• Nelfinavir Recalled Outside the U.S.
• FDA Announces Changes to Enfuvirtide and Efavirenz Product Labels
• Two New Antiretroviral Classes Nearing Approval
• Other Investigational Drugs
• Boosted Atazanavir Monotherapy
• Natural HIV Entry Inhibitor Discovered
• Benefits of Early Treatment
• Treatment Interruption
• Cardiovascular and Metabolic Complications
• Influence of Drug Class
• Lipoatrophy Treatment
• Medical Marijuana for Peripheral Neuropathy Pain
• New HIV/HCV Coinfection Guidelines
• Liver Disease in HIV/HCV-Coinfected People
• Entecavir Active Against HIV
• Longer Tuberculosis Treatment for People With HIV
• Cancer Rates in People With HIV
• Revised Gonorrhea Treatment Guidelines
• Safety of Antiretroviral Drug Exposure During Pregnancy
• Circumcision Reduces HIV Infection Risk
• Ushercell Microbicide Study Halted
• Could Serosorting Increase HIV Transmission Risk?
• HIV Prevention and Testing in Prisons

Nelfinavir Recalled Outside the U.S.

On June 6, Roche announced an immediate recall of its entire stock of nelfinavir (Viracept) tablets and powder in countries outside the United States, Canada, and Japan due to a chemical impurity discovered in some of these formulations. The recall affects all nelfinavir manufactured at Roche's factory in Switzerland. Pfizer, which manufactures a slightly different formulation of nelfinavir for sale in the United States, announced that its product was not contaminated and is not affected by the recall; nelfinavir marketed by Japan Tobacco is also unaffected. The problem was discovered after patients reported that their nelfinavir emitted a foul odor. Some batches of the drug were found to be contaminated with methane sulfonic acid ethylester (ethyl mesylate), a byproduct of the manufacturing process that is usually present in minute quantities (less than three parts per million); in the worst contamination cases, the level reached 2300 parts per million. Roche attributed the problem to "human error" rather than deliberate tampering. The contamination is a concern because the chemical is a genotoxic agent that has been shown to cause cancer in rats that ingested large amounts. Roche recommended that patients in the affected regions contact their health care providers as soon as possible to discuss switching to a nelfinavir alternative.

FDA Announces Changes to Enfuvirtide and Efavirenz Product Labels

On January 31, the FDA announced changes to the product label for efavirenz (Sustiva), reflecting new information from studies of drug interactions with the anti-tuberculosis medication rifampin; the antifungal drugs itraconazole (Sporanox), ketoconazole (Nizoral), and voriconazole (Vfend); the lipid-lowering agents atorvastatin (Lipitor), pravastatin (Pravachol), and simvastatin (Zocor); calcium channel blockers, including bepridil (Vascor) and diltiazem (Cardizem, Tiazac); and the antipsychotic drug pimozide (Orap). The contraindications section was revised to state that efavirenz should not be used concurrently with bepridil or pimozide. If administered together, efavirenz doses should be lowered and maintenance doses of voriconazole should be increased. For details, see the complete efavirenz package insert on the Bristol-Myers Squibb Web site at www.bms.com.

Two New Antiretroviral Classes Nearing Approval

A major highlight of this year's Retrovirus conference was the presentation of promising data on two experimental agents, Pfizer's CCR5 antagonist maraviroc (Celsentri) and Merck's integrase inhibitor raltegravir (Isentress; formerly MK-0518). For the first time in years, two new classes of antiretroviral agents -- both of which target novel steps in the HIV lifecycle -- are nearing the end of the development pipeline. This offers new hope for treatment-experienced individuals who have HIV that is resistant to existing drug classes, and the best outcomes may be obtained by using the new drugs together. John Mellors, MD, of the University of Pittsburgh characterized the findings as the most exciting development since the advent of protease inhibitors in the mid-1990s.

Maraviroc

On June 20, Pfizer announced that the U.S. Food and Drug Administration (FDA) has issued a letter granting approvable status to maraviroc. This status change does not mean that maraviroc is approved for sale; further details must be resolved before the agency can approve the drug for marketing. The FDA's Antiviral Drug Advisory Committee unanimously voted to recommend accelerated approval in April, but the panel requested additional data on use of the drug in women and people of color.

The recommendation was based on data from the MOTIVATE-1 and MOTIVATE-2 studies, as presented at the Retrovirus conference (abstracts 104aLB, 104bLB). These two identical Phase IIb/III trials included heavily treatment-experienced subjects with triple-class antiretroviral resistance. MOTIVATE-1 included 601 participants in North America, while MOTIVATE-2 included 475 subjects in Europe, Australia, and the U.S. Participants in all study arms were generally similar; about 90% were men, the median CD4 cell count was 150-180 cells/mm³, and the mean HIV viral load was about 65,000 copies/mL.

Subjects with CCR5-tropic HIV were randomly assigned to receive oral maraviroc at doses of 150 mg once or twice daily or else placebo, in combination with an optimized background regimen. About 40% also took enfuvirtide (Fuzeon; T-20); 62%-76% had two or fewer other active drugs in their regimens.

After 24 weeks, virological response rates were about twice as high in the maraviroc arms compared with the placebo arms; 45.6%-48.5% of patients in the maraviroc twice-daily arms and 40.8%-42.2% in the maraviroc once-daily arms achieved viral loads below 50 copies/mL, compared with 20.9%-24.6% in the placebo arms. Among individuals with no active background drugs, 29%, 18%, and 3%, respectively, achieved virological suppression; participants who also received enfuvirtide had a better response. CD4 cell counts increased from baseline by 102-112 cells/mm³ in the maraviroc arms, compared with 52-64 cells/mm³ in the placebo arms.

Adverse events were similar in the maraviroc and placebo arms, with about 5% of participants discontinuing treatment prematurely. However, there were no signs of significant liver toxicity, which had led to the abandonment of another CCR5 inhibitor candidate, aplaviroc. More subjects in the maraviroc arm experienced a shift in HIV coreceptor usage from CCR5-tropic to CXCR4-tropic or dual/mixed-tropic virus (for an explanation of HIV coreceptor tropism, see Drug Watch in the Winter 2007 issue of BETA).

Pfizer is conducting another Phase III trial of maraviroc in treatment-naive patients. The drug is currently available to eligible patients through an international expanded access program (see www.maraviroceap.com for more information).

Raltegravir

On June 27, Merck announced that the FDA has accepted a New Drug Application (NDA) for the company's investigational HIV integrase inhibitor raltegravir and granted the drug priority review status. This status is designed to speed approval of experimental agents that address unmet medical needs. The FDA is expected to review and act on the NDA within six months.

Raltegravir, which prevents HIV from inserting its genetic material into host cells, is the first drug in its class to be considered for approval. Data included in the NDA submission support the use of raltegravir in combination with other antiretroviral drugs for treatment-experienced patients with evidence of continued HIV replication despite ongoing antiretroviral therapy.

In two presentations at the conference, researchers described results from BENCHMRK-1 and BENCHMRK-2 (abstracts 104aLB, 104bLB), also identical Phase IIb/III trials involving heavily treatment-experienced and drug-resistant patients; BENCHMRK-1 included 350 participants in Europe, Asia, and Peru, while BENCHMRK-2 included 349 subjects in North and South America. Here, too, participants in the study arms were similar, with about 90% men, a mean CD4 cell count of about 150 cells/mm³, and mean HIV viral loads of 30,000-50,000 copies/mL.

