1. Home>
2. Specialty Care>
3. HIV/AIDS>
4. Meeting Report

Meeting Report from the XV International AIDS Conference

The XV International AIDS Conference was held in Bangkok, Thailand from July 11 to 14, under the banner "Access for All." The conference was the largest thus far, attended by more than 17,000 delegates, and included nearly 5000 accepted presentations in poster and oral format, plenary sessions, scientific symposia, and a new track in which international and national policy leaders presented and debated their treatment and prevention programs.

Arguably, Bangkok was a more appropriate venue than Yokohama, Japan, the previous Asian venue for the conference (in 1994). Thailand has not only a substantial HIV epidemic but also an impressive record of successful HIV prevention and research. As with Durban, South Africa, this bustling city where one tour book proclaimed "everything goes but the traffic" accommodated most of the conference's logistical challenges.

Conference highlights covered below include progress in preventing perinatal transmission (with a glimmer of hope for avoiding NNRTI resistance), a new take on the induction-maintenance approach to therapy, some early follow-up on several treatment-interruption studies, and reports of the usual incremental progress in new drug development. Our focus, as always, is on presentations likely to affect clinical practice now or in the near future.

ACCESS IN THE DEVELOPING WORLD

For the first time, we seem to be on the verge of obtaining sufficient funds to combat HIV in Africa and other resource-constrained areas. Seven years ago, US$200 million was available for resource-poor countries for HIV treatment and prevention, whereas by 2003, $5 billion had become available. The money now flows from several sources, including the now-functioning Global Fund for AIDS, Tuberculosis and Malaria; the World Bank; President Bush's Emergency Program for AIDS Relief (PEPFAR); the European Union; and several philanthropic foundations, most notably, the Bill and Melinda Gates Foundation.

Two years ago, the World Health Organization unveiled its "3 by 5" initiative to obtain antiretroviral therapy for at least 3 million HIV-infected people in the developing world by 2005. At the time, about 50,000 people in developing countries had access to antiretroviral therapy; this figure is now estimated to have increased to between 100,000 and 150,000 people. At this rate of increase, the 3 by 5 goal would not be met until 2009. In a plenary session, Dr. James Kim from the WHO expressed the belief that achieving the goal in time remains feasible. More realistic was the view expressed by UNAIDS Executive Director Dr. Peter Piot - in a plenary session and in closing remarks - that 3 by 5 is an important goal but should not be taken literally. Piot and others were energized by the rapid influx of funds from multiple sources but cautioned that this flood has created a confusing mosaic of programs, with overlap in some program elements and geographic areas and large gaps in others. The great new challenge is to coordinate and focus these resources in collaboration with national and international agencies.

Of growing concern is the rapid spread of the epidemic in Eastern Europe and parts of Asia. Those of us who lived through the difficult early days of the U.S. epidemic are familiar with the scenario: Government authorities deny the extent of the problem, while HIV circulates in populations that are already stigmatized and either despised or ignored, predominately drug users. Indeed, Thailand - for all of its deservedly touted success in curbing the sexual transmission of HIV by creatively promoting condom use - has addressed the epidemic in drug users with a "War on Drugs" that has reportedly resulted in the extralegal killing of thousands of citizens. In addition, substance-abuse treatment is still largely unavailable in Eastern Europe and most of Asia.

Beyond these immediate challenges of allocating resources and confronting new growth areas in the epidemic, governments and policy makers need to continue to develop funding sources and sustainable infrastructure for prevention, care, and treatment that will carry us through the years to come. As Piot pointed out, the response to the pandemic to date has been largely reactive - rather than proactive. In the current maturing phase, we must apply long-term thinking and strategy and plan in terms of decades, not years. Research into potential cures and vaccines remains of utmost importance, but, realistically, we have to prepare for the presence of HIV for the indefinite future. -- Gerald H. Friedland, MD

