The 18th International AIDS Conference was in beautiful (but hot) Vienna, Austria, and for the first time ever, HIV prevention truly took center stage. The biggest news by far was the CAPRISA 004 study (see JW AIDS Clin Care Jul 26 2010), which was presented in detail on the same day the results were published in Science. By demonstrating the efficacy of a tenofovir microbicide gel in preventing HIV infection, investigators gave us the first-ever clearly positive microbicide study, thus ending years of disappointing research. Almost as interesting as the results was the rapturous response of the assembled audience, with spontaneous cheers erupting as results were presented, and a standing ovation taking place at the end. A colleague of mine remarked that the reaction was similar when investigators first presented the results of BW 002, the first placebo-controlled study of AZT showing that the drug saved lives. One gets the impression that we are at a similar place in HIV prevention science, with some early positive signals but still quite far to go before we have widely effective interventions.

Below, our physician-editors offer perspective on the most clinically relevant findings described at the conference. Next year’s meeting (the International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention) will take place in Rome from July 17 through 20.

— Paul E. Sax, MD

Start Antiretroviral Therapy Early with active Tuberculosis

— Abigail Zuger, MD

Initiation of antiretroviral therapy (ART) is often delayed in patients with newly diagnosed AIDS and tuberculosis (TB), to avoid immune reconstitution inflammatory syndrome (IRIS) and a tangle of possible drug interactions and adverse effects. However, results from the CAMELIA trial in Cambodia [Blanc FX et al. Abstract THLBB106] add to the growing body of evidence indicating that earlier ART is better.

A total of 661 HIV-infected, ART-naive patients with smear-positive TB and CD4 counts <200 cells/mm³ (median, 25 cells/mm³) were randomized to initiate efavirenz + d4T + 3TC either 2 weeks or 8 weeks after starting a standard four-drug regimen for TB. At 1 year, overall mortality was 34% lower in the early-ART group than in the late-ART group, and a multivariate analysis confirmed that late receipt of ART was an independent predictor of death. IRIS occurred more than twice as often in the early-ART group but was manageable in most patients. Among survivors, the virologic efficacy of ART was excellent in both groups.

These results confirm and extend the findings of two other important trials: ACTG A5164, which showed the importance of early ART during treatment for opportunistic infections other than TB (JW AIDS Clin Care Jun 22 2009), and SAPiT, which showed that integrating ART into TB treatment is better than deferring it until after TB treatment is complete (JW AIDS Clin Care Feb 24 2010). The analysis of early versus late ART in SAPiT is still pending, but one wonders whether simultaneous initiation of HIV and TB treatment could reduce early mortality rates even further.

A “CAScade” of Data on When to Start ART

— Charles B. Hicks, MD

Until findings become available from the randomized prospective clinical trial known as START, results from large cohort studies offer the best guide for when to begin ART.

A total of 9455 HIV seroconverters who enrolled in the CASCADE study between January 1996 and May 2009 were analyzed in 161 sequential nested cohorts, comparing patients who initiated ART in a given month with those who deferred ART that month [Jonsson Funk M et al. Abstract THLBB201]. Patients were grouped according to CD4 count: 0–49, 50–199, 200–349, 350–499, and 500–799 cells/mm³. Follow-up averaged about 5 years per patient (representing a total of 52,268 person-years), and the primary endpoint was a new AIDS-defining diagnosis (812 cases were identified during the study period) or death (544 patients died).

Among patients with CD4 counts <500 cells/mm³, starting ART was found to significantly reduce the risk for an AIDS-defining event or death, although the likelihood of reaching an endpoint during the first year of follow-up was very low for patients on the higher end of the spectrum (CD4 counts of 350 to 499 cells/mm³). No benefit was seen from starting ART at CD4 counts 500 cells/mm³.

Notably, for patients with CD4 counts of 350 to 499 cells/mm³, the number needed to treat (NNT) to prevent one case of AIDS or death within 3 years was 34; when death was the only endpoint considered, the NNT was 71. For patients with CD4 counts of 200 to 349 cells/mm³, the NNTs were 21 and 74, respectively. Such relatively low NNTs give credence to the notion of starting therapy earlier and support current recommendations for ART initiation in HIV-infected patients with CD4 counts 500 cells/mm³.

Treatment as Prevention

— Abigail Zuger, MD

One reason often cited for treating HIV-infected patients at higher CD4-cell counts is the theory that more people receiving treatment means more transmission events averted. Data from the shape of the HIV epidemic in Denmark, described by Susan Cowan, seem to support this idea [Abstract MOAC0103].