Participants were randomly assigned to receive either 400 mg oral raltegravir twice daily or placebo, in addition to an optimized background regimen. Results at 16 weeks were presented for all subjects, and 60% had 240-week data available. Further raltegravir study data were published in the April 14, 2007, issue of The Lancet.

In the two studies combined, 61%-62% of patients in the raltegravir arms achieved virological suppression below 50 copies/mL, compared with 33%-36% in the placebo arms. Among subjects with no other active drugs in their regimens, 61% achieved viral loads below 400 copies/mL with raltegravir, compared with only 5% taking placebo. Among subjects who started enfuvirtide and darunavir (Prezista) at the same time, 98% achieved virological suppression with raltegravir. CD4 cell gains were about 85 cells/mm³ in the raltegravir arms and 30-40 cells/mm³ in the placebo arms.

While the FDA considers Merck's NDA, raltegravir is available to qualified patients through an expanded access program (see www.benchmrk.com/secure/earmrk/earmrk.html or call 1-877-EARMRK1).

Other Investigational Drugs

Researchers also recently presented data on other experimental agents further back in the development pipeline.

Elvitegravir

Andrew Zolopa, MD, of Stanford University (abstract 134LB) presented 24-week results from a trial of Gilead Science's experimental oral integrase inhibitor elvitegravir (formerly known as GS-9137 and JTK-303). In this Phase II study, treatment-experienced patients receiving the highest dose of elvitegravir (125 mg once daily) boosted with 100 mg ritonavir (Norvir) experienced greater virological suppression than those receiving a boosted protease inhibitor (PI), all in combination with an optimized background regimen. After 24 weeks, 36% of patients in the 125-mg elvitegravir arm achieved HIV RNA levels below 50 copies/mL, versus 27% in the comparator boosted-PI arm. CD4 cell increases were similar across all arms. Elvitegravir was well tolerated, with few participants discontinuing the study due to adverse events.

TMC278

TMC278 is a novel non-nucleoside reverse transcriptase inhibitor (NNRTI) developed by Tibotec that appears to have a higher genetic barrier to resistance than currently approved drugs in its class. Anton Pozniak, MD, of Chelsea and Westminster Hospital (abstract 144LB) presented preliminary results from an ongoing study of 368 treatment-naive patients randomly assigned to receive one of three once-daily doses of TMC278 (25, 75, or 150 mg) or else 600 mg efavirenz; both drugs were combined with either AZT/3TC (Combivir) or tenofovir/emtricitabine (Truvada). In a 48-week analysis of time to loss of virological response (TLOVR), there were no statistically significant differences in efficacy among the treatment arms. Patients taking TMC278 were slightly more likely to experience nausea, but less likely to develop neuropsychiatric side effects or skin rash.

Vicriviroc

Phase II data on another experimental oral CCR5 antagonist, Shering-Plough's vicriviroc, were published in the June 2007 issue of AIDS and in the July 15, 2007, Journal of Infectious Diseases. In the first study, 48 HIV positive participants were randomly assigned to receive 10, 25, or 50 mg twice-daily vicriviroc monotherapy or placebo for 14 days. HIV viral load decreased significantly in all vicriviroc arms, and virological suppression persisted for two to three days after drug discontinuation. In the three ascending dose groups, 45%, 77%, and 82% of subjects, respectively, achieved at least a 1-log reduction in HIV RNA. Although adverse events were common, they occurred with similar frequency in the vicriviroc and placebo arms (72% and 62%, respectively). The researchers concluded that the 50- and 100-mg total daily doses were most effective, and that the drug's long half-life suggests it may be used once daily. The second study (ACTG 5211) included 118 treatment-experienced subjects with CCR5-tropic HIV who experienced virological failure while receiving ritonavir-containing regimens; the median CD4 cell count was 146 cells/mm³ and the median viral load was 36,380 copies/mL. Participants were randomly assigned to add vicriviroc at doses of 5, 10, or 15 mg, or else placebo, to their failing regimens for 14 days. At week 24, mean decreases in HIV RNA were significantly greater in the three vicriviroc dose groups (1.51, 1.86, and 1.68 logs, respectively), compared with the placebo arm (0.29 log). The incidence of severe (grade 3/4) adverse events was similar across all arms. However, malignancies occurred in six subjects receiving vicriviroc, compared with two receiving placebo. The researchers concluded that "vicriviroc demonstrated potent virologic suppression through 24 weeks" in treatment-experienced patients, but the higher rate of cancer remains a concern.

More Evidence of Darunavir Efficacy

Tibotec's second-generation PI darunavir (marketed as Prezista), approved in June 2006, continues to demonstrate efficacy at 48 weeks in treatment-experienced patients with drug resistance, according to a study published in the April 7, 2007, issue of The Lancet. Bonaventura Clotet, MD, and colleagues reported combined data from the ongoing international POWER 1 and POWER 2 trials, which together included more than 230 subjects who had experienced treatment failure using nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs), NNRTIs, and other PIs. Participants received 600 mg darunavir boosted with 100 mg ritonavir twice daily or else a boosted comparator PI, all combined with an optimized background regimen. After 48 weeks, in an intent-to-treat analysis, 45% of subjects taking darunavir achieved undetectable viral loads (below 50 copies/mL), compared with 10% of those taking other boosted PIs. As expected, those with more active drugs in their background regimens responded better, as did those who started enfuvirtide and darunavir at the same time. The darunavir arm also experienced a larger mean CD4 cell count increase (102 vs 19 cells/mm³). Overall, side effects -- mostly mild or moderate -- occurred with similar frequency in the two arms, but fewer subjects in the darunavir group discontinued for this reason.

Darunavir was not associated with any particular severe adverse events, although subjects in this arm were more likely to experience elevated triglycerides and to develop herpes simplex infection (possibly a manifestation of immune response syndrome in patients with pre-existing latent herpes infection). Darunavir did not cause severe liver toxicity, even in individuals coinfected with hepatitis B or C.

Clotet characterized the results as "very unexpected and very impressive," given the extent of the study participants' drug resistance. The POWER trials are continuing, and Tibotec is also conducting studies of darunavir in treatment-naive individuals.

Boosted Atazanavir Monotherapy

Maintenance monotherapy using a single boosted PI has been studied as a way to reduce side effects and improve adherence to antiretroviral therapy. However, according to a study in the April 1, 2007, Journal of Acquired Immune Deficiency Syndromes, boosted atazanavir (Reyataz) monotherapy may be a risky strategy. Swedish researchers conducted a pilot trial to assess the feasibility of 300 mg/100 mg atazanavir/ritonavir monotherapy in patients on stable combination antiretroviral regimens who had maintained HIV viral loads below 20 copies/mL for at least 12 months. Although the study was intended to recruit 30 patients, it was terminated early after five cases of virological failure were observed among the 15 patients enrolled to date. The investigators concluded that "Ritonavir-boosted atazanavir as maintenance monotherapy in HIV-1 infection might not be as potent as conventional antiretroviral therapy."