PERINATAL TRANSMISSION

A compelling problem facing the developing world, where the inexpensive single-dose nevirapine regimen for preventing perinatal transmission is the most widely available form of HIV treatment, is the high rate of NNRTI resistance that occurs after such exposure to this agent. Resistance emerges so easily because nevirapine has a low genetic barrier to resistance and because NNRTIs have a pharmacokinetic "tail" (the prolonged half-life elimination) that may persist up to 3 weeks in some patients, essentially providing an extended period of subtherapeutic monotherapy. It was expected that such exposure might be problematic for the future treatment of women who received single-dose nevirapine, and this hypothesis was confirmed by data presented by Lallemant from the Perinatal HIV Prevention Trial (PHPT) (see ACC August issue and N Engl J Med 2004 Jul 15; 351:217). In this study, more than 600 pregnant women received AZT starting at 28 weeks of gestation and were randomized to no additional therapy or to a single intrapartum dose of nevirapine. The infants in this latter group were further randomized to an additional single dose of nevirapine or no nevirapine within the first week of life. Predictably, the addition of nevirapine was more effective at preventing perinatal transmission; the rates of transmission were 6.5% for the AZT-alone group and 1% to 2% in the groups receiving AZT + single-dose nevirapine. However, women with resistance mutations after receipt of a single-dose of nevriapine, who subsequently started on a nevirapine-containing generic 3-in-one formulation, had poorer responses to antiretroviral therapy than those who had not received single-dose nevirapine. Even women who had had no detectable mutations after single-dose nevirapine had poorer responses than those who had not previously been exposed to this agent. These data suggest that minor variants that are not detected by standard genotyping after exposure to single-dose nevirapine will emerge and be associated with later treatment failure in women who subsequently receive this agent.

In 157 women and 21 infants with NNRTI resistance after single-dose nevirapine exposure - defined genotypically as K103N, G190A, Y181C, or V106A/M - different patterns of mutation persistence were seen [ThOrB1353]. Y181C tended to persist in infants, whereas K103N persisted in mothers. Of note, higher baseline viral loads and lower baseline CD4-cell counts were associated with the persistence of resistance mutations. Cressey documented measurable nevirapine levels in women up to 21 days after single-dose administration [ThOrB1352], underscoring the problem of the extended half-lives of nevirapine and efavirenz, as previously described at the eleventh Retrovirus Conference.

Relative success in covering the NNRTI "tail" with dual-nucleoside therapy was arguably the most important resistance data presented at Bangkok. McIntyre reported results from an interim analysis of 61 mothers out of 156 enrolled (projected total, 300) in the TOPS study (Treatment Options Preservation Study) [LBOrB09]. HIV-infected pregnant women were randomized to a single intrapartum dose of nevirapine (n=18) or to the intrapartum addition of AZT + 3TC for a total of either 4 (n=20) or 7 (n=23) days; all newborns received the same treatment as their mothers. The primary endpoint was the development of NNRTI resistance in the mothers within the first 6 weeks postpartum. All 61 subjects after 6 weeks of follow-up for resistance had fully susceptible virus at baseline and all were infected with clade C. Dual-NRTI therapy lasting for 4 to 7 days after single-dose nevirapine was highly successful in preventing NNRTI resistance: 50% (9 of 18) of women who received only nevirapine had evidence of resistance, compared with only 5% (1 of 20) and 13% (3 of 23) of the women in the 4- and 7-day NRTI arms, respectively. The development of resistance for the two dual-nucleoside groups together was 9.3% (P=0.001), but clearly the optimum duration of triple-drug coverage (2 NRTIs to cover the NNRTI "tail") has not yet been determined. No 3TC or NRTI resistance was detected. The nevirapine-only arm was closed on the basis of these results, but the study is ongoing. -- Judith Feinberg, MD

CLINICAL TRIALS OF APPROVED ANTIRETROVIRALS

Bangkok was long on the mechanics and politics of scaling-up clinical care and providing antiretrovirals in the developing world - but short on the topics traditionally of interest to researchers and clinicians from industrialized countries. Basic science submissions were so few in number that drug development presentations got moved into Track A, usually reserved for laboratory research in HIV virology and immunology. There was a similar paucity of new clinical trials data - these presentations were largely devoted to longer-term follow-up results of older trials.

Researchers presented an intent-to-treat analysis of the 96-week data from the TORO 1 and 2 trials, the studies that led to FDA approval of the first fusion inhibitor, enfuvirtide (T-20) [MoOrB1058]. Compared with an optimized regimen alone, T-20 plus optimized therapy provided a durable response at the 2-year mark in the proportion of patients with a CD4-count increase of at least 50 cells/mm³ as well as the proportions with viral loads less than 400 copies/mL and less than 50 copies/mL. Similar data based on as-treated analyses of the NEAT study (unboosted amprenavir twice daily) [TuPeB4506] and SOLO study (ritonavir-boosted amprenavir once daily) [TuPeB4507] showed persistent benefits at 96 weeks and no unexpected toxicities or resistance patterns.

The CLASS study compared an NRTI (d4T), an NNRTI (efavirenz), and a boosted PI (amprenavir/ritonavir) in 291 antiretroviral-naive individuals who all received a nucleoside backbone of 3TC + abacavir [TuPeB4544]. This was a strategy study that evaluated outcomes at 96 weeks, regardless of whether subjects were on their initial randomized regimen or on a subsequent regimen after switching for toxicity or virologic failure. Although the NNRTI arm was superior at most interim measurements, there was no statistically significant difference among the three arms in an intent-to-treat analysis at 96 weeks.

