Table of Contents

• Introduction

• Top 10 HIV Clinical Developments of 2007:

  1. Newsworthy New Drugs
     

  2. Presenting for Care Too Late and With Too Few CD4+ Cells
     

  3. Myocardial Infarctions: The Role of PIs and NNRTIs
     

  4. Metabolic Results From the A5142 Study -- Assumptions Gone Wild
     

  5. Sperm Washing Works
     

  6. Screening for Abacavir Hypersensitivity
     

  7. The Sad Failure of a Promising HIV Vaccine
     

  8. Crystal Meth Is Evil
     

  9. Survival With HIV: Great Expectations
     

  10. Risk of Cancer Is Related to Immune Function

• Runners Up

  • The Kids Are Not Alright: HIV and Street Youth in St. Petersburg

  • Aging and HIV

  • Underestimating HIV Incidence in the United States

• References

Introduction

In previous years, this annual "top 10" edition of HIV JournalView has looked back on the critical successes and failures of HIV research during the previous 12 months. This time, however, I will buck tradition and, instead of reviewing studies that have already changed the way we manage HIV infection, I will look at the prescient studies that point to an approaching shift in our long-established paradigms.

The articles and presentations summarized below provide a "crystal ball" glimpse at our evolving responses to an ever-changing pandemic. The studies focus on new drugs to treat HIV, new strategies to minimize treatment-related toxicities, and new challenges to containing HIV's spread. Collectively, these studies suggest we are at the dawn of a new day in HIV disease management -- a day on which we shed our old assumptions and reflect anew on the emerging possibilities and challenges that lie before us. Looking closely at the top HIV clinical developments of 2007, it is clear that the year to come will not be business as usual.

1. Newsworthy New Drugs

The last two years have been unprecedented in the history of HIV medicine. Never before have there been so many new drugs from old and new drug classes approved at one time.

The release of any one of these agents would have been major news for both clinicians and patients, but the availability of two new drugs with wildly novel mechanisms of action, as well as two new agents from existing antiretroviral classes, all capable of suppressing drug-resistant virus, have many heads a-spinning.

Reverberations from the introduction of these newest antiretrovirals will continue to unsettle the HIV treatment landscape for the foreseeable future as the critical mass of new agents now makes it possible to creatively craft treatment regimens with an excellent chance of being able to control HIV long term in patients who have had one or two HIV cocktails too many. These agents have helped many patients to suppress the virus for the very first time and they have given many a new lease on life.

Refreshingly, these new drugs are true advances in HIV therapeutics and not just "me too" medications that add little except another line on the HIV medication charts on our clinic walls.

Raltegravir (MK-0518, Isentress), maraviroc (MVC, Selzentry, Celsentri), etravirine (TMC125, Intelence) and darunavir (TMC114, Prezista) -- their awkward syllables are now rolling off our tongues (as is their spelling off our prescribing pens). New and impressive data were released in 2007 for these drugs, proving their effectiveness and presaging their eventual embrace by clinicians and patients, particularly treatment-experienced patients, eager for better therapeutic options. The results of the clinical trials involving these new drugs have been reported widely. Those hungering for the details can satisfy themselves easily at TheBodyPRO.com. The major findings related to each drug, as well as their implications for the future of HIV management, are discussed below.

Raltegravir (MK-0518, Isentress)

A review of:
Results of BENCHMRK-1, a phase III study evaluating the efficacy and safety of MK-0518, a novel HIV-1 integrase inhibitor, in patients with triple-class resistant virus. David Cooper, José Gatell, Jürgen Rockstroh, Christine Katlama, Patrick Yeni, Adriano Lazzarin, Joshua Chen, Robin Isaacs, Hedy Teppler, Bach-Yen Nguyen, for the BENCHMRK-1 Study Group. In: Program and abstracts of the 14th Conference on Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, Calif. Abstract 105aLB.

and

Results of BENCHMRK-2, a phase III study evaluating the efficacy and safety of MK-0518, a novel HIV-1 integrase inhibitor, in patients with triple-class resistant virus. Ron Steigbigel, Princy Kumar, Joseph Eron, Mauro Schechter, Martin Markowitz, Mona Loufty, Jing Zhao, Robin Isaacs, Bach-Yen Nguyen, Hedy Teppler, for the BENCHMRK-2 Study Group. In: Program and abstracts of the 14th Conference on Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, Calif. Abstract 105bLB.

and

Efficacy of raltegravir, an HIV integrase inhibitor, in combination with regimens containing enfuvirtide, darunavir, or tipranavir in patients with triple-class resistant virus: combined results from BENCHMRK-1 and BENCHMRK-2. Princy N. Kumar, David A. Cooper, Ron T. Steigbigel, Jing Zhao, Hedy Teppler, Bach-Yen Nguyen, for the BENCHMRK-1 and BENCHMRK-2 Study Groups. In: Program and abstracts of the 11th European AIDS Conference; October 24-27, 2007; Madrid, Spain.

and

Rapid onset and durable antiretroviral effect of raltegravir (MK-0518), a novel HIV-1 integrase inhibitor, as part of combination ART in treatment HIV-1 infected patients: 48-week data. Martin Markowitz, Bach-Yen Nguyen, Eduardo Gotuzzo, Fernando Mendo, Winai Ratanasuwan, Colin Kovacs, Hong Wan, Lucinda Gilde, Robin Isaacs, Hedy Teppler, and the Protocol 004 Part II Study Team. In: Program and abstracts of the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; July 22-25, 2007; Sydney, Australia. Abstract TUAB104.
View slides: Download PowerPoint

Raltegravir, which is an inhibitor of the HIV-1 integrase enzyme, has been studied in patients with extensive antiretroviral treatment experience and, to a lesser extent, in individuals naive to HIV therapy. The BENCHMRK trials are a pair of identically designed, placebo-controlled, randomized investigations that were conducted across the globe.^1,2 The aim of the studies was to examine the efficacy and safety of raltegravir (400 mg twice daily) when it was added to an optimized background therapy chosen by the investigator, based on both patient clinical history and drug resistance testing. The control arms of the trials received only an optimized background.

Collectively, 699 participants who had HIV drug resistance to each of the original three antiretroviral classes were enrolled and randomized in a 2:1 manner to the study drug versus placebo. This was an antiretroviral-experienced cohort with a median of one decade of HIV treatment under their belts. One quarter to one third of the participants had one or no drugs predicted to be active against their virus, based on baseline HIV genotypic resistance testing.

The early results of BENCHMRK-1 and BENCHMRK-2, which were presented separately at an evening late breaker session at the 14th Conference on Retroviruses and Opportunistic Infections (CROI 2007), demonstrated an impressively potent response to raltegravir -- far exceeding the expectations of veteran observers of such, so-called "salvage studies," where a 30% rate of viral suppression is considered a success.^1,2 For the sake of expediency, I will summarize the combined results of the two BENCHMRK studies, which were presented at the 11th European AIDS Conference in October 2007.³

Twenty-four week data were available for most of the participants and at this time, 63% of the participants who had been assigned raltegravir, compared to only 34% of those who had been assigned placebo, achieved a plasma HIV RNA level below 50 copies/mL (P < .001).

Not surprisingly, in both study arms, the virologic response increased in relation to the number of active agents used by the patients. When patients received two or more fully active drugs, they had remarkably high rates of success. Specifically, at least 68% of the participants saw their viral load fall to below 50 copies/mL by 24 weeks when their regimen contained a minimum of two of the following agents to which they were previously naive: raltegravir, enfuvirtide (T-20, Fuzeon) and/or darunavir + ritonavir (RTV, Norvir).

Click to enlarge

This is an astonishing rate of viral suppression among such a treatment-experienced cohort of patients. A follow-up analysis of the 48-week data from these studies, which was presented at the 15th Conference on Retroviruses and Opportunistic Infections (CROI 2008), demonstrated continued suppression of viremia at these rates.^4,5 Even among those with baseline drug resistance testing predicting no active agents in the optimized background, raltegravir was found to produce greater rates of virologic suppression that persisted during the first year of treatment. As would be expected, given the virologic data, the CD4+ cell count increased at week 24 by a mean 84 cells/µL with active drug, compared to 37 cells/µL with placebo.

Treatment discontinuation rates were low (less than 2%). Raltegravir-containing regimens appeared to be well tolerated in patients relative to those who were assigned to receive placebo. Although moderately severe creatine phosphokinase (CPK) and lipase levels were relatively more common in patients who were taking raltegravir, these developed in few of the participants.

Large comparative trials focusing on the use of raltegravir higher up the antiretroviral food-chain are being designed and implemented. One double-blind, dose-ranging study by Martin Markowitz et al found nearly identical rates (87% to 88%) of viral suppression to below 50 copies/mL at 48 weeks between the raltegravir 400 mg twice daily and efavirenz (EFV, Sustiva, Stocrin) arms among treatment-naive patients (n = 198) who were also receiving tenofovir (TDF, Viread) and lamivudine (3TC, Epivir).⁶

A more rapid early decay in viral load in the patients taking the raltegravir-based regimen than in those taking the efavirenz-based regimen produced a fair share of "oohs and ahs," and is interesting to think about, but the real-world significance remains unclear.

Drug-related adverse effects were practically non-existent with raltegravir, while efavirenz-assigned patients experienced the usual central nervous system (CNS) disturbances. Lipoprotein levels also appeared to remain untouched by raltegravir, in contrast to the rise in triglyceride and LDL (low-density lipoprotein) cholesterol levels seen with efavirenz.

