Nearly 4,000 participants gathered February 8-11, 2004, for the 11th Conference on Retroviruses and Opportunistic Infections, the major annual U.S. scientific meeting on HIV. The conference opened with Stephen Lewis, United Nations Special Envoy for HIV/AIDS in Africa, calling on the developed world to devote more funding to AIDS relief efforts in poor countries. Some new epidemiological information was presented -- including a report of an unexpected HIV outbreak among college men in North Carolina -- but the Retrovirus conference was primarily devoted to basic science and treatment.
For the 2004 Retrovirus conference program and abstracts, see www.retroconference.org. For more complete conference coverage, see:
Nevirapine Resistance After a Single Dose
HIV can develop resistance to nevirapine (Viramune) after just one dose, which has important implications for the use of single-dose nevirapine monotherapy to prevent mother-to-child HIV transmission in resource-poor settings.
Gonzague Jourdain, M.D., from Harvard School of Public Health (abstract 41LB) reported results from a Thai study of more than 1,800 HIV-positive pregnant women who received AZT (zidovudine, Retrovir) with or without single-dose nevirapine. Women and infants who received the combined regimen had a vertical transmission rate of just 2%, comparable to rates seen in the U.S. and Europe. Following delivery, about 25% of the women began taking triple-drug regimens containing nevirapine. A random subset of 90 women received genotypic resistance tests 10 days after giving birth; resistance mutations (e.g., K103N, Y181C, G190A) were detectable in 18%. After six months of treatment, only 34% of women who had taken nevirapine monotherapy during delivery and had resistance mutations had viral loads below 50 copies/mL, compared with 53% of women who had taken intrapartum nevirapine but did not have resistance mutations, and 75% who did not receive nevirapine during delivery. In a second study of more than 600 HIV-positive mothers in South Africa, presented by Neil Martinson from Johannesburg (abstract 38), about 39% of HIV-positive mothers and about 42% of infants who were infected despite use of nevirapine developed resistance to the drug.
HIV easily develops cross-resistance to non-nucleoside reverse transcriptase inhibitors (NNRTIs), and nevirapine resistance can also limit future use of efavirenz (Sustiva) and possibly other NNRTIs yet to be developed. Some research suggests that nevirapine resistance may be short-lived. In the Thai study, mothers who started a nevirapine-based regimen more than six months after delivery responded better than those who started therapy sooner; however, in the South African study, nevirapine-resistant HIV persisted for at least nine months. Because the single-dose nevirapine regimen is inexpensive, convenient, and effective in reducing mother-to-child transmission, experts are not calling for an end to its use in areas with limited resources. "Single-drug therapy should not be withheld if no other alternative exists," said Elaine Abrams, M.D., from Columbia University in the session's concluding remarks. Wherever possible, however, HIV-positive pregnant women should receive combination antiretroviral therapy as appropriate for their viral load and CD4 cell count.
Atazanavir Still Looks Good at 48 Weeks
Following up on 24-week data presented at last summer's International AIDS Society meeting showing that atazanavir (Reyataz) boosted with ritonavir (Norvir) is nearly as effective as lopinavir/ritonavir (Kaletra), Margaret Johnson, M.D., from the Royal Free Hospital in London (abstract 547) reported that boosted atazanavir continued to suppress HIV at 48 weeks. The BMS 045 study included 358 treatment-experienced participants with resistant virus randomly assigned to receive boosted atazanavir, atazanavir plus saquinavir (Invirase or Fortovase), or lopinavir, along with tenofovir DF (Viread) plus a nucleoside reverse transcriptase inhibitor (NRTI). After 48 weeks, 46% of subjects receiving lopinavir had viral loads below 50 copies/mL, compared with 38% receiving boosted atazanavir and 26% receiving atazanavir/saquinavir. Participants in the boosted atazanavir and lopinavir/atazanavir arms had CD4 cell increases of about 11 5 cells/mm3, while the third arm saw of gain of 72 cells/mm3. While lopinavir appears slightly more effective, atazanavir is less likely to cause gastrointestinal problems or blood lipid (fat) abnormalities.
In related news, recent small studies have shown that atazanavir can be used to boost blood levels of amprenavir (Agenerase) and saquinavir -- a potential benefit for individuals who cannot tolerate ritonavir. Also, a case report series published in the April 9, 2004 issue of AIDS suggests that beyond causing fewer lipid abnormalities itself, atazanavir may help reverse dyslipidemia (blood lipid abnormalities) and lipodystrophy (body shape changes) associated with use of other protease inhibitors (PIs). Within 12 weeks of switching to atazanavir, two individuals experienced reduced dorsocervical fat pad ("buffalo hump") size and one experienced decreased waist size; lipid levels declined in all three, and the subjects maintained viral loads below 50 copies/mL. "We propose that in patients with lipodystrophy syndrome, switching to atazanavir from established [PIs] could lead to a reversal of the metabolic alterations and ... rapid regression of pre-existing body fat accumulations," the authors concluded.
Treatment During Acute Infection
There appears to be little benefit to starting antiretroviral therapy in the months immediately following HIV infection, according to a presentation by Bruce Walker, M.D., from Harvard Medical School (abstract 24). Walker presented final data from a small cohort of individuals who began therapy during acute, or primary HIV infection (PHI) followed by supervised treatment interruptions (STIs). Researchers have hypothesized that such interruptions might help spur the immune system to fight HIV. Earlier data from the study looked promising, with all eight initial subjects demonstrating continued viral suppression after one or two STIs. (Subjects were restarted on therapy if their viral loads stayed above 5,000 copies/mL for three weeks, or ever increased above 50,000 copies/mL.) With 14 subjects now enrolled and after an average of five years of follow-up, however, it appears that the ability to control HIV without treatment is short-lived. While 11 subjects maintained virological control for at least 90 days, that figure dropped to six subjects after one year, and to three subjects after three years. Most participants experienced gradual increases in viral load and declines in CD4 cell count. HIV-specific CD8 cell responses increased three-fold during the first STI, less during the second and third interruptions, and not at all after subsequent breaks; however, increased CD8 response did not correlate with improved virological control. In addition, two reports based on data from the French PRIMO study (abstracts 396 and 397) also failed to demonstrate benefits from early therapy and STIs. In an editorial review in the March 26, 2004 issue of AIDS, Walker and coauthors concluded, "Based on the currently published data, there is no clear evidence that patients with access to antiretroviral therapy have any greater clinical benefit if therapy is introduced immediately during or prior to their seroconversion illness. ... [T]here is currently no evidence from these studies to suggest that therapy during PHI results in a reduction in clinical progression compared with use of effective therapy in later disease."