Gathe presented final 48-week data from an open-label pilot study of lopinavir/ritonavir as single-agent therapy for antiretroviral-naive patients [MoOrB1057]. Thirty individuals with a mean baseline viral load of 262,000 copies/mL (57% with >100,000 copies/mL) and a mean baseline CD4 count of 170 cells/mm³ (70% with <200 cells/mm³) were treated with lopinavir/ritonavir 400/100 mg twice daily if they weighed 70 kg or less; subjects who weighed more than 70 kg received an enhanced dose of 533/133 mg twice daily. Intensification for virologic nonresponse was permitted with either tenofovir + 3TC or with saquinavir. Eight patients discontinued study participation for various reasons, and two underwent intensification. In an intent-to-treat analysis, 67% of patients achieved a viral load <400 copies/mL (60% achieved a viral load <50 copies/mL). The success of this proof-of-concept study has led to the design of randomized trials to evaluate ritonavir-boosted single-agent PI therapy (lopinavir/r or atazanavir/r) in previously untreated individuals. -- Judith Feinberg, MD

NEW DRUGS

The conference included several industry-supported oral and poster presentations of experimental agents in both existing and new drug classes.

Protease Inhibitors: The new drug closest to FDA approval is tipranavir, long known to be active in vitro against PI-resistant viruses. However, problems with formulation and drug-drug interactions have slowed its development. Only relatively recently has the appropriate dose been chosen for clinical use: 500 mg (two capsules) of tipranavir twice daily with 200 mg of ritonavir twice daily. The relatively high dose of ritonavir is needed because tipranavir induces ritonavir metabolism. Sharon Walmsley presented encouraging data from study BI 1182.51 [WeOrB1236]. The 315 patients selected for this study were triple-class experienced and unable to participate in other tipranavir studies because they had 3 or more PI mutations (at position 33, 82, 84, or 90). They were randomized to an optimized regimen combined with one of 4 treatment arms: (1) tipranavir/ritonavir (control), (2) lopinavir/r, (3) amprenavir/r, or 4) saquinavir/r. At the end of two weeks, tipranavir/r was added to each PI, so that all 4 arms received the investigational drug. At baseline, these patients had extensive genotypic and phenotypic PI resistance, with phenotypic fold-changes to the standard PIs varying from 41- to 102-fold. The three key findings of this study were that tipranavir treatment induced a two-week viral-load decline of 1.15 log copies/mL, versus 0.2 to 0.4 log copies/mL for the other PIs; that two weeks later, when the tipranavir was added to the other PIs, those groups also experienced a 1 log viral-load reduction; and, finally, that tipranavir strongly induced metabolism of the other PIs, reducing trough concentrations of saquinavir, amprenavir, and lopinavir by 84%, 51%, and 45%, respectively. Although the viral-load reductions were transient - consistent with the high-level resistance at baseline and the presence of usually only one active drug - a subset of patients who also received T-20 had a more sustained response. A tipranavir expanded-access program is open now, and FDA approval is likely either late this year or sometime in 2005.

NRTIs: Updating his presentation from the recent Retrovirus Conference, Robert Murphy presented data on the investigational NRTI Reverset, which shows activity in vitro against many NRTI-resistant viruses [MoOrB1056]. In 30 treatment-naive patients, maximum viral-load reductions were 1.0 to 2.6 log during 10 days of double-blind monotherapy. In addition, data derived from a treatment-experienced cohort were presented for the first time. Of 10 experienced subjects, eight received active drug and two received placebo. Four of the eight had three or more thymidine-associated mutations (including 41, 210, and 215), and five of the eight were resistant to 3TC. After 10 days of monotherapy, a 0.8 log viral-load reduction was observed in the active treatment group. (It is of interest that one of the placebo patients also had a significant viral load reduction, reportedly due to enhanced adherence to his background regimen.) These results confirm that Reverset is active both in vitro and in vivo against NRTI-resistant viruses, but perhaps with some reduction in antiretroviral activity. Obviously, given the small size of the study and its short duration, the full antiretroviral activity and safety information on this drug await the results of larger clinical studies, which are ongoing.

Another NRTI in development is SPD754, which is also active against many NRTI-resistant viruses. In a small monotherapy study in three patients with varying degrees of NRTI resistance, viral-load reductions were 0.65 to 1.35 log copies/mL after only days [WePeA5642]. As reported at the Retrovirus Conference, and repeated here, there is an intracellular drug interaction between this agent and 3TC, leading to reduced concentrations of SPD754, and thus likely contraindicating coadministration of these two drugs.

