News Briefs
Conference Coverage
Introduction
The 12th Conference on Retroviruses and Opportunistic Infections, one of the two major annual meetings addressing HIV/AIDS, took place February 22-25 in Boston.
| On the Web |
|
For more complete coverage of this and other recent conferences, see:
|
This year's conference did not feature any major breakthroughs, but included many oral and poster presentations in areas such as HIV pathogenesis, antiretroviral regimens, side effects of therapy, and experimental agents. Diverging from the usual format, the conference organizers hastily arranged a special session to discuss a case of multidrug-resistant, possibly rapidly progressing HIV that garnered extensive media attention in the weeks before the meeting (see
news item below). Due to the amount of information presented, this conference summary is necessarily incomplete; for more in-depth reports, see the web sites listed
above. For news related to HIV/HCV and HIV/HBV, see the coinfection item following the conference report. For Retrovirus conference coverage and other recent news related to HIV/AIDS in women -- including data on the use of single-dose nevirapine (Viramune) to prevent mother-to-child HIV transmission -- see "
Women's Research Roundup" in this issue.
Treatment Regimens and Strategies
Looking at first-line anti-HIV regimens, data from the large international INITIO study (abstract 165LB) showed that a three-drug regimen consisting of efavirenz (Sustiva, Stocrin) plus two nucleoside reverse transcriptase inhibitors (NRTIs) is superior to two other first-line regimens. The 915 subjects were randomly assigned to receive one of three regimen sequences: 1) ddI (didanosine, Videx)/d4T (stavudine, Zerit)/efavirenz followed by AZT (zidovudine, Retrovir)/3TC (lamivudine, Epivir)/abacavir (Ziagen)/nelfinavir (Viracept) in the case of treatment failure; 2) ddI/d4T/nelfinavir followed by AZT/3TC/abacavir/nelfinavir; or 3) a four-drug regimen of ddI/d4T/efavirenz/nelfinavir. After more than three years of follow-up, in an intent-to-treat analysis, rates of undetectable viral load (below 50 copies/mL) were 74%, 62%, and 62%, respectively. There were no significant differences in terms of CD4 cell count increases, progression to new AIDS-defining illnesses or death, or rates of serious adverse events. Starting with a four-drug regimen provided no additional benefit, but was associated with more side effects. In total, more than one-third of the subjects changed their NRTI backbones, with most switching from ddI/d4T to AZT/3TC.
As the number of antiretroviral drugs has grown, it has become impossible to test all potential combination regimens against each other in clinical trials. In lieu of this, John Bartlett, MD, and colleagues (abstract 586) performed a meta-analysis (an analysis that includes data from multiple trials) of studies looking at three-drug regimens for first-line therapy. This analysis included data from 64 trials with a total of 10,559 subjects. On the whole, regimens containing either a ritonavir (Norvir)-boosted protease inhibitor (PI) or a non-nucleoside reverse transcriptase inhibitor (NNRTI) performed better than regimens based on unboosted PIs or triple-NRTI regimens (undetectable viral load rates of 64%, 63%, 44%, and 51%, respectively). However, boosted PI regimens produced greater increases in CD4 cell count (209 cells/mm3) than NNRTIs (180 cells/mm3), unboosted PIs (178 cells/mm3), or triple-NRTI regimens (150 cells/mm3).
Triple-NRTI regimens have gotten bad press recently due to several studies showing that they may not be potent enough to durably suppress HIV. However, a combination of AZT/3TC/tenofovir DF (Viread) appears to work well in some individuals. An open-label French study of 36 treatment-naive subjects (abstract 599) found that 90% had viral loads below 50 copies/mL after six months on this regimen, as did 69% after 12 months. During the first year, four subjects (11%) experienced virological failure; two of these stayed on the AZT/3TC/tenofovir regimen and maintained low viral loads. Despite recent recommendations to avoid triple-NRTI regimens, the authors concluded that AZT/3TC/tenofovir "should be further evaluated."