Maraviroc (MVC, Selzentry, Celsentri)

A review of:
Efficacy and safety of once-daily (QD) compared with twice-daily (BID) maraviroc plus optimized background therapy (OBT) in treatment-experienced patients infected with CCR5-tropic-HIV-1: 24-week combined analysis of the MOTIVATE 1 and 2 studies. Roy M. Gulick, Elna van der Ryst, Harry Lampiris, Gerd Fätkenheuer, François Raffi, Jacob Lalezari, John F. Sullivan, Margaret Tawadrous, Irina Konourina, James Goodrich, Howard Mayer. In: Program and abstracts of the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; July 22-25, 2007; Sydney, Australia. Abstract WEPEB116LB.

and

Efficacy of maraviroc in combination with at least one other potent new antiretroviral drug: 24-week combined analysis of the MOTIVATE 1 and 2 studies. Elna van der Ryst, David Cooper, Irina Konourina, Michael Saag, James Goodrich, Margaret Tawadrous, P. Simpson, John F. Sullivan, Mike Westby, Howard Mayer. In: Program and abstracts of the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; July 22-25, 2007; Sydney, Australia. Abstract WEPEB115LB.

and

A multicenter, randomized, double-blind, comparative trial of a novel CCR5 antagonist, maraviroc versus efavirenz, both in combination with Combivir (zidovudine [ZDV]/lamivudine [3TC]), for the treatment of antiretroviral naive patients infected with R5 HIV 1: week 48 results of the MERIT study. Michael Saag, Prudence Ive, Jayvant Heera, Margaret Tawadrous, Edwin DeJesus, Nathan Clumeck, David Cooper, Andrej Horban, Lerato Mohapi, Horacio Mingrone, Gustavo Reyes-Teran, Sharon Walmsley, Frances Hackman, Elna van der Ryst, Howard Mayer. In: Program and abstracts of the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; July 22-25, 2007; Sydney, Australia. Abstract WESS104.
View slides: Download PowerPoint

The word "CCR5" looks like something that should have been written in red on the outside of a Soviet-era space craft. In fact, just a few years ago, no one had even heard of HIV co-receptors. Now, patients come into the clinic asking if they are "R5" or "X4" (okay, some patients).

HIV co-receptors were discovered not long after it became clear to a few people tinkering with HIV in the lab that simply having the virus bind to a CD4+ receptor does not permit entry into the cytoplasm.⁷ Then, in a wonderful example of how basic science translates to clinical drug development, then to product marketing, a cottage industry was born.

Maraviroc is the first drug to act as an antagonist to the CCR5 co-receptor. This is the co-receptor that HIV -- which has a predilection for lymphocytes -- requires to obtain entry into the T cell. Like raltegravir, maraviroc has been studied mostly in treatment-experienced patients who have few remaining antiretroviral options.

MOTIVATE 1 and 2 are identical twin trials of maraviroc (at two doses: 150-mg daily or 150-mg twice a day) versus placebo.^8,9 The trials included 1,076 patients with HIV drug resistance and/or experience with drugs from each of the three initial antiretroviral classes. All participants received an optimized background of antiretrovirals. Dosage adjustments of maraviroc were made at entry because of known interactions between this drug and many other antiretrovirals.

In each trial, patients were screened prior to study entry to determine if their virus was predominantly CCR5 tropic -- if it wasn't, the patients were excluded. Importantly, 44% of the patients screened for this trial had dual/mixed HIV populations or CXCR4-tropic virus -- these are HIV strains that can enter CD4+ cells without using the CCR5 co-receptor and, therefore, are unaffected by this co-receptor blocker. Further, some people who were originally determined to have only CCR5-tropic virus when screened for the study had dual/mixed or CXCR4-tropic virus later found in their blood when tests were conducted at the study initiation.

Click to enlarge

At 24 weeks of therapy, the viral load was below 50 copies/mL in 48% of participants who had been assigned to maraviroc (twice daily) versus 24.6% of participants who had been assigned to placebo (P < .0006). The once-daily maraviroc arms did slightly less well than the twice-daily dosing of the drug.

A mantra to be repeated for all of these salvage drugs is "the more active drugs used, the more active the regimen." As in the BENCHMRK trials, patients who harbored virus with resistance to a greater number of antiretrovirals had lower response rates.

Of the patients who experienced virologic failure, CXCR4- or dual/mixed-tropic virus was detected at the time of failure in about 60% of patients who had been treated with maraviroc compared to 6% of those who were on placebo. Sequence analysis of baseline blood specimens from a subset of 20 of these patients indicates that all harbored undetected CXCR4-tropic virus at baseline, suggesting that rather than switching from CCR5-using virus to CXCR4-using virus, these patients had the CXCR4-using virus all along, but it had not been detected by the tropism assay.

Such salvage studies are messy in terms of detecting drug-related toxicities since a variety of agents are included in patients' background regimens. However, in these trials, there was no strong toxicity signal to indicate any serious problem with maraviroc. The detection of several malignancies among patients who were receiving another investigational CCR5 antagonist, vicriviroc (SCH 417690, SCH-D), has led to concerns regarding the development of cancers with drugs of this class. In the MOTIVATE trials, only 11 malignancies were reported to have developed on study, six in the placebo-assigned patients. Whether there is any association between co-receptor antagonists and malignancies will be best addressed during post-marketing surveillance.

The observation that, during the natural history of HIV, the viral swarm generally starts out being CCR5-tropic with CXCR4-tropic virus emerging during later stages of the disease has led many researchers to consider the use of maraviroc as initial therapy for HIV infection. This was done in the MERIT trial by Michael Saag et al, which pitted maraviroc against efavirenz in 721 antiretroviral-naive patients who were also assigned to zidovudine/lamivudine (AZT/3TC, Combivir).¹⁰ The study found that the CCR5 antagonist was "not non-inferior" to the non-nucleoside reverse transcriptase inhibitor (NNRTI) in suppressing plasma HIV to below 50 copies/mL after 48 weeks of therapy.

In English, this means that the success rate of maraviroc fell short of the boundaries that were established when the study was designed that defined what would be considered a statistically meaningless difference between the two study groups. In the case of this study, if the difference in virologic responses between the two arms was less than 10%, then the study treatment would be considered non-inferior compared to the comparator treatment.

Click to enlarge

In the MERIT study, 65.3% of the patients who were on maraviroc, compared to 69.3% of those on efavirenz, had a viral load below 50 copies/mL at 48 weeks -- results that look very similar; however, the lower boundary of the 97.5% confidence interval of the difference between the two arms was -10.9%, which is just below the 10% cutoff.

However, a recent analysis from CROI 2008 looking at viral tropism at study entry and response to study regimen suggests that the subpar response to maraviroc compared to efavirenz was driven by the inclusion of patients with dual/mixed virus at treatment initiation (i.e., individuals who screened R5-tropic, but by study entry had dual/mixed virus detected).¹¹ Among those who remained R5-tropic between screening and study entry, there was no difference in viral suppression rates. However, the difference in the proportion in each arm who discontinued study treatment due to lack of efficacy (11.9% for maraviroc and 4.2% for efavirenz), is not fully explained by the small number of patients with dual/mixed-tropic virus at study entry who were assigned the CCR5 antagonist.

Perhaps an indication of a real disparity in the potency of these two agents is that the difference Michael Saag et al found in virologic responses between the efavirenz-based regimens and maraviroc-based regimens widened among patients who had a baseline viral load of 100,000 copies/mL or greater.¹⁰ Interestingly, when the data were cut by geographic region, there was no difference in response by treatment arm among participants from North America and Europe, although maraviroc did not do as well among patients enrolled from Australia, South Africa and Argentina, a correlation between success with maraviroc and possibly the direction toilet water spins when flushed that should be investigated further.

CD4+ cell count increases and lipid profiles favored maraviroc, which was overall much better tolerated than efavirenz.

Darunavir (TMC114, Prezista)

A review of:
Efficacy and safety of darunavir-ritonavir at week 48 in treatment-experienced patients with HIV-1 infection in POWER 1 and 2: a pooled subgroup analysis of data from two randomised trials. Bonaventura Clotet, Nicholas Bellos, Jean-Michel Molina, David Cooper, Jean-Christophe Goffard, Adriano Lazzarin, Andrej Wöhrmann, Christine Katlama, Timothy Wilkin, Richard Haubrich, Calvin Cohen, Charles Farthing, Dushyantha Jayaweera, Martin Markowitz, Peter Ruane, Sabrina Spinosa-Guzman, Eric Lefebvre, on behalf of the POWER 1 and 2 study groups. The Lancet. April 7, 2007;369(9568):1169-1178.

and

Efficacy and safety of darunavir-ritonavir compared with that of lopinavir-ritonavir at 48 weeks in treatment-experienced, HIV-infected patients in TITAN: a randomised controlled phase III trial. José Valdez Madruga, Daniel Berger, Marilyn McMurchie, Fredy Suter, Denes Banhegyi, Kiat Ruxrungtham, Dorece Norris, Eric Lefebvre, Marie-Pierre de Béthune, Frank Tomaka, Martine De Pauw, Tony Vangeneugden, Sabrina Spinosa-Guzman, on behalf of the TITAN study group. The Lancet. July 7, 2007;370(9581):49-58.

and

Efficacy and safety of darunavir/ritonavir versus lopinavir/ritonavir in ARV treatment-naive HIV-1-infected patients at week 48: ARTEMIS. Edwin DeJesus, Roberto Ortiz, Homayoon Khanlou, Evgeny Voronin, J. Van Luzen, Jaime Andrade-Villanueva, J. Fourie, Sandra De Meyer, Melissa Haley, Eric Lefebvre, Carline Vanden Abeele, Sabrina Spinosa-Guzman. In: Program and abstracts of the 47th Interscience Conference on Antimicrobial Agents and Chemotherapy; September 17-20, 2007; Chicago, Illinois. Abstract H-718b.
View slides: Download PowerPoint

Although this protease inhibitor (PI) was approved in mid-2006, darunavir only became a real contender in an increasingly crowded antiretroviral drug class because of a pair of clinical trials in 2007. Darunavir had already earned its salvage therapy bona fides in a series of trials known as POWER 1 and 2.¹² Results from these trials demonstrated that darunavir could outperform the older PIs, when it is included in a regimen administered to patients who had been already exposed to the three original antiretroviral classes and who had virus that had developed PI resistance.