NNRTIs: Although NNRTIs have been a relatively challenging class for new drug development - efavirenz, the youngest drug in this class, was approved by the FDA in 1998 - data were presented on at least three new NNRTIs that are moving forward in clinical studies. A pharmacokinetic study of the experimental NNRTI capravirine showed no drug-drug interactions with atorvastatin when coadministered with lopinavir/ritonavir [TuPeB4630]. In addition, no alteration in antiretroviral exposure was noted when saquinavir was added to a regimen that included lopinavir/ritonavir and capravirine [TuPeB4631]. The clinical studies of capravirine in NNRTI-experienced patients are ongoing.

Although several years have passed since the first encouraging in vivo data on TMC125 in both treatment-naive and -experienced patients were reported, the formulation has been improved, and it is again in clinical trials. An in vitro study of site-directed mutants with TMC125 resistance found that only 3 single mutations could produce >10-fold resistance (y181l, y181v, and f227c); one double and four triple mutations also led to resistance [WeOrA1271]. These researchers found that the prevalence of these triple mutations from current NNRTI use in the ViroLogic database was <2%, suggesting (but not proving) that TMC125 will be active against viruses resistant to current NNRTIs.

Finally, phase I data were presented on the investigational NNRTI 695634G, a prodrug of the active compound 678248X [TuPeB4480]. In this 10-day study in uninfected controls, the drug caused no significant adverse effects and did not affect the QT interval; a pilot study in NNRTI-experienced patients is currently enrolling.

Entry Inhibitors: Three posters on the CCR5 antagonist UK-427,857 were presented, including further description of its antiviral activity [TuPeB4479], a pharmacokinetic analysis showing higher plasma levels when given without food [TuPeB4489], and a study demonstrating no clinically significant effect on the QTc interval in healthy subjects [TuPeB4605]. Phase III studies of this agent in treatment-naive and -experienced patients are just getting started.

In earlier stages of development are CCR5 antagonist 873140 [WeOrA1231], two CXCR4 antagonists, AMD 070 [TuPeB4475] and KRH-2731-5HCI [LbOrA01], and a humanized monoclonal antibody to CCR5 PRO 140 [WeOrA1230]. None of these presentations included data on anti-HIV effects in humans, but all appear promising in vitro, and all are orally bioavailable except the monoclonal antibody. It seems likely that this drug class will soon have several other members besides enfuvirtide (T-20).

Novel Mechanisms: The company Panacos Pharmaceuticals had three abstracts on their agent PA-457, the first-in-class maturation inhibitor. PA-457 blocks HIV maturation, leading to defective core formation and noninfectious virions [WePeA5643, WePeA5644, WeOr1276]. Oral administration of three different doses demonstrated a dose-dependent antiretroviral effect. Larger dose-finding studies are underway. -- Paul E. Sax, MD

STRUCTURED TREATMENT INTERRUPTION AND OTHER NOVEL TREATMENT STRATEGIES

Ever since the presentation by Tony Fauci at the Durban AIDS Conference in 2000, structured treatment interruption (STI) has held a special place at the International AIDS Conferences because, if viable, such a strategy would allow less-developed countries to administer antiretroviral therapy at a lower cost than with continuous treatment. This year's conference featured several abstracts on treatment-interruption studies of varying designs and rationales.

A randomized study in Spain compared continuous versus CD4-cell-count and viral-load-driven antiretroviral therapy for 201 patients who had (1) virologic suppression for at least one year; (2) a CD4 count of at least 500 cells/mm³; and (3) a CD4 nadir >50 cells/mm³ [TuPeB4567]. Therapy was resumed for members of the interruption group if the CD4 count fell to <350 cells/mm³, the viral load increased to >100,000 copies/mL, or any opportunistic infection (OI) occurred. At the end of 96 weeks, there were no OIs in the interruption group, although 12 patients developed high virologic rebound and symptoms consistent with acute HIV infection. Predictors of remaining off therapy were a CD4 nadir >200 cells/mm³, pre-therapy viral load <100,000 copies/mL, and younger age. The study is ongoing.

In the Staccato Trial [WeOrB1283], 74 patients were randomized to continuous therapy or one of two intermittent strategies - CD4-driven or 7 days on/7 days off. The 7-days-on/7-days-off strategy was associated with high rates of virologic failure and therefore was stopped. The CD4-driven approach, however, is ongoing with no differences in clinical outcome compared with continuous therapy.

A different approach was taken in the FOTO (Five On/Two Off) study, preliminary data from which were presented by Cohen [TuPeB4575]. Twenty patients with virologic suppression and CD4 counts >200 cells/mm³ changed to a 5-days-on, 2-days-off treatment schedule (usually stopping therapy on the weekends). After a median 42 weeks of follow-up, virologic breakthrough occurred in only two patients, both of whom were receiving lopinavir/ritonavir. There were no breakthroughs in patients treated with efavirenz or nevirapine. The study is continuing, and there is a plan for a larger trial in Africa comparing this approach with continuous therapy.

