The 12th Conference on Retroviruses and Opportunistic Infections (CROI) was held in Boston, February 22-25, 2005, with David D. Ho, MD serving as Chair of the Scientific Program Committee. Workshops, symposia, plenary sessions, oral abstracts, and poster presentations provided the attendees with the latest updates on a variety of topics. In this month's Snapshots, we provide an overview of some of the material covered at the meeting.
HIV CARE OUTCOMES IN THE UNITED STATES AND GLOBALLY
Dr. Rochelle Walensky of Harvard Medical School estimated the cumulative survival benefits of highly active antiretroviral therapy (HAART) and opportunistic infection prophylaxis in the United States from 1989 (when only zidovudine was available) to 2003 (when 20 agents were available). She reported that 2 million years of life have been saved as a direct and attributable result of progress in HIV care and that HAART can lengthen the lifespan of patients with AIDS by nearly 15 years; this kind of survival benefit exceeds that which can be achieved for patients with other chronic diseases including chemotherapy for breast cancer, bypass surgery for coronary artery disease, and bone marrow transplantation for non-Hodgkin's lymphoma. In addition, the use of antiretroviral therapy (initially zidovudine monotherapy) has almost eliminated vertical HIV transmission for those pregnant women who are identified and treated. She concluded that efforts should be aimed at promoting and financing routine, voluntary HIV screening for all adults in the United States, noting that approximately 30% of the 900,000 persons infected with HIV in this country are unaware of their infection. Data emerging from developing countries are also encouraging. Francois Dabis from France presented an analysis of clinical databases from 4 continents comparing low-income and high-income countries. Immunologic and virologic responses appeared to be similar for HIV-positive individuals. However, mortality was higher in the developing countries, possibly because a larger proportion of patients in the 12 countries examined presented with severe opportunistic infections. Unfortunately, the World Health Organization's "3 by 5" mandate calling for 3 million people living with HIV/AIDS in developing countries to receive antiretroviral therapy (lamivudine/stavudine/nevirapine) by 2005 will, in all likelihood, fall short of its goal. Only 700,000 of the 5.8 million people worldwide currently in need of care receive HAART.
EARLY RESISTANT HIV INFECTION
Following the February 11, 2005 announcement by the New York City Department of Health and Mental Hygiene and the accompanying media frenzy, the details of the New York City patient who was diagnosed with early infection with a highly resistant HIV strain associated with rapid progression were of significant interest to CROI attendees (poster 973B and oral presentation by Dr. David Ho). The case report detailed a gay male in his late 40s infected within 4 to 20 months of his December 2004 diagnosis (+ detuned HIV antibody) with a CD4 lymphocyte count of 80. Strikingly, resistance testing demonstrated 21 mutations and phenotypic sensitivity to only efavirenz (although he had 2 NNRTI mutations) and enfuvirtide. His virus was subtype b, exhibited dual tropism (for the chemokine co-receptors CCR5 and CXCR4), and demonstrated a replication capacity of 136% (typically multidrug resistant strains lose replicative fitness). Speculation on the contribution of his having participated in high-risk (anal) sex with multiple partners while using methamphetamine added to the report's complexity. The phenomena of transmitted drug resistance and rapid disease progression were well documented before this report, but the case has alerted healthcare workers and patients that HIV infection continues to be a serious disease which can be largely prevented by avoiding risky behavior. The ongoing attempt to identify linked cases (the patient had some 100 contacts several months prior to his suspected "acute retroviral syndrome" in November 2004) may help clarify whether this case is indeed a new "super AIDS bug."