It was the study called TITAN in the darunavir trial trilogy that put this new PI in a new light. TITAN was an open-label study by José Valdez Madruga et al that was designed to examine the effectiveness and tolerability of ritonavir-boosted darunavir, compared to that of lopinavir/ritonavir (LPV/r, Kaletra), in patients who did not have extensive antiretroviral experience.¹³

Participants were required to have a viral load greater than 1,000 copies/mL while on any highly active antiretroviral therapy (HAART) regimen and be lopinavir (LPV) naive. As such, there was some variety as far as antiretroviral experience in this cohort of 595 patients. Almost half of the patients were triple-class experienced, but the vast majority continued to demonstrate phenotypic susceptibility to at least two nucleoside reverse transcriptase inhibitors (NRTIs) (92%) and at least four PIs (82%). All participants were randomized to twice-daily darunavir 600 mg + ritonavir 100 mg versus twice-daily lopinavir/ritonavir, along side an optimized background that did not include enfuvirtide. The soft-gel formulation of lopinavir/ritonavir was used by most of the patients who were assigned to this agent.

At 48 weeks, 71% of those who had been assigned to darunavir + ritonavir-based regimens, compared to 60% of those who had been assigned to lopinavir/ritonavir-based regimens, achieved a viral load below 50 copies/mL (P = .005).

Subset analyses were conducted looking at a variety of baseline variables and their influence on the outcomes. Most any way you slice the virologic results, darunavir, at the very least, tended to outperform lopinavir/ritonavir -- albeit the power of some of these analyses was limited by sample size.

Importantly, even with the exclusion of the 58 participants who had at least a 10-fold change at baseline in susceptibility to lopinavir/ritonavir, more participants who were receiving darunavir (70%) achieved a viral load below 50 copies/mL compared to those who were receiving lopinavir/ritonavir (63%). As far as tolerability, there was little difference seen between the two study groups.

The next trial to move darunavir up a notch was the ARTEMIS trial by Edwin DeJesus et al.¹⁴ Reaching for the golden ring of the U.S. DHHS (Department of Health and Human Services)¹⁵ assignment to the "preferred treatment" box for initial HIV therapy, the makers of darunavir studied the drug at a more user-friendly once-daily dosing of 800 mg with 100 mg of ritonavir against the stalwart standard of lopinavir/ritonavir in 689 treatment-naive patients. All participants also received tenofovir/emtricitabine (TDF/FTC, Truvada) and most of those assigned to lopinavir/ritonavir ended up on the Meltrex formulation of the drug.

Click to enlarge

A very impressive 84% of participants who were assigned to the darunavir + ritonavir arm and 78% who were assigned to lopinavir/ritonavir achieved a viral load below 50 copies/mL at 48 weeks of treatment (difference = 5.6%; 95% confidence interval [CI]: -0.1 to 11.3; P < .001). The a priori boundary for non-inferiority was ±12%, therefore, darunavir + ritonavir was non-inferior to lopinavir/ritonavir in this study.

Importantly, when stratified by baseline plasma HIV RNA, the virologic response rates among patients who had a viral load greater than 100,000 copies/mL were superior with darunavir + ritonavir compared to lopinavir/ritonavir (79% versus 67%, respectively; P < .05).

Both agents were well-tolerated, although treatment discontinuation due to adverse events was more common among patients who were assigned to lopinavir/ritonavir (7% versus 3%) and was often related to gastrointestinal events -- also more frequent with lopinavir/ritonavir. Rash, a potential adverse effect of darunavir, occurred in 3% of those assigned to this agent (and 1% of the lopinavir/ritonavir group).

Etravirine (TMC125, Intelence)

A review of:
Efficacy and safety of TMC125 (etravirine) in treatment-experienced HIV-1-infected patients in DUET-1: 24-week results from a randomised, double-blind, placebo-controlled trial. José Valdez Madruga, Pedro Cahn, Beatriz Grinsztejn, Richard Haubrich, Jacob Lalezari, Anthony Mills, Gilles Pialoux, Timothy Wilkin, Monika Peeters, Johan Vingerhoets, Goedele de Smedt, Lorant Leopold, Roberta Trefiglio, Brian Woodfall, on behalf of the DUET-1 study group. The Lancet. July 7, 2007;370(9581):29-38.

and

Efficacy and safety of TMC125 (etravirine) in treatment-experienced HIV-1-infected patients in DUET-2: 24-week results from a randomised, double-blind, placebo-controlled trial. Adriano Lazzarin, Thomas Campbell, Bonaventura Clotet, Margaret Johnson, Christine Katlama, Arend Moll, William Towner, Benoit Trottier, Monika Peeters, Johan Vingerhoets, Goedele de Smedt, Benny Baeten, Greet Beets, Rekha Sinha, Brian Woodfall, on behalf of the DUET-2 study group. The Lancet. July 7, 2007;370(9581):39-48.

and

Christine Katlama, José M. Gatell, Jean-Michel Molina, Monika Peeters, Johan Vingerhoets, Brian Woodfall. Pooled 24-week results of DUET-1 and DUET-2: efficacy of TMC125 (etravirine; ETR) in treatment-experienced HIV-1-infected patients. In: Program and abstracts of the 11th European AIDS Conference; October 24-27, 2007; Madrid, Spain.

It is no easy feat to compete for attention when flashy new drugs boasting novel mechanisms of action are being studied and approved, particularly when you are a member of a small antiretroviral class dominated by one of the most popular drugs to treat HIV. However, etravirine just may turn out to be the little NNRTI that could.

Claiming fame as the first in a new generation of NNRTIs that can suppress virus resistant to older drugs of this class, etravirine was tested in a duo of trials, appropriately named the DUET trials.^16,17 In these studies, more than 1,200 patients who had a history of at least one NNRTI and at least three PI resistance mutations at screening were randomized to receive etravirine or placebo plus a background regimen that had to include darunavir + ritonavir and at most two NRTIs. Therefore, these were studies of etravirine + darunavir + ritonavir + two NRTIs versus darunavir + ritonavir on top of an optimized background therapy.

Click to enlarge

Pooled results from the DUET trials were presented at the 11th European AIDS Conference in October 2007.¹⁸ They showed that, after 24 weeks, 59% of patients who were receiving etravirine and 41% in the comparator arm experienced a viral load drop to below 50 copies/mL (P < .0001) (combined data of DUET 1 and 2). Significantly, the baseline mutations that were associated with a decreased response to etravirine were identified and included V90I, A98G, L100I, K101E/P, V106I, V179D/F, Y181C/I/V, and G190A/S. Patients who had three or more of these mutations at baseline had a reduced response to etravirine. Yet, these mutations did not equally affect drug activity. Many of them typically occur in concert with other damaging mutations. For example, detection of G190S or V179F should lead to low expectations for activity for etravirine, while most patients harboring virus with G190A or Y181 may respond well virologically. The K103N mutation appears to have a limited effect on etravirine's antiretroviral activity.

As would be expected, given the virologic results, CD4+ cell count increases were greater in the etravirine arm. Rash was observed with etravirine, developing in 17% of patients who were randomized to the drug, compared to 9% of patients who were on placebo plus darunavir + ritonavir.

The Bottom Line

Even before these four new antiretrovirals were released, the possibilities they offered were being sensed by clinicians who had prescribed practically every possible permutation of antiretroviral combinations. The new drugs present a credible threat to the formulaic mindset of HAART: two NRTIs + PI or NNRTI. In 2008 we will learn how each of these upstarts becomes incorporated into practice.

Raltegravir has justly earned the lion's share of attention and thus far has been a relatively clean drug that more clinicians are itching to use. Its graduation to early therapy, especially as part of a second-line regimen following the failure of efavirenz/tenofovir/emtricitabine (EFV/TDF/FTC, Atripla), has progressed beyond the pre-contemplation stage and will certainly be used in this manner even while the clinical trials are on the drawing board.

In the case of darunavir, it is difficult not to imagine an increasing role for this agent in patients with little or no resistance, given the results of ARTEMIS¹⁴ and TITAN.¹³ Clinicians embraced atazanavir (ATV, Reyataz) + ritonavir with remarkably little comparative study among patients who were treatment naive. When a 400-mg tablet of darunavir becomes available, this drug will also offer once-daily dosing convenience. Certainly, darunavir + ritonavir will continue to be weighed against the co-formulation advantage of lopinavir/ritonavir and clinician and patient comfort with this well-known entity.

