Single drug substitution from stavudine to abacavir or tenofonvir for children and adolescents on combination art: describing the impact on virological outcomes

Abstract

Abstract: 1 Background: In an era geared towards HIV treatment optimization, data regarding the impact of 2 single drug substitutions on viral suppression in routine care settings is critical. This study 3 compares viral suppression in children and adolescents who switched from stavudine to either 4 abacavir or tenofovir, to those who remained on stavudine. 5 Methods: This retrospective cohort study examined routine clinical information from children 6 and adolescents <19 years initiated on stavudine-containing ART prior to 2010, at a large public 7 sector paediatric ART clinic in South Africa. In children who switched, observation time (T0) 8 was set at the time of switching from stavudine to either abacavir or tenofovir. In those who did 9 not switch, T0 was set at the first visit after 1 April 2010. Incidence of viral rebound (>400 10 copies/ml) and virological failure (VF, two consecutive VL>1000 copies/ml) were calculated, 11 and Kaplan-Meier survival curves for first viral rebound as well as VF were plotted. Cox 12 proportional hazard models were used to estimate the association between viral rebound and 13 switching. 14 Results: The 723 children in the cohort contributed 14882.7 person-years (pys) of follow-up 15 time. The median age at cohort entry was 9.3 years and 47.4% were male. Of these, 566(78%) 16 had a switch from stavudine and 157(22%) did not. At 6 months, the incidence of viral rebound 17 in children who switched was 6.2/100pys (95%CI: 3.7-10.6), compared to 14.4/100pys (95%CI: 18 8.1-26.8) in the non-switch group (P value =0.02). Those who switched were significantly less 19 likely to experience rebound (adjusted hazards ratio: 0.39, 95%CI: 0.17-0.91) at 6 months after 20 T0 but this was not sustained at 12 months (adjusted hazards ratio: 0.89, 95%CI: 0.49-1.60). 21 Although more children had VF in the switch group by database closure (OR 1.88, 95%CI: 1.02-22 3.48), there was no significant difference in incidence of virological failure (switch group: 0.61 23 (0.49-0.78); non-switch group: 0.42 (0.25-0.73); p-value=0.40). Similarly, no significant 24 difference in incidence of VF was seen on survival analysis. 25 Conclusion: There was no significant difference in viral rebound or virological failure between 26 groups. Switches to more palatable drugs with fewer side effects may improve adherence and 27 viral suppression in the short term.