The DART study in Africa (abstract 22) also looked at the same triple-NRTI regimen. In a cohort of 200 symptomatic Ugandan subjects with CD4 cell counts below 200 cells/mm3, 51% achieved viral loads below 50 copies/mL and 68% below 400/copies/mL in an intent-to-treat analysis after 24 weeks; the median CD4 cell increase was 88 cells/mm3. While better outcomes would likely be achieved using PIs or NNRTIs, the researchers concluded that triple-NRTI regimens "are highly relevant in resource-limited settings."
T-20 (enfuvirtide, Fuzeon), one of the few drugs that effectively suppress HIV in people with extensive resistance to the three major antiretroviral drug classes, is itself susceptible to resistance. Cecilia Cabrera and colleagues (abstract 718) studied 15 heavily treatment-experienced subjects. All initially experienced reductions in viral load after starting T-20, but HIV RNA increased soon after week 4 in 13 of the subjects. Mutations in the gp41 envelope protein developed by week 2-4 in all subjects. However, Steven Deeks, MD, and colleagues (abstract 680) reported that after 22 subjects with T-20-resistant HIV stopped taking the drug, they experienced modest viral load increases, leading the authors to conclude that T-20 continues to have "persistent low-level activity" even when resistance mutations are present. Finally, a 30-person study by George Beatty, MD, and colleagues (abstract 581) found that interrupting HAART prior to starting salvage therapy with T-20 did not lead to improved virological response at 24 weeks compared with immediate initiation of T-20, and that baseline susceptibility to T-20 did not predict treatment outcome. [Ed. note: Dr. Beatty is a member of BETA's Scientific Advisory Committee.] Together, these studies indicate that when T-20 failure occurs, it happens relatively early (4-8 weeks), but that individuals who do well on the drug can expect sustained response.
Along with all the study data about specific antiretroviral regimens, researchers presented some "big picture" results. An analysis of more than 6,800 subjects in the EuroSIDA cohort by Christian Holkmann-Olsen and colleagues (abstract 601a) revealed that HAART is effective across the board, regardless of baseline CD4 cell count or viral load. During 22,766 person-years (PY) of follow-up, the researchers recorded 889 instances of new AIDS-defining illness or death (125 deaths). They found that the incidence of AIDS or death for any given CD4 cell count or viral load category was "similar regardless of specific drugs being used." In comparison with indinavir (Crixivan) -- for which clinical endpoint data are available (most newer drugs were approved on the basis of short-term laboratory marker data) -- a wide variety of regimens (two NRTIs plus either an NNRTI or a boosted or unboosted PI, or abacavir plus two other NRTIs) reduced the risk of AIDS-defining illness and death to a similar degree.
Cardiovascular Risk and HAART
As has become typical at HIV conferences, many presentations were devoted to adverse events associated with antiretroviral therapy, especially metabolic side effects. One such side effect, elevated lipid levels, is a concern because it may increase the risk of cardiovascular disease. Based on the latest data from the large D:A:D study -- which includes more than 23,000 HIV positive subjects -- Jens Lundgren, MD, (abstract 62) reported that use of HAART appears to double the risk of myocardial infarction (MI; heart attack). The researchers reached this conclusion based on an analysis of data from various studies assessing arterial intima-media thickness (a marker for atherosclerosis, or "hardening of the arteries") or presence of cardiovascular disease. The increased risk associated with HAART was similar in magnitude to that seen in tobacco smokers. Other modifiable (e.g., elevated lipid levels, hypertension) and unmodifiable (e.g., age, sex, family history) risk factors also contributed to heart attack risk. In contrast to some previous research, neither lipodystrophy (body shape changes) nor degree of immunosuppression were linked to increased MI risk.
In a related study, Wafaa El-Sadr, MD, and colleagues (abstract 42) analyzed how MI risk changes over the course of anti-HIV treatment. Data from the same cohort (median age 39 years; 76% men) were collected through February 2004. During 76,577 PY of observation, 277 first MIs were reported. The MI incidence rate was 1.39 per 1,000 PY among those not exposed HAART, 2.53 per 1,000 PY among those on HAART for less than one year, and 6.07 per 1,000 PY among those on HAART for six or more years. The researchers calculated that for each additional year on HAART, the risk of MI increased by 1.17-fold. A similar association between antiretroviral therapy and MI risk was seen in both sexes and all age groups. Interestingly, the authors noted that elevated blood lipids explained "part but not all" of the association between HAART and increased MI risk.
