IDWeek 2014: NNRTIs and Protease Inhibitors Both Good for First ART, Channeling Affects Choices


Non-nucleoside reverse transcriptase inhibitors (NNRTIs) and boosted protease inhibitors work equally well for people starting HIV treatment for the first time, with similar viral suppression, CD4 cell gains, and disease progression, according to a large meta-analysis presented at IDWeek 2014 last week in Philadelphia. A related study shed light on factors affecting choice of initial antiretroviral regimen.

ART Meta-analysis

As described in the first presentation, Alvaro Borges from the University of Copenhagen and colleagues with the METART Global Consortium performed a systematic review and meta-analysis of randomized clinical trials (RCTs) of first-line antiretroviral therapy (ART) containing NNRTIs and ritonavir-boosted protease inhibitors.

Some prior studies have suggested that NNRTIs may enable more rapid viral suppression while protease inhibitors may lead to faster CD4 T-cell recovery and have a high barrier to drug resistance, but data have not been consistent.

The investigators searched established medical databases such as PubMed to identify randomized clinical trials comparing NNRTI-based versus boosted protease inhibitor-based initial ART regimens. NNRTI recipients used either efavirenz (Sustiva, also in the Atripla coformulation) or nevirapine (Viramune), while most protease inhibitor recipients used lopinavir/ritonavir (Kaletra) or ritonavir-boosted atazanavir (Reyataz); newer drugs were not well-represented.

A meta-analysis was performed using intent-to-treat results. The primary endpoint was progression to AIDS or death, with secondary endpoints including AIDS alone, death alone, and treatment discontinuation, as well as virological suppression and CD4 cell gains at 48 weeks. The researchers calculated risk ratios (RRs) and mean differences (MDs) in outcomes, as appropriate. Due to data limitations, not all trials were included in every analysis.


"In a comprehensive meta-analysis of RCTs, we found no difference in clinical outcomes between subjects randomized to NNRTI- or protease inhibitor/ritonavir-based initial ART," the researchers concluded.

They added that a meta-analysis of individual patient data is warranted to explore differences in specific adverse event profiles and dynamics of viral suppression and CD4 cell recovery associated with particular NNRTIs and protease inhibitors.

Channeling Factors

In the second study, Michael Saag from the University of Alabama at Birmingham and colleagues looked at factors that affect how providers select initial antiretroviral drug classes for specific patients, for example, whether doctors prescribe certain drugs for patients with co-existing conditions or at risk for specific side effects.

The researchers useddata from the Centers for AIDS Research Network of Integrated Clinical Systems, or CNICS, an electronic medical records database integrating clinical data from 8 prominent HIV clinics at academic medical centers (University of Alabama at Birmingham, University of California at San Francisco and San Diego, University of North Carolina, University of Washington, Case Western Reserve, Harvard, and Johns Hopkins).

This analysis included 1215 patients who started combination ART consisting of 3 or more drugs -- containing a NNRTI, boosted protease inhibitor, or the integrase inhibitor raltegravir (Isentress) -- between July 2009 and December 2012.

More than 80% of participants were men, more than half were white, one-third were black, 19% were Hispanic, and the median age was 38 years. Two-thirds were gay/bisexual men and 14% were injection drug users. One-quarter started ART with a CD4 count <200 cells/mm3, indicating advanced disease, while a similar proportion started with >500 cells.mm3. A majority had more than 2 co-morbid conditions, the most common being depression (34%), substance use (33%), other psychiatric conditions (26%), high blood pressure (23%), and hepatitis C virus coinfection (16%). About 60% were on public insurance such as Medicaid, nearly one-third had private insurance, and 14% were uninsured.

NNRTI regimens were most commonly used (53%), while 37% of regimens contained a boosted protease inhibitor and 9% contained raltegravir. Factors suspected of being associated with regimen choice were explored individually and in 3 separate multivariate logistic regression models.


"The choice of initial regimen is associated with several demographic and clinical factors, which is sometimes referred to as 'channeling'," the researchers concluded. "As more choices for HIV therapy become available, factors that impact initial regimen selection will likely become even more heterogeneous over time," they predicted.



A Borges, A Lundh, B Tendal, et al. Systematic Review and Meta-Analysis of Randomized Controlled Trials (RCTs) Comparing Initial Non-Nucleoside Reverse Transcriptase Inhibitor (NNRTI)- versus Ritonavir Boosted Protease Inhibitor (PI/r)-based Anti-Retroviral Therapy (ART). IDWeek 2014. Philadelphia, October 8-12, 2014. Abstract 536.

M Saag, A Westfall, S Cole, et al. Factors Associated with the Selection of Initial Antiretroviral Therapy: Real-world Channeling. IDWeek 2014. Philadelphia, October 8-12, 2014. Abstract 1555.