Earlier this month, the fluoronated 3TC agent, FTC (Emtriva) received FDA approval for the treatment of HIV in adults. The latest once-daily nuke on the block (now that Gilead has acquired the drug from Triangle Pharmaceuticals), lends itself handsomely to a fixed dose co-formulation with Gilead's other QD offering: tenofovir (Viread). FTC is also in Phase III clinical studies for the treatment of chronic hepatitis B. Excerpts of TAG's position paper on FTC (a.k.a. emtricitabine), prepared by Antiviral Project Director, Rob Camp, appear below.

FTC is a once-daily (QD) thiacytidine nucleoside analogue developed by Triangle Pharmaceuticals. The compound, 2'-3'-dideoxy-5-fluoro-3'-thiacytidine, was licensed from Emory University in 1996. It comes in 200 mg capsules and should be stored at room temperature. There are no food restrictions with FTC administration. FTC has been developed quite slowly. What may have been the most interesting thing about this drug, i.e., once daily dosing, is less interesting now that 3TC is licensed as such.

People with the M184I/V mutation associated with 3TC (Epivir) will not benefit from FTC, given that the M184 mutation confers high-level resistance to FTC. For this reason, it is important that persons with detectable viral loads who plan to switch from 3TC to FTC have genotypic testing performed to determine whether the M184V mutation is present. (Of course, a patient's treatment history is also extremely important.) If 3TC has failed in the past, the 184 mutation is archived, thus rendering FTC ineffective.

How effective is FTC for the indication in question?
In all studies to date, including the FTC-302 study (which was stopped due to liver issues, later shown not due to FTC), this drug has been seen as a competent first-line cytidine analogue, the main difference being that it is dosed once a day. Interestingly, Triangle/Gilead conducted head to head comparison studies with other nucleoside analogues. Unlike abacavir where the development philosophy was to demonstrate its superiority to a combination of two nucleoside analogues (AZT and 3TC), the FTC development strategy involved comparing it directly with d4T and 3TC, utilizing a "non-inferiority" design.

What are the real side effects?
Again, FTC deserves a thumbs up. Although new and more horrific side effects are discovered over time with most drugs, because FTC is so similar to 3TC, the side effect profile is probably well-characterized. Although neither drug was ever studied in monotherapy for extended lengths of time, 3TC has shown over time that headache, malaise/fatigue, nausea, sleep interference and dizziness are the main side effects, all hovering around 5%. Lab abnormalities do not reach 2% at 24 weeks. FTC side effects should be very similar.

Are things not being said?
Triangle could have probably been a bit more imaginative in their trial designs, but there seem to be only a handful of questions unaddressed:

• One issue not looked at yet FTC's effect in pregnancy.
• Pediatrics trials done are being repackaged and resubmitted at a later date. When and why?
• There also was a recent charge from the FDA that the Abbott production facility where FTC was being produced was not maintaining standards of quality. The Abbott facility is no longer producing FTC.
• The dose ranging study is only in press now. Why was it submitted so late?
• Liver toxicity may take longer to see than the observation period of these registration studies.
• Trials should have been stratified for hepatitis B serostatus.

Efficacy studies
In FTC-301, antiretroviral-naive HIV-infected people with viral loads gt; 5000 copies/mL were randomized to receive FTC once daily or d4T twice daily and matching placebo, as well as open-label Videx EC (ddI) and Sustiva (efavirenz). Study endpoints included measurement of viral failure, the frequency of genotypic mutations at time of failure and the CD4+ change from baseline between the two groups. The primary measure of analysis for this study was the percentage of patients with plasma HIV-1 RNA < 50 copies/mL at Week 48. The study was powered to be able to demonstrate equivalence between FTC and d4T within a difference of 12.5% as well as to detect differences between treatment groups of 15%.

Five hundred and seventy-one people were randomized and received at least one dose of randomized study medication. At baseline, the median viral load was 4.9 log copies/mL and the median CD4+ count was 288 cells/mm3. The proportion of people having viral failure through Week 48 was 5.3% in the FTC group and 12.7% in the d4T group (p < 0.01). The mean increase from baseline to Week 48 in CD4+ was significantly greater in the FTC group (+153 cells/mm3) than the d4T group (+120 cells) (p < 0.05).

Also measured was efficacy failure—defined as virologic failure, death, progression to CDC class C event, or loss to follow-up—which occurred in 18% of people in the d4T group and 9% of people in the FTC group through Week 48 (p < 0.01). In terms of the proportion of people with undetectable viral loads at Week 48 using an intent-to-treat analysis (non-completer equals failure): 80% in the FTC group and 67% in the d4T group had viral loads below 400 copies/mL (p < 0.001), and for plasma viral load < 50 copies/mL, 74% FTC, compared to 58% d4T, got there after 48 weeks of treatment (p < 0.0001).