Importantly, even though the MI risk doubled with HAART, the absolute number of heart attacks was still very small. Lundgren emphasized that the reduction in HIV-related illness and death attributable to the use of antiretroviral therapy greatly outweighs the small additional risk of MI.
Switching Drugs to Reduce Lipoatrophy
The ACTG 5116 study looked at the safety and efficacy of treatment simplification in individuals with well-controlled HIV. Margaret Fischl, MD, and colleagues (abstract 162) looked at 236 subjects who had viral loads below 200 copies/mL, had no phenotypic evidence of drug resistance (i.e., when their HIV was combined with anti-HIV drugs in the lab), and had been on three- or four-drug PI- or NNRTI-based first regimens for at least 18 months. Subjects were randomly assigned to switch to an NRTI-sparing regimen of lopinavir/ritonavir (Kaletra) plus efavirenz (Arm 1) or to substitute efavirenz if they were taking a PI while staying on their NRTIs (Arm 2); most subjects (78%) who were taking two NRTIs used AZT/3TC. After 110 weeks, in an intent-to-treat analysis, 66% of subjects in Arm 1 had viral loads below 50 copies/mL, compared with 74% in Arm 2. In addition, 17% of subjects in Arm 1 discontinued due to adverse events (mainly elevated triglyceride levels), compared with 5% in Arm 2.
Results of this study suggest that a PI-sparing regimen including NRTIs is superior to a regimen that excludes NRTIs, but the side effect profiles of the two regimens tell a somewhat different story. Lipoatrophy (loss of peripheral fat in the face and limbs) presents a particular concern for individuals taking NRTIs, especially d4T. Pablo Tebas, MD, (abstract 40) presented data from 62 subjects in the ACTG-5125s substudy of ACTG 5116. After 48 weeks, limb fat increased significantly in the NRTI-sparing arm, but decreased in those who stayed on NRTIs. Among a subset of 46 subjects followed for a median of 104 weeks, those in the NRTI-sparing arm continued to gain limb fat while those in the NRTI arm continued to lose it. However, triglyceride and total cholesterol levels increased much more in the NRTI-sparing arm. Tebas concluded that the benefits of NRTI-sparing regimens on body fat distribution must be balanced against their inferior virological potency. This study was not powered to detect differences among specific NRTIs (only about one-quarter were taking d4T).
Switching to an NRTI-sparing regimen can reduce lipodystrophy, but it may not be necessary to exclude all drugs in this class; studies suggest that omitting only thymidine analogs (d4T and AZT) may accomplish the same goal. Robert Murphy, MD, (abstract 45LB) reported results from ACTG 5110s. In this study, 101 subjects with lipoatrophy were randomly assigned either to replace a thymidine analog (AZT 24%; d4T 76%) with abacavir, or to switch to an NRTI-sparing regimen of lopinavir plus nevirapine; 77 subjects switched immediately, while 24 controls stayed on their baseline regimen for 24 weeks and then switched. After 24 weeks, based on computed tomography (CT) scans, subjects in both the thymidine-sparing and the all-NRTI-sparing arms experienced significantly increased subcutaneous abdominal fat; visceral abdominal fat decreased in the abacavir group. Thigh fat increased significantly (by 8%) in the all-NRTI-sparing arm but not in the abacavir arm. Subjects in all groups maintained good virological control, but CD4 cell counts increased more in the all-NRTI-sparing arm. Elevated lipids were seen in the lopinavir arm and hypersensitivity reactions were seen in the abacavir arm.
For individuals who wish to stay on an NRTI-containing regimen without thymidine analogs, is it better to switch to abacavir or tenofovir? In the British RAVE study, Graeme Moyle, MD, and colleagues (abstract 44L) randomly assigned 105 subjects with lipoatrophy to switch from a thymidine analog (AZT 33%; d4T 67%) to either abacavir or tenofovir. After 48 weeks, DEXA scans showed that limb fat increased significantly, and by similar amounts, in both the abacavir and tenofovir groups. Subjects in both arms maintained good virological suppression. Looking at lipid profiles, triglycerides, total cholesterol, and LDL "bad" cholesterol decreased more in the tenofovir arm. More subjects discontinued in the abacavir arm due to hypersensitivity reactions; no differences in renal function or bone density were detected. Subjects who switched from AZT showed improvements in anemia. The researchers concluded, "While both agents maintain virological suppression, [tenofovir] is associated with fewer treatment discontinuations and greater improvements in lipid parameters than [abacavir]."