Another potential strategy for reducing the cost of treatment is the "induction-maintenance" approach. Although abandoned several years ago when clinical trials showed high rates of virologic failure when triple-therapy was reduced either to just indinavir or AZT + 3TC, there is increased interest in this strategy using newer agents (especially boosted PIs) due to their improved pharmacokinetics. Arribas presented preliminary data on the OK study ("Only Kaletra"), in which 42 patients with virologic suppression were randomized to continue their lopinavir/ritonavir + dual NRTI regimen ore switch to lopinavir/ritonavir alone[TuPeB4486]. At the end of 24 weeks, three patients in the simplification arm experienced virologic rebound, but again achieved virologic suppression with resumption of the NRTIs; no lopinavir resistance was detected. Interestingly, these 3 patients had a achieved an undetectable viral load relatively recently before entering the study. Other studies of lopinavir/ritonavir maintenance therapy are ongoing. Markowitz presented data on the ES40013 Study, in which 282 patients treated with Trizivir (AZT + 3TC + abacavir) + efavirenz for 48 weeks were randomized to continue that regimen or receive Trizivir alone [LbOrB14]. No significant differences were seen between the two groups in the likelihood of treatment failure. There was a slightly higher rate of virologic rebounds in the Trizivir-alone arm, and there was a higher rate of lipid disturbances and CNS toxicity in the Trizivir +efavirenz arm. Finally, a novel approach to treatment interruption was taken by Castagna and colleagues in the E-184V study [WeOrB1286]. Fifty patients with CD4 counts >500 cells/mm³ and virologic failure who harbored the M184V mutation associated with 3TC resistance were randomized either to stop all of their antiretrovirals or to stop all antiretrovirals except 3TC. At the end of 24 weeks, those remaining on 3TC alone had a higher CD4-cell count, a lower viral load, and a lower measured replication capacity than those who stopped all medications. This study adds to a body of data suggesting ongoing beneficial effects from 3TC therapy, even with high-level resistance.

In summary, although the "autoimmunization" hypothesis and long-cycle treatment interruptions have fallen by the wayside, the CD4-driven approach to intermittent therapy remains under active investigation. Clearly, proving the safety of this approach will require carefully conducted studies with lengthy follow-up, such as the ongoing CPCRA SMART study. -- Paul E. Sax, MD

METABOLIC COMPLICATIONS

There continues to be intensive investigation in the area of metabolic complications, and some important new data did emerge in Bangkok. We are starting to see long-term results from randomized trials where the primary endpoint is the evaluation of currently popular antiretroviral regimens, but that have included objective assessments of body composition, markers of insulin resistance, and measures of glucose and lipid levels as secondary endpoints. These rigorous investigations will enable us to compare the long-term risk for metabolic complications across different treatment strategies and will provide insights into the relationship between early changes in metabolic parameters and future risk for changes in body composition. For example, in the Gilead 903 study, significantly greater increases in triglyceride and LDL cholesterol levels were seen out to 144 weeks in the d4T-treated subjects compared to those receiving tenofovir [TuPeB4538]. Interestingly, a substudy demonstrated that the development of hypertrigylceridemia was a greater problem in men receiving d4T than for women [MoOrB1083]. Another substudy using DEXA scans to evaluate body composition demonstrated a significant increase in limb fat between week 96 and week 144 among tenofovir-treated subjects. Total limb fat increased from 7.9 to 8.6 kg during these two time points, while stavudine treated subjects showed a non-significant decline from 5.0 to 4.5 kg of limb fat during this same interval. The absence of baseline DEXA measurements precludes a complete assessment of the changes in limb fat over the course of this study; however, the results suggest that the regimen of tenofovir + 3TC + efavirenz evaluated in this trial may be less likely to cause a loss of limb fat (lipoatrophy), at least out to 144 weeks. Finally, bone mineral density at the hip and spine were evaluated in this trial by DEXA scans. Women receiving tenofovir had a greater decreases in BMD from baseline compared to those on d4T; however, the magnitude of the decrease was small, and no difference in fracture rates were observed during the follow-up of the trial.

Shlay and her CPCRA colleagues performed simple assessments of body-fat using anthropometric measures to estimate limb and trunk fat within a randomized trial comparing the nucleoside backbone abacavir + 3TC to ddI + d4T combined with either a PI or an NNRTI [ThOrB1360]. Subjects in both arms of the study initially gained weight and limb fat; however, by month 32, those randomized to the combination of ddI + d4T had lost limb fat, while those in the thymidine-sparing NRTI-backbone arm maintained limb fat at or above baseline levels. These results confirm earlier reports from the metabolic substudy of ACTG 384, suggesting that regimens containing ddI + d4T were associated with a greater loss of limb fat when compared with patients treated with AZT + 3TC. Current treatment guidelines no longer recommend the use of ddI + d4T as a preferred initial NRTI backbone, in part due to the greater risk of toxicity with this combination.