Promising HIV/HCV Coinfection Data
Coinfection with HIV and hepatitis C virus (HCV) received considerable attention at the Retrovirus conference. Douglas Dieterich, M.D., from Mt. Sinai School of Medicine presented eagerly awaited results from the Roche APRICOT study of 868 coinfected subjects in 19 countries (abstract 112). Participants were randomly assigned to receive standard interferon plus ribavirin, Pegasys brand pegylated interferon plus placebo, or Pegasys plus ribavirin, all for 48 weeks. Most participants were white men on HAART with well-controlled HIV. Overall, 40% of the participants treated with Pegasys/ribavirin achieved a sustained virological response (SVR; undetectable HCV viral load at the end of a 24-week post-treatment follow-up period) -- the highest SVR rate yet seen in a coinfected population -- compared with just 12% of those receiving standard interferon/ribavirin. Among those with genotype 1 HCV (which is harder to treat), the corresponding SVR rates were 29% and 7%; among those with genotypes 2 or 3, the SVR rates were 62% and 20%, respectively.
At the same session, Raymond Chung, M.D., from Massachusetts General Hospital (MGH) presented final results from ACTG A5071 (abstract 110). In this trial, 133 participants received either standard interferon or Pegasys for 48 weeks; both groups also received daily ribavirin, starting with lower than normal doses to reduce side effects. Most participants were men and about half were African American. After 72 weeks (48 weeks of therapy plus 24 weeks of follow-up), 27% in the Pegasys/ribavirin arm and 12% in the standard interferon/ribavirin arm had undetectable HCV viral load. Among those with genotype 1, the corresponding SVR rates were 14% and 6%; among those with genotypes 2 or 3, the SVR rates were 73% and 33%, respectively. Notably, while the end-of-treatment and SVR rates were similar in the standard interferon arm, the response rate declined dramatically from week 48 to week 72 in the Pegasys arm. Chung suggested that the lower initial dose of ribavirin may have contributed to the higher relapse rate.
Finally, Christian Perronne, M.D., presented results from the French RIBAVIC trial (abstract 117LB). This study compared Peg-Intron brand pegylated interferon to standard interferon, both with ribavirin. Most participants were men and 40% had advanced liver disease. Overall, 27% of participants receiving Peg-Intron/ribavirin had sustained undetectable HCV viral load after 72 weeks, compared with 19% of those taking standard interferon/ribavirin. Among those with genotype 1 HCV, the SVR rates were 15% and 5%; in those with genotypes 2 or 3, the corresponding rates were 45% and 40%. About 40% of subjects in both groups stopped treatment prematurely, and about 30% experienced severe side effects.
It is unclear why the SVR rates were so much higher in APRICOT compared with the other two trials, but there were some important differences in the study populations. Subjects in all three trials had well controlled HIV with median CD4 cell counts of 400-500 cells/mm3; at least 80% were on HAART. However, ACTG 5071 included more African Americans, a group that responds less well to interferon therapy. RIBAVIC included more participants with advanced liver disease, another "hard to treat" population, and drop-out rates were higher than in the other two trials.
Efavirenz Side Effects in People of Color
Data from substudy A5097s of the ACTG 5095 trial presented by Heather Ribaudo of Harvard School of Public Health (abstract 132) revealed that people of color clear efavirenz more slowly than whites. The substudy included about 200 participants (53% white, 32% black, 12% Hispanic); about 80% were men. Black and Hispanic subjects cleared the drug about 30% slower than whites. Similarly, Stephen Taylor, M.D., from the University of Birmingham in the U.K. (abstract 131) presented data from the STOP study showing that efavirenz reaches higher concentrations and persists longer after drug discontinuation in black women compared with white men.
Helping to explain this effect, David Haas of Vanderbilt University (abstract 133) reported data from another ACTG 5095 substudy showing that a gene variant seen most often in blacks is associated with slower efavirenz clearance and thus higher drug concentrations in the body. The gene controls expression of the CYP2B6 enzyme in the liver, which plays a role in drug processing. Each individual has one of three genotypes: T/T, G/T, or G/G. The T/T genotype (indicating two copies of the variant gene) was seen in 20% of blacks and just 3% of whites. Efavirenz concentrations were nearly three times as high in people with the T/T genotype, and slightly higher in those with the G/T combination, than in those with the G/G pairing. In this study, the T/T genotype was associated with adverse central nervous system side effects (e.g., bizarre dreams, depression). In related news, Lucia Gallego and colleagues from Madrid reported in the February 1, 2004 issue of Clinical Infectious Diseases that higher blood concentrations of efavirenz are associated with sleep disturbances such as insomnia and waking during the night. Together, these studies suggest that therapeutic drug monitoring may help physicians determine appropriate individualized levels of efavirenz.