Natural HIV Entry Inhibitor Discovered

In the April 20, 2007, issue of Cell, German researchers reported on the discovery of a naturally occurring human peptide that blocks HIV entry into cells. After screening a comprehensive "library" of more than one million small peptides generated from human blood filtrate (the residue left after kidney dialysis), they found a 20-residue amino acid sequence that reduced HIV infection by 99%. The sequence -- dubbed "virus inhibitory peptide," or VIRIP -- blocks HIV entry by interacting with the gp41 fusion peptide on the viral envelope. VIRIP was shown to inhibit some 60 strains of HIV-1, including those resistant to current antiretroviral drugs, and the researchers were able to increase its potency 100-fold by changing a few amino acids. VIRIP appeared non-toxic even at high doses and did not encourage the emergence of drug-resistant virus in cell cultures. The authors suggested that their discovery might lead to the development of a new type of antiretroviral drug (which would probably have to be injected, like enfuvirtide) and possibly a microbicide to prevent infection. German biotechnology company Viro Pharmaceuticals is currently conducting animal studies to determine whether VIRIP is safe enough to test in humans.

Benefits of Early Treatment

Since the advent of potent combination antiretroviral therapy in the mid-1990s, researchers have debated the value of early treatment. Proponents of the "hit early, hit hard" philosophy suggested that starting therapy at the earliest stages of infection might lower the viral load "set-point" (level of stabilization), but as the long-term toxicities of therapy became more apparent, experts began to favor delaying therapy until there was evidence of disease progression. Two studies presented at the Retrovirus conference provided further data on early therapy.

In the first study, Dutch researchers looked at individuals with primary HIV infection in the Amsterdam Cohort Study and the ATHENA cohort (abstract 124LB). Out of 332 subjects with primary HIV infection, 64 started HAART within six months of infection, of whom 32 then stopped treatment. After interrupting therapy, the viral load setpoint was 0.6 log copies/mL lower in patients who started treatment early compared with those who did not do so. However, there was no difference in the rate of CD4 cell decline between the two groups.

In the second study, researchers analyzed participants in two German cohorts (abstract 125LB). Out of 200 subjects with primary HIV infection, 144 started treatment immediately after infection and 56 remained untreated. Untreated subjects had a lower median first viral load measurement than those who started and then interrupted early therapy. But one year after seroconversion, the untreated subjects had a median viral load of 52,880 copies/mL, compared with 38,056 copies/mL for treated patients 12 months after they stopped therapy. In addition, while treated patients experienced a CD4 count increase of 60 cells/mm³ from baseline after stopping therapy, untreated subjects had a median decrease of 87 cells/mm³.

Taken together, these studies indicate that early therapy is associated with lower HIV viral loads and possibly a slight benefit in terms of immune function. But a related study of the Johns Hopkins HIV Clinical Cohort, reported in the February 1, 2007, issue of Clinical Infectious Diseases, showed that CD4 count at the time of treatment initiation influenced long-term immune recovery. In this study of 655 participants followed on treatment for up to six years, subjects across all baseline CD4 cell levels experienced significant increases during the first four years on HAART, but then reached a plateau. After six years, median CD4 cell counts were 493 cells/mm³ among patients with baseline counts of 200 or less, 508 cells/mm³ among those with baseline counts of 201-350, and 829 cells/mm³ among those with baseline counts above 350. Only 12% of patients who started treatment with 200 cells/mm³ or less eventually attained 750 cells/mm³ or greater, compared with 21% of those with baseline counts of 201-350 and 46% of those with baseline counts above 350. The researchers concluded that only individuals with baseline CD4 cell counts above 350 cells/mm³ returned to nearly normal levels after six years of follow-up.

In a similar analysis reported in the June 1, 2007, Journal of Acquired Immune Deficiency Syndromes, researchers found that people who started treatment early in the ATHENA cohort were more likely to eventually attain CD4 cell counts of 800 cells/mm³ or more. In this study, 20% of subjects who started treatment with CD4 counts below 50 attained at least 800 cells/mm³ after seven years on HAART, compared with 26% of those with 50-200 cells/mm³, 46% of those with 200-350 cells/mm³, 73% of those with 350-500 cells/mm³, and 87% of those with counts above 500 cells/mm³. While even patients who started treatment with very low CD4 cell counts experienced robust immunological responses, their recovery was slower and did not reach the same levels as those who started therapy earlier. The benefits of early treatment were most apparent in individuals aged 50 years or older. The authors concluded that it may be beneficial to start therapy earlier than current guidelines recommend, particularly in older patients.

In an accompanying editorial, Evan Wood, PhD, and Julio Montaner, MD, of the British Columbia Centre for Excellence in HIV/AIDS wrote that with the availability of simpler and safer antiretroviral regimens, it may be time to re-evaluate the ideal time to start therapy.

Treatment Interruption

Even as some researchers are encouraging earlier therapy, others are continuing to explore treatment interruption in an effort to reduce drug toxicities, inconvenience, and cost. As reported in the Winter 2007 issue of BETA, data from the large SMART study suggested that interruption of antiretroviral therapy in individuals with CD4 cell counts between 250 and 350 cells/mm³ was associated with a greater risk of disease progression and death (data were published in the November 30, 2006, New England Journal of Medicine). Current U.S. government treatment guidelines do not recommend starting therapy until the CD4 cell count falls below 350 cells/mm³. In two recent studies, investigators looked at whether people who started therapy "too soon" -- with CD4 counts above this level -- could safely stop treatment.

In ACTG study 5170 (reported in the May 15, 2007, Journal of Infectious Diseases), 167 HIV positive subjects with CD4 cell counts above 350 cells/mm³ underwent treatment interruption. After stopping therapy, the initial mean CD4 cell decrease was 20 cells per week during the first eight weeks, decreasing to two cells per week for the remainder of the follow-up period. By week 96, 17 patients had CD4 counts that had fallen to 250 cells/mm³ or less, and 46 had resumed antiretroviral therapy. Four were diagnosed with AIDS-defining events, all of whom still had CD4 counts above 350 cells/mm³. "Disease progression after treatment interruption was low in this cohort," the authors concluded. Similarly, as reported in the April 1, 2007, Journal of Acquired Immune Deficiency Syndromes, Dutch researchers evaluated the safety and efficacy of HAART discontinuation in the ATHENA cohort. Out of 71 enrolled participants with CD4 counts above 350 cells/mm³, 46 elected to interrupt therapy and 25 continued on HAART. After 48 weeks, the median viral load in the discontinuation group stabilized at about the pretreatment level (4.55 logs) and the median CD4 count remained above the pretreatment level (563 cells/mm³); no AIDS-defining conditions or deaths occurred after treatment interruption. Although the authors said that "HAART can safely be interrupted in patients with a high CD4 T-cell nadir," they added that discontinuation led to no improvement in quality of life and presented a potential risk of HIV transmission to sexual partners. "We would not actively advise stopping treatment in patients who started treatment too early according to current guidelines," the authors concluded.

Cardiovascular and Metabolic Complications

Numerous studies looking at cardiovascular and metabolic complications related to HIV infection and its treatment were presented at the Retrovirus conference or published in recent journal articles.

