20th International AIDS Conference - Melbourne, Australia

Abstract

TUAB0103 - Oral Abstract Session


Switching from first antiretroviral therapy regimen while virologically suppressed is associated with increased risk of subsequent virologic failure

Presented by Marina Klein (Canada).

M. Hull1,2, A. Cescon1, J. Raboud3,4, M.B. Klein5, S. Walmsley3,4, E. Ding1, S. Shurgold1, C. Tsoukas5, J.S. Montaner1,2, N. Machouf6, M.R. Loutfy4,7, C. Cooper8, A.N. Burchell4,9, R.S. Hogg1,10, CANOC Collaboration


1BC Centre for Excellence in HIV/AIDS, Vancouver, Canada, 2University of British Columbia, Vancouver, Canada, 3University Health Network, Toronto, Canada, 4University of Toronto, Toronto, Canada, 5McGill University, Montreal, Canada, 6Clinique Médicale l'Actuel, Montreal, Canada, 7Women's College Research Institute, Toronto, Canada, 8The Ottawa Hospital, Division of Infectious Diseases, Ottawa, Canada, 9Ontario HIV Treatment Network, Toronto, Canada, 10Simon Fraser University, Burnaby, Canada

Background: First-line combination antiretroviral regimens (ART) are safe and durable in a majority of individuals, however many may switch ART for tolerability reasons. Outcomes of ART switches are presumed similar to those of first-line regimens. We evaluated factors associated with regimen switch for non-virologic failure and the association between switching and subsequent virologic failure.
Methods: ARV-naïve individuals initiating ART in CANOC, a multi-site Canadian observational cohort, between 2005-2012 were eligible. Factors associated with regimen switch with suppressed viral load (2 viral loads < 50 copies/mL, > 1 month apart) were assessed using multivariable logistic regression. Durability of subsequent regimen was assessed using a Cox proportional hazards model, adjusted for age, gender, province, injection drug use (IDU) history, and baseline CD4 count. Virologic failure was defined as viral load > 1000 copies/mL.
Results: Of 2797 eligible individuals, 13% were women, 12% IDU, and the median age at baseline was 40 (IQR=33-47). Overall 986 individuals switched from first-line ART regimen with VL < 50 copies/mL (incidence 13.5/100 person-years[PY]), of which 601 switched ART on ≥ 2 occasions with VL < 50 copies/mL (incidence 36.8/100PY). Switches were more common with NNRTI than PI-based regimens (incidence 47.4/100PY vs. 35.9/100PY), however switches within class were more common for PI vs. NNRTI-based regimens (26.7/100PY vs. 13.9/100PY). Factors associated with a single switch from first-line ART included residence in Ontario (adjusted Odds Ratio[aOR] 3.08; 95% CI 2.23-4.24) or Quebec (aOR=3.04; 95% CI 2.15-4.29) vs. British Columbia, and longer duration of ART (aOR=1.47; 95% CI 1.38-1.57). Switching ≥ 2 times was less likely in Ontario (aOR=0.35; 95% CI 0.27-0.45) and Quebec (aOR=0.46; 95% CI 0.35-0.60), among men (aOR=0.53; 95% CI 0.39-0.71), and was associated with longer duration of ART (aOR=1.85; 95% CI 1.74-1.97). In a confounder model switching from first regimen was associated with increased risk of virologic failure (adjusted Hazard Ratio[aHR] 2.70; 95% CI 1.94-3.76).
Conclusions: Switches from first ART while virologically suppressed were common, with higher rates of switches from NNRTI-based regimens. Switching from first ART regimen while suppressed was associated with increased risk of subsequent virologic failure.


Back to the Programme-at-a-Glance