Home Search Abstracts View Session E-mail Abstract Author


Session 95 Poster Abstracts
Predictors of ART Discontinuation, Virologic Response, and Outcomes
Session Day and Time: Tuesday, 1 - 4 pm
Poster Hall


528    
Response to Combination ART: Variation by Age
Caroline Sabin and the Collaboration of Observational HIV Epidemiological Research Europe (COHERE) Study
Royal Free and Univ Coll London Med Sch, UK

Background:  Despite 10 years’ experience of combination ART (cART), initial treatment responses in children, adolescents, and older HIV-infected persons, groups which now contribute significantly to the HIV epidemic, have been poorly documented.

Methods:  COHERE contains information from 33 European cohorts/collaborations. Data on immunological and virological responses were pooled from all ART-naïve persons starting cART from 1998 to 2006; eligible patients had ≥1 CD4 count and viral load pre-cART and over follow-up. Time to confirmed (2 consecutive) viral load <50 copies/mL and CD4 increase >100 cells/mm3 from pre-cART levels were analysed using Kaplan-Meier and Cox regression, censoring at last visit. Age strata are shown in the table; other covariates were gender, country of origin, year of cART start, pre-cART CD4 and viral load, AIDS and initial regimen. Sensitivity analyses considered alternative viral load response thresholds.

Results:  We included 49,921 individuals (29% female, 39% non-European origin) whose age ranged from 1 day to 87 years; 223, 184, 219 and 201 were in the 4 youngest age groups, 2693, 1656, and 1613 were in the 3 oldest age groups. Pre-cART CD4 was highest in young children and dropped with increased age (see the table). Frequency of CD4 and viral load monitoring in the first year of cART was similar in all groups. The table shows 12-month unadjusted responses to cART. In multivariable analyses, the probability of a viral load response was lower in those aged 6 to 12 (relative hazard 0.87, 95%CI 0.74 to 1.02; p = 0.09) and 13 to 17 (0.78, 0.65 to 0.94; p = 0.01) years compared to those aged 30 to 39 years, and higher in those aged 50 to 54 (1.24, 1.19 to 1.30; p <0.0001), 55 to 59 (1.24, 1.17 to 1.32; p <0.0001), and ≥60 (1.18, 1.12 to 1.26; p <0.0001) years. Compared to those aged 30 to 39 years, the chance of a CD4 response was higher in younger individuals (6 to 12 years: 1.61, 1.39 to 1.88; p <0.0001; 13 to 17 years: 1.09, 0.92 to 1.29; p = 0.33; 18 to 29 years, 1.05, 1.02 to 1.08; p = 0.01) and was reduced in those aged ≥60 years (0.93, 0.87 to 0.98; p = 0.01).

Conclusions:  Responses to cART were generally good at all ages. The better viral load responses in older individuals were associated with poorer CD4 responses, possibly due to age-related immune impairment. Conversely, CD4 responses were best in young children, although the possibility of a poorer viral load response may increase the risk of acquiring resistance. These factors should be balanced against the possibility of adverse events when deciding on the timing of cART initiation in these groups.