The recently reported CASTLE study comparing lopinavir/ritonavir versus atazanavir + ritonavir, which found these two PIs to be similar in efficacy for initial HIV treatment, provides additional information that clinicians can use when selecting among these excellent therapeutic options.¹⁹ A proposal for a super-study pitting darunavir + ritonavir, lopinavir/ritonavir and raltegravir against one another is working its way through the U.S. AIDS Clinical Trials Group (ACTG) and, although unlikely to yield data for a few years, will further refine our appreciation of the relative merits of first-line therapies.

What the world needs now is not only love, but an NNRTI that can mop up after efavirenz fails. The drum beat for TMC278 (rilpivirine), a once-a-day second-generation NNRTI, drowned out the buzz regarding etravirine, which, at two pills twice-a-day plus a host of drug-drug interactions, was a bit of an ugly duckling. But, with the delay in the start of a large treatment-naive study of TMC278, etravirine is looking pretty swan-like right now. It can be considered a "cherry-on-top" drug that can help boost a salvage regimen that is dependent on two agents to do the viral suppression heavy lifting. It is also not too far-fetched to consider etravirine's use as an initial treatment for those patients who would prefer an NNRTI, but who cannot take efavirenz. However, its pill count and drug-drug interactions may hold this agent back.

Lastly, where maraviroc will fit in remains unclear; although the drug clearly provides an answer, no one is completely sure of the question. CCR5-tropic virus predominates early in the HIV course, but the drug has been best studied in treatment-experienced patients with advanced disease. More than half of the patients for whom the drug is currently indicated have CXCR4- or dual/mixed-tropic virus detectable on tropism assay, and some who seem not to may still have it as illustrated by the post-hoc analysis of the virologic response during the MERIT trial.¹⁰

The one initial study of maraviroc in treatment-naive patients raises some concerns about potency, especially in those with high baseline viral loads, but the drug's second-place photo-finish seems to have been a result of the inclusion of those with non-R5-tropic virus -- a problem that extends to real world use of this assay and drug. Add to this the exorbitant cost of the tropism assay (clocking in at almost $2,000 a pop!) and you are asking for trouble.

But not all is doom and gloom in maraviroc-land. Adam Smith's metaphoric "invisible hand" will likely lead other companies to offer tropism testing and a subsequent lowering of the shocking and disturbing current cost of the test. Further, maraviroc is a ritonavir-free drug with a low pill count -- enough said. More work will need to be done to help find the question to the maraviroc answer; meanwhile, we lay the two grand on the table, spin the wheel and watch it go around and around.

2. Presenting for Care Too Late and With Too Few CD4+ Cells

A review of:
Immune status at presentation to care did not improve among antiretroviral-naive persons from 1990 to 2006. Jeanne C. Keruly, Richard D. Moore. Clinical Infectious Diseases. November 15, 2007;45(10):1369-1374.

An issue of never-ending debate in the HIV treatment community centers on the question of when is the best time to initiate antiretroviral therapy. Yet, in fact, for most patients who present for HIV care, the question of when to start is absolutely moot. The average CD4+ cell count of those initiating care in the United States is 187 cells/µL according to some studies,²⁰ which is well below the 350 cells/µL threshold established in the U.S. DHHS guidelines for initiating HIV therapy.¹⁵

Further, about one quarter of the estimated one million persons with HIV infection in this country are unaware they are infected.²¹ Many of these people will not learn of their HIV-positive status until they have incurred severe damage to their immune systems and become seriously ill -- typically with preventable conditions.

With billions of dollars being spent via the Ryan White Comprehensive AIDS Resources Emergency (CARE) Act and other initiatives to improve access to care for those with HIV infection,²² a reasonable expectation is that more people should be getting tested, diagnosed and treated. However, this report by Jeanne C. Keruly and Richard D. Moore from Baltimore, Md., warns that whatever it is that we have been doing to identify people who are unknowingly living with HIV before they develop advanced disease and opportunistic infections is failing.²³

Tapping into the wellspring of the Johns Hopkins HIV clinical database, the investigators examined the characteristics of 3,348 patients who presented for HIV care at the Johns Hopkins AIDS clinic from 1990 to 2006. Over this period, the investigators found that the initial median CD4+ cell count at presentation fell significantly.

Click to enlarge

Among those who entered care during 1990 to 1994, the median initial CD4+ cell count was 371 cells/µL, but by 2003 to 2006, this had fallen to 276 cells/µL (P < .01).

This decline in median CD4+ cell count at presentation was observed among men, women, injection drug users (IDUs), blacks and whites, but not among men who have sex with men (MSM). Interestingly, during this period, the absolute number and proportion of new patients identifying their risk as MSM or IDU declined, while there was an increase in the number of patients infected heterosexually.

In a multivariate analysis, older age, being male and black race were independently associated with a lower CD4+ cell count on presentation, while injection drug use was associated with higher CD4+ cell counts. Paradoxically, IDUs were also found to have the longest time from diagnosis to clinical presentation.

The time from the diagnosis of HIV to presentation for HIV care decreased from 271 days in 1990 to 1994 to 196 days in 2003 to 2006 for the cohort as a whole, as well as among the subgroups of men, whites and MSM. Time from diagnosis to entry into care did not change significantly for women and heterosexuals.

The Bottom Line

These are sobering data that raise a number of concerns, if not an outright call to action. First, despite an increasing wealth of resources dedicated to enhancing access to HIV care over the 16 years of the study, patients presenting for medical attention are sicker now than they were in the early 1990s -- arriving to care with CD4+ cell counts similar to those reported among patients entering care in some underdeveloped countries in Africa. Thus, the funding that has led to dramatic improvement in the survival of HIV-infected people through HIV treatment has been much less capable of delivering that care to people at earlier stages of their disease.

Second, there were important disparities in disease severity when first accessing care among the racial, gender and risk category subgroups. Blacks presented with more advanced disease than non-blacks as did men compared to women. These troubling, and all too familiar, results are evidence of racial and, in the case of HIV testing, gender differences in utilization and access to care. That MSM were able to enter care at higher CD4+ cell counts may be explained by socioeconomic and other factors that were not explored in this analysis.

Third, the results highlight the tragic limitations of U.S. HIV testing policies, especially in areas where HIV may be more endemic. Although Baltimore may not be representative of the United States, Johns Hopkins is Maryland's largest HIV care provider and serves an urban population that is similar to the population in other HIV clinics across the country. This population is underinsured, disenfranchised and often poverty-stricken. As such, people who are served by this public hospital are at heightened risk for HIV and their HIV screening rates should be exemplary rather than lackluster.

Lastly, as pointed out in an accompanying editorial by Miguel Goicoechea and Davey Smith of the University of California, San Diego, this study shamefully illustrates the dramatic failings of the health care system in the United States -- where the ability to access quality care is correlated with the ability to pay for that care.²⁴

HIV testing, while generally free, is not hassle-free and requires an individual to appreciate his or her risk and be motivated to get tested (and return for the result). Stigma, competing needs and fear are obstacles to HIV testing. Discovering one's serostatus may not be considered a priority for men and women who are struggling to survive despite poverty, substance abuse and/or mental illness.

Hope comes in the form of the U.S. Centers for Disease Control and Prevention (CDC)'s Revised Recommendation for HIV Testing of Adults, Adolescents, and Pregnant Women in Health Care Settings.²⁵ These sensible recommendations advocate HIV testing of most persons in the United States who are accessing health care, and for changing standing state laws to abandon the written informed consent impediment to HIV screening and permit verbal opt-out consent. These changes are welcome and, if adopted by emergency rooms, urgent care centers and primary care clinics, will lead to early diagnosis of HIV -- affording an opportunity to prevent further immune destruction and the unwitting transmission of the virus to others. All that is left to do is to "just do it."

3. Myocardial Infarctions: The Role of PIs and NNRTIs

A review of:
Class of antiretroviral drugs and the risk of myocardial infarction. The D:A:D Study Group. The New England Journal of Medicine. April 26, 2007;356(17):1723-1735.

and

Cardiovascular risks of antiretroviral therapy. James H. Stein. The New England Journal of Medicine. April 26, 2007;356(17):1773-1775.

More than a few patients, innocently leafing through waiting room copies of Positively Aware or Poz, come across articles linking HIV therapies to heart attacks. Panicked, they enter their clinicians' offices apprehensively, worried that their HIV cocktails are laced with the stuff clogged arteries are made of. Unfortunately, differing conclusions reached by a host of studies looking at the risk of cardiovascular disease (CVD) and HIV and its therapies has left this subject confusing and unclear.

More than any of these studies, the large Data Collection on Adverse Effects of Anti-HIV Drugs (D:A:D) collaborative has convinced us that HIV therapies can indeed increase the risk of CVD. Collecting data prospectively from over 26,000 HIV-infected participants in 11 cohorts across Europe, the United States and Australia, the study had previously found that the incidence of myocardial infarctions increased during combination HIV therapy by approximately 17% per year.²⁶ This stepwise increase in myocardial infarction incidence with each year of treatment exposure was a powerful counterpoint to earlier studies -- some retrospective and usually with fewer participants -- finding no link between antiretrovirals and heart attacks.