The World Health Organization continues to struggle with its guidelines for the prevention of mother-to-child transmission in resource-poor areas-a committee will revisit the issue this spring in Geneva. The use of single-dose nevirapine (SD-NVP) has been widespread because of its availability, low cost, and effectiveness; however, reports of NVP resistance, potential limitations of future treatment options, and better results using combination regimens are forcing a re-examination of recommendations. By standard genotyping, SD-NVP selects NVP-resistant variants in 30% to 50% of mothers. Using plasma samples from South African women in an allele-specific RT-PCR assay (for 103N and 181C), Sarah Palmer from the National Cancer Institute demonstrated that nonnucleoside reverse transcription inhibitor (NNRTI)-resistant variants can still be detected in the majority (69% one to four months after treatment discontinuation) of women after standard genotyping becomes negative. Despite NVP resistance and the potential for increased HAART virologic failure, Daniel Westreich from The University of North Carolina at Chapel Hill presented a model that suggested that the long-term survival from the effect of NVP resistance might be small if and when there is 100% access to HAART in developing countries. Francois Dabis reported on combination therapy in a cohort of pregnant women from the Cote d'lvoire; the women started zidovudine (ZDV) + lamivudine (3TC) at ≥32 weeks of gestation, received an extra dose of ZDV + 3TC and SD-NVP at the onset of labor, then ZDV + 3TC for 3 days postpartum (neonates received ZDV for 7 days and SD-NVP on Day 2). The transmission rate was only 4.7% at 6 weeks, and only 1.14% of women developed NVP resistance. An Italian study (presented by Leonardo Palombi) named DREAM (Drug Resource Enhancement against AIDS and Malnutrition) demonstrated that using HAART regimens in Mozambique pregnant women (on average for 73.9 days) helped protect mothers' health, lower HIV transmission (4.1% and 6.1% at 1 and 6 months, respectively), and avoided a high rate of drug-resistant mutations (but still 15%).
SALVAGE ANTIRETROVIRAL AGENTS
There is a need for effective salvage therapy in highly experienced patients failing currently available HAART. Tipranavir is the first nonpeptide protease inhibitor (PI) and has good activity against PI-resistant viruses. Its New Drug Application was submitted to the FDA on October 25, 2004, and its Expanded Access Programme was announced December 1, 2004. The RESIST-1 and RESIST-2 studies are ongoing phase 3 multinational, open-label trials comparing tipranavir/ritonavir with ritonavir boosted comparator protease inhibitors. Patients were heavily experienced (exposed to 3 or more antiretroviral classes but with 2 or less PI mutations at positions 33, 82, 84, or 90 [referred to as "UPAMS" or "PRAMS" for universal protease or protease resistant associated mutations]), on average, had used 12 antiretrovirals prior to study entry, and had viral loads >1000 m/L. Treatment response was defined as a ≥1 log decrease in viral load from baseline and was enhanced in patients also receiving enfuvirtide (the fusion inhibitor T20). At Week 24, the treatment response in patients using enfuvirtide was 25.8% in the comparator arm compared with 58.2% in the tipranavir arm. Response decreased to 16.9% in the comparator arm and 34.9% in the tipranavir arm without enfuvirtide. The protease inhibitor TMC114 will be boosted with ritonavir (600/100 mg BID) in phase 3 studies after the report of results from a 24-week interim analysis of two 96-week multinational phase 2 dose-finding trials. Richard Haubrich concluded that TMC114/r demonstrated "unprecedented" efficacy in 3 class-experienced patients with limited treatment options. TMC114/r resulted in a 97% reduction in viral load (mean change from baseline was −1.85 log10) compared with 16% (mean viral load change of −0.27 log10) with currently available protease inhibitors. As with tipranavir, patients who received enfuvirtide as part of their regimen achieved the best virologic response. Additional salvage agents highlighted at the conference included TMC278 (a novel NNRTI active against currently available NNRTI-resistant strains) and PA-457 (the first maturation inhibitor).
NOVEL CLASS OF ANTIRETROVIRAL AGENTS-SURFACE CO-RECEPTOR INHIBITORS
A symposium convened by Barbara Weiser and Cecelia Cheng-Mayer and several abstracts described the developing area of chemokine receptor blockade as an antiretroviral modality. Due to the critical role of the CD4 positive, T lymphocyte CCR5 co-receptor in HIV infection, this is an attractive cellular target for drug inhibition. A series of small molecules have been developed that appear to inhibit a wide range of isolates of HIV by this mechanism. Despite their apparent extracellular locus of action, drug pharmacokinetics do not appear to adequately explain their pharmacodynamics. High affinity for, and prolonged occupancy of, receptor sites may explain this discrepancy and result in infrequent dosing of some of these agents. Among multiple resistance mechanisms to these compounds is the possible selection of HIV strains capable of using the alternative CXCR4 co-receptor. Emergence of such strains is of concern as they may be associated with rapid progression of clinical HIV infection (formerly called "syncytium inducing"). Current and planned investigations will assess the potential for this complication. CCR5 inhibitor compounds under study include SCH-417690, UK-427857, and GW-873140. All have demonstrated 1.5 log drops in HIV RNA when used alone in phase 1 clinical investigations. Ongoing research is aimed at defining the clinical efficacy and safety of these agents and their utility as components of a multiclass antiretroviral regimen.