Heart Disease and MI Incidence

At the conference, two research teams presented further data on such complications in the SMART study. Of the 5472 total participants, 79 (1.4%) developed major cardiovascular events, including death, non-fatal heart attacks, silent myocardial infarctions (MIs), strokes, coronary heart disease (CHD), and related medical problems. Andrew Phillips, MD, and colleagues (abstract 41) showed that subjects in the intermittent therapy arm who took NRTI-only regimens had a somewhat higher risk of cardiovascular events compared with those taking PI-based regimens, while those taking NNRTI-based regimens -- especially ones containing nevirapine (Viramune) -- had approximately twice the risk. However, in the continuous therapy arm, there was a small increase in the risk of cardiovascular events per additional year of exposure to PI-based, but not NNRTI-based, regimens. The researchers concluded that there was a "borderline significant excess risk of cardiovascular disease" among patients receiving intermittent therapy. Looking at body shape changes in a substudy of 275 subjects, Fehmida Visnegarwala and colleagues (abstract 803) found that limb fat increased after 12 months in the intermittent therapy arm, but decreased in the continuous therapy group; in addition, there were significant reductions in blood lipids in the intermittent compared with the continuous therapy arm.

Researchers from Kaiser Permanente health plan (abstract 807) analyzed data from a cohort of about 5000 HIV positive men and about 43,000 HIV negative men followed from 1996 through 2006; they also reported data from a surveillance cohort of all HIV positive plan members aged 35-64 years and age- and sex-matched HIV negative members. In the first comparison, age-adjusted rates of CHD and MI were 6.1 and 3.7 per 1000 person-years (PY), respectively, among HIV positive participants, compared with 2.9 and 2.2 per 1000 PY among HIV negative individuals. Both CHD and MIs were more common in patients who used PIs, and the risk increased with each additional year of PI exposure. In the surveillance cohort, MI risk increased with the introduction of HAART in 1996, but then leveled off after 1999. In this comparison, the increased risk of MI associated with HIV was especially pronounced among women.

Larger Risk Increase in Women

As reported in the April 24, 2007, online edition of the Journal of Clinical Endocrinology and Metabolism, Steven Grinspoon, MD, and colleagues found that the link between HIV and cardiovascular disease was especially strong in women. This analysis included 3851 mostly HAART-treated HIV positive participants (about 30% women) and 1,044,589 HIV negative individuals (about 60% women) receiving care at Massachusetts General Hospital (MGH) and Brigham and Women's Hospital in Boston between 1996 and 2004. Overall, rates of acute MI were significantly higher in HIV positive compared with HIV negative individuals (11.13 vs 6.98 cases per 1000 PY). HIV positive subjects were significantly more likely to have hypertension, diabetes, and dyslipidemia. After adjusting for these factors, as well as age, sex, and race, HIV positive individuals were nearly twice as likely to experience acute MIs. However, separate analyses by sex showed that the difference in MI rates between HIV positive and HIV negative men was no longer statistically significant, but HIV positive women had nearly three times the MI rate of women without HIV.

Ischemic Heart Disease

MIs are often preceded by ischemic heart disease, which involves reduced blood flow in the coronary arteries that supply the heart muscle. As reported in the June 15, 2007, issue of Clinical Infectious Diseases, researchers analyzed data collected between January 1995 and December 2004 from 3953 HIV positive participants in the nationwide Danish HIV Cohort Study, as well as a population-based control group of nearly 374,000 HIV negative individuals. HIV positive patients who had not started HAART were slightly more likely than control subjects to be hospitalized for ischemic heart disease, but the difference did not reach statistical significance. Soon after starting HAART, the risk became "substantially higher," and the authors estimated that the increase was equivalent to that associated with smoking one to four cigarettes per day; after this initial jump, however, the risk did not increase further during the first eight years on therapy. Since the risk did not rise with additional time on HAART, the researchers suggested that mechanisms besides antiretroviral-associated lipid elevation might be responsible.

Influence of Drug Class

A study by Judith Shlay, MD, and colleagues, reported in the April 15, 2007, Journal of Acquired Immune Deficiency Syndromes, looked at metabolic parameters and body composition in 422 previously treatment-naive participants in the CPCRA Flexible Initial Retrovirus Suppressive Therapies (FIRST) study; one-third started PI-based regimens, one-third started NNRTI-based regimens, and the remainder started regimens that included both drug classes. After a median follow-up period of five years, triglyceride and low-density lipoprotein (LDL or "bad") cholesterol levels increased by similar amounts in patients taking PI-based or NNRTI-based regimens, but were higher in those whose regimens included both classes; high-density lipoprotein (HDL or "good") cholesterol increased significantly more in patients taking NNRTI-based regimens. Glucose levels, insulin levels, and insulin resistance increased to a similar extent in all three groups, and changes in body composition did not differ based on treatment strategy.

Traditional Risk Factors

Experts continue to debate the extent to which cardiovascular complications in people with HIV are related to antiretroviral therapy as opposed to traditional risk factors. As reported in the March 1, 2007, issue of Clinical Infectious Diseases, Kristin Mondy, MD, and colleagues retrospectively analyzed data from 471 HIV positive patients and matched HIV negative individuals. The overall prevalence of metabolic syndrome was similar in both groups (25.5% vs 26.5%), but the HIV positive subjects had a significantly smaller waist circumference, lower body mass index, lower HDL cholesterol levels, higher triglyceride levels, and lower glucose levels. Framingham 10-year cardiovascular risk scores were similar in the two groups, and the type or duration of antiretroviral therapy was not an independent risk factor for metabolic syndrome. "The prevalence of metabolic syndrome is high among HIV-infected persons, but not higher than the prevalence among HIV-uninfected persons," the authors concluded, adding that "Traditional risk factors play a more significant role in the development of metabolic syndrome than do HIV treatment-associated factors."

The latest data from the large international D:A:D study were reported in the April 26, 2007, New England Journal of Medicine. This ongoing study includes 23,437 HIV positive participants, mostly in Europe. Through February 2005, 345 patients had MIs during 94,469 PY of observation. The incidence of MI increased from 1.53 per 1000 PY in patients not exposed to PIs to 6.01 per 1000 PY in those who received PIs for more than six years, mainly due to dyslipidemia. In contrast, NNRTI use was not associated with an increased MI risk. In an accompanying editorial, James Stein, MD, stressed that the magnitude of the increased heart attack risk associated with PI use was low compared with other known cardiovascular risk factors, suggesting that "perhaps more effort should be spent assisting our patients with smoking cessation and the prevention of diabetes, rather than our focusing so intently on the dyslipidemic effects of antiretroviral therapy."

Effect Of Diet

In another study reported at the Retrovirus conference, researchers from MGH (abstract 813) evaluated the relationship between diet and metabolic parameters in 356 HIV positive individuals without wasting syndrome (197 men and 159 women) and 162 HIV negative control subjects; about 90% of the HIV positive subjects were on HAART, including about two-thirds taking PIs. One-third of the HIV positive individuals met the criteria for metabolic syndrome, compared with about one-quarter of the control subjects. HIV positive participants consumed more daily calories overall, although the difference did not reach statistical significance. However, the HIV positive group consumed significantly more total dietary fat, saturated fat, and cholesterol. Saturated fat consumption was significantly associated with elevated triglyceride levels after controlling for PI use and other factors. These results suggest that patient education and diet modification could play an important role in managing metabolic complications in this population.