At the time of the original D:A:D report,²⁷ many clinicians assumed that any excess risk of CVD seen in this study was due to exposure to PIs, since their use was widespread among the treated study participants. With additional follow-up time and increased use of NNRTIs, the D:A:D investigative group examined the relative relationships between PI- and NNRTI-based regimens and the risk of myocardial infarction.

Their latest analysis included 23,437 patients, 345 of whom had a myocardial infarction during 94,469 person-years of follow-up.²⁸ About two-thirds of the cohort had been previously exposed to PIs and one-third to NNRTIs.

Click to enlarge

Similar to their earlier observations, the investigators found that the risk of myocardial infarction increased by 16% per year of exposure to combination HIV therapy (adjusted relative rate, 1.16 95% CI 1.09 to 1.23). After adjusting for a number of CVD risk factors, but not lipoprotein levels, patients who were receiving PIs had the same overall relative rate of myocardial infarction of 16% per year (adjusted relative rate, 1.16 95% CI 1.10 to 1.23, P < .001) while people who had been exposed to NNRTIs had a 5% risk per year (adjusted relative rate 1.05 95% CI 0.98 to 1.13, P = .17). When lipids, hypertension and diabetes mellitus -- traditional risk factors for CVD -- were introduced to the model, the relative risk fell to 10% for PIs (P = .002) and 0% for NNRTIs (P = .92).

Nadir (and most recent) CD4+ cell counts and peak (and most recent) HIV RNA levels were not found to be associated with a patient's risk of myocardial infarction.

The Bottom Line

These important data, from a carefully designed and conducted observational study, suggest that any added risk of myocardial infarction during HIV therapy rests with PIs. It is clear that treatment with an agent in this antiretroviral class increased CVD risk. However, a take-home lesson that is often missing from the sleek and pithy coverage of this study was that the incidence of myocardial infarction was low (0.6% per year) among those exposed to PIs for more than six years and includes some patients who undoubtedly would have developed an infarct regardless of HIV therapy or even HIV infection. Further, other more traditional CVD risk factors were operative in this population and found to have a greater impact on risk relative to HIV treatment exposure.

Therefore, the contribution of HAART to CVD may be fairly minimal for patients without significant risk, whereas for HIV-infected people who are at high risk for CVD, adding an additional risk of a PI has to be seriously considered.

Whether NNRTIs are exonerated from contributing to myocardial infarction risk is less clear. As discussed in an accompanying editorial by University of Wisconsin cardiologist James H. Stein, the incidence rates of myocardial infarction according to years of exposure to each drug class overlap.²⁹ A trend in infarcts in the patients who are NNRTI-experienced increased up to the fourth year and then suddenly decreased. Further follow-up of this group may provide a better sense of the association between NNRTIs and heart attacks.

Adding a new twist on the HAART and CVD question, the D:A:D investigators recently reported a link between myocardial infarction and the use of the NRTI abacavir (ABC, Ziagen).³⁰ Sure to be a top 10 story of 2008, these results sparked considerable discussion and spawned retrospective analyses to explore this relationship.

While a link between PIs and CVD does not tax the imagination and is supported by other data, including cohort study results, the finding of an association between abacavir and heart attacks in this study is, as of yet, isolated and unexplained. Until more information becomes available, this provocative and important finding will continue to be a source of some worry for me and my patients, but it will not prompt a change in my treatment recommendations or my focus on reducing non-HAART-related CVD risk factors.

Much of the fallout from the published D:A:D study results linking PIs and CVD may have been attenuated somewhat by the equally important results of the SMART Study, during which an HIV treatment interruption was linked to increased rates of CVD compared to the rates seen while maintaining HIV therapy.³¹ Coupled, these studies paint a picture that includes a role for HAART in CVD that is second only to the adverse cardiac effects of HIV itself. Therefore, stopping HAART appears to be worse for cardiovascular health than taking and staying on HIV therapy.

4. Metabolic Results From the A5142 Study -- Assumptions Gone Wild

A review of:
Metabolic outcomes of ACTG 5142: a prospective, randomized, phase III trial of NRTI-, PI-, and NNRTI-sparing regimens for initial treatment of HIV-1 infection. Richard H. Haubrich, Sharon Riddler, Gregory DiRienzo, Lauren Komarow, William Powderly, Kevin Garren, T. George, James Rooney, John Mellors, Diane Havlir. In: Program and abstracts of the 14th Conference on Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, Calif. Abstract 38.

It's a mystery why bookmakers in Vegas have not been attracted by the astonishing frequency of the unexpected in major HIV clinical research studies in recent years. Case in point: the metabolic results of the ACTG 5142 study by Richard H. Haubrich et al.³² What odds would Louie, my uncle Henry's bet taker when I was a kid in New York, have placed on a big win for lopinavir/ritonavir in a contest of fat and lipids with efavirenz? Who in their right mind would have wagered a mortgage payment on efavirenz placing behind this PI when it comes to limb fat loss?

Prior to the release of the lipid and body fat data from this large head-to-head study of efavirenz and lopinavir/ritonavir in treatment-naive patients, we knew the following:

1. nucleosides, especially stavudine (d4T, Zerit), produce loss of subcutaneous fat from the limbs and face;
   
2. PIs might potentiate stavudine's peripheral fat wasting effects;
3. both PIs and efavirenz could lead to increases in trunk fat; and
4. stavudine, PIs and NNRTIs tend to raise triglyceride and LDL cholesterol levels (ritonavir-boosted PIs and NNRTIs also increase HDL [high-density lipoprotein] cholesterol).

Since those 10 minutes it took for the metabolic outcomes from A5142 to be presented in Los Angeles at the 14th Conference on Retroviruses and Opportunistic Infections (CROI 2007), nothing in the HIV metabolics world has been the same.

Click to enlarge

Click to enlarge

In those 10 minutes, we learned that:

• Zidovudine (AZT, Retrovir) also produced peripheral fat wasting at rates that are between those that have been observed with stavudine (worse) and tenofovir (better).
• When co-administered with an NRTI, efavirenz led to greater peripheral fat wasting than what was produced by lopinavir/ritonavir.
• When efavirenz and lopinavir/ritonavir were paired in an NRTI-sparing regimen, peripheral fat increased.
• Efavirenz- and lopinavir/ritonavir-based regimens similarly increased trunk fat.
• Efavirenz- and lopinavir/ritonavir-based regimens that included NRTIs produced similar increases in LDL and HDL cholesterol levels, but increases in triglyceride levels were greater with lopinavir/ritonavir.
• If you really want a patient to increase his or her lipoprotein and triglyceride levels, prescribe efavirenz and lopinavir/ritonavir together.

The Bottom Line

For HIV metabolic-niks, ACTG 5142 was a landmark study. Sure, some smart-alecky researchers will claim that they were not too surprised by these results since they remembered that lipids rose during the DuPont 006 study of efavirenz (this is the study on which the U.S. Food and Drug Administration [FDA] based its approval of efavirenz, back when we referred to efavirenz as DMP-266).³³ Others note that in the BMS 034 study of efavirenz versus atazanavir, visceral fat increased in both groups.³⁴

Even with all the retrospective chatter, the findings from this study have all the HIV metabolic experts scratching their heads. Among the questions sparked by this study:

• Is efavirenz, a drug not known to be toxic to mitochondria, directly contributing to fat loss?
• If so, is it exerting an effect independent of mitochondria?
• Why did the NRTI-sparing arm not experience fat loss?
• Is PI therapy protective against apoptosis of adipocytes?

In BMS 089, a study of atazanavir with and without ritonavir, in patients also receiving extended release stavudine, patients taking ritonavir were found to have experienced less fat loss of the limbs than those who had been receiving unboosted atazanavir.³⁵ (How ironic would it be if metabolic bad-boy ritonavir is found to actually protect against this most notorious of body shape changes!)

Not insignificantly, A5142 also made it clear that zidovudine can cause peripheral fat loss, which supports the DEXA (dual energy X-ray absorptiometry) results from the GS 934 trial, a study of zidovudine/lamivudine versus tenofovir/emtricitabine in treatment-naive patients who were also receiving efavirenz.³⁶

Beyond its specific scientific findings, as well as the spawned inquiries, the A5142 investigation emphasizes how blindsided we can be when we hold our assumptions too dear. We and our patients have assumed for years that most of the HAART-associated metabolic ills are caused by PIs. In fact, we are learning that this class of drugs holds no monopoly on such complications.

The major lesson learned from this trial is to look at the data before we leap to conclusions. Another lesson: A5142 represents clinical research at its finest. Sharon Riddler et al conducted a well designed study tackling a question that no one else would take on (this is only the second head-to-head, large clinical trial of a ritonavir-boosted PI versus efavirenz).

5. Sperm Washing Works

A review of:
Safety and efficacy of sperm washing in HIV-1-serodiscordant couples where the male is infected: results from the European CREAThE network. Louis Bujan, Lital Hollander, Mathieu Coudert, Carole Gilling-Smith, Alexandra Vucetich, Juliette Guibert, Pietro Vernazza, Jeanine Ohl, Michael Weigel, Yvon Englert, Augusto E. Semprini, for the CREAThE network. AIDS. September 2007;21(14):1909-1914.