Metabolic side effects associated with HAART remain a major cause for concern, but studies of long-term complications continue to yield frustratingly inconsistent data. On the cardiovascular front, the latest analysis of data from the D:A:D (Data Collection on Adverse Events of Anti-HIV Drugs) trial, collected from more than 23,000 HIV-positive individuals, revealed a cardiovascular event rate of 5.5 per 1,000 person years (PY), including 121 heart attacks and 30 strokes (abstract 737). The heart attack rate among subjects on HAART was only slightly higher than that seen in the Framingham Heart Study, a long-term study of cardiovascular risk factors in the HIV-negative population. However, risk increased with longer duration of antiretroviral therapy. In the D:A:D study, HAART was not associated with hypertension (high blood pressure) (abstract 75), but in the Women's Interagency HIV Study (WIHS) use of HAART -- and increased duration of therapy -- were independently associated with new-onset hypertension (abstract 741). Uchenna Iloeje, M.D., from Bristol-Myers Squibb (abstract 736) reported that in a subset of the HIV Outpatient Study (HOPS) cohort, PI use was associated with an increased risk of cardiovascular disease (9.8 per 1,000 PY in the PI group vs. 6.5 per 1,000 PY in the non-PI group), along with traditional risk factors such as older age, tobacco use, hypertension, and hyperlipidemia (elevated blood lipids). With PIs looking problematic, it is reassuring that T-20 (enfuvirtide, Fuzeon) was not associated with metabolic abnormalities or body fat changes after 48 weeks (abstract 715); this is not surprising, since T-20 works by a different mechanism than other antiretroviral drugs and does not interfere as much with normal cellular functioning.
Todd Brown, M.D., from Johns Hopkins (abstract 73) presented data from the ongoing prospective MACS study showing that HIV-positive men using HAART had higher rates of hyperglycemia (high blood sugar; fasting glucose 110 mg/dL or higher) and diabetes (fasting glucose 126 mg/dL or higher) compared with HIV-negative men. The study included 5,622 gay or bisexual men, about 85% white. Overall, the risk of prevalent (existing) and incident (new onset) fasting hyperglycemia was 2-3 times greater, and the risk of prevalent and incident diabetes was 4-5 times greater, in HIV-positive men on HAART compared with HIV-negative men. In an incidence analysis of 765 subjects (after excluding those with hyperglycemia at study entry), about 19% of HIV-positive men on HAART developed new-onset hyperglycemia, compared with about 9% of HIV-positive men not on HAART, and about 11% of HIV-negative men; the corresponding incidence rates for frank diabetes were about 11%, 5%, and 3%. Use of any PI, d4T (stavudine, Zerit), or efavirenz was associated with an increased risk of hyperglycemia. In addition, men with lower nadir (lowest ever) CD4 cell counts were more likely to develop blood glucose abnormalities.
Deaths Due to Non-AIDS-Defining Illnesses
With HAART causing dramatically lower rates of opportunistic illnesses (OIs), an increasing proportion of deaths in people with HIV are now due to non-AIDS-defining illnesses. Frank Palella, M.D., of Northwestern University (abstract 872) presented the latest analysis of data from the HOPS cohort, showing that while the rate of death due to OIs fell from 23 per 100 PY in 1996 to 6 per 100 PY in 2002, mortality due to nonopportunistic causes rose during the same period by 45% among individuals on HAART for two years, and by 70% among those on HAART for seven years. "If someone takes [antiretroviral therapy], they will live longer and when death occurs, it will not be due to an AIDS-related condition," Palella concluded.
One area where this shift is evident is the increased rate of nonopportunistic cancers in people with HIV. AIDS-defining malignancies such as Kaposi's sarcoma (KS) and invasive cervical cancer have declined in the HAART era. For example, in the May 10, 2004 online edition of Cancer, researchers with the EuroSIDA trial reported that in a study of nearly 10,000 HIV-positive subjects, the rate of KS declined 39% from 1994 to 2003; the decrease was especially marked in participants with higher CD4 cell counts and those with a longer duration of HAART use. But rates of some other types of cancer are on the rise. Based on a retrospective analysis of data from more than 12,000 HOPS participants collected between 1992 and 2000, Pragna Patel, Ph.D., from the Centers for Disease Control and Prevention (CDC) (abstract 81) found that rates of lung, head/neck, and anorectal cancer, Hodgkin's lymphoma, and malignant melanoma were higher in people with HIV than in the general population. The incidence of other common cancers (e.g., breast, colon, prostate) was not significantly different in the HIV-positive and HIV-negative populations. The likelihood of developing a nonopportunistic cancer was correlated with nadir CD4 cell count. This suggests that immune suppression plays a role in the development of cancer, for example by allowing oncogenic (cancer-causing) viruses such as human papillomavirus (HPV) and Epstein-Barr virus (EBV) to proliferate.
Back to the Drawing Board for HIV Vaccines
After several HIV vaccine studies have shown little or no benefit, Ronald Desrosiers, M.D., from Harvard Medical School (presentation 109) suggested that it might be time to go back to the laboratory to learn more about how the immune system responds to the virus. According to Desrosiers, none of the dozen or so current HIV vaccine candidates have much likelihood of success, due to the present "inability to solve some fundamental scientific questions." At the same session, Dennis Burton, M.D., from the Scripps Research Institute (presentation 108) said that the standard vaccine strategy of mimicking the human immune system -- for example, encouraging heightened immune cell activity and/or antibody production -- is not the best approach, since the immune system itself is not very good at controlling HIV. Desrosiers suggested that more research on basic immunology is indicated before spending more time and money on large human trials of current vaccine candidates.
Superinfection Rate 5%
In a small study of men who have sex with men in San Diego and Los Angeles, researchers observed an HIV superinfection rate of 5% per year, higher than previously assumed. (Superinfection refers to subsequent infection with a new strain of the virus in a person who is already HIV positive.) Davey Smith, M.D., from the University of California at San Diego (UCSD) (abstract 21) detected three cases of superinfection out of a total of 78 men by matching HIV pol gene sequences. All three men were exposed to their second strain through sexual activity. None were on antiretroviral therapy; two men were initially infected with drug-resistant strains and then superinfected with a wild-type (nonmutated) strain, while the third was superinfected with drug-resistant virus. Studies have shown that infection with more than one strain of HIV can lead to more rapid disease progression; indeed, in this study the men experienced viral load increases and CD4 cell decreases after becoming superinfected. The results suggest that even two individuals who are already HIV positive should consider safer-sex precautions.