Tenofovir also received support from two Spanish studies. In the LIPOTEST study (abstract 860), 53 subjects with well-controlled HIV and lipoatrophy switched from d4T to tenofovir; other drugs in their regimens did not change. After 18 months, facial fat thickness increased significantly and blood cholesterol decreased slightly. In addition, lactic acid levels decreased significantly and mitochondrial DNA (mtDNA) in peripheral blood mononuclear cells increased slightly; both elevated lactic acid and reduced mtDNA are signs of mitochondrial toxicity. Virological control of HIV was maintained and no serious adverse side effects were reported.
In a study by A. Milinkovic and colleagues (abstract 857), 56 subjects with well-controlled HIV were randomly assigned to continue on d4T, reduce their d4T dose from 40 to 30 mg twice daily, or substitute tenofovir for d4T; other drugs were not changed. After six months, limb fat increased significantly in the dose-reduction and tenofovir arms, but continued to decrease in the full-dose d4T arm. Triglyceride and total cholesterol levels decreased somewhat in the dose-reduction arm and significantly in the tenofovir arm, but increased in the full-dose d4T arm.
Taken together, these studies suggest that substituting tenofovir or abacavir for thymidine analogs can improve lipodystrophy without sacrificing virological control. New fixed-dose combination pills have made tenofovir (in Truvada) and abacavir (in Epzicom) as convenient as Combivir and Trizivir, both of which contain AZT. D4T is clearly the major contributor to lipoatrophy; for this reason, the drug was demoted from "preferred" to "alternative" status in a recent revision of the U.S. federal HIV treatment guidelines. However, AZT also appears to play a role in lipoatrophy, and further studies will show whether it deserves to meet the same fate.
New Anti-HIV Agents
Researchers at the Retrovirus conference presented new data on a plethora of experimental agents at different stages of the drug development process. Two presentations offered the latest data on Boehringer Ingelheim's new PI, tipranavir (Aptivus), which was granted accelerated approval by the Food and Drug Administration (FDA) in June (see
news item below). Another PI in the pipeline, Tibotec's TMC-114, generated perhaps the most excitement at the conference. This and other investigational agents further along in the development process are discussed in "
Drug Watch" in this issue.
Data were also presented for several agents further back in the development pipeline. One that generated some excitement was PA-457, the first HIV maturation inhibitor. PA-457 works by interfering with a protein needed to construct functional viral progeny; without it, new viral particles are incomplete and noninfectious. In a placebo-controlled single-dose study (abstract 159), 24 treatment-naive or treatment-experienced HIV positive men received PA-457 at doses of 75, 150, or 250 mg, or placebo. The 250 mg dose reduced HIV viral load by 0.51 logs and PA-457 appeared to suppress HIV for several days. A ten-day study is underway and a Phase II trial should start by the end of the year. (For an explanation of logs, see "About Logs" in this issue.)
Susan Little (abstract 161) presented the first results using Merck's integrase inhibitor, L-870810, in 30 treatment-naive and treatment-experienced HIV positive subjects. Ten-day monotherapy with 200 and 400 mg doses reduced HIV viral load by 1.73 and 1.77 logs, respectively; 20% and 38%, respectively, had viral loads below 400 copies/mL. Although L-870810 was well tolerated in this study and no major toxicities have been detected in humans, development has been halted due to toxicity in dogs; however, research on the related agent L-870812 is continuing. Data were also presented on BioAlliance's candidate integrase inhibitor, FZ41 (abstract 547). GW-873140, a CCR5 coreceptor blocker, was tested in a study of 31 HIV positive and eight HIV negative subjects (abstract 77). Good viral suppression was achieved after 10 days. Researchers also found that GW-873140 appears to continue to block receptors for as long as 10 days after discontinuation, suggesting that infrequent dosing may be possible. This agent is due to enter Phase III trials later this year.
