Is less than 95% adherence to NNRTIs good enough to lead to viral suppression?

Adherence to nonnucleoside reverse transcriptase inhibitor (NNRTI) regimens does not need to be perfect to push a viral load below 400 copies/mL, according to results from a study of indigent San Franciscans presented at the 12th Conference on Retroviruses and Opportunistic Infections in Boston, Massachusetts. But shaky adherence to a single (unboosted) protease inhibitor (PI) regimen greatly lowered the chance of viral suppression in this group.

These conclusions from a study conducted by David Bangsberg, MD, at the University of California, San Francisco, and his colleagues come with important caveats:

• HIV experts agree that the viral load must lie below 20 to 50 copies/mL to shut down evolution of resistant virus.
• Even nearly perfect adherence—94% to 100% by two measures—failed to guarantee a load below 400 copies/mL in the study population recruited from homeless shelters, soup kitchens, and low-rent hotels.
• Recent work by another research group found resistance most likely with good, but imperfect, adherence similar to levels reported by Dr. Bangsberg and colleagues.

Research from the days of single-PI therapy set 95% adherence as the standard needed to stop viral replication judged by the 400-copy cutoff (Patterson). Because recent trials found better viral suppression with the NNRTI efavirenz than with unboosted indinavir or nelfinavir, Dr. Bangsberg and colleagues compared adherence rates and viral suppression in 56 people taking single PIs (64% nelfinavir and 27% indinavir) and 53 taking an NNRTI (62% nevirapine and 38% efavirenz).

Gauging adherence by unannounced pill counts in everyone and by electronic medication monitoring in 65 people (60%) revealed equivalent overall adherence rates among people taking an NNRTI (70%) or a PI (71%). But after a median 9.1 months of follow-up, most people on NNRTI therapy had a viral load below 400 copies/mL even with adherence as low as 54%, while substantially fewer PI takers had viral loads that low if their adherence was shaky.

The drop off in viral suppression among people with good but imperfect adherence (74% to 93%) reflects recent findings presented by Richard Harrigan, PhD, at the University of British Columbia, Vancouver BC, Canada, and his colleagues. Dr. Harrigan and colleagues determined that 80% to 89% adherence independently raised the risk of resistance 4.15 times when compared with

In Dr. Bangsberg and colleagues’ study, 4 factors independently predicted a viral load below 400 copies/mL:

• NNRTI vs PI therapy: odds ratio 7.75
• Adherence: odds ratio 1.63
• Nadir CD4+ cell count: odds ratio 1.48
• Months on regimen: odds ratio 1.28.

The balance between viral suppression and resistance to NNRTIs is precarious. Because both efavirenz and nevirapine have long half-lives, a missed or delayed dose may not allow an immediate surge in viral replication. But if HIV does escape NNRTI control, resistant virus emerges swiftly—and a single NNRTI mutation can defeat an NNRTI regimen. In Dr. Harrigan and colleagues’ study, first-line treatment with an NNRTI independently raised the odds of multiclass drug resistance 1.84 times (P = .001).

“While NNRTI potency may lead to viral suppression at moderate levels of adherence,” Dr. Bangsberg cautions, “near complete adherence improves the probability of durable viral suppression for all regimen types.”

This content licensed to aidsmap by iMedOptions, publishers of http://clinicaloptions.com. Copyright iMedOptions, LLC, 2005

For more coverage of CROI from Clinical Care Options for HIV, including news reports, detailed Capsule Summaries and PowerPoint slides of the key studies, and analysis from our panels of leading experts, visit http://clinicaloptions.com/croi