This small pilot study of only six patients was notable because it advanced a new concept that might turn out to be useful in the management of patients on a failing antiretroviral regimen.

From a practical viewpoint, in patients with virologic failure on antiretroviral therapy, it would be advantageous to know if an agent in the regimen still possessed significant activity, or was just taking up space. In this setting, genotypic and phenotypic resistance assays are certainly helpful, but not always predictive of clinical response to therapy. Genotypic testing has limitations because some mutational patterns are complex and difficult to interpret. Phenotypic analysis does not entirely correlate with clinical outcomes and minor variants may be missed. For both tests, an intermediate range sensitivity of an agent produces uncertainty in antiretroviral agent selection.

The clever idea put forth in this study was that a short, single-agent, discontinuation in a failing regimen might quickly determine if that medication was contributing to the antiretroviral activity of the regimen.

In this trial, subjects with a viral load of more than 5,000 copies/mL on a standard antiretroviral regimen stopped one antiretroviral and continued the remainder of the regimen. Resistance testing was done prior to discontinuation and frequent viral load testing was done in the period immediately after the patient stopped a medication in order to determine the dynamics of viral load changes.

Interestingly, three of the subjects had stavudine (d4T, Zerit) discontinued for two weeks and during this time interval there was a significant increase in viral load. Of note, two of these three patients had multiple reverse transcriptase mutations that predicted probable stavudine resistance. Two of these three subjects elected to restart stavudine and viral load levels promptly returned to baseline.

In a similar manner, one subject discontinued didanosine (ddI, Videx) and two stopped efavirenz (EFV, Sustiva). All had baseline genotypic mutations suggesting viral resistance and there were no significant changes in viral load during the short treatment interruption. So with these agents and typical mutations (K103N and Y188L for efavirenz; M41L, L210W and T215Y for didanosine), the short, single-agent interruption confirmed a lack of activity.

The authors' main conclusion was that a single drug interruption might be useful for identifying combinations of agents that maintain activity in heavily treated patients.

The results of this pilot study are quite interesting and will likely be expanded to test other agents in a comparable manner. The frequency of viral load testing over a short period of time represents a practical limitation, but as more experience with this approach accumulates, it may turn out that less intense testing may be possible. Although this pioneering strategy is clearly in the experimental realm at this time, it may have important implications for the clinic once it is tested further.

Reference

Abstract: A Short-Duration, Single-Drug Interruption Determines the Activity of Individual Drugs in Patients on Failing Antiretroviral Therapy (Poster 623)
Authored by: F. Maldarelli, S. Palmer, M. Kearney, A. Wiegand, D. Rock, J. Falloon, R. T. Davey, A. Powers, S. Vogel, A. Pau, R. Dewar, J. A. Metcalf, J. Mellors, J. Coffin
Affiliations: HIV Drug Resistance Prgm, NCI, NIH, DHHS, Frederick MD; NIAID/CCMD Clin, NIH, DHHS, Bethesda, MD; Lab of Immunoregulation, NIAID, NIH, DHHS, Bethesda, MD; SAIC, Frederick, MD; Univ of Pittsburgh, PA