More Conference News in Brief
Michael Kozal, M.D., from Yale University (abstract 35LB) reported that a small proportion of HIV-positive men who have sex with men (about 7% in his study) may be responsible for a large proportion of sexual transmissions. Peter Chin-Hong, M.D., from the University of California at San Francisco (UCSF) (abstract 845) reported that individuals with drug-resistant HIV were at least as likely as those with susceptible virus to have unprotected sex; factors associated with unprotected sex included younger age, less education, use of sildenafil (Viagra), and depression. John Mellors, M.D., from the University of Pittsburgh (abstract 39) reported that low-level resistance due to "minority variants" (those that comprise less than 25% or so of the HIV in the body) may not be detected by standard genotypic resistance tests. Finally, Cheryl Jay, M.D., from UCSF (abstract 496) reported results from a small study showing that smoked marijuana relieves pain due to peripheral neuropathy, a possible side effect of certain NRTIs. After smoking three marijuana cigarettes per day for seven days, 10 out of 16 subjects reported that their average daily neuropathy pain decreased by 30% or more.
A Food and Drug Administration (FDA) advisory group voted unanimously on March 25 to recommend conditional approval of a new treatment for facial wasting (lipoatrophy) in people with HIV. Sculptra, manufactured by the French company Aventis and marketed by Dermik Laboratories, is an injected substance (polylactic acid) that causes the body to produce collagen (a fibrous protein) to fill in areas of lost fat. Results from studies to date (mostly in white men) have been impressive, with most treated individuals reporting a high degree of satisfaction. The therapy is not permanent, however, and repeated injections may be needed. Common side effects include pain at the injection site, temporary bruising, swelling, and small nodules (lumps) under the skin. To reduce the risk of adverse outcomes, Sculptra should be administered only by trained practitioners. Polylactic acid -- under the name New-Fill -- has been approved in Europe since 1999 for cosmetic use. The pending U.S. approval would be only for HIV-related facial fat loss; advisory panel members expressed concern that once approved, Sculptra could be used off-label for other cosmetic purposes such as reducing wrinkles.
Also in late March the FDA approved the first rapid oral HIV antibody test. The new version of the OraQuick test, manufactured by OraSure Technologies of Bethlehem, PA, uses a sample of oral fluid taken with a swab from around the gums. Results can be read in about 20 minutes. The rapid OraQuick test was previously approved only for use with blood samples. The new test is more than 99% accurate; a positive result should be confirmed with a Western blot test. The rapid oral test will allow individuals to receive their results during a single session. When using older antibody tests, many people never returned to obtain their results, which typically took about two weeks. The test, expected to cost $12 to $15, will be available at some 40,000 approved medical laboratories, and federal officials have approved a waiver to use the test in other settings such as doctors offices, community clinics, and mobile vans. The rapid oral HIV test is not approved for home use.
In welcome news for the 30-40% of HIV-positive people coinfected with HCV, the FDA in early April approved two generic versions of ribavirin, an antiretroviral drug used in combination with standard or pegylated interferon to treat chronic hepatitis C. The generic drugs will be marketed by Sandoz (a subsidiary of Novartis) and Three Rivers Pharmaceuticals in partnership with Pharmaceutical Resources, Inc. Patient advocates have long called for the approval of generic ribavirin, which has been held up for years by patent lawsuits. But many were disappointed at the announced cost of the new versions. Generic drugs are typically priced at a fraction of the cost of their brand-name equivalents. Both Sandoz and Three Rivers, however, set prices for their generic ribavirin at about $10 per capsule -- in between the average wholesale prices for Schering-Plough's Rebetol (about $11) and Roche's Copegus (about $6). Schering countered by announcing its own generic version of ribavirin, priced to undercut the two newcomers. The first manufacturer(s) of new generic drugs receive exclusive marketing rights for six months, but more companies are expected to enter the generic ribavirin market by the end of the year, thus driving down prices.
As mentioned briefly in the Winter 2004 edition of BETA, Roche and Trimeris announced in January that they have halted clinical trials of T-1249, a fusion inhibitor that was touted as a more potent second-generation successor to T-20. Like T-20, T-1249 must be administered by injection. The two companies said they remain committed to developing new fusion inhibitors that are more effective and easier to administer -- a process expected to take years, according David Reddy, Ph.D., Roche's head of HIV research. T-1249 (again like T-20) is a peptide that has proven difficult to manufacture; Roche said it was not confident it could produce T-1249 on a large scale. Advocates speculate that Roche abandoned T-1249 in part because sales of T-20 have not met expectations, largely due to its inconvenient twice-daily administration and exceedingly high price (about $20,000 per year). In related news, Roche announced that as of late April, T- 20 would be available through retail and specialty pharmacies. Previously, due to supply limitations, the drug had to be ordered through a single mail-order distributor. Roche also said it plans to launch a nursing support program to assist individuals in preparing and administering the drug. Finally, a new set of international consensus guidelines for the use of T-20 were published in the May 21, 2004 issue of AIDS. The consensus panel said that successful treatment is most likely if T-20 is started when the CD4 cell count is above 100 cells/mm3 and is used as part of a third or fourth regimen in conjunction with one or two other drugs to which HIV remains sensitive. But T-20 may also benefit heavily treatment-experienced individuals, including those taking few or no other active drugs.
In late January Boehringer Ingelheim issued a "Dear Doctor" letter warning of the risk of hepatotoxicity (liver toxicity) associated with the use of nevirapine (Viramune). "Severe, life-threatening, and in some cases fatal hepatotoxicity, including fulminant and cholestatic hepatitis, hepatic necrosis [tissue death] and hepatic failure, has been reported in patients treated with Viramune," reads the company's revised package insert. Studies have shown that the risk of liver damage is about three times higher in women (including pregnant women) than in men. The risk of hepatotoxicity is especially high in women with more than 250 CD4 cells/mm3 and men with more than 400 cells/mm3. People with existing elevated liver enzyme (ALT and AST) levels and those coinfected with hepatitis B or C are also at higher risk. Although the absolute risk of severe hepatotoxicity is small, the insert states that "patients with signs or symptoms of hepatitis must discontinue Viramune and seek medical evaluation immediately"; such symptoms may include fatigue, loss of appetite, nausea, abdominal pain, and jaundice. "It is essential that patients be monitored intensively during the first 18 weeks of therapy with Viramune to detect potentially life-threatening hepatotoxicity or skin reactions," the company advises. If these adverse events occur, the drug should not be restarted. Complete nevirapine prescribing information is available at www.viramune.com.