Genotypic analysis was performed on all of the 49 individuals with confirmed virologic failure through Week 48 (35 d4T, 14 FTC). Of the 35 genotypic evaluable volunteers failing d4T, 34 (97%) had mutations in the HIV polymerase gene as compared to 71% (10/14) from the FTC subgroup (p=0.019). The M184V mutation was observed only in the FTC subset, 43% (6/14), while the thymidine analogue mutations ("TAMs") were observed in 7% (1/14) of the FTC group and 20% (7/35) of the d4T group. Although these results show that FTC was statistically superior to d4T, it is worth reminding that the combination ddI + d4T is not regularly recommended.

In another "non-inferiority" study, the French ANRS 099 or ALIZE, people receiving a protease inhibitor-based regimen with plasma HIV-RNA level < 400 copies/mL were randomized to continue their regimen (C) or to switch to once-daily combination (5 pills per day) of FTC, ddI, and efavirenz (once daily). Intent-to-treat on available data, on treatment on available data and intent-to-treat with (missing= failure) analyses were conducted.

A total of 355 people were randomized; 86% were male with a median age of 41, a median duration of PI use of 35 months, and a median CD4 count of 540 cells/mm3. Of course, today, at 540 CD4s, these people would probably be offered a treatment interruption of some sort instead of rolled into another new treatment protocol of 48 weeks. The proportion (98%) of people with virologic success at Week 48 was 92% for the "C" group (continue) and 94% for the FTC group (ITT).

There was a statistically significant difference in the proportion of people having plasma HIV-1 RNA < 50 at Week 48 (95% on once daily vs. 87% on C, p=0.01). Median CD4 count increase was similar between groups. Rates of treatment discontinuations were "low" (12.4% and 10.1%, no statistical difference). A significant increase in median fasting HDL cholesterol levels was observed in the QD group as compared to the C group. Other metabolic parameters remained similar between groups throughout the 48 weeks of the study. People who simplified to this once daily regimen had an improved outcome compared to subjects remaining on more the complex protease inhibitor-based regimens.

FTC-303 was a randomized, 48-week, open label equivalence trial in which people with HIV-1 RNA <400 copies/mL either continued their 3TC regimen or switched 3TC 150 mg twice daily to FTC 200 mg once daily while continuing the other medications in their regimen. People with plasma HIV-1 RNA <400 copies/mL at Week 48 were offered FTC as part of their HAART regimen in protocol FTC-350.

Out of 294 people originally randomized to FTC in study 303, 227 (77%) had HIV-1 RNA <400 copies/mL at Week 48. Of these, 215 continued in rollover study FTC-350; 152 of 294 (51%) maintained suppression of HIV-1 RNA < 400 copies/mL and 139 (47%) <50 copies/mL through Week 120 (2.3 years). Through a median follow-up of four years, the probability of virologic failure (gt;400 copies/mL) was 11% while prescribed FTC.

FTC was well tolerated throughout the study. The probability of treatment limiting adverse events was <13% at year 4 of follow-up. The annualized rate of the most common treatment-emergent Grade 3 or Grade 4 laboratory abnormalities was 7% for creatine kinase and 5% for hypertriglyceridemia, with the annualized rate of all other laboratory abnormalities < 2% after 4 years of follow-up. Asymptomatic and transient elevations in CPK accounted for more than 2/3 of the overall Grade 4 adverse events. Everyone continued on the background regimen that was part of the stable 3TC regimen at the start of FTC-303. This regimen included AZT or d4T as well as an NNRTI or PI.

Studies to be done

• A pharmacokinetic assessment in people with different stages of hepatic impairment.
• Are there plans to do AZT+FTC studies due to neurological issues seen with AZT+3TC?
• Is there a clinical benefit to keeping the 184V mutation active?
• In a treatment interruption setting, should FTC be stopped before the other drugs, not unlike 3TC and the NNRTIs?
• ACTG studies 5015 (accelerated disease and aging) and 5073 (a DOT study) are fully enrolled and should be reporting results soon.

Expanded access
There has been no expanded access for FTC. We would like to express our regret at the sponsor's unwillingness to test this drug in real life prior to approval. Even an open-label pseudo-expanded access scheme (where the issue is not the life and death of patients) would have been helpful in better defining some of the outstanding questions as enumerated in this report. ¤

The author would like to thank Francois Houyez for his collaboration on this paper.