Management of Elevated Lipids

With regard to management of metabolic complications, another Kaiser Permanente study reported at the Retrovirus conference (abstract 814) found that lipid-lowering therapy may not work as well in people with HIV. While LDL declined to a similar extent in HIV positive and HIV negative individuals treated with lipid-lowering agents (primarily statins), those with HIV experienced smaller decreases in total cholesterol and triglyceride levels; HIV positive patients also were less likely to attain recommended blood lipid levels.

In more promising news, a French study reported in the March 1, 2007, Journal of Acquired Immune Deficiency Syndromes found that fish oil (which contains N-3 polyunsaturated, or omega-3, fatty acids) reduced triglyceride levels in HIV positive people on HAART. This study included 122 HIV positive participants on antiretroviral therapy who still had elevated triglyceride levels after a four-week diet. Subjects were randomly assigned to receive Maxepa fish oil capsules or placebo capsules for eight weeks, followed by an open-label phase during which all participants received fish oil. Median triglyceride levels decreased by 25.5% in the fish oil group, but rose by 1% in the placebo group. At week 8, triglyceride levels normalized in 22.4% of subjects in the fish oil arm compared with 6.5% in the placebo arm. Subjects in the fish oil arm also experienced a slight decline in total cholesterol, compared with a small increase in the placebo arm.

Lipoatrophy Treatment

In late December, the FDA approved Radiesse, a new injectable therapy for facial fat loss (lipoatrophy) in people with HIV. Manufactured by BioForm Medical, Radiesse contains a synthetic material that stimulates collagen production. Approval was based on a prospective study of 100 subjects with HIV-associated facial lipoatrophy who had been receiving HAART for at least three years; 94% were men and 56% were Caucasian. Three months after injection, mean cheek thickness increased by more than 2 mm. The most common adverse events were temporary swelling, bruising, redness, and pain at the injection site. Radiesse is not a permanent treatment, but is expected to last at least several months, and possibly as long as a few years.

In other lipoatrophy news, a study in the June 15, 2007, Journal of Infectious Diseases showed that the thiazolidinedione drug rosiglitazone (Avandia), usually used to treat type 2 diabetes, provided no benefit in improving fat loss in HIV positive people on HAART. The trial included 96 mostly male subjects randomly assigned to receive either 4 mg per day rosiglitazone or placebo for 24 weeks. At week 24, median changes in arm, leg, trunk, and total body fat did not differ significantly between the two groups. Individuals who switched off AZT (zidovudine; Retrovir) or d4T (Zerit), however, experienced more pronounced fat gains in the rosiglitazone arm. In related news, a meta-analysis of more than 40 prior studies, published in the June 14, 2007, New England Journal of Medicine, suggested that rosiglitazone may increase the risk of heart attacks. The FDA issued an advisory to health care providers, but did not ask manufacturer GlaxoSmithKline (GSK) to take any specific action; GSK contested the findings and issued a statement that included data from studies attesting to the drug's safety.

Medical Marijuana for Peripheral Neuropathy Pain

A study reported in the February 13, 2007, issue of Neurology provided the first evidence from a randomized controlled trial demonstrating the therapeutic benefits of medical marijuana. Donald Abrams, MD, and colleagues from the University of California at San Francisco assessed the effect of smoked cannabis on pain related to peripheral neuropathy, a potential side effect of certain antiretroviral drugs. Between May 2003 and May 2005, 50 adults with painful HIV-associated sensory neuropathy were randomly assigned to smoke either cannabis cigarettes containing 3.56% tetrahydrocannabinol or similar placebo cigarettes with the cannabinoids removed three times daily for five days. Participants then rated their pain and underwent controlled pain testing using heat and other stimuli.

Smoking the first cannabis cigarette reduced chronic pain by a median of 72%, compared with 15% for placebo cigarettes. Over five days, smoked cannabis reduced daily pain by a median of 34%, compared with 17% for placebo cigarettes. Half the individuals who smoked marijuana experienced at least a 30% reduction in pain, compared with 24% of subjects in the placebo group. Cannabis also reduced experimentally induced pain, but had little effect on sensitivity to heat stimuli. The authors concluded that "Smoked cannabis was well tolerated and effectively relieved chronic neuropathic pain from HIV-associated sensory neuropathy. The findings are comparable to oral drugs used for chronic neuropathic pain."

New HIV/HCV Coinfection Guidelines

In the May 31, 2007, issue of AIDS, an international panel of experts led by Vincent Soriano, MD, published updated guidelines for the management of hepatitis C virus (HCV) coinfection in HIV positive individuals, reflecting an improved understanding of concurrent infection and its treatment since the previous guidelines were issued in 2004. HIV/HCV-coinfected individuals tend to experience more rapid liver disease progression and typically do not respond as well as HCV-monoinfected individuals to treatment with pegylated interferon (Pegasys or PegIntron) plus ribavirin.

The panel issued recommendations in 11 areas. With regard to who should receive treatment for hepatitis C, the authors recommended that coinfected individuals should be considered for therapy regardless of alanine aminotransferase (ALT) level, since the coinfected population is more likely to develop significant fibrosis despite normal ALT. They also wrote that "liver biopsy is not mandatory for considering the treatment of chronic HCV infection," given that new noninvasive methods "accurately predict fibrosis in most cases."

In terms of treatment optimization, studies have shown that weight-based ribavirin (1000 mg/day for patients weighing under 75 kg and 1200 mg/day for those over 75 kg) is superior to a fixed dose of 800 mg/day. While some studies have tested higher doses of pegylated interferon in coinfected subjects, the benefits have not been established and the panel recommended that such patients should receive the standard doses used for HCV-monoinfected individuals. The panel further advised that all coinfected patients should receive 48 weeks of combination therapy regardless of genotype, although they acknowledged that 24 weeks may be adequate for those with genotypes 2 or 3, while slow-responding individuals with genotypes 1 or 4 might benefit from extended therapy (60-72 weeks). As is the case for HCV-monoinfected individuals, coinfected patients who do not achieve early virological response by week 12 or who still have detectable HCV RNA at week 24 should be advised to stop treatment. For HIV positive individuals with acute HCV infection, the panel recommended 24-week treatment with pegylated interferon plus weight-based ribavirin, after waiting 12 weeks to allow for possible spontaneous clearance.

The panel noted that investigational drugs -- including new types of interferon and directly targeted agents such as HCV protease and polymerase inhibitors -- offer the prospect of improved outcomes, and recommended that "Trials exploring the efficacy and safety of these drugs in coinfected patients should be prioritized, without waiting for the final results of Phase III trials conducted in HCV-monoinfected individuals."

Looking at antiretroviral therapy in coinfected individuals, the panel noted that the major drug interaction concerns are concurrent use of ribavirin with ddI (Videx) or AZT. They recommended that ddI and ribavirin "should never be used" together, since both can cause mitochondrial damage, while the combination of AZT and ribavirin "should also be avoided when possible," since both can cause anemia. Finally, the panel discussed the various mechanisms by which antiretroviral drugs can cause liver toxicity. Despite the complexities of antiretroviral therapy in coinfected patients, the panel concluded that the benefits of HAART outweigh the risks. "Since severe immunosuppression accelerates HCV-related liver fibrosis progression, it may be advisable to start HAART without unnecessary delays in coinfected patients and even consider earlier initiation of treatment," they wrote.