On March 23 the U.S. Department of Health and Human Services (DHHS) again updated its guidelines for the treatment of adults and adolescents with HIV. For treatment-naive individuals, the newly approved protease inhibitor fosamprenavir (Lexiva) -- with or without ritonavir -- has been added as a component of alternative PI-based regimens, while the older, less potent amprenavir has been removed. Unboosted indinavir (Crixivan) has also been removed as a component of initial regimens. Abacavir (Ziagen) plus 3TC (lamivudine, Epivir) has been added as an alternative double-NRTI backbone. In terms of safety, new information has been added regarding nevirapine-associated hepatotoxicity (see "New Nevirapine Warning," above), as well as a new table of dosing recommendations for individuals with liver or kidney dysfunction. The federal government's guidelines for the use of antiretroviral agents in children with HIV were also recently updated to include information on fosamprenavir. The revised adult and pediatric guidelines are available at www.aidsinfo.nih.gov/guidelines.
The DHHS guidelines list a variety of preferred and alternative antiretroviral regimens for first-line therapy, but recent reports suggest that one initial three-drug combination is superior to its competitors.
Two articles in the December 11, 2003 issue of the New England Journal of Medicine (NEJM) reported data from ACTG study 384, the most extensive head-to-head comparison of consecutive regimens to date. The study included 980 treatment-naive participants in the U.S. and Italy (approximately 80% men and 45% white) randomized to receive one of six different three- or four-drug combinations:
Subjects in the three-drug arms were switched to a new regimen if they experienced virological failure or discontinued due to toxicity; the primary study endpoint was length of time until failure of the second regimen. Participants were followed for a median of 2.3 years.
Gregory Robbins, M.D., of Harvard Medical School and colleagues found that among the 620 subjects started on a three-drug regimen, initial use of AZT/3TC/efavirenz produced the most durable HIV suppression. At the end of follow-up, 90% of subjects who started with this regimen still had undetectable viral loads, compared with 60-70% of subjects who started with one of the other three-drug regimens. Robert Shafer, M.D., of Stanford University Medical Center and colleagues compared the same 620 subjects with an additional 360 participants who received both efavirenz and nelfinavir plus either AZT/3TC or d4T/ddI. The four-drug regimens suppressed HIV longer than any of the three-drug regimens except AZT/3TC/efavirenz, which was comparable. As has come to be expected, the d4T/ddI backbone was associated with higher rates of adverse side effects than AZT/3TC. In an editorial in the same issue, Paul Skolnik, M.D., of Boston University said that the results do not show that NNRTI-based regimens are necessarily better than PI-based regimens for first-line therapy, but starting with AZT/3TC/efavirenz allows the generally more potent PI class to be saved for future use.
While physicians and people with HIV are eager to adopt simpler regimens, there are risks in making things too simple. In the April 29, 2004 issue of NEJM Roy Gulick, M.D., and colleagues from the ACTG A5095 study team reported that the triple-NRTI regimen of AZT, 3TC, and abacavir (the three drugs in the Trizivir combination pill) was inferior to regimens containing AZT, 3TC, and efavirenz, with or without abacavir. In this study of 1,147 subjects, 21% of those taking AZT/3TC/abacavir and 11% of those taking the efavirenz-based regimens experienced virological failure after a median follow-up of 32 weeks. (The difference was so great that the triple-NRTI arm was suspended following an interim review of the data.) This trial and others suggest that once-daily NRTI-only regimens may not be sufficiently potent to keep HIV under control for most people (see "News Briefs," BETA, Winter 2004).
At the February Retrovirus conference, however, Richard Elion, M.D., from George Washington University (abstract 53) presented results indicating that a once-daily, four-NRTI regimen consisting of Trizivir plus tenofovir can effectively suppress HIV in treatment-naive individuals. (Unlike the other three drugs in this regimen, AZT has not been shown to be effective when used once daily.) This interim analysis from study COL40263 included data from 88 subjects who had taken the four-NRTI regimen for at least eight weeks. At 24 weeks 67% achieved viral loads below 50 copies/mL. While this rate is higher than those seen with triple-NRTI regimens, it is lower than those achieved with regimens that include the best available NNRTIs or PIs.
On May 16 DHHS Secretary Tommy Thompson announced a new FDA program that will allow rapid approval of fixed-dose combination (FDC) and copackaged anti-HIV medications for use in developing countries under the Presidential Emergency Plan for AIDS Relief (PEPFAR). The new process will enable companies to submit existing data from past studies and peer-reviewed literature, rather than conducting new trials, and will waive the usual $500,000 application fee. Applicants will be required to show that the drugs in an FDC retain their bioavailability and do not interact in a detrimental manner. Thompson suggested that the expedited process could take as little as 2-6 weeks.
Some advocates contend that the new FDA process is wasteful duplication of effort, since the World Health Organization (WHO) already prequalifies FDCs for HIV therapy. The new process applies to both brand name and generic manufacturers; however, FDCs containing generic components could not be distributed in the U.S. or other developed countries due to patent laws. The WHO has qualified generic d4T/3TC/nevirapine combination pills produced by the Indian companies Ranbaxy and Cipla, as well as versions of d4T/3TC and AZT/3TC. (However, Cipla's generic 3TC and AZT/3TC were removed from the WHO's approved products list in late May due to poor documentation at an independent laboratory used by Cipla.) The only combination pills currently available in the U.S., Combivir (AZT/3TC) and Trizivir, contain drugs patented by the same company (GlaxoSmithKline).