Hardwired in our brainstems is the desire to eat, drink (especially double espresso lattes) and reproduce. HIV infection does not quell these primal directives and as more of our patients enjoy the prolonged AIDS-free survival afforded by HAART, many will want to also enjoy their very own "Mini Me."

Due to interventions to reduce mother-to-child transmission, 95% of children born to HIV-infected women will be uninfected.³⁷ However, for HIV-infected men in the United States, the reproductive options have been limited, causing many to make the difficult decision to father a child even while placing their uninfected partners at risk of infection.

A purportedly safer option is sperm washing -- a process wherein motile spermatozoa are isolated from the semen of an HIV-infected man for use in artificial insemination or some other assisted-reproduction technique. In the United States, the high cost of this procedure and the difficulty of finding clinics that perform it for HIV-infected men have made this an unrealistic option for many.

In addition, there have been few data regarding the safety vis-à-vis HIV transmission of this procedure.

A report, published in AIDS by Louis Bujan et al of eight European centers where sperm washing is available, details the outcomes of 533 pregnancies achieved with this method between 1989 and 2003.³⁸

The exact procedure that was used varied somewhat from center to center, but all involved segregating the spermatozoa from seminal fluid and potentially HIV-infected semen cells. This was followed by the testing of the washed samples for HIV genomes. The assisted-reproduction techniques included intrauterine insemination, in vitro fertilization and intracytoplasmic sperm injection.

Click to enlarge

A total of 1,036 serodiscordant couples underwent the procedure and in 84% intrauterine insemination was used. Pregnancy was known to have resulted in 533 (51.4%) women, resulting in 410 deliveries and 463 live births (29 twins and 13 triplets). Six months after the procedure, 967 women had a negative HIV antibody test; the results of post-procedure testing among the remainder of the women were not known. Therefore, there was no known case of female seroconversion following the procedure.

The Bottom Line

These results from a large number of European couples, albeit retrospective, indicate that sperm washing is effective for procreation -- a finding that will hearten many serodiscordant couples for years to come. The ability of the procedure to prevent female HIV infection is suggested, but not conclusive given the limitations of this study -- specifically, a significant proportion of women were lost to follow-up or did not have a follow-up HIV test and the sample size was small given the low risk of transmission.

Nonetheless, these are encouraging findings that will hearten the many couples interested in the procedure. Missing from the discussion section of this paper was any mention of the monetary costs involved. In the United States, sperm washing is only offered at specialized fertility centers and the price can be in the thousands of dollars -- this is in addition to the added cost of the assisted-reproduction techniques, which can be between $6,000 to $15,000 for in vitro fertilization, more than $400 for intrauterine insemination and more than $1,000 for intracytoplasmic sperm injection. In addition to its hefty, if not prohibitive, price tag, couples looking at some of these procedures need to consider a risk that exceeds that of HIV transmission: triplets! There were 13 sets of triplets seen out of 410 deliveries -- a rate of 3%. Of course, if the women are given no fertility drugs, there will be little risk of multiple births.

6. Screening for Abacavir Hypersensitivity

A review of:
PREDICT-1: a novel randomised prospective study to determine the clinical utility of HLA-B*5701 screening to reduce abacavir hypersensitivity in HIV-1 infected subjects (study CNA106030). Simon Mallal, Elizabeth Phillips, Giampiero Carosi, Jean-Michel Molina, Cassy Workman, Janez Tomažič, Eva Jägel-Guedes, Sorin Rugina, Oleg Kozyrev, Juan Flores Cid, Phillip Hay, David Nolan, Sara Hughes, Arlene Hughes, Daren Thorborn, Alastair Benbow. In: Program and abstracts of the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; July 22-25, 2007; Sydney, Australia. Abstract WESS101.

and

High sensitivity of HLA-B*5701 in whites and blacks in immunologically-confirmed cases of abacavir hypersensitivity (ABC HSR). Michael Saag, Rukmini Balu, Philip Brachman, Cindy Brothers, Britt-Nicole Stancil, Mike Mosteller, Paul Wannamaker, Denise Sutherland-Phillips, Elizabeth Phillips, Simon Mallal, Mark Shaefer. In: Program and abstracts of the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; July 22-25, 2007; Sydney, Australia. Abstract WEAB305.
View slides: Download PowerPoint

Since its approval in 1998, abacavir has been dogged by a 5% to 10% risk of hypersensitivity reaction (HSR) that has kept this NRTI from being a preferred first-line agent.³⁹ (Funny how a skull and crossbones label warning and a risk of life-threatening reactions on reintroduction can really put the kibosh on the enthusiasm for a medication.)

But, what if it were possible to identify patients who are not at risk for abacavir HSR? Would clinicians then warm up to this well-studied antiretroviral? It remains to be seen whether we're in for a new wave of abacavir-love for the first time since those carefree days when zidovudine/lamivudine/abacavir (AZT/3TC/ABC, Trizivir) was given as first-line therapy.

Click to enlarge

Two important studies have examined a strategy of screening patients for the human leukocyte antigen (HLA) variant B*5701. This specific allele is found worldwide with varying frequencies and has been determined to be associated with risk of an abacavir HSR. The use of genetics to tailor medical therapy is a dream that originated with the discovery of genetic variability in drug metabolism over 40 years ago. I still recall attending grand rounds at a large medical school to which I was applying where the presenter predicted that soon the cover of hospital charts would be labeled with patient specific codes based on their genotype analysis of medication metabolism. More than a couple of decades later, paper medical charts are becoming extinct and we have not made much headway in incorporating our rapid understanding of our genome into clinical practice. HIV is beginning to change that.

Simon Mallal et al rigorously and prospectively examined HLA-B*5701 screening in 1,650 individuals who were starting treatment with regimens containing abacavir in Australia, the United Kingdom, France, Italy, Germany, Spain and Romania.⁴⁰ In this trial, known as PREDICT-1, participants were randomized to two study arms:

Click to enlarge

• A control arm that started an abacavir-containing regimen with subsequent monitoring for evidence of HSR (n = 847).
• A screening arm that prior to the initiation of an abacavir-containing regimen underwent HLA-B*5701 testing (n = 803). Those without this allele were permitted to take an abacavir-containing regimen. Those who were screened and were found to have the allele were given a non-abacavir-based regimen.

Most of the participants were white, 12% were of African heritage and three quarters were male. Importantly, participants in the control arm also had HLA-B*5701 testing done, although these results were not available to them or their clinicians. Further, patients who had a clinically detected HSR underwent abacavir skin patch testing to determine if an immunological reaction to the drug could be elicited.

Click to enlarge

HLA-B*5701 testing was positive in 5.8% of the controls and 5.7% of the screening arm participants. During the course of the study, 7.8% of the controls and 3.4% of the screening arm participants had HSR diagnosed clinically (odds ratio = 0.4 95% CI 0.25 to 0.62; P < .0001). Better yet, when patients with clinically diagnosed HSR underwent skin patch testing, the proportion with "confirmed" HSR shot down to 2.7% in the controls and 0.0% in the screening arm (odds ratio = 0.03 95% CI 0 to 0.18; P < .0001). Note that the skin patch was tested on 100 abacavir-tolerant study participants who were HLA-B*5701 negative and none had a positive skin test. Therefore, not one patient who was screened for HLA-B*5701 and found to be negative later developed an immunologically confirmed HSR (negative predictive value = 100%).

Being HLA-B*5701 positive does not guarantee that HSR will develop and 25 of the 48 patients in the control arm with this allele did not develop (confirmed) HSR -- yielding a positive predictive value of 48%.

A second, retrospective case-control study by Michael Saag et al examined the utility of HLA-B*5701 screening among racially diverse groups of HIV-infected persons, since the frequency of this allele can vary with race.⁴¹ Known as the SHAPE Study, the investigators looked at 130 white patients and 69 African-American patients who were found to have clinical HSR and 202 white patients and 206 African-American patients who received abacavir-containing therapy without incident.

Click to enlarge

HLA-B*5701 positivity among the abacavir-tolerant groups was found to be 4% in white patients and 1% in the African-American patients. Of the white patients who had suspected HSR, 44% were later determined to be HLA-B*5701 positive while only 14% of African-American patients with HSR were. Confirmation with skin patch testing reduced the rates in both groups even further. Validating the skin patch technique, all those with a positive patch test were also HLA-B*5701 positive. Therefore, the sensitivity of HLA-B*5701 testing for HSR -- when skin patch testing was done -- was 100% for both white and African-American patients.

The Bottom Line

The incorporation of testing for HLA type to predict abacavir HSR is another example of the truly innovative nature of the researchers who are working in HIV medicine. The community of men and women dedicated to fighting HIV (along with patient activists) has helped to create a more responsive drug approval process, has rapidly developed new antivirals and has implemented clinical use of measures of viremia and drug resistance to guide therapy. Now, after much speculation regarding the potential of pharmacogenomics, we in the field of HIV medicine are looking at our patients' genotypes and acting on what we find.

Both the PREDICT-1 and SHAPE investigations provide impressive evidence that HLA-B*5701 is valuable when considering whether to prescribe abacavir to an individual. With this testing strategy, few patients will be wrongly denied access to this drug and many more that were at high risk for developing HSR will be able to avoid this often serious adverse event.

Fortunately, the test is reasonably priced, although one or two commercial laboratories have shamefully refused to unbundle testing for this specific allele from a more expense battery of tests looking at other HLA types. Some clinicians have questioned whether it is necessary to screen African Americans, given the relatively low frequency of the allele in this population.