Risk for Subclinical Atherosclerosis: PI Vs. NNRTI: Non-invasive techniques are increasingly being used to evaluate the impact of specific antiretroviral regimens on the risk of atherosclerosis in prospective studies. Ierone and colleagues reported the preliminary results of an ongoing trial examining three different measures of subclinical atherosclerosis among subjects receiving either PI- or NNRTI-based therapy: Carotid intima media thickness measured by ultrasound, flow-mediated dilation in the brachial artery also measured by ultrasound and coronary artery calcification measured by CT scan [TbOrB1355]. Each of these tests has been previously shown to correlate with the risk for future coronary events in subsets of the general population. The investigators evaluated subjects who were receiving therapy with either nevirapine (n=40), efavirenz (n=40), or a PI (n=40). There were potentially confounding imbalances within the treatment groups: significantly more women than men were in the nevirapine group, and the PI-treated subjects had significantly lower CD4-cell counts and were less likely to smoke or have diabetes compared with patients in the NNRTI groups. No difference in baseline carotid IMT or brachial reactivity was found between PI and the combined NNRTI treated subjects. Of note, coronary calcium scores were significantly higher in the PI-treated compared with the NNRTI-treated subjects; risk factors for the presence of coronary calcium included age, duration of HIV infection and presence of hypertension. In addition, preliminary one-year follow-up data on half of the nevirapine- and PI-treated subjects suggested extremely high rates of progression of carotid IMT at one year (0.1mm-0.18 mm per year). The absence of an HIV-negative control group and the imbalances in the distribution of cardiac risk factors (possibly due to differences in prescribing patterns for subjects at high risk for CAD) make these results difficult to interpret. Despite these limitations, the finding that longer duration of HIV infection (independent of age) was associated with a higher risk of coronary calcification warrants further study. Future studies to evaluate the impact of specific antiretroviral regimens on the risk for subclinical atherosclerosis will be strengthened if they are embedded in trials where the assignment of antiretroviral regimens is randomized.

Gynecomastia: Case reports and small series have previously described gynecomastia in men as a potentially HIV-related metabolic complication. In a well designed case-control study, a investigators in Spain performed comprehensive evaluations on 44 consecutive subjects presenting for care who were found to have gynecomastia and then compared them to a control group matched for age, BMI, CD4-cell count and mean time on potent combination antiretroviral therapy [ThOrB1357]. The main finding was that hypogonadism, both primary and secondary, was much more common among those with gynecomastia compared with controls. After controlling for hypogonadism, there were no significant associations between specific antiretroviral agents and the presence of gynecomastia. Multivariate analyses also identified the presence of Hepatitis C coinfection and lipoatrophy as factors associated with gynecomastia. These results highlight the importance of evaluating testosterone levels in patients who present with gynecomastia.

Osteonecrosis: For years, the relationship between HIV infection, antiretroviral therapy and the risk of osteonecrosis has been an area of concern. Case series and clinic-based case-control studies have suggested that prior use of steroids, long-term HIV infection, and perhaps use of PI-based therapy appeared to increase the risk for this complication, but large scale studies had until now been lacking. The French Hospital Database investigators examined their cohort of nearly 60,000 HIV infected people and performed a detailed review to identify possible cases of osteonecrosis [ThOrB1358]. They first used ICD-9 codes followed by chart review to find cases of osteonecrosis and to identify possible risk factors for this condition. Between January 1996 and December 2002, 122 cases of osteonecrosis were identified in this cohort. Importantly the rate of ostenecrosis appeared to increase with longer duration of antiretroviral therapy. In the group unexposed to potent combination therapy, the rate was 2.6 per 10,000 person years and increased in a linear fashion with each additional year of combination therapy up to a maximum of 17.5 cases per 10,000 py in those with at least 60 months of such therapy. Independent of duration of therapy, lower CD4-cell counts were also found to increase the risk for osteonecrosis. Ongoing analyses from this cohort are planned to examine whether specific antiretroviral agents further increase the risk for osteonecrosis. -- Judith Currier, MD, MsC

OPPORTUNISTIC INFECTIONS

The OI research presented was largely focused on tuberculosis. Many descriptive and epidemiologic studies from Asia, Africa, and Latin America emphasized that TB is the leading cause, worldwide, of morbidity and mortality in people living with HIV. Even Nelson Mandela, in his opening remarks to the attendees, emphasized the importance of considering both diseases together. In a plenary session, Sow presented an excellent summary of the current knowledge and unresolved questions surrounding the epidemiology and clinical care of HIV/TB-coinfected patients.