No one yet makes a fixed-dose pill containing AZT/3TC/efavirenz. But following the administration's recent announcement, Bristol-Myers Squibb, Gilead Sciences, and Merck & Co. announced that they plan to seek approval for an FDC comprised of efavirenz, tenofovir, and emtricitabine (FTC, Emtriva; an NRTI similar to 3TC). Gilead also announced that the FDA granted priority review for its two-drug tenofovir/emtricitabine combination pill, which the company says could be approved by September. In addition, GlaxoSmithKline and Boehringer Ingelheim said they were negotiating a copackaged antiretroviral combination, most likely nevirapine plus Combivir.
Much recent HIV research has focused on cardiovascular complications in individuals receiving antiretroviral therapy. According to a report in the January issue of Stroke, people with AIDS are more likely than the general population to have strokes (cerebrovascular accidents) at young ages. John Cole, M.D., of the University of Maryland at Baltimore and colleagues examined the medical records of 557 individuals aged 15-44 who had strokes between 1988 and 1991; 386 had ischemic strokes (blocked blood flow in the brain) and 171 had intracerebral hemorrhages (bleeding in the brain). Twelve stroke patients (2.2%), six with each type, were found to have an AIDS diagnosis. After adjusting for age, sex, and race, the relative risk of stroke was 17.8 times greater for people with AIDS. This study, based on events that occurred prior to the advent of HAART, suggests that HIV itself appears to increase the risk of stroke (e.g., by promoting clot formation or causing blood vessel damage), independent of the impact of antiretroviral therapy.
Another study, reported in the January 2004 issue of American Heart Journal, found that HIV-positive individuals are more likely than their HIV-negative counterparts to experience myocardial infarctions (MIs, heart attacks), and at younger ages. In a study of 690 HIV-positive subjects (mostly men), Philip Varriale, M.D., and colleagues from Cabrini Medical Center found that 29 individuals (about 4%) were diagnosed as having had an acute MI; 22 of these (about 76%) were younger than 55 years of age and more than half were taking PIs. Among the MI patients, more than 75% of those below age 55 and about 70% of those above this age had either one or no cardiovascular risk factors, leading the authors to suggest that HIV itself, rather than metabolic complications associated with HAART, contributed to coronary artery disease. The authors hypothesized that HIV, by causing endothelial (blood vessel lining) injury, may "initiate the inflammatory process of early atherosclerosis," eventually resulting in MI as coronary arteries progressively narrow and deprive the heart of oxygen.
Priscilla Hue, M.D., and colleagues from UCSF reported in the April 2004 issue of Circulation that people with HIV have a higher risk of atherosclerosis (hardening and clogging of the arteries), associated with classic cardiovascular risk factors such as older age, elevated cholesterol levels, tobacco use, and high blood pressure. This study included 148 HIV-positive individuals (average age 45) treated with PI-based HAART for a median 3.3 years, and 63 HIV-negative control subjects. Using ultrasonography, the researchers determined that HIV-positive subjects had increased thickness of the carotid artery intima-media (inner and middle layers), and that the rate of thickening was more rapid compared with control subjects. Arterial plaque buildup was seen in 45% of HIV-positive participants vs. 24% of uninfected controls. The researchers also found that atherosclerosis was worse in subjects with the lowest nadir CD4 cell counts, suggesting that HIV itself, or greater immune suppression, has a deleterious effect on blood vessels. In another ultrasound study, published in the April 30, 2004 issue of AIDS, Paolo Maggi and colleagues with the Italian PREVALEAT Group found that about 52% of HIV-positive subjects treated with PIs had evidence of atherosclerotic lesions in their carotid arteries, compared with about 15% of PI-naive subjects taking NNRTIs, and about 14% of those not on antiretroviral therapy or receiving only NRTIs. The authors suggested that "a periodic ultrasonographic study of the vascular wall should be included in the follow-up of HIV-infected patients."
Physicians are increasingly prescribing adjunct medications to control metabolic manifestations such as hyperlipidemia and diabetes that may increase cardiovascular risk in HIV-positive individuals on HAART. At the Retrovirus conference, James Sosman, M.D., from the University of Wisconsin (abstract 77) reported results from a placebo-controlled study showing that pravastatin (Pravachol) decreased levels of LDL "bad" cholesterol and improved endothelial function. In terms of body fat changes, Donald Kotler, M.D., of the STARS Trial Investigator Group (abstract 80) reported that low maintenance doses of recombinant human growth hormone kept off abdominal fat after initial higher-dose therapy (see "HIV and Hormones" in this issue).
Studies to date have produced inconsistent data concerning the benefits of rosiglitazone (Avandia), a drug used to treat diabetes. Andrew Carr, M.D., and colleagues with the ROSEY Study Group compared rosiglitazone with placebo in 180 HIV-positive, HAART-treated subjects (98% men) with fat wasting in the limbs (lipoatrophy), but without diabetes. The results were presented at the Retrovirus conference (abstract 79) and published in the February 7, 2004 issue of The Lancet. After 48 weeks, no differences in peripheral fat gain or other body composition parameters were seen in the two arms. Limb fat did increase somewhat in the rosiglitazone arm, but also increased to a similar degree in the placebo group. Rosiglitazone was associated with elevated total cholesterol, LDL cholesterol, and triglyceride levels (not usually seen in HIV-negative diabetics using the drug), improved fasting insulin levels (even though no subjects in this study were diabetic), and decreased levels of the liver enzyme alanine transaminase (ALT) -- likely because the drug improved hepatic steatosis (fatty liver). "Rosiglitazone cannot be recommended for the treatment of HIV lipoatrophy in adults receiving antiretroviral therapy," the researchers concluded, "even though it has insulin-sensitizing effects in this population."
But a small study published in the May 18, 2004 issue of the Annals of Internal Medicine suggests that rosiglitazone may help improve metabolic abnormalities associated with HAART. Colleen Hadigan, M.D., and colleagues from MGH and Harvard Medical School treated 27 HIV-positive individuals with lipoatrophy and elevated insulin levels (an indicator of insulin resistance) with rosiglitazone or placebo. After three months, subjects receiving rosiglitazone showed improved insulin sensitivity, decreased fatty acid levels, increased adiponection (a hormone produced by fat cells), and increased subcutaneous leg fat. However, they also had increased levels of total and LDL cholesterol. Given the inconsistent results of these two studies, more research is needed to determine whether rosiglitazone can help reduce cardiovascular risk factors in individuals receiving antiretroviral therapy.