Liver Disease in HIV/HCV-Coinfected People

Several past studies have shown that HIV/HCV-coinfected people experience more rapid liver fibrosis progression compared with HCV-monoinfected individuals. However, other research suggests that this may not be the case for coinfected patients with well-preserved immune function and high CD4 cell counts. A French study presented at the annual European Association for the Study of the Liver meeting in Barcelona in April, for example, found that HIV positive subjects with normal CD4 counts did not have worse fibrosis progression compared with HCV-monoinfected individuals, and their HCV-specific immune responses were not impaired.

There are also conflicting data regarding the influence of antiretroviral therapy on liver disease progression. At the Retrovirus conference, Spanish researchers (abstract 935) reported on an analysis of liver biopsy results from 213 coinfected patients (58% with genotype 1 HCV) who had not received HCV treatment. At the time of biopsy, the subjects had well-controlled HIV (73% with HIV RNA below 50 copies/mL) and the median CD4 count was 460 cells/mm³. Based on an estimated median time of 21 years since HCV infection, the subjects as a group had been off HAART for about 17 years and on HAART for about four years since they acquired HCV. Most subjects had mild-to-moderate (METAVIR stage F1-F2) fibrosis. Only 0.5% had absent or minimal (stage F0) fibrosis, 19% had stage F3 fibrosis, and 12% had severe bridging fibrosis or cirrhosis (stage F4). The rate of fibrosis progression decreased with additional time on HAART, while the odds of having lower (F0-F2) versus higher (F3-F4) fibrosis stages increased. The investigators concluded that "HAART reduces the fibrosis progression rate and the development of bridging fibrosis and cirrhosis in HIV/HCV-coinfected patients."

As discussed in a review by Stephen Shafran, MD, in the April 15, 2007, Journal of Acquired Immune Deficiency Syndromes, these findings confirm data from 11 previous cohort studies showing that HAART is associated with a reduced rate of HCV-related liver disease progression, four of which also demonstrated a reduction in liver-related mortality. However, treatment of chronic HCV remains more difficult in coinfected compared with HCV-monoinfected patients. According to Shafran's analysis, anti-HCV therapy with pegylated interferon plus ribavirin produces sustained response in up to 40% of coinfected patients, but only about 10% of this population are considered suitable candidates for therapy. Thus, he concluded, "Although offering HCV therapy to the few eligible HIV/HCV-coinfected patients is important, early initiation of HAART in coinfected patients has a greater public health impact in reducing liver-related mortality."

The importance of HCV treatment was underlined by another recent study showing that coinfected individuals may still experience fibrosis progression despite use of effective antiretroviral therapy. As reported in the April 1, 2007, issue of the same journal, researchers used the noninvasive AST-to-platelet ratio index (APRI) to assess fibrosis progression in 673 HIV positive individuals without liver complications at baseline; 540 had HIV alone and 133 had also had HCV. At baseline, the coinfected patients had a higher median APRI score than HIV monoinfected subjects (0.59 vs 0.33). Over a median follow-up period of 4.6 years, coinfected patients were four times more likely to develop liver complications (4.5% vs. 1.1%), and did so in a shorter amount of time (2.85 vs. 3.96 years) compared with HIV monoinfected subjects. Unexpectedly, however, HAART use was associated with greater progression of APRI scores in both HIV-monoinfected and coinfected subjects. A possible explanation is that drug-related liver toxicity can cause elevation of one factor in the index, the liver enzyme AST (aspartate aminotransferase). "There was clearly a complex relationship between HAART and fibrosis," the researchers wrote. "While a higher CD4 cell count and better HIV control were indeed associated with lower rates of fibrosis progression as expected, HAART was additionally associated with increased progression in the APRI scores." Despite the association between antiretroviral therapy and higher APRI scores, the authors noted that this did not translate into an increased rate of adverse liver-related outcomes over time, leading them to suggest that while HAART may accelerate progression to moderate-to-severe fibrosis, it may have less of an impact on further progression to end-stage liver disease. "Our findings highlight the need to treat HCV infection specifically," they concluded, "because immune restoration from HAART alone cannot be relied on to improve outcomes in HCV coinfection."

Entecavir Active Against HIV

Researchers at the Retrovirus conference (abstract 136LB) reported that entecavir (Baraclude), which is approved for the treatment of chronic hepatitis B virus (HBV) infection, is also active against HIV and can select for a mutation that confers resistance to certain antiretroviral drugs. Since most approved and investigational anti-HBV drugs (including adefovir [Hepsera], 3TC [Epivir], emtricitabine [Emtriva], and tenofovir [Viread]) have anti-HIV activity, entecavir -- which was previously believed to have no such effect -- was recommended for treating hepatitis B in HIV/HBV-coinfected patients who do not yet require antiretroviral therapy.

But this recommendation has now changed, since investigators described three HIV positive individuals not on antiretroviral therapy who experienced a 1-log drop in HIV RNA after starting hepatitis B treatment with entecavir, indicating that the drug was active against HIV, as well. Entecavir also suppressed HIV replication in laboratory tests, and in vitro resistance assays demonstrated the emergence of the M184V mutation, which confers resistance to 3TC and emtricitabine. Bristol Myers-Squibb, the manufacturer of entecavir, did not detect evidence of anti-HIV activity in its own studies of the drug. Nevertheless, based on the recent findings, the company issued a letter to health care providers in February, stating that the risk of developing HIV resistance "cannot be excluded" when entecavir is used in HIV/HBV-coinfected patients not receiving HAART. In April, the Department of Health and Human Services released a supplement to its HIV treatment guidelines, recommending that "For HBV/HIV-coinfected patients, entecavir should not be used for the treatment of HBV infection without concomitant treatment for HIV."

Longer Tuberculosis Treatment for People With HIV

Tuberculosis (TB) is the leading cause of death in people with HIV worldwide. An extremely drug-resistant form of the disease (XDR-TB) was described in South Africa last year and has now been reported in several countries -- including a recent case in an American man who was quarantined after taking an international airline flight. Standard treatment for TB is a combination of antibiotics taken for six months. While current guidelines indicate that this regimen is adequate for both HIV negative and HIV positive individuals, a study in the June 1, 2007, American Journal of Respiratory and Critical Care Medicine indicated that HIV positive patients are more prone to relapse and may benefit from longer treatment.

Investigators retrospectively reviewed data from 700 TB patients (264 HIV positive, 315 HIV negative, and 121 untested) reported to the San Francisco Tuberculosis Control Program between 1990 and 2001. The overall relapse rate among HIV positive subjects was 9.3 per 100 PY, compared with 1.0 among HIV negative and unknown-serostatus individuals. Among patients with HIV, those who received the standard six-month course of therapy were significantly more likely to relapse than those treated for longer durations, as were those who received intermittent rather than daily therapy. While some experts recommend that antiretroviral therapy should be deferred until TB treatment is completed in order to avoid drug interactions and immune reconstitution syndrome, the study found that concurrent use of HAART was associated with a lower TB relapse rate, more rapid conversion to negative TB sputum smears and cultures, and improved survival.