In a seven-day dose-ranging study in 27 treatment-naive subjects (abstract 160), Tibotec's new NNRTI, TMC-278, appeared highly active against HIV that was resistant to approved NNRTIs. The median viral load decrease was about 1.2 logs, which was similar across doses (25, 50, 100, and 150 mg daily). No adverse events were detected and a Phase II trial is now starting (see "Open Clinical Trials" in this issue).
Other experimental agents to keep an eye on include Pfizer's CCR5 antagonist UK-427,857, now named maraviroc (abstracts 96, 663); Tibotec's nucleotide-competing reverse transcriptase inhibitor known as Compound-1 (abstract 156); Takeda's TAK-652, which replaces TAK-779 and TAK-220 (abstracts 541, 542); Boehringer Ingleheim's BILR-335BS (abstracts 557, 558); and NCI UIC-02031, a new nonpeptidic PI (abstract 562).
Unusual New York AIDS Case Remains an Enigma
On February 11 New York City Department of Health and Mental Hygiene officials announced that they had discovered a case of multidrug-resistant, apparently rapidly progressing HIV in a gay man in his forties who reported crystal methamphetamine use and multiple sexual partners. The man last tested HIV negative in May 2003, was believed to have been infected in October 2004 (though possibly earlier), and developed symptoms suggestive of acute retroviral syndrome (fatigue, fever, sore throat) in early November 2004. He then tested HIV positive with an unusual multidrug-resistant strain of the virus on December 16, by which time he showed signs of advanced immunosuppression. By January his CD4 cell count had fallen to 28 cells/mm
3.
The announcement set off a flurry of press coverage and discussion among medical professionals and HIV prevention workers. Some criticized New York officials for raising the alarm too soon based on a single case. In a rare move, organizers convened a special symposium to discuss the case at the February Retrovirus conference.
Genetic analysis showed that the man had a strain of subtype B HIV that had not been seen previously. Initial genotypic resistance testing indicated his HIV was resistant to the three major classes of antiretroviral drugs. Drug resistance in newly infected individuals is not uncommon, but resistance to all three classes in treatment-naive individuals is rare (see "Transmission of Drug-Resistant HIV," below). It was later determined that the man's virus was susceptible to efavirenz as well as T-20. The man's HIV strain was also found to be dual-tropic (able to use both CCR5 and CXCR4 coreceptors to enter cells) and syncytium-inducing (causing fusion of cells). Dual-tropic, syncytium-inducing virus, which is associated with rapid disease progression, is not unknown in newly infected individuals, although it is more often seen later in the course of disease. The unusual -- though not singularly unique -- aspect of the New York case is the co-occurrence of multidrug resistance and apparently rapid progression. Details of the genetic analysis were presented at the Retrovirus conference by David Ho, MD (abstract 973B) and published in the March 19, 2005 issue of The Lancet.
Experts emphasized that the New York man's rapid progression did not necessarily herald a "super strain" of HIV. It has long been known that a small proportion of individuals naturally progress rapidly to AIDS, and a variety of factors -- including genetic variations and methamphetamine use -- may have contributed to the early onset of immunosuppression in this case. Since February little new information has come to light. Despite extensive testing of more than a dozen of the New York man's sexual contacts (many of his 100 or more partners were anonymous) and retrospective analysis of stored blood samples, no similar strains have been detected; however, the investigation is ongoing. New York health officials issued a press release on March 29 stating that they were still examining data from several individuals who might be infected with related HIV strains.
Transmission of Drug-Resistant HIV: More or Less?
Concern about a growing incidence of drug-resistant (DR) HIV transmission is complicated by inconsistent data from recent small studies. According to Andrew Leigh Brown, PhD, of the University of Edinburgh (speaking at the special symposium at the Retrovirus conference; see previous item) several studies during the HAART era appear to show one of two trends: either an increase in transmission of DR HIV from 1996 to around 2000, followed by a stable or declining rate; or no increase in DR virus transmission. These outcomes might be explained by data indicating that DR HIV is less "fit," or transmissible, than wild-type (nonmutated) virus.
A picture of transmission patterns of DR HIV based on geography emerged at the 3rd European HIV Drug Resistance Workshop, held March 30 - April 1 in Athens. Reported rates of DR