HOME
Return to Table of Contents
DOES HIV-1 CLADE EFFECT DURATION OF INITIAL HIGHLY ACTIVE ANTIRETROVIRAL THERAPY (HAART)?
N Pant Pai, T Murphy, ZS Saeed, RG Lalonde, MB Klein
Montréal, QC
Objective: Both host and viral factors may modify responses to antiretroviral treatment in patients infected with different HIV-1 clades. We evaluated whether viral clade influences the durability of initial HAART.
Methods: We conducted a retrospective cohort study at a university HIV clinic between 1998-2006. All naïve patients with at least 2 clinic visits were included. Patients were grouped according to country of HIV acquisition (Canadians/Haitians: clade B; Africans/Asians: non-B) based on the clade distribution among 241 patients with genotypes. Our primary outcome was time elapsed until stopping/changing first HAART for any reason estimated using Kaplan Meier and Cox proportional hazards models.
Results: Overall, 453 patients were followed for a median of 1.79 years (792 person-years). 312 (69%) were subtype B and 141(31%), non-B. More patients with clade B infections were male (85% vs. 50%) with higher baseline HIV RNA (4.47 vs. 4.18 log copies/ml). Baseline CD4 (192 vs.184 cells/ul) and HAART regimens were similar: PI (52% and 49%), NNRTI (36% and 40%), triple nucleoside (8% and 11%). Overall, 67% of subtype B failed vs. 50% of non-B (p<0.001). The median time on first HAART was longer in non-B patients at 1.23 years (95% CI; 0.94-1.48) vs. 1.06 years (95% CI: 0.89-1.19); log rank p<.05. However, the in final adjusted model there was no significant difference in failure rates according to subtype: non- B: HR, 1.11 (95% CI, 0.88-1.41). Important predictors of failure were: HCV co-infection (HR 2.52; 95% CI, 1.46, 4.37) and recent calendar period of HAART initiation. A minority (11%) experienced virologic failure in both groups with the major reasons for treatment change being intolerance or patient choice.
Conclusion: We found that HAART is equally effective in clade B and non-B infections. Other factors such as HCV co-infection are far stronger predictors of treatment failure.