Of the roughly 5250 HIV-infected individuals in Denmark, about half are men who have sex with men (MSM). More than 80% of these men are taking antiretrovirals, and 86% have undetectable viral loads. Surveys have repeatedly shown that unsafe sexual practices are on the rise among HIV-infected MSM in Denmark. However, the HIV transmission rate in this population seems to be falling, from about 11 transmission events per 100 HIV-infected MSM in 1995, to 7 in 2001, to fewer than 5 now. The only logical explanation, Dr. Cowan suggested, is that the risk associated with unsafe sex in this population has declined as more and more individuals are rendered relatively noninfectious with treatment.

TMC278 (Rilpivirine) ShowS Promise and possibly problems

— Judith Feinberg, MD

A phase II study of the second-generation nonnucleoside reverse transcriptase inhibitor (NNRTI) TMC278 (rilpivirine) suggested that the drug is both safe and effective in treatment-naive patients. Now, data from a pooled analysis of two manufacturer-supported phase III trials (ECHO and THRIVE) indicate that the drug is noninferior to efavirenz in this population [Cohen C et al. Abstract THLBB206].

Between the two studies, 686 patients were randomized to TMC278 (25 mg once daily) and 682 to efavirenz. Eighty percent of patients received tenofovir/FTC as the nucleoside backbone, and the rest received AZT/3TC or abacavir/3TC. The primary endpoint was the proportion of patients with viral loads <50 copies/mL at 48 weeks in an intent-to-treat, time-to-loss-of-virologic-response analysis.

Both groups had high rates of success (84.3% in the TMC278 group and 82.3% in the efavirenz group, thus demonstrating noninferiority) but for different reasons. The TMC278 group had a lower rate of discontinuation due to adverse events (2% vs. 7%), but the efavirenz group had a lower rate of virologic failure (5% vs. 9%) and also a lower rate of resistance among those who failed: NNRTI resistance mutations emerged in 54% of efavirenz recipients (most commonly, K103N) versus 63% of TMC278 recipients (most commonly, E138K, conferring resistance to etravirine), and resistance to nucleoside reverse transcriptase inhibitors (NRTIs) emerged in 32% versus 68%, respectively. In the TMC278 group, patients with baseline viral loads >100,000 copies/mL appeared to have somewhat higher rates of virologic failure than patients with baseline viral loads below that threshold — a difference that was not seen in the efavirenz group. Mean CD4-count increases were 192 cells/mm³ in the TMC278 group and 176 cells/mm³ in the efavirenz group.

On the basis of these data, the manufacturer of TMC278 has submitted a New Drug Application to the FDA and is also in the process of developing a single tablet that contains TMC278 + tenofovir/FTC. This regimen would be a reasonable alternative for treatment-naive patients who want a once-daily regimen but who cannot take efavirenz (due to concerns about central nervous system adverse effects or teratogenicity) and do not want to start a boosted protease inhibitor (PI). Whether the use of TMC278 will be limited by its somewhat higher rate of virologic failure despite meeting criteria for noninferiority awaits further analysis of the predictors of virologic failure.

NRTI-Sparing Strategies: Not Ready for Prime Time

— Paul E. Sax, MD

Current treatment guidelines recommend that initial antiretroviral regimens contain two NRTIs, but interest remains in NRTI-sparing combinations as well. Researchers evaluated the safety and efficacy of such combinations in three industry-supported, prospective, comparative trials.

• In the SPARTAN study, patients received either raltegravir + unboosted atazanavir (at a dose of 300 mg twice daily — twice the usual dose) or ritonavir-boosted atazanavir + tenofovir/FTC [Kozal MJ et al. Abstract THLBB204].
• In the PROGRESS study, patients received lopinavir/ritonavir + either raltegravir or tenofovir/FTC [Reynes J et al. Abstract MOAB0101].
• In the third study, patients received ritonavir-boosted atazanavir + either maraviroc (150 mg once daily — half the usual recommended dose) or tenofovir/FTC [Mills A et al. Abstract THLBB203].

All three trials were relatively small (the largest, PROGRESS, had 206 patients) and hence can be considered pilot studies at best. In addition, although all the NRTI-sparing regimens appeared to show virologic activity comparable to that of the standard-of-care regimens, they also had limitations that would make one reluctant to choose them unless NRTIs were contraindicated.

• In the SPARTAN study, the raltegravir + atazanavir group had a higher rate of grade 4 hyperbilirubinemia and more resistance during treatment failure than the control group. These differences prompted the study sponsors to stop the trial early.
• In the PROGRESS study, lopinavir/r + raltegravir was associated with somewhat higher lipid levels than the control regimen. It also has the disadvantage — compared with other, preferred initial antiretroviral regimens — of being a twice-daily regimen with a relatively high pill burden.
• In the third study, the use of boosted atazanavir + maraviroc required a pretreatment viral tropism assay that is currently expensive, has a slow turnaround time, and is not a routine part of baseline evaluations.

