Responses to highly active antiretroviral therapy and clinical events in patients with a low CD4 cell count: late presenters vs. late starters

Objective We investigated whether adverse responses to highly active antiretroviral therapy (HAART) associated with late HIV presentation are secondary to low CD4 cell count per se or other confounding factors. Methods A longitudinal analysis of the UK Collaborative HIV Cohort (CHIC) Study of individuals starting HAART in 1998-2007 was carried out, comparing late presenters (presenting/starting HAART at a CD4 count <200 cells/mu L) with late starters (presenting at a CD4 count > 350 cells/mu L; starting HAART at a CD4 count <200 cells/mu L), using 'ideal starters' (presenting at a CD4 count > 350 cells/mu L; starting HAART at a CD4 count of 200-350 cells/mu L) as a comparator. Virological, immunological and clinical (new AIDS event/death) outcomes at 48 and 96 weeks were analysed, with the analysis being limited to those remaining on HAART for > 3 months. Results A total of 4978 of 9095 individuals starting first-line HAART with HIV RNA > 500 HIV-1 RNA copies/mL were included in the analysis: 2741 late presenters, 947 late starters and 1290 ideal starters. Late presenters were more commonly female, heterosexual and Black African. Most started nonnucleoside reverse transcriptase inhibitors (NNRTIs); 48-week virological suppression was similar in late presenters and starters (and marginally lower than in ideal starters); by week 96 differences were reduced and nonsignificant. The median CD4 cell count increase in late presenters was significantly lower than that in late starters (weeks 48 and 96). During year 1, new clinical events were more frequent for late presenters [odds ratio (OR) 2.04; 95% confidence interval (CI) 1.19-3.51; P=0.01]; by year 2, event rates were similar in all groups. Conclusion Amongst patients who initiate, and remain on, HAART, late presentation is associated with lower rates of virological suppression, blunted CD4 cell count increases and more clinical events compared with late starters in year 1, but similar clinical and immunological outcomes by year 2 to those of both late and ideal starters. Differences between late presenters and late starters suggest that factors other than CD4 cell count alone may be driving adverse treatment outcomes in late-presenting individuals.