Cancer Rates in People With HIV

Studies have produced conflicting evidence regarding the occurrence of malignancies among HIV positive people in the HAART era. While combination antiretroviral therapy can improve immune function and may thereby reduce the risk of AIDS-defining cancers such as non-Hodgkin's lymphoma (NHL) and Kaposi's sarcoma (KS), patients have more opportunity to develop non-AIDS-related malignancies as they live longer thanks to effective treatment.

Researchers at the Retroviruses conference (abstract 84) presented the latest data on rates of fatal cancer in the D:A:D study, which includes more than 23,000 HIV positive patients followed since 1999. Overall, rates of both AIDS-defining and non-AIDS-defining cancers were significantly lower in 2004-2005 compared with 1999-2001. Out of 1246 total deaths reported up to the time of the analysis, 305 were due to any type of cancer. About two-thirds of these (63%) were due to non-AIDS-defining malignancies (for a rate of 1.79 per 1000 PY) and 37% were due to AIDS-defining cancers (1.05 per 1000 PY). Among the fatal AIDS-defining malignancies, there were 82 cases of NHL, 28 cases of KS, and two cases of cervical cancer. The most common fatal non-AIDS-defining cancers were lung cancer (62 cases), gastrointestinal cancer (41 cases), hematological malignancies (20 cases), and anal cancer (20 cases).

The risk of death due to either AIDS-defining or non-AIDS-defining cancers increased with lower CD4 cell counts, but patients with fatal non-AIDS-defining cancers had a higher median CD4 count at the time of death (211 vs 75 cells/mm³) and a higher median nadir CD4 count (87 vs 30 cells/mm³) than those who died from AIDS-defining malignancies. The investigators concluded that among HIV positive individuals with access to potent antiretroviral therapy, deaths due to non-AIDS-defining cancers are now more common than those due to AIDS-defining malignancies. Since lower CD4 counts are associated with a significantly increased risk of fatal cancer, they suggested that prevention of advanced immunodeficiency should be a key strategy for preventing cancer-related deaths in this population, along with attention to known risk factors such as tobacco smoking and chronic hepatitis B (a risk factor for liver cancer).

In a related study reported in the July 2007 issue of Clinical Infectious Diseases, researchers found that people with HIV appear to be at significantly higher risk for lung cancer, independent of smoking. This study looked at a cohort of 2086 injection drug users in Baltimore followed since 1988; about 75% were HIV positive at study entry and a further 16% seroconverted during follow-up. Most subjects (84%) reported that they smoked, and the proportion of smokers was similar in the HIV positive and HIV negative groups; 45% reported smoking 20 cigarettes (one pack) or more per day, and 10% said they smoked two packs or more. Overall, lung cancer mortality increased during the HAART era. Over 19,835 PY of observation, the investigators identified 27 lung cancer deaths, 14 of them among HIV positive individuals. All but one of the patients with lung cancer (96%) smoked, consuming a mean 1.2 packs per day. While smoking remained the major risk factor for lung cancer, HIV infection itself was associated with about a 3.5-fold increased risk after adjusting for age, sex, smoking status, and calendar period. Pre-existing lung disease (particularly noninfectious diseases and asthma) demonstrated a trend toward increased lung cancer risk, but use of illicit drugs was not associated with an increased risk. In contrast with the D:A:D findings, CD4 cell count and HIV viral load were not strongly associated with an increased risk of lung cancer.

Revised Gonorrhea Treatment Guidelines

In April, the Centers for Disease Control and Prevention (CDC) issued updated guidelines for the treatment of gonorrhea due to an increase in drug-resistant cases nationwide. Gonorrhea is the second most common sexually transmitted infection in the U.S., with an estimated 700,000 new cases per year; like other sexually transmitted diseases that cause genital inflammation and ulceration, it can increase the risk of contracting and transmitting HIV. As reported in the April 13, 2007, issue of Morbidity and Mortality Weekly Report, a survey of men seen at sexual health clinics in 26 areas found that fluoroquinolone-resistant gonorrhea was reported at all but one of the surveillance sites, covering all regions of the U.S. Among heterosexual men, fluoroquinolone-resistant strains accounted for 6.7% of all gonorrhea cases reported during the first half of 2006, up from 0.6% in 2006. The increase among men who have sex with men during the same period was even more dramatic: from 1.6% to 38.3%. Resistant gonorrhea accounted for at least 20% of all reported cases in Philadelphia, Honolulu, Long Beach, San Diego, and San Francisco.

According to the revised guidelines, gonorrhea should no longer be treated with fluoroquinolone antibiotics, including ciprofloxacin (Cipro), levofloxacin (Levaquin), or ofloxacin (Floxin). Instead, it should be treated with cephalosporins, such as injected ceftriaxone (Rocephin); oral formulations, including cefpodoxime (Vantin) or cefuroxime (Ceftin), may be appropriate alternatives. Spectinomycin (Trobicin) may be used for patients who are allergic to cephalosporins, but this drug is not yet available in the U.S. In 2000, the CDC advised against using fluoroquinolones to treat gonorrhea acquired in Asia or the Pacific Islands. The recommendation was extended to California in 2002, and then to all men who have sex with men in 2004. It now applies to all cases of gonorrhea in any population anywhere in the U.S. "We are running out of options," said John Douglas of the CDC. "There are currently no new drugs for gonorrhea in the drug development pipeline."

Safety of Antiretroviral Drug Exposure During Pregnancy

While the prophylactic use of antiretroviral agents such as AZT and nevirapine has reduced the risk of mother-to-child HIV transmission to less than 2% in developed countries, there have been persistent concerns about the effects of drug exposure on babies. In an analysis of data from the Women and Infants Transmission Study reported in the March 1, 2007, Journal of Acquired Immune Deficiency Syndromes, researchers examined the medical records of 2527 infants born to 2353 HIV positive U.S. women between 1990 and 2004. Birth defects were identified in 90 infants, for an overall rate of 3.56 per 100 live births, which was not significantly different from the rate for babies born to HIV negative mothers. Birth defect rates were 4.05 among infants with no antiretroviral exposure, 3.19 for those with first-trimester exposure, and 3.54 for those with second- or third-trimester exposure. The only type of birth defect observed more often among babies born to women with first-trimester exposure was hypospadias, a male genital abnormality. The authors concluded that the "data were reassuring," although they acknowledged that information regarding newer antiretroviral agents was limited.

In another study, published in the March 12, 2007, issue of AIDS, researchers conducted a meta-analysis of past trials of antiretroviral exposure during pregnancy; 13 prospective cohort studies and one retrospective study met the inclusion criteria. The analysis showed that use of antiretroviral therapy during pregnancy did not increase the risk of premature delivery overall. However, use of PI-based regimens and starting combination therapy before pregnancy or during the first trimester were associated with a slight increase in premature births, which may have been related to more advanced HIV disease in such women.