On a broader policy level, a study from the WHO estimated an annual added cost of $250 million for integrating HIV care into TB programs in the 34 countries receiving support from the Global Fund [TuOrd11209]. About 50% of the total was attributable to the cost of antiretroviral medications; another 30% was due to administrative costs; and only 20% of the cost was taken up by counseling and testing programs. This suggests that with further reductions in antiretroviral costs, the strategy would become substantially more cost-effective.

Several interesting studies using innovative preventive and therapeutic strategies were presented. In South Africa, pilot studies integrating antiretroviral therapy into existing directly observed therapy (DOT) TB programs were explored both in urban [MoPeB3355] and rural [MoOrB1014] areas. In these preliminary studies, the strategy appeared feasible and effective. The proper dose of efavirenz when combined with rifampin for TB treatment was examined in a study from Thailand in which coinfected patients receiving rifampin-based TB therapy were randomized to receive efavirenz at either 600 mg or 800 mg combined with 2 NRTIs. [MoOrB1013]. No significant differences were found in efavirenz plasma levels or HIV viral loads and CD4 counts. However, debate over 600-mg versus 800-mg doses will persist, as the study examined only one mid-dosage time point and Thai patients' average weight is lower than that of many populations. An interesting and potentially very important trial that compared prophylactic isoniazid versus placebo from birth among infants and children born with HIV in South Africa demonstrated a significant reduction in early mortality among infants receiving isoniazid [LbOrB12]. -- Gerald H. Friedland, MD

ADHERENCE

Just a few years ago, it was difficult to get adherence onto the agendas of major HIV meetings, but many adherence studies were presented this year, and adherence issues formed a major part of the plenary talk on antiretroviral therapy given by Dr. Ruxrungtham. Two issues were of particular interest: adherence in resource-poor settings and interventions to improve adherence.

Data documenting high levels of adherence in Africa were presented, further dispelling fears about inadequate adherence in the developing world [WeOr1323, WePeB5829]. However, two studies raise questions about the universality and sustainability of such high adherence levels as antiretrovirals become more generally available worldwide. In Brazil, where universal antiretroviral access has been available for several years, one study indicated that only 50% of patients from clinics in Rio de Janeiro were fully adherent (>95% of doses taken) as determined by self-report [WeOrB1319]. Nonadherence was significantly associated with perceptions of side effects, inadequate information about medications, and insufficient interaction with prescribing doctors (who, in a separate questionnaire could predict adherence in only 40% of patients). In a study among Senegalese patients who had been on antiretrovirals for more than 3 years, adherence was very high initially, but declined over time [WeOrB1320]. Both of these studies underscore the continued need for close attention to adherence and for well-documented strategies for enhancing and maintaining high levels of compliance with antiretroviral regimens.

Another presentation reported a meta-analysis of the world's literature on randomized controlled trials to enhance HIV-medication adherence [WePeB5819]. Of 13 RCTs published as of January 2003, 6 demonstrated significant improvements in adherence; in 3 of these studies, the gains were maintained at follow-up, but no study demonstrated an intervention benefit on virologic or clinical indicators. Clearly, research on interventions to improve adherence is still early in its development.

Several new rigorously executed and encouraging intervention trials were presented at the conference. In one RCT, a nurse/health-educator team performed home- or community-based visits to support adherence, with significant improvement in recipients' adherence and virologic measures compared with controls [WePeB5783]. In another RCT, modified directly observed therapy (DOT) in out-of-treatment drug users showed impressive adherence benefits, virologic benefit at 6 months, and continued immunologic benefit at 12 months after transition to self-administration [ThPpB2093]. Finally, in the largest RCT of an adherence intervention yet performed, patients were randomized into four groups: (1) training from a nurse schooled in adherence coaching ("medication manager"), (2) receipt of an electronic-reminder bracelet, (3) receipt of both coaching and bracelet, or (4) receipt of standard support for adherence [LbOrB15]. Each intervention was compared to the number 4, which served as the control. After almost three years of implementation and follow-up, sustained adherence and immunologic and virologic benefit was seen with the medication manager while the electronic reminder alone actually performed worse than the control population. -- Gerald H. Friedland, MD

PREVENTION

Unlike past conferences, where prevention coverage was evenly spread over the duration of the meeting, this year's presentations on the topic were given primarily on one day. In keeping with the "Access for All" theme, a strong focus on women emerged at the meeting, particularly on woman-controlled HIV prevention interventions. However, few data of any substantial interest were presented on particular candidate vaccines or microbicides.