Finally, in the February 20, 2004 issue of AIDS, Susan Driscoll, NP, and colleagues from MGH reported that a combination of exercise and use of the antidiabetes drug metformin (Glucophage) significantly improved cardiovascular risk factors in individuals taking HAART. In this study, 37 participants with lipodystrophy and insulin resistance were randomly assigned to receive either metformin alone or metformin in conjunction with a program of thrice-weekly aerobic exercise and weight training for 12 weeks. Subjects in the metformin plus exercise arm experienced greater benefit than those receiving metformin alone in the following areas: decreased blood pressure, reduced abdominal fat, improved fasting insulin levels, increased muscle mass, and improved exercise capacity; improvements in lipid profile, however, were similar in the two groups.
Although the rate of HIV-associated dementia or HIV-related cognitive-motor complex (formerly called AIDS dementia complex) has decreased markedly since the advent of HAART, cognitive impairment still appears to be more common among HIV-positive individuals over age 50 than among HIV-negative individuals in the same age bracket. Cognitive impairment in older HIV-positive people was addressed in several reports published in a January 2004 AIDS supplement devoted to HIV and aging.
Mariana Cherner, Ph.D., from UCSD and colleagues reported that HIV-positive individuals over age 50 were more likely to have neuropsychological impairment than those under age 35 (64% vs. 54%), even though the older group had lower viral load levels in the blood and cerebrospinal fluid (CSF) and were more likely to be taking HAART. Victor Valcour, M.D., and colleagues with the NeuroAIDS Specialized Neuroscience Research Program reported preliminary data from an ongoing study of cognitive ability in older HIV-positive people. Among the first 47 enrolled subjects over age 50, 56% retained normal cognitive function, compared with 88% of the first 32 subjects aged 20-40. The researchers suggested that cognitive impairment in older HIV-positive individuals may be attributable to a synergy between HIV-associated dementia and other types of dementia such as Alzheimer's disease, or may be due to vascular (blood vessel) pathology. In another study, James Becker, Ph.D., and colleagues from the University of Pittsburgh found that alcohol or drug use or dependence and higher HIV viral load were predictive of cognitive disorders, while a higher level of education appeared to be protective.
Looking at mechanisms underlying cognitive problems in 46 HIV-positive, treatment-naive subjects and 58 HIV-negative subjects, Thomas Ernst, Ph.D., and Linda Chang, M.D., from the University of California at Los Angeles found that HIV-positive individuals showed notable differences in brain anatomy and chemistry. "In the basal ganglia, HIV infection appeared to induce neuronal damage or loss beyond that observed in normal aging," the authors concluded. "In the frontal white matter, HIV infection seemed to exacerbate glial [brain support cell] activation beyond that observed in normal aging." In related research, Yan Xu from Thomas Jefferson University and colleagues reported in the April 21, 2004 issue of the Proceedings of the National Academy of Sciences that certain proteins (e.g., gp120) produced by HIV appear to accelerate the death of central nervous system (brain and spinal cord) neurons, possibly leading to HIV-related encephalopathy. This research team previously showed that HIV was associated with neuronal apoptosis (programmed cell death). In this laboratory study, they introduced HIV-infected T cells and macrophages, or immune cells from which virus had been removed, to neuron cell cultures. HIV-infected T cells induced neuronal apoptosis, but once the virus was removed, these immune cells themselves did not have a deleterious effect. Macrophages induced apoptosis whether or not they contained HIV, but the infected macrophages were significantly more neurotoxic. The authors concluded that HIV and its proteins have a direct neurotoxic effect, in addition to triggering immune cells to produce cytokines that kill neurons.
Finally, in the April 15, 2004 issue of the Journal of AIDS (JAIDS), Kevin Robertson, Ph.D., from the University of North Carolina at Chapel Hill and colleagues reported that HAART improved neurological functioning in people with HIV even though many antiretroviral drugs have a limited ability to cross the blood-brain barrier. In this prospective longitudinal study, 48 HIV-positive subjects received neurological and neuropsychological examinations immediately before and six months after starting HAART. Viral load decreased in both the blood and CSF after the initiation of antiretroviral therapy, and neurological and neuropsychological functioning improved significantly. "[D]espite the poor central nervous system penetration of most of these agents," the authors concluded, "there is satisfactory short-term improvement in both central nervous system viral burden and nervous system function with HAART."
The latest results from the WIHS cohort indicate that many women with advanced HIV disease benefit from HAART despite having very high viral loads or very low CD4 cell counts before starting therapy. In the present analysis, which included data from 1,132 HIV-positive women followed for a median of about four years, Kathryn Anastos, M.D., of Montefiore Medical Center and colleagues examined how well viral load levels and CD4 cell counts before and after starting HAART predict eventual clinical outcome; results were reported in the February 17, 2004 issue of the Annals of Internal Medicine. The researchers found that women whose CD4 cell counts remained below 200 cells/mm3 after starting HAART were more than twice as likely to die from any cause, and nearly 50 times more likely to die of AIDS-defining illnesses, compared with women whose CD4 cell counts rose above 350 cells/mm3 with treatment. Likewise, women with HIV viral loads above 10,000 copies/mL while on HAART were more than three times as likely to die of any cause than women with viral loads below 80 copies/mL. For women with CD4 cell counts between 200 and 350 cells/mm3 and/or viral loads between 80 and 10,000 copies/mL after starting HAART, death rates were similar to those seen in the higher CD4 cell bracket and the lower viral load bracket. This study indicates that women's viral load levels and CD4 cell counts after starting HAART are more accurate predictors of HIV disease progression than pretreatment measurements. "Pre-HAART CD4 cell count and HIV-1 RNA level were not predictive of clinical outcomes if adjusted for values attained after HAART initiation," the authors concluded. "[E]ven advanced immune suppression can be overcome with HAART" if therapy increases the CD4 cell count above 200 cells/mm3 and pushes viral load below 10,000 copies/mL.
