Abstract

E-mail Abstract Author Add To Itinerary Session Search Abstracts Program

Session 61 Poster Session Antiretroviral Chemotherapy in Previously Treated Individuals Session Time: 4:30-6:30 pm Room 4E-F

419-W. The Effect of Different HAART Regimens on Slope of Viral Rebound in HIV-Infected Patients Failing Therapy N. Press*, R. Woods, J. Raboud, P. R. Harrigan, and J. G. S. Montaner for the INCAS AVANTI and CNAAB3005 Study Teams Canadian HIV Trials Network, Vancouver, BC, Canada Background: The purpose of this study was to compare the characteristics of virologic rebounds in HIV-1-infected patients on various HAART regimens. Methods: Individual patient data were compared from the triple regimen arms of 4 completed clinical trials (INCAS: AZT/ddI/NVP, n=51; AVANTI-2: AZT/3TC/IND, n=52; AVANTI-3: AZT/3TC/NFV, n=53; and CNAAB3005: AZT/3TC/ABC, n=262). HIV-1 RNA levels were performed in prospectively collected samples after completion of each trial, and thus were not available for use in patient management. Rebound was defined in patients who achieved viral suppression as 2 consecutive HIV-1 RNA values above the lower limit of quantification (LLOQ for INCAS, AVANTI-2, AVANTI-3 was 50 copies/mL, and for CNAAB3005 was 400 copies/mL). The triple drug regimens from the 4 clinical trials were compared for predictors of rebound, time to rebound and slope of rebound, using a logistic regression model, Cox proportional-hazards model and linear regression model, respectively. Results: The NVP-containing regimen had the steepest slope of viral rebound (p=0.006) using the LLOQ of each trial. Even when 400 copies/mL was used as the LLOQ for all the trials, the NVP-containing regimen maintained a trend for steepest slope of rebound (p=0.065). A NVP-containing regimen did not influence time to rebound (p=0.63). Non-adherence was a predictor of viral rebound (OR 3.1, p=0.0001), a shorter time to rebound (RR 3.66, p=0.0001), and a steeper slope of rebound (p=0.04). Occurrence of a single detectable HIV-1 RNA (blip), baseline CD4 count and baseline viral load were not predictive of rate of viral rebound or slope of rebound. Conclusions: Patients taking NVP-containing regimens showed a steeper slope of rebound than patients taking PI or ABC-containing regimens. The non-NVP regimens all contained 3TC, hence differences in early virologic rebound could be related to the emergence of the M184V mutation vs the K103N or Y181C mutation in the NVP regimen.

©2002 9th Conference on Retroviruses and Opportunistic Infections