In the aggregate, NRTI-sparing approaches still lack a tangible benefit over the standard initial NRTI-containing therapies that treatment guidelines recommend.

572 Remains on Track

— Helmut Albrecht, MD

S/GSK1349572 (or 572, as it is often called) is a promising, once-daily, second-generation integrase inhibitor. It has shown excellent antiretroviral activity at low doses, does not require boosting, and, at least in vitro, remains active against many raltegravir-resistant isolates. Now, two industry-supported studies have validated the high expectations for this drug in raltegravir-resistant and treatment-naive patients.

In the VIKING study, 27 patients with documented raltegravir treatment failure added 572 (50 mg once daily) to their regimens in place of raltegravir — and then, on day 11, had their background regimens optimized [Eron J et al. Abstract MOAB0105]. Favorable virologic response was seen in all 18 patients who had resistance due to the N155H and Y143H mutational pathways, compared with only 3 of 9 patients who had both a mutation at the Q148 position and at least one other associated mutation. There was a strong correlation between baseline susceptibility and antiviral response.

In the SPRING-1 study, 205 treatment-naive patients were randomized to 572 (at 10, 25, or 50 mg once daily) or efavirenz, each in combination with an investigator-chosen, fixed-dose, dual-nucleoside backbone (67% tenofovir/FTC, 33% abacavir/3TC) [Arribas J et al. Abstract THLBB205]. As in all studies of integrase inhibitors to date, time to virologic suppression was significantly shorter with all doses of 572 than with efavirenz. Efficacy was independent of the nucleoside backbone, and 572 was extremely well tolerated. On the basis of these findings, the 50-mg dose was selected for upcoming phase III studies in integrase inhibitor–naive patients.

Overall, 572 has the potential to set a new standard for integrase inhibitors. Although only a relatively small number of patients received this drug together with abacavir/3TC in SPRING-1, the regimen was still effective enough to warrant further evaluation, which is gratifying since it would represent an option for a new one-pill, once-daily, fixed-dose combination. For treatment-experienced patients, the Q148 mutational pathway could be 572’s Achilles heel. To avoid the accumulation of complex Q148-associated mutation patterns — and thus preserve potential future activity of 572 — we should ensure that patients do not remain on failing raltegravir or elvitegravir regimens for longer than is absolutely necessary.

Raltegravir Resistance in a Treatment-Naive Patient

— Paul E. Sax, MD

In a small pilot study of abacavir/3TC + raltegravir, a treatment-naive patient turned out to have integrase inhibitor–resistant virus at baseline — the first time this phenomenon has been reported.

The study (SHIELD) was aimed at evaluating how raltegravir performs in initial regimens that include NRTIs other than tenofovir [Young B et al. Abstract THPE0112]. All 35 participants received abacavir/3TC + raltegravir after testing negative for both HLA-B*5701 and hepatitis B surface antigen. At week 48, the median CD4-count increase was 247 cells/mm³, and 91% of patients had achieved virologic suppression.

More important than these results, though, were the additional data obtained on a single patient who experienced virologic failure [Young B et al. Abstract TUPE0163]. This patient was known to have baseline resistance to PIs and NNRTIs. However, when his stored baseline sample was evaluated retrospectively, it also showed the major integrase mutation Q148H. Phenotype testing demonstrated a 150-fold loss of susceptibility to raltegravir.

Transmitted resistance to integrase inhibitors is likely to be rare now, given the generally infrequent occurrence of raltegravir resistance. However, clinicians should consider supplementing baseline genotype tests with an integrase genotype if the patient has evidence of resistance mutations in other drug classes and is likely to start using raltegravir.

Does the Human Papillomavirus Vaccine Prevent Cancer in Men?

— Abigail Zuger, MD

The quadrivalent human papillomavirus (HPV) vaccine has won great acclaim for the prevention of cervical cancer in women and genital warts in men. New data from a manufacturer-supported, postmarketing trial give some sense of what we can expect from the vaccine among HIV-uninfected MSM [Jessen H et al. Abstract THLBB101].

More than 4000 HIV-uninfected men (ages 16 to 26; 85% heterosexual) in 18 countries were randomized to receive three doses of vaccine or placebo. During 3 years of follow-up, vaccination led to a 90% reduction in the incidence of genital warts and other HPV-associated external genital lesions. It was also associated with an 86% reduction in the incidence of persistent HPV infection.

Anoscopy results were reported for about 400 of the 600 MSM enrolled in the trial: Anal intraepithelial neoplasia related to HPV types 6, 11, 16, and 18 was diagnosed in 2.6% of the vaccine group versus 11.5% of the placebo group — a 78% relative difference. No cases of anal cancer occurred in either group.

Whether the vaccine retains similar efficacy in an HIV-infected population, of course, awaits further study.

Published in Journal Watch HIV/AIDS Clinical Care August 9, 2010