Sexually transmitted diseases (STDs) other than HIV are an increasing concern in the gay community, with rates of infections such as syphilis, chlamydia, and gonorrhea increasing among gay and bisexual men -- especially young men -- in several U.S. cities. While many STDs can be treated, drug-resistant pathogens are a growing problem. In April the CDC recommended a change in the standard treatment for gonorrhea in men who have sex with men; the new guidelines were published in the April 30, 2004 issue of Morbidity and Mortality Weekly Report. According to the agency, fluoroquinolone antibiotics such as ciprofloxacin (Cipro) should no longer be used as first-line treatment for gay and bisexual men due to the rise of drug-resistant Neisseria gonorrhoeae. Drug-resistant gonorrhea is common in Asia, and first appeared in the U.S. on the West Coast about four years ago; since 2002 the CDC has recommended that fluoroquinolones should not be used to treat gonorrhea in Hawaii and California. The nationwide prevalence of fluoroquinolone-resistant gonorrhea among men who have sex with men doubled between 2002 and 2003, and now accounts for about 5% of cases. However, resistance rates are considerably higher in some cities (e.g., an estimated 20% in San Francisco and 12.5% in New York City). CDC officials called the rise in drug-resistant gonorrhea in this population "alarming," since it suggests these men are engaging in unprotected sexual activity; moreover, individuals with gonorrhea can more easily transmit and contract other STDs, including HIV.
Instead of ciprofloxacin, the CDC now recommends cephalosporin antibiotics such as ceftriaxone (Rocephin) or cefixime (Suprax) for men who have sex with men, regardless of their sexual self-identification. Unfortunately, while ciprofloxacin is given as a single oral dose, ceftriaxone must be administered by injection. Cefixime tablets have been unavailable in the U.S. since the manufacturer's patent expired in 2002, although they might be found for sale online. A new tablet formulation produced by the Indian company Lupin was approved this past February and should be on the market soon (a liquid suspension is currently available). Ciprofloxacin remains the treatment of choice for heterosexuals, except in Hawaii and California, because fluoroquinolone-resistant gonorrhea is still uncommon (less than 1%) in this population.
Several recent journal articles have examined the deleterious effects of methamphetamine ("crystal") use in people with HIV, including a review of patterns of use and impact on HIV disease progression by Anthony Urbina, M.D., and Kristina Jones published in the March 15, 2004 issue of Clinical Infectious Diseases. Research has shown that crystal use is associated with unprotected sex, and use of shared needles to inject the drug also contributes to HIV transmission. Laboratory and animal studies have shown that methamphetamine accelerates retrovirus replication, and the drug appears to impair immunological function. Urbina and Jones concluded that "methamphetamine abuse represents a new challenge in HIV treatment and prevention."
In the December 15, 2003 issue of JAIDS, Dianne Langford, Ph.D., and colleagues from UCSD reported that methamphetamine worsens damage to brain cells in individuals with HIV-related encephalitis, a progressive condition characterized by cognitive and motor impairment. The researchers studied the brains of 77 individuals who had died of AIDS. The brains of methamphetamine users showed greater loss of specific subsets of neurons (especially in the frontal cerebral cortex), more neuronal degeneration, a reduced number of synapses, and greater proliferation of support cells called microglia. Surprisingly, however, a review of the subjects' medical records revealed that fewer methamphetamine users than nonusers had been diagnosed with HIV-related encephalitis.
The same week in the Journal of Infectious Diseases, Ronald Ellis, M.D., and colleagues (also with the UCSD team) reported results from a study of 230 men showing that current methamphetamine users had significantly higher HIV blood viral loads than either previous users or those who had never used the drug (nonusers); past users responded as well as nonusers. Among men on HAART, 39% of current users, 59% of past users, and 62% of nonusers had undetectable blood viral loads. CSF viral loads were also somewhat higher in current users. Notably, the difference was apparent only in methamphetamine users receiving HAART, leading the researchers to suggest that methamphetamine use may reduce the effectiveness of antiretroviral therapy. The observed difference may be due to poorer adherence in methamphetamine users, although reported adherence was similar in all groups. Methamphetamine may also affect concentrations of antiretroviral medications; interactions between street drugs and anti-HIV therapy have not been well studied. For more information on harm reduction for gay and bisexual men who use methamphetamine, see www.tweaker.org.
According to a study by Travis Porco, Ph.D., and colleagues from UCSF published in the January 2, 2004 issue of AIDS, use of HAART may reduce HIV infectivity by 60% or more. Effective anti-HIV therapy decreases viral load, and research has shown that people with low or undetectable viral loads are less likely to transmit the virus to their sexual partners; mutant drug-resistant HIV strains selected by treatment may also be less transmissible than wild-type virus. Using data from 534 initially seronegative gay or bisexual men aged 18-29 enrolled in the San Francisco Young Men's Health Study, the researchers created a mathematical model showing that the per-partner probability of HIV transmission decreased from 0.120 before the advent of HAART to 0.048 after HAART came into widespread use -- a decline of 60%. Over the same period, however, rates of self-reported unprotected sexual activity increased, from an average of 0.6 unprotected receptive anal sex partners during 1994-1995 to 1.3 during 1997-1999. Thus, while antiretroviral therapy may play a role in preventing HIV transmission, this effect may be offset by increases in unprotected sex, resulting in an overall stable or rising rate of new infections. "Use of HAART is a potentially important HIV prevention tool," the researchers concluded, but "one that is likely to succeed ... only if accompanied by continued emphasis on avoidance of exposure."
Liz Highleyman (firstname.lastname@example.org) is a freelance medical writer based in San Francisco.
This article was provided by San Francisco AIDS Foundation. It is a part of the publication Bulletin of Experimental Treatments for AIDS. Visit San Francisco AIDS Foundation's Web site to find out more about their activities, publications and services.