Extracellular agents: Attachment and entry inhibitors

With the exception of T-20 (Fuzeon), all FDA-approved anti-HIV drugs act inside HIV infected cells by interfering with reverse transcriptase or the HIV protease enzymes. The third major drug target that has been successfully exploited is HIV gp41—a molecule on the virus' surface that changes its shape in a specific way to allow viral fusion.

Why are investigators looking here, outside the cell? HIV drugs that work inside the cell can be efficiently neutralized by some cells, using primitive, innate self-defense mechanisms such as efflux pumps which sense toxins and eject them out of the cell. This kind of "cellular resistance" may be an important reason for viral persistence and evolution in people on seemingly potent combination therapies. Extracellular antiretroviral therapy might circumvent this problem. More immediately, any drug working on a separate part of the virus' life cycle would have little or no potential for cross-resistance with the current drug regimens.

Entry inhibitor: T1249
T-1249's development is roughly three years behind T-20. And as there are no orally available fusion inhibitors in current development, where we are in 2003 is where we are likely to be for the foreseeable future.

T1249 is another injectable Roche peptide, similar in design to T-20 but binding to gp41 just downstream of T-20's binding site. It has been tested in phase I/II trials for activity against T-20-resistant viruses.

In one small study, T1249 was given to 23 people with both genotypic and phenotypic resistance to T-20 for 10 days. 19/23 (79%) people had at least a 0.5 log drop in HIV RNA at day 11, which is in the range of non-significant. People failing T-20 for 24-48 weeks appeared to have better responses (7/7 people achieved gt;1 log decrease in HIV RNA; median day 11 decrease in HIV RNA -1.6 log) than those who had been failing for gt;48 weeks (8/17 achieved gt;1 log decrease in HIV RNA; median day 11 decrease in HIV RNA -0.94 log). Frequency of injection sites reactions was not reported.

This was an interim analysis. The full data set from this study was expected to be presented at the IAS meeting in Paris.

Extracellular agents: CD4 receptor antagonists

Attachment inhibitor: PRO 542
Progenics' Pro 542 is sort of rsCD4 revisited. A soluble CD4 receptor, Pro 542 binds to and neutralizes gp120 before binding can occur. The CD4 receptor region is integrated into an immunoglobulin molecule to form a protein synthesized via monoclonal antibody technology.

Results from phase I/II clinical trials of Pro 542 involving HIV-infected adults and children were published in 2000. In the adult phase I study, volunteers were treated with a single intravenous infusion of Pro 542 at doses of 0.2-10 mg/kg. Pro 542 was well tolerated, and no dose-limiting toxicities were identified. Transient HIV-RNA decreases were reported after single-dose administration.

In the phase I/II study, enrolling 18 HIV-infected children, Pro 542 was evaluated by single and multidose intravenous infusions. The drug was well tolerated, and, as seen in the adults, dose proportionality was observed in terms of AUC and serum concentrations. Decreases of approximately 0.7 log copies/mL in plasma HIV-RNA levels were seen in four of six children treated with four weekly 10 mg/kg doses. After two weeks of treatment, three children had sustained reductions in serum HIV-RNA; the other children had rebounded to baseline levels.

A second set of phase II clinical trials, which are also being conducted in adults and children, were kicked off in 2000. These studies include people with HIV resistance to current antiretroviral options. The drug is currently being evaluated in an improved formulation for subcutaneous administration.

Entry inhibitor: TNX-355
TNX-355 (formerly HU5A8) a humanized IgG4 anti-CD4 domain that acts by binding to the CD4 receptor, and thus keeps HIV from entering after binding. It has demonstrated potent anti-HIV-1 activity in vitro, and studies in rhesus macaques and human peripheral blood lymphocytes indicated that it is not immunosuppressive. It is broadly inhibitory across all clades of HIV and inhibits both CCR5 and CXCR4 tropic virus.

The antibody attaches to the side of the CD4 receptor in a way that does not interfere with its regular function as a chemokine receptor. Nor does it interfere with HIV's docking with the CD4 receptor, but it does block HIV from taking further steps in the process of entering the cell.

Does it also then block the entrance of other normal agents? Based upon these preliminary data, a study was performed to determine the safety and preliminary anti-HIV activity of a single dose of TNX-355 in HIV-infected people.