ADHERENCE TO ANTIRETROVIRAL THERAPY
CCTG 578 randomized 530 patients in a 1:2:2 distribution to one of two adherence interventions or usual clinical care. Interventions included 5 weekly cognitive, behavioral counseling sessions with or without a 2-week pretreatment period taking placebo doses. Using a standard of adherence to 90% of doses over 12 weeks by electronic measure, the study demonstrated a benefit of some intervention over none (82% vs. 72%), without showing a benefit between interventions. Adherence studies (abstracts 600, 626, 627, and 629) from Brazil, Uganda, Nigeria, and India all supported the role of caregiver and family support in optimizing adherence. Using unscheduled pill counts and electronic vial caps in 110 patients, Bangsberg et al, found that nonnucleoside reverse transcriptase inhibitor (NNRTI) therapy was frequently successful virologically at less than 95% adherence, in contrast to single PI-based regimens. The best chance of virologic suppression in all regimens was seen with near-complete adherence. A retrospective chart review by Kyle Petersen of 78 patients taking boosted and unboosted atazanavir (ATV) led researchers to conclude that an increase of ≥0.3 mg/dL total bilirubin at first follow-up visit was associated with adherence to and virologic suppression with the drug (sensitivity 81%, specificity 61%). Bartlett and co-workers' updated meta-analysis of factors associated with response in clinical trials involving treatment-naive patients (1994-2004) demonstrated an apparent association of pill burden with virologic outcome, however, it was not statistically significant on multivariate regression analysis.
Despite recent advances, antiretroviral therapy continues to be complicated by clinically significant drug-drug interactions. ATV demonstrates pH-dependent solubility and requires acid for absorption. Sangeeta Agarwala from the Bristol-Myers Squibb group formally presented data on the interaction of omeprazole and ATV. Large reductions in ATV AUC of 72% to 78% were demonstrated and not corrected with either higher ATV dosing or the concomitant use of cola drinks. The conclusion was to avoid proton pump inhibitors with ATV. The use of antacids or histamine H2-receptor antagonists with ATV could also affect ATV absorption and spacing of agents would be important. Because many women use depomedroxyprogesterone (Depo-provera®) as contraception, it was reassuring that a group headed by Susan Ellen Cohn demonstrated no significant interactions with NNRTIs or nelfinavir. Concurrent therapy for HIV and tuberculosis has been difficult because of drug-drug interactions. Induction of cytochrome P-450 hepatic oxidation systems by rifampin dramatically reduces protease inhibitor exposure. Similar to previous studies with lopinavir, Burger et al reported substantially reduced ATV exposures when coadministered with rifampin. They also noted that a dose increase of either ATV or rifampin or both was insufficient to overcome this effect and cautioned against the concurrent use of these 2 drugs. Efavirenz concentrations decrease only slightly (≤20%) under the influence of rifampin and investigations suggest that EFV 600 mg or possibly 800 mg daily will provide sufficient exposure to both agents. Dose-adjusted rifabutin should be employed in place of rifampin in patients taking concurrent protease inhibitors. Rodriguez-Novoa and colleagues confirmed the importance of CYP 2B6 codon 516 status on efavirenz pharmacokinetics, noting association with serum concentrations and neurologic adverse effects in 111 Caucasian patients (7% TT homozygotes). Haas and co-workers sought to associate CYP 2B6 allele status (G516T) with pharmacokinetics and predicted duration of efavirenz exposure post discontinuation in 152 patients. Despite substantial interpatient variability, longer duration of exposure was confirmed in homozygous TT allele subjects (19 days) versus heterozygous or wild-type (GG) subjects (7-9 days), confirming that conservative strategies of efavirenz discontinuation should be employed. The use of efavirenz in opioid-dependent patients may be limited by its metabolic induction effects, causing withdrawal in patients maintained on methadone. A pharmacokinetic study presented by McCance-Katz found that induction effects of efavirenz on buprenorphine were insufficient to cause opioid withdrawal. Further work may confirm the usefulness of efavirenz in opioid-dependent patients concurrently maintained on buprenorphine.
