Boosted protease inhibitor (PI)-based antiretroviral therapy (ART) in children with perinatally acquired HIV (PaHIV) infection typically includes lopinavir co-formulated with ritonavir (LPV/r), which is the preferred PI according to the 2009 Paediatric European Network for the Treatment of AIDS (PENTA) guidelines [1]. PIs adversely affect lipid parameters, and data from adult studies suggest that alternative once-daily boosted PIs may have more favourable lipid profiles compared with LPV/r [2,3]. While a switch from PIs to nonnucleoside reverse transcriptase inhibitors (NNRTIs) has been shown to improve lipid profiles in paediatric cohorts [4], data are lacking regarding the optimal PI for children who have previously failed NNRTI-based regimens.A retrospective case note audit of a single-centre cohort of paediatric patients switching from suppressive LPV/r-to ritonavir-boosted atazanavir (ATV/r)-or darunavir (DRV/r)-based ART was performed. Retrospective data were collected at two time-points; 2-7 and 11-18 months post-switch. The primary endpoint was change in mean total cholesterol. Secondary endpoints were other lipid changes, viral suppression, CD4 count and liver function.In total, 26 patients suppressed on PI-based ART switched from LPV/r to ATV/r or DRV/r. Three patients were excluded from the analysis because of incomplete lipid data. Fifteen of 23 patients (65%) and eight of 23 (35%) switched to DRV/r and ATV/r, respectively. Twelve of 23 patients (52%) were male and 23 of 23 (100%) were black African. The median age at switch was 14 years [interquartile range (IQR) 12-15 years] and the median duration of LPV/r therapy was 4 years (IQR 2.25-6.75 years).Mean total cholesterol was 4.83 mmol/L [95% confidence interval (CI) 4.36-5.30 mmol/L] pre-switch. At 2-7 months post-switch, mean cholesterol reduced to 4.43 mmol/L (95% CI 4.08-4.77 mmol/L). Analysis of variance (ANOVA) showed a significant reduction in serum cholesterol of 0.41 mmol/L (P < 0.05). At 11-18 months post-switch, mean cholesterol remained reduced at 4.37 mmol/L (95% CI 4.03-4.71 mmol/ L), a significant reduction of 0.46 mmol/L (P < 0.05) compared with pre-switch cholesterol, although the difference between the first and second time-points was not significant (Fig. 1). There were no significant changes in triglycerides or high-density lipoprotein (HDL) cholesterol.Viral suppression was maintained in all patients apart from one who stopped ART following a family bereavement. No significant change in CD4 count was seen in any patient throughout the study period. No reported side effects resulted in discontinuation of therapy; however, a significant increase in mean total bilirubin was noted in those switched to ATV/r [11 (standard deviation 5) to 51 (standard deviation 23) μmol/L]. This is a known, previously reported side effect of ATV/r [5]. This is the first report of a sustained reduction in mean total cholesterol associated with a switch from LPV/r to an alternative once-daily boosted PI in PaHIV infection. Licensed paediatric dosing is ...

No abstract