A human study of TNX-355 enrolled five sequential cohorts of six HIV-infected people who received single IV doses of TNX-355 in an open-label dose-escalation study. They had a mean baseline CD4 count of 354 cells/mm3 and viral load of 5.1 log copies/mL. All were HAART-experienced and almost 2/3 of the people were on failing HAART regimens at study entry.

Duration of complete CD4 cell coating with TNX-355, ranged from 1-2 days at 1 mg/kg to 15-27 days at 25 mg/kg, which correlated with the day of viral load nadir. No significant adverse events were reported.

The single dose did not generate any natural antibody reaction. Peak reduction in viral load occurred out to days 14 and 21. The study team is in discussions with researchers who do kinetic modeling to try to understand the process. Further assessment of therapeutic potential awaits data from longer duration trials; a phase 1b multiple-dose study is planned.

One advantage of TNX-355 may be the possibility of administering it on a schedule of once a week to once every three weeks. Macaques with multiple exposures to the monoclonal antibody developed antibodies to it that blocked activity of the drug. This issue, not seen in the single-dose human study, is one that will be closely monitored as the drug moves into phase II trials.

Extracellular agents: Co-receptor antagonists

Co-receptor antagonist: SCH-C
Schering C (SCH-C) is one of several small-molecule agents that have been studied as potential antagonists of CCR5.

Earlier, there had been some concern regarding QT prolongation seen in individuals receiving the highest dose of SCH-C. These observations led the FDA to put the SCH-C development program on hold. However, the hold on development has since been lifted.

Three doses have been selected for further evaluation—25 mg, 50 mg, and 100 mg, all administered twice daily—and additional safety reviews will be conducted. One clinical trial had 12 people with SCH-C 25 mg twice daily and 12 people with SCH-C 50 mg twice daily. After ten days, the average reduction in HIV-RNA was approximately 0.7 log copies/mL in the 25 mg twice daily group and 1.1 log copies/mL in the 50 mg twice daily group.

There were no discontinuations because of adverse events, the most common being headaches and altered taste. Some QT prolongation was observed—the mean increase was 11.5 msecs after ten days of treatment—an observation that originally had led the FDA to put clinical development of SCH-C on hold.

SCH-D, Schering-Plough's second CCR5 antagonist, has been shown to be more potent than SCH-C in vitro.

Co-receptor antagonist: UK-427,857
This small Pfizer molecule is is a prototype CCR5 antagonist for the treatment of HIV. It is already in phase I dose-ranging trials in humans. Thus far, it appears that this compound has potency against isolates that utilize CCR5 for entry; however, it has no activity against CXCR4-tropic viral isolates. It is non-competitive with respect to chemokine binding. While it binds the receptor reversibly, it has a long binding half-life, which may lead to advantageous pharmacodynamics.

Integrase Inhibitors

Integrase inhibitor: S-1360
S-1360 is a diketo acid being developed by Shinogi Pharmaceuticals and GlaxoSmithKline. Preliminary data from early preclinical and clinical studies were reported at the 2002 Retrovirus conference and at the XIV International AIDS Conference in Barcelona. According to in vitro data reported at the Seattle meeting, S-1360 is synergistic with all of the available antiretrovirals and has potency that is on a par with lamivudine. In animal models, S-1360 was found to be 70% to 80% bio-available and had a half-life ranging from one to two hours.

As for clinical data, 18 HIV-negative study volunteers received single doses of S-1360. In all 18 people, the plasma half-life ranged from 7.7 to 16 hours, meaning that once-daily dosing is possible. Phase I and II studies of S-1360, in HIV-infected people, are under way.

Integrase inhibitors: L-870,812, L-870,810
Merck's twin contenders, L-870,812 and L-870,810, are napthyridine carboxamides compounds, which have demonstrated potent antiretroviral activity in vitro. The 812 compound has an oral bio-availability of 64%, and the 810 an oral bio-availability of 49% (both in rhesus macaques).

L-870,812 has been tested in macaques infected with a recombinant SIV/HIV virus. SHIV-RNA was reduced by 1 to greater than 3 log in the treated laboratory animals, and 4/6 macaques experienced an SHIV-RNA decrease to undetectable levels. (Samples collected from the two macaques that did not achieve maximal SHIV-RNA suppression had evidence of an N155H mutation in the integrase gene.)

Despite the structural differences between the Shinogi and Merck integrase inhibitors, a report at the 2003 Retrovirus meeting noted a significant potential for cross-resistance between these two integrase inhibitors, which are both currently undergoing clinical development. A conflicting report, however, by investigators from the Rega Institute, concluded precisely the opposite: saying that the resistance profiles are in fact distinct. Guess we'll just have to wait and see. ¤