THERAPEUTIC DRUG MONITORING
Therapeutic drug monitoring in antiretroviral therapy remains controversial among many HIV clinicians. Proposed clinical uses of antiretroviral serum concentrations include adherence assessment and evaluation of altered drug clearance or interaction. Difficulty in interpretation may arise when these factors covary in the same patient. In a controlled trial of adherence interventions (CCTG 578), serum concentrations of NNRTIs and PIs correlated well with electronic adherence measures. In the same study, a panel of experts evaluated antiretroviral concentrations drawn 2 weeks following initiation of an antiretroviral regimen. Dose adjustments (usually increases) were recommended in 67 (38%) of 177 level sets evaluated. No assessment of outcomes was reported. Adherence measures were not predictive of the need to adjust regimens. Using an HPLC method, Gonzalez de Requena and colleagues sought to define the therapeutic range for ATV. They evaluated 38 naive and PI-experienced patients from the ATV expanded access program, finding that ATV trough concentrations ranging from 150 to 850 ng/mL were associated with the highest probability of virologic response and the lowest chance of unconjugated hyperbilirubinema. No assessment of adherence was reported. In another study (abstract 642), serum antiretroviral concentrations were measured weekly in 10 virologically suppressed patients to determine intraindividual variability in levels. No assessment of concurrent adherence was reported. Substantial variability was noted. Investigators were unable to correlate this variability with transient elevations in HIV RNA ("blips"). Finally, researchers at UCSF (abstract 744) reported correlating a liquid chromatography-mass spectrophotometry tandem assay for LPV and RTV in hair samples with virologic response in 46 patients.
Body-fat abnormalities can be found in 40% to 50% of HIV-infected patients. Fat redistribution after starting antiretroviral therapy is evident in 20% to 35% of patients within 12 to 24 months. Limb fat increases for the first few months then progressively declines. This lipoatrophy (particularly involving the face, buttocks, trunk, and limbs ["vein cabling"]) has been most commonly associated with the use of the thymidine analog nucleoside reverse transcriptase inhibitors (NRTI) stavudine and zidovudine. Therapy has been frustrating. A randomized, placebo-controlled Canadian trial (abstract 854) of rosiglitazone (4 mg once daily) over 24 weeks failed to show a benefit. However, several studies demonstrated that switching out the thymidine analog (to abacavir, tenofovir, or a nucleoside-sparing regimen) was beneficial; unfortunately, improvement was slow. Poly-l-lactic acid (PLA) is now commercially available (SculptraTM in the United States since 2004 and New-FillTM elsewhere since 1999) and is a biocompatible, biodegradable material which stimulates collagen formation, not fat. Anne Mijch, from an Australian team, reported that subcutaneous injections of PLA produced durable improvement over 6 months (in other studies up to 2 years) in facial appearance in 74% of individuals with moderate to severe facial atrophy. Spiral CT scan provided a quantifiable measure of severity and improvement. Although the report detailed few adverse effects and high patient satisfaction, the need for serial sessions (3-7) and high cost (potentially in the thousands of dollars) will limit the use of this therapy.