As reported in the May 11, 2007, issue of AIDS, another study analyzed data from 1037 HIV-uninfected children (born between 1991 and 1992) in PACTG study 219/219C. Overall, 20 children had evidence of possible mitochondrial toxicity (damage to energy-producing structures within cells). While there was no overall association between in utero NRTI exposure and mitochondrial dysfunction, children with mitochondrial problems were about 10 times more likely to have been exposed to AZT or 3TC during the third trimester.

Circumcision Reduces HIV Infection Risk

In December 2006, the National Institutes of Health halted two adult male circumcision trials in Kenya and Uganda after an interim analysis showed that the intervention reduced the risk of HIV infection by about half. Results of the Uganda trial were presented at the Retrovirus conference (abstracts 155aLB, 155BLB), and data from both studies were published in the February 24, 2007, issue of The Lancet.

In both trials, young HIV negative heterosexual men (2784 in Kenya and 4996 in Uganda) were randomly assigned to undergo the procedure either immediately or after a waiting period; all participants received condoms and HIV prevention counseling. In the Kenya study, after a median 24 months of follow-up, there were 47 new HIV infections among uncircumcised men, compared with 22 among circumcised men. The two-year HIV incidence rate was 2.1% in the circumcised group and 4.2% in the control group, representing a risk reduction of 53% in an intent-to-treat analysis; in an as-treated analysis, the risk reduction was 60%. In the Uganda study, there were 43 new HIV infections among uncircumcised men and 22 among circumcised men, for an overall risk reduction of 53%, or 60% in an as-treated analysis. In both studies, circumcision did not appear to increase high-risk sexual behavior.

"Circumcision is the most potent intervention in HIV prevention that has been described," said Kevin DeCock, MD, director of the World Health Organization (WHO) HIV/AIDS Programme, though he emphasized that it should not be considered a replacement for other HIV prevention methods, such as condoms. WHO and UNAIDS officials met in March to discuss the implications of these studies. Adding a further note of caution, data presented at that meeting showed that circumcision may actually increase the risk of HIV transmission to women if men resume sexual activity before their surgical wound has completed healed. Following the meeting,

WHO and UNAIDS recommended that adult male circumcision be made available as part of a comprehensive prevention program in developing countries heavily affected by HIV/AIDS. They also stressed the importance of performing circumcision in a safe and sanitary manner, in light of the shortage of medical personnel and lack of health care infrastructure in many developing countries.

The benefits of adult male circumcision in developed countries with lower HIV prevalence remains open to debate. In a statement responding to the Kenya and Uganda studies, the San Francisco AIDS Foundation emphasized that circumcision does not provide absolute protection, since it "may reduce, but does not eliminate, risk of HIV infection." The statement noted that the recent studies do not provide answers regarding men who have sex with men, anal intercourse, or circumcision of newborns or children. In conclusion, the agency wrote, "Circumcision is a personal decision that should be made in consultation with providers, pediatricians, and others." For the full statement, see www.sfaf.org/policy/statements/circumcision.html.

Ushercell Microbicide Study Halted

In January, researchers announced the discontinuation of a study of an experimental vaginal microbicide, Ushercell, due to worrisome early trial data. Ushercell, or cellulose sulfate gel, produced by Polydex Pharmaceuticals, was being developed by the nonprofit health agency CONRAD. The product had advanced through 11 Phase I and II clinical trials involving more than 500 women and about 50 men, with no evidence of safety problems. But preliminary results from a Phase III study of more than 1300 women in Benin, South Africa, Uganda, and India showed that women using Ushercell had a higher rate of HIV infection compared with women using a placebo gel. These data led the trial's Independent Data Monitoring Committee to stop the study. As a precaution, Family Health International halted a similar trial of Ushercell in Nigeria, although a higher rate of HIV infection was not observed in that study. "While the findings are unexpected and disappointing, we will learn scientifically important information from this trial that will inform future HIV prevention research," said principal investigator Lut Van Damme, who spoke about the CONRAD trial discontinuation at the Retrovirus conference, but did not disclose actual data about the number of infections in the Ushercell and placebo arms. Three other microbicides are currently in advanced clinical development. As reported in the February 19, 2007, issue of AIDS, researchers found that PRO2000 appeared to reduce vaginal inflammation in a study of 24 HIV negative women randomly assigned to receive either the microbicide or a placebo gel for 14 days. These are promising data, given that an earlier microbicide candidate, nonoxynol-9, was suspended after studies showed that it damaged the vaginal and rectal mucosal linings.

Could Serosorting Increase HIV Transmission Risk?

Some harm reduction proponents, including officials with the San Francisco Department of Public Health, have proposed that serosorting -- whereby people have unprotected intercourse only with others of the same HIV status -- may help reduce the risk of transmission. But as reported in the May 31, 2007, issue of AIDS, a mathematical model by researchers from the University of California at San Diego suggests that serosorting may increase the risk of HIV transmission under some circumstances.

Newly infected individuals often have very high viral loads, and one recent study found that about half of all new infections may be transmitted during acute or primary infection. Such individuals may have become infected since their last test, or may have been tested during the "window period" before the body has had a chance to produce enough antibodies to be detected. Thus, they may inadvertently misinform prospective sex partners that they are HIV negative. According to the model, the benefits of serosorting decrease as the proportion of recently infected individuals in a population of potential sex partners increases. "The effectiveness of a serosorting strategy for HIV prevention depends on the accuracy of individuals' serostatus disclosures," the researchers wrote. "The effectiveness of serosorting on the basis of mutual disclosure of perceived HIV status is a flawed strategy for reducing sexual transmissions of HIV when it does not consider the prevalence of recent HIV infections in specific populations." For more on serosorting, see the feature article in the Winter 2007 issue of BETA.

HIV Prevention and Testing in Prisons

This past January, Congresswoman Barbara Lee (D-California) reintroduced HR 178, also known as the Justice Act of 2007, a bill that would require federal, state, and county correctional facilities to provide comprehensive HIV education and counseling to inmates. The legislation also would allow community-based organizations to distribute condoms to inmates, who would be permitted to possess them without punishment; while a few jurisdictions -- including San Francisco -- allow prisoners to possess condoms, most states regard them as contraband. "Ninety-five percent of inmates come back into our community," Lee said at a press conference. "People need to understand that when a prisoner is infected, we all are affected."

In a related measure, Representative Maxine Waters (D-California) in April introduced HR 1943, the Stop AIDS in Prison Act of 2007, which would require federal correctional facilities to offer all inmates HIV testing at the beginning of their sentences and prior to release (individuals would be allowed to decline the test). Although some state prisons now require routine HIV testing of all inmates, federal facilities currently do not. The bill would also require HIV education for all inmates and treatment for those who test positive.

At the state level, Assemblyman Mervyn Dymally (D-Los Angeles) introduced AB 66, a bill that would require that all inmates be offered HIV testing 30-60 days after entry into a state prison and again 30-60 days prior to release; treatment plans would be developed for those who test positive. Finally, in March, the California Assembly Committee on Public Safety approved the Inmate and Community Public Health and Safety Act (AB 1334), a measure introduced by Assemblyman Sandre Swanson (D-Oakland) that allows nonprofit and health care agencies to distribute condoms in state correctional facilities and specifies that condom possession may not be used as evidence of illegal activity.

Liz Highleyman (liz@black-rose.com) is a freelance medical writer and editor based in San Francisco.