Since race is less well defined as a biological categorization rather than a social construct, and since one cannot tell by looking at skin color whether HLA-B*5701 is present, there is no acceptable reason not to screen all persons in the United States for this HLA type before prescribing abacavir. On the other hand, there is little rationale for mass screening for the allele, say of all HIV-infected persons entering clinical care, as we do for G6PD. The cost of the HLA-B*5701 test is an order of magnitude greater than that looking for G6PD deficiency and the majority of patients tested will receive no benefit from the result -- unless of course they own stock in the commercial laboratories that process it. It is difficult to imagine a situation in which abacavir needs to be administered emergently to a patient and common sense suggests that screening should be reserved for those who are about to receive the drug -- a sane approach endorsed in the Jan. 29, 2008 update to the U.S. DHHS adult and adolescent HIV treatment guidelines.¹⁵

7. The Sad Failure of a Promising HIV Vaccine

A review of:
STEP trial: efficacy analyses. Michael Robertson. In: Program and abstracts of the HIV Vaccines Trial Network Conference; November 7-9, 2007; Seattle, Wash.
View slides: Download PDF

and

Efficacy results from the STEP study (Merck V520 Protocol 023/HVTN 502): a phase II test-of-concept trial of the MRKAd5 HIV-1 Gag/Pol/Nef trivalent vaccine. Michael Robertson, Devan Mehrotra, Dan Fitzgerald, Ann Duerr, Danny Casimiro, Julie McElrath, Dale Lawrence, Susan Buchbinder. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 88LB.
View slides: Download PDF

It has become clear over the years that we are not going to treat, safe sex, circumcise or abstain our way out of the HIV pandemic. Therefore, enormous hope has been pinned on developing an effective vaccine to prevent the acquisition of HIV infection. The HIV vaccine story is a long one and has been marked by fits of exuberant optimism and stunning defeats. The latest chapter in this saga portrays yet one more disappointment.

The STEP Study by Michael Robertson et al examined the ability of a promising vaccine containing an attenuated adenovirus type 5 (Ad5) as a vector plus synthetic gag, pol and nef HIV genes to prevent HIV acquisition in a randomized, placebo-controlled trial of 3,000 HIV-uninfected individuals who were considered to be at high risk for HIV because of their sexual behaviors. The trial participants hailed from the United States, Canada, Brazil, the Caribbean and Australia. The study was conducted by the makers of the vaccine and the federally supported HIV Vaccine Trials Network.

While one primary outcome of the study was to determine if the vaccine reduced HIV infections, another major endpoint was to gauge whether the vaccine had an effect on the viral load levels of individuals who did become HIV infected despite vaccination.

In September 2007 an independent safety monitoring board released the results of a planned interim analysis of the efficacy of the vaccine among 1,500 participants who were expected to have the best response to the vaccine due to low baseline levels of immunity to the adenovirus vector.⁴²

The vaccine did not prevent HIV acquisition. Among the 741 people who received all three planned doses of the vaccine, 24 cases of HIV were detected. Among the 762 who received placebo, 21 cases of HIV were detected. Similar results were seen among participants who had two doses of the vaccine or placebo. These findings led to a halt to further vaccinations in this study as well as in a trial of this vaccine taking place in South Africa, and in a small phase 1 domestic U.S. study.

Almost all of the HIV infections occurred in men and all but one of these men were MSM. Importantly, in further analyses of the STEP data, there was a trend toward more HIV infections among individuals who had pre-existing immunity to Ad5.⁴³ It had been thought that the adenovirus vector would elicit an immune response in people without pre-existing immunity to this virus, which would in turn enhance the response to the HIV genes incorporated into the vaccine.

The interim results suggest that participants with greater baseline immunity to Ad5 were actually more likely to acquire HIV infection if vaccinated.

The table below shows the rate of HIV infection among participants who were MSM, stratified by study assignment and baseline Ad5 immunity level.

The reason for this result is not yet clear, but may indicate that among people who had a prior history of adenovirus infection and subsequent brisk immune response to the virus, vaccination led to immune activation and thus an increased susceptibility to HIV infection.

Further, there was no difference in viral load set point observed between people who acquired HIV despite vaccination and individuals who became HIV-infected after they received the placebo.

The Bottom Line

The failure of this vaccine to prevent HIV infection was a profound disappointment -- particularly to the many people who had worked so hard to bring the world a realistic way to prevent HIV infection. The suggestion that the vaccine may actually increase susceptibility to HIV infection in some individuals was further salt in the wounds of these researchers, clinicians, advocates and patients.

After the interim STEP Study results were announced, similar trials of this vaccine were halted and the brakes were applied to studies of vaccines using a similar model. The HIV Vaccine Trials Network -- on the cusp of launching ambitious domestic HIV vaccine initiatives when the results became available -- is now looking at retooling these protocols in light of the STEP Study observations.

Clearly, additional analyses from the rubble of the STEP Study may help guide future vaccine efforts. It remains to be determined whether this or similar vaccines have any effectiveness in women or heterosexuals and whether there are factors, other than Ad5 immunity level, that predict response. In a fascinating post hoc analysis of the STEP Study presented by Michael Robertson et al at CROI 2008, circumcision was found to protect against HIV infection, reducing the risk of infection even among individuals who had Ad5 immunity at baseline.⁴⁴

While the HIV vaccine effort will continue -- as it has despite previous adversity -- the failure of this promising vaccine, combined with the recently reported rising rates of HIV transmission in the United States⁴⁵ and Eastern Europe⁴⁶ and the lack of efficacy of any vaginal microbicide, point to a long overdue need to seriously re-examine other HIV prevention interventions. Improving our ability to diagnose people with HIV infection, especially during acute infection when HIV may be more likely to be transmitted, must become a priority. Additionally, pre-exposure prophylaxis with antiretrovirals (i.e., PrEP) -- as the last man standing -- is becoming increasingly attractive and further study of this strategy is underway.

We need a renewed effort to determine effective ways to reduce high-risk behavior. Millions of dollars have gone into studying behavioral interventions to reduce the spread of HIV. It is time for this investment to yield dividends. The approaches that have been demonstrated to be effective should be expanded and those that are ineffective, abandoned.

The HIV Prevention Trials Network, a sibling of the HIV Vaccine Trials Network, is taking a lead in developing initiatives to increase HIV testing and reduce risk behavior. Together, these efforts may give us the time we need to contain this epidemic as we wait the long wait for a vaccine to come to our rescue.

Ultimately, the failure of prevention interventions sends a strong message that, at least in the United States, there is no valid excuse to not implement the U.S. CDC recommendations on HIV screening in health care settings.²⁵ Pediatric HIV has become nearly extinct following the implementation of HIV testing being offered by health care providers to pregnant women. Health care providers can lead the way in identifying those who are unknowingly living with HIV infection so that they can be counseled and treated to prevent further transmission. Every time an unknowingly HIV-infected individual passing through an emergency room or primary care clinic is missed, so too is the opportunity to protect not only this individual's health, but also that of their partners. Our health care system has to pull its head out of the sand and recognize the powerful potential of its clinicians to be an intervention to stop the spread of a virus that infects one out of every 300 of their patients.²¹

8. Crystal Meth Is Evil

A review of:
Tripling of methamphetamine/amphetamine use among homeless and marginally housed persons, 1996-2003. Moupali Das-Douglas, Grant Colfax, Andrew R. Moss, David R. Bangsberg, Judith A. Hahn. Journal of Urban Health. December 27, 2007.

and

Frequent methamphetamine use is associated with primary non-nucleoside reverse transcriptase inhibitor resistance. Grant Nash Colfax, Eric Vittinghoff, Robert Grant, Paula Lum, Gerald Spotts, Frederick M. Hecht. AIDS. January 11, 2007;21(2):239-241.

and

Bupropion for the treatment of methamphetamine dependence. Ahmed M. Elkashef, Richard A. Rawson, Ann L. Anderson, Shou-Hua Li, Tyson Holmes, Edwina V. Smith, Nora Chiang, Roberta Kahn, Frank Vocci, Walter Ling, Valerie J. Pearce, Michael McCann, Jan Campbell, Charles Gorodetzky, William Haning, Barry Carlton, Joseph Mawhinney, Dennis Weis. Neuropsychopharmacology. June 20, 2007.

A story: A few years back, the devil and his demons were sitting around a fire during their quarterly business meeting. The expansion of the HIV pandemic project they had been working on had hit a roadblock or two. The big guy listened as a minion reported that HIV was still thriving, but that in some sectors the spread of the virus may have been starting to wane -- for example, among MSM in the United States. In fact, among that subset of people, the rate of new HIV infections had reached a plateau -- and was even starting to decline.⁴⁷

As he scanned the U.S. CDC slides included in the presentation, the devil's eyes narrowed. His fists clenched. The bonfire in front of him flared. Glaring at the circle of demons, he demanded that a new plan be formulated at once to reverse this unacceptable trend.

One demon tremulously spoke up: "What if we could create something that people took? Some kind of drug, and when they take it, it increases their risk behavior?"

Everyone looked at him blankly. The devil snarled, "Yeah, right -- like there's any way we could possibly improve on crack cocaine."

The demon's face turned ashen. He stammered and wrung his hands.

Then it came to him. "What if the drug also made people really horny?" he asked.