COMPLICATIONS OF ANTIRETROVIRAL THERAPY
One symposium covered cardiovascular complications of antiretroviral therapy. Updates on the D:A:D (The Data Collection on Adverse Events of Anti-HIV Drugs) study were presented by Wafaa El-Sadr expanding the database through February 2004 from the original publication in The New England Journal of Medicine in November 2003. This observational study, begun in 1999, now includes over 23,000 patients with more than 75,000 patient-years of follow up. The cohort experienced 277 myocardial infarctions with a 79% mortality rate. Although the risk of myocardial infarction was low, there was a demonstrated increased risk associated with antiretroviral therapy. Compared with controls with no therapy, the relative risk for a myocardial infarction was 1.94 after 1 to 2 years and increased to 4.38 after more than 6 years. Additional risk factors were also present in the population including smoking (47%), dyslipidemia (42%), lipoatrophy (23%), and hypertension (6%). Two studies (abstracts 872 and 873) suggested that HAART initiation (particularly with NNRTIs and lopinavir/ritonavir) was associated with increasing blood pressure. These findings support closely monitoring patients with risk factors for cardiovascular disease, encouraging lifestyle changes (diet, exercise, smoking cessation) and treating hypertension and dyslipidemia. The use of (n−3) polyunsaturated fatty acids (2 g of Maxepa® omega-3 fish oil concentrate softgels TID) significantly decreased triglycerides in antiretroviral-treated HIV-infected patients with good tolerance and with an absence of drug interactions. As part of the ATV expanded access program in Spain (abstract 850), patients with hyperlipidemia on primarily boosted protease inhibitors were switched to ritonavir boosted ATV. Significant improvements in plasma lipids were demonstrated without an increased risk of virologic failure. A series of presentations described genetic predispositions to peripheral neuropathy (association with mitochondrial haplotypes), abacavir hypersensitivity, nevirapine hepatotoxicity, and efavirenz central nervous system toxicity; unfortunately, screening for these associations is not available clinically.
Andrea James from the FDA presented a review of recent clinical trials which included abacavir and concluded that there was a higher rate of grade 3-4 hypersensitivity reactions (5% vs. 2%) with once daily dosing (EpzicomTM).
HIV AND HEPATITIS COINFECTION
One quarter to one third of HIV-infected persons in the United States are also infected with hepatitis C virus (HCV); increasingly, as HIV infection is better controlled, these patients are being diagnosed with chronic liver disease including cirrhosis and end-stage liver disease. Because HCV infection therapy with peginterferon and ribavirin combinations in this setting can result in sustained virologic responses of 11% to 62% (depending on genotype, dosing, race, etc.), more data are needed to identify patients who are treatment candidates. By using paired liver biopsies, Mark Sulkowski from Johns Hopkins University reported that within 3.5 years significant fibrosis developed in more than 25% of patients with minimal damage on their initial study. This finding emphasizes the importance of monitoring coinfected patients more closely for liver disease progression, perhaps repeating liver biopsies every 3 years. Although liver biopsy has been the gold standard to grade and stage hepatic damage, noninvasive tests such as serum biochemical markers (FibroSURE or FIBROSpect) and transient elastography (FibroScan) are being evaluated in the coinfected population. In an update on the APRICOT (AIDS PEGASYS Ribavirin International Coinfection Trial) study, another noninvasive test called the FIB-4 Index (Age [years] × AST/platelets × √ALT) was predictive of hepatic fibrosis and might have avoided liver biopsy in over 70% of patients. There was a disturbing report presented by Marie-Laure Chaix from France of 12 cases of homosexually transmitted HCV infection and the failure of standard interferon therapy to prevent chronic hepatitis (perhaps secondary to 10/12 being genotype 4). Although hepatitis B virus (HBV) infection occurs in 10% of HIV-infected patients, no therapy to date has been effective in clearing HBV infection. However, the antiretroviral agents lamivudine, emtricitabine, and tenofovir disoproxil are excellent inhibitors of HBV, and tenofovir was shown to be not inferior to adefovir dipivoxil (approved for HCV monoinfected patients) in lowering HBV DNA in ACTG A5127 as reported by Marion Peters. In another study, entecavir (a selective inhibitor of HBV polymerase approved by the FDA 3/29/05 and called BaracludeTM) resulted in rapid and effective reductions in HBV DNA and normalization of ALT within 24 weeks of therapy in patients with lamivudine-resistant HBV strains. Because entecavir has NO HIV activity, it can be used to treat coinfected patients without the potential for the development of HIV resistance. Of note, 5% of some cohorts are tri-infected (HIV/HBV/HCV).