 is a small molecule that inhibits the ability of HIV to bud from infected cells.
A European group presented preliminary laboratory data on a new class of potential inhibitors of the HIV-1 integrase, pyranodipyrimidines18, which may offer new hope in the search for a suitable agent against this elusive target.
Lastly, there was quite a buzz about RNA interference. Dr. Judith Lieberman19 reported her results from experiments in which she introduced synthesized pieces of RNA into macrophages infected with HIV and observed significant reductions in HIV replication. As opposed to previous ideas regarding the need to use gene therapy to deliver small pieces of RNA into the cell, in her work, the RNA was readily taken up by liver cells. This exciting report, while above the heads of most of us attending the conference, may be the first step in a completely new way to think about treating HIV, and perhaps a variety of other illnesses.
Patients, especially those on their third or more antiretroviral regimen, frequently ask about new drugs in development. Many say they feel as if they are traveling on a highway that is being constructed as they move along. The fear that they will hit the end of the road prompts them to hope that new drugs are going to be there to pave the way toward long term survival. At this conference there was evidence that the future holds out the promise of easier and more convenient HIV therapies that will be effective in treating drug resistant strains of the virus. As hopeful as this all was, one could not help wondering while listening to the polished presentations and reading the slick posters, whether such innovative medications will be any more available to the millions of people on this planet who need them. If they were, now, that would truly be a breakthrough.
This article was provided by The Body PRO. Copyright © Body Health Resources Corporation. All rights reserved.