Everyone stared. The devil leaned closer.

"What if," the demon continued, gaining confidence, "we made a drug that causes them to think only about sex when they take it?"

Mutters of approval rumbled around the circle. "That's it, sex!" exclaimed a demon. "A drug that'll make them feel good -- real good," chimed in another, "and get them higher than heroin or crack. The best high you can get, and we'll make it last for hours -- no, days!"

The circle of demons began to cackle as the mantra spread throughout the room: Better than crack, better than crack, better than crack. They patted each other on their scaly backs, and the devil smiled.

Click to enlarge

Click to enlarge

Sounds like fiction, but it's difficult to imagine a more diabolical method for fanning the flames of the HIV epidemic in the United States than the use of crystal methamphetamine. Producing a euphoria that leaves cocaine in its own dust and rendering more mundane pleasures -- such as that age-old standby, the orgasm -- little more than a yawn, this wicked little drug makes one feel simultaneously hypersexual, gorgeous and invincible. Inside the mind of a meth user, it's Spring Break, Carnival and Mardi Gras all at the same time, and you are too sexy for your shirt. Not a single gray cell is pondering safe sex.

Methamphetamine abuse is not new.⁴⁸ In one form or another, the drug has been used in waves since the end of World War II. Since then, the popularity of this stimulant has been increasing and there is emerging evidence of a role for its abuse in the spread of HIV infection.⁴⁹

Contemporary methamphetamine use was first noted in Hawaii and on the West Coast. It then spread to the Midwest and eastward. U.S. government efforts to reduce access to crystal methamphetamine by restricting the sale of over-the-counter cold medicines and busting meth labs have largely resulted only in a shift to the importation of crystal meth from south of the border.⁵⁰

The nexus between crystal meth and HIV is most apparent in MSM. The drug increases sexual desire⁵¹ and in studies of such men, it has been associated with unprotected anal intercourse.⁵² In particular, the incidence of receptive anal sex has been found to at least double with the use of crystal meth, possibly as a consequence of the drug's ability to simultaneously enhance libido and also, over time, cause impotence.⁵³ Thus, men who are typically "on top" may opt to be "on bottom" during intercourse while on crystal meth.

Furthermore, use of crystal meth is linked to a higher number of sex partners among men who use it -- more so for MSM.⁴⁹ Unsurprisingly, crystal meth use among gay men has been found to increase the risk of sexually transmitted infections, including syphilis and HIV.

Three reports published last year offer instructive lessons on the intersection between methamphetamine abuse and HIV and are described below. Each builds a case for methamphetamine use being the most significant new obstacle to curtailing HIV transmission in the United States.

On the epidemiological side of the problem, researchers Moupali Das-Douglas et al from San Francisco examined temporal trends in methamphetamine use among over 2,300 homeless and marginally housed individuals -- a highly vulnerable population with high rates of mental health disorders that can serve as a bellwether for general trends in substance abuse.⁵⁴

Click to enlarge

Between 1996 and 2003 methamphetamine use tripled among the homeless or near homeless from 5.7% to 15.1%. Among MSM, the rate of methamphetamine use during the preceding 30 days doubled from 12.3% to 27.6%.

Among heterosexual men and women, rates of methamphetamine use were lower (3% to 5%) at the earlier time point, but tripled over the course of the study.

The most dramatic increase in the use of the drug was among HIV-infected persons -- jumping from 4.4% in 1996 to 21.7% by 2003. As has been observed clinically, the route of administration of the drug shifted away from injection to smoking, snorting and inhaling.

A separate and very concerning study also from San Francisco found that frequent methamphetamine use was associated with acquisition of NNRTI-resistant virus.

Looking at 300 MSM within 12 months of their HIV diagnosis, Grant N. Colfax from the San Francisco Department of Public Health et al found that 83 (28%) of the men had used methamphetamine during the 30 days prior to study enrollment; 12% of the cohort reported at least weekly use of the drug.⁵⁵

Drug-resistant virus prior to the initiation of HIV therapy was detected in 77 (26%) of the men -- a remarkably high proportion. In a multivariable analysis, controlling for multiple sex partners, race, ethnicity, other illicit drug use and prior antiretroviral exposure (i.e., post-exposure prophylaxis), frequent methamphetamine use among these MSM almost quadrupled the odds of having primary NNRTI resistance (odds ratio = 3.9, 95% CI 1.3, 11.4). The authors plausibly speculate that methamphetamine use in the source individual leads to suboptimal drug adherence and subsequent cultivation of resistance mutations, especially to the NNRTIs, which have a lower genetic barrier to resistance than PIs. This resistant virus is then passed on to HIV-uninfected sex partners during methamphetamine-fueled sex.

While the above two studies documented the effect of methamphetamine use on HIV, this study looked at a possible treatment for methamphetamine addiction. It is well known in mental health circles that frequent crystal methamphetamine use carries a poor prognosis.⁵⁶ In fact, there are few proven pharmacological treatments for stimulant abuse and none that are approved by the U.S. FDA for this purpose. Further, relapse rates among people who have been treated are high.

Bupropion (Zyban) is an antidepressant used to treat nicotine dependence that, like stimulants, occupies the presynaptic dopamine transporter responsible for reuptake of this neurotransmitter. As such, it may compete with methamphetamine for this transporter, alleviating symptoms of withdrawal while also potentially treating the patient's underlying depression.

The efficacy of bupropion for treating methamphetamine dependence was studied by Ahmed M. Elkashef from the U.S. National Institute on Drug Abuse et al in a randomized, placebo-controlled trial at five sites across Hawaii, California and the Midwest.⁵⁷ Participants in this federally funded study were individuals who were seeking treatment for their methamphetamine abuse. The 156 individuals enrolled were randomized to either sustained-release bupropion 150 mg twice a day or matching placebo for 12 weeks.

All participants also received a standardized, cognitive-behavioral group intervention delivered over 90 minutes, three times a week. At these visits, participants were queried about their substance abuse using a standardized questionnaire and urine was collected for toxicology. The primary outcome was the difference between arms in the proportion of participants who were methamphetamine-free for a week.

The cohort was largely male and white. Half of the participants had used methamphetamine more than 18 of the previous 30 days. During the trial, loss to follow-up was high, with approximately 50% of the participants in each group dropping from the study prematurely.

Among the people who completed the trial, there was a trend for a greater proportion of participants who had been assigned bupropion to have a methamphetamine-free week, compared to the placebo-assigned participants (54% versus 44%, P = .09).

In a subgroup analysis, bupropion was found to have a significant effect compared to placebo in men who used methamphetamine less frequently (less than 18 of the prior 30 days) at baseline (56% versus 40%, P = .0001). The medication was well tolerated.

The Bottom Line

Despite the clear contribution of methamphetamine use to the spread of HIV infection in the United States, remarkably little new data about this troubling relationship was published in 2007 and few are considering a relationship between the increasing incidence of HIV infection among young men (see below) and the direct and indirect effects of this treacherous drug.

The available data warn that as methamphetamine use climbs and extends into disparate populations, it will leave HIV in its wake. Just as crack cocaine fueled the spread of HIV in the United States, the much more dangerous and insidious crystal meth is now driving HIV infection, especially among many MSM across the country.

The above paper by Moupali Das-Douglas et al examining methamphetamine use among homeless people in San Francisco, while not looking specifically at MSM, makes it very clear how prevalent methamphetamine use is and how closely linked it is with HIV infection.⁵⁴

That methamphetamine abuse not only facilitates the spread of the virus, but also complicates HIV treatment because of transmitted drug resistance is strongly suggested by Grant N. Colfax and colleagues.⁵⁵ These are ominous results that warn HIV providers to brace for a wave of patients addicted to methamphetamine who may be difficult to keep on antiretroviral therapy, if not in care.

Lastly, the bupropion trial provides a bittersweet result.⁵⁷ More than half the participants did not complete the intensive treatment program and even the low-bar outcome of one week free from methamphetamine was difficult for most of the trial participants to achieve. The people who did best in this trial had less self-reported baseline use of the drug and, for such individuals, this may be a treatment option.

Methamphetamine abuse among people at risk for or with HIV infection, almost impossibly, may have us longing for the days when crack was our major substance abuse concern. It is dangerous to underestimate and remain unconcerned about the role of methamphetamine in the epidemiology of HIV in the United States. To do so is to ignore the sweep of this drug across the country and courts wishful thinking.

9. Survival With HIV: Great Expectations

A review of:
Survival of persons with and without HIV infection in Denmark, 1995-2005. Nicolai Lohse, Ann-Brit Eg Hansen, Gitte Pedersen, Gitte Kronborg, Jan Gerstoft, Henrik Toft Sørensen, Michael Væth, Niels Obel. Annals of Internal Medicine. January 16, 2007;146(2):87-95.

HIV is no longer the death sentence it once was, at least in regions where HIV therapies are abundant. Yet, for many people, HIV antiretrovirals offer only a reprieve, as they will still die from HIV, albeit after a longer period of time.

How long a period of time? This is the question most HIV-infected patients want an answer to. They are hungry for numbers and we eagerly oblige them, feeding them CD4+ and CD8+ cell counts, viral loads, blood pressure, lipid levels, etc.

Vague answers about living to get gray hair and Social Security don't cut it. The once handy reply that the average patient survives 10 years from the time of diagnosis has become outdated and was never all that useful, given patients can present eight months after infectio