An entire plenary session was devoted to expanding options for prevention. Quarraisha Abdool Karim discussed the opportunities and challenges in reducing heterosexual HIV transmission. To reduce the probability of sexual exposure to HIV, male and female condoms are the most well-established options. Other promising prevention options include postexposure antiretroviral prophylaxis, control of herpes simplex virus 2 infection, and male circumcision, as well as vaccines and microbicides. Although all of these approaches have great potential, they remain unproven.

In her plenary presentation, Zeda Rosenberg outlined the mechanisms of action of the various classes of microbicides and of antiretrovirals that might be used as microbicides (such as UC-781 and tenofovir). She expressed renewed optimism that an effective microbicide will be developed within the next 5 to 10 years. Jose Esparaza, in his plenary address, described the challenges posed by the lack of correlates of protection for vaccine development and described the efforts of the HIV Vaccine Enterprise -- the newly created coordinating body for HIV vaccine research. Siripon Kanshana, the final speaker in the prevention plenary session, presented encouraging evidence from several studies in Thailand that combining peripartum AZT with a single dose of nevirapine during labor is effective in reducing mother to child transmission (see also resistance section).

Access to clean syringes for injection-drug users is still poor in many countries, but there are signs that the situation is improving. Data from China and Vietnam showed that distribution of clean needles and syringes, through their voucher programs, and community education on drug-related dangers were broadly accepted across key community sectors, including the police [TuOrC1113]. HIV transmission has not stopped completely in these countries, but the number of new infections has decreased [WePeC6045]. Although still in the testing and acceptability phases, Colon showed that new drug-preparation devices, such as needle-less syringes, hold promise for reducing HIV transmission among injection drug users in many settings [TuOrC1112].

Access to testing has emerged as a key barrier to both prevention and treatment. Several innovative approaches to increasing the uptake and acceptability of testing are under investigation, including ways of reducing time-to-results [TuPeC4737, WePeC6037], tailoring access times to clients' needs, recruitment of couples [TUPeC4923], and introducing voluntary counseling and testing in detoxification centers [WePeC5999].

Several presentations focused on prevention efforts in men who have sex with men. Although a study performed in a San Francisco cohort of men who have sex with men indicated that risky sexual behavior has decreased [WePeC6088], several studies in developed and developing countries have shown an increase in HIV incidence rates, particularly among young men. Several speakers highlighted the need to shift the focus of prevention strategies in this group.

Risk for HIV in prison has also been a topic of intensive investigation, with data emphasizing the need for cooperation between health and corrections authorities to improve access to care, reduce stigma, increase prevention efforts, and improve quality of life for HIV-infected inmates [C10267]. In some countries, nongovernmental organizations are collaborating with prisons to establish prevention programs that include education and even harm-reduction strategies, a model that has proved mutually beneficial [WeOrE1295].

Several reports described prevention efforts aimed at youth. Akers described an HIV/AIDS awareness program among university students that demonstrated the need for new, more engaging strategies to keep the attention of young people, correct their mistaken beliefs, and motivate them to want to increase their involvement in HIV prevention [TuPeC4783]. Several reports focused on broadcast programs aimed at delivering prevention messages via mass media. One such study described a project in Nigeria that combines the performing arts with comprehensive life-skills education including the proper use of condoms and contraceptives and information about HIV/AIDS and other sexually transmitted diseases, with the aim of changing sexually risky behavior [TuPeC4734].

Overall, the conference focused more on prevention than previous meetings. This new emphasis, combined with new rollout programs on antiretroviral therapy, should provide substantial benefit for the regions and populations most affected by the epidemic. -- Salim S. Abdool Karim, MD

CONCLUSION

For the first time in the history of the HIV/AIDS pandemic, The XV International AIDS Conference was exactly on target in its location, orientation, and focus. Some attendees expressed disappointment at the relatively limited number of basic science presentations. Although treatment breakthroughs and vaccine successes are the most sought-after news, worldwide expansion of access to the potent therapy and proven prevention strategies that are already available is a challenge of at least equal importance. Thus, the "Access to All" theme and the recognition that this is a political and economic as well as medical issue were appropriate. Reflective of this, in addition to clinicians, scientists, and community members, the meeting was attended by heads of state and leaders at cabinet, government, and international agency levels. And though the cameo appearances of entertainment stars lent little to the content or gravity of the meeting or to worldwide needs, their involvement could bring resources and attract more attention to HIV prevention and treatment issues. Many promises were made, including Thailand Prime Minister Thaksin Shinawatra's declaration that treatment will be available for all in Thailand. If these and other promises are kept, the world community will view the Thailand meeting as a watershed in the struggle against AIDS.

— The ACC Editorial Board

Published in AIDS Clinical Care September 1, 2004