Greater CD4 increases from 12 weeks with nevirapine could have caused more serious immune reconstitution inflammatory syndrome (IRIS) [20]. However, week 4 and 12 viral load

changes were similar in the two groups, and the clinical events diagnosed were not predominantly IRIS-type events. Furthermore, there was no clear association between developing clinical events and rapidity of viral load changes, nor did the difference between nevirapine and abacavir vary with pre-ART CD4 cell count, both strong predictors of IRIS [21,22], nor was there evidence that the differences between groups were restricted to the first 6 months on ART when IRIS events would be likely to predominate. Emergence of HIV resistance mutations is described separately [5] but, given that higher viral load suppression was substantially and significantly greater in the nevirapine group, it is unclear how any find more differences in resistance accumulation Z-VAD-FMK concentration or the relative fitness cost of NNRTI and NRTI mutations could have increased overall disease progression relative to the abacavir group. If suboptimal virological potency of abacavir was driving results, either

more ART modifications or more clinical events (or both) would be expected in the abacavir group – neither was observed. While patients initiating ART with advanced HIV disease and low CD4 cell counts may be at higher risk of opportunistic infections, it is unclear why their risk would be greater on nevirapine- than abacavir-containing regimens. Furthermore, we found no evidence to suggest that absolute levels of CD4 and HIV RNA had different prognostic values for clinical outcomes in the nevirapine and abacavir groups, Chloroambucil and after adjustment for time-updated CD4 cell count, haemoglobin and weight, differences between randomized groups were similar to those obtained in unadjusted analyses. Without stored cells, we are unable to explore whether the quality rather than the quantity of CD4 immune restoration differed between the abacavir and nevirapine groups. The only published data appear to be

those of a small study in children simplifying from boosted protease inhibitor to triple NRTI therapy, which demonstrated increased functionality [23]; whether this would be similar in ART-naïve adults in NORA is unclear. Nevertheless, our findings highlight the importance of close follow-up and high-quality clinical management (including primary and secondary prophylaxis/treatment for opportunistic infections) for patients initiating ART with advanced disease. The timescale for changes in prognostic markers to influence clinical outcome could differ; i.e. inferior 48-week immunological/virological response in the abacavir group might lead to poorer clinical outcome only later on. We noted a trend towards superior weight gain with nevirapine at 48 weeks but not before; and a nonsignificant (P=0.