Recent advances in pharmacovigilance of antiretroviral therapy in HIV-infected and exposed children

Abstract

Antiretroviral therapy (ART) has greatly improved the survival of HIV-infected children. However, ART is associated with immediate and long-term adverse events. Pharmacovigilance systems, although imperfect, have been developed in many high-income countries (HICs), but coverage in low- and middle-income countries (LMICs) is poor and uneven. This review covers the recent advances in the understanding of adverse events following perinatal ART exposure, including surveillance from birth cohorts; we also describe the adverse events of antiretroviral drugs among HIV-infected children, focussing particularly on those relevant to LMICs, where more than 90% of HIV-infected children live. ART is largely safe in both HIV-infected and HIV-exposed uninfected children, in whom no significant increase in birth defects has been noted. Among HIV-infected children, toxicity to some drugs may be less frequent than in adults, possibly related to immature immune systems in younger children. As per WHO guidelines, many countries are moving from stavudine-based to zidovudine-based or abacavir-based fixed-dose combination (with nevirapine/lamivudine) paediatric mini-pills. However, reassuring data are emerging about short-term stavudine use in LMICs, as this remains an important first-line regimen for young children, as well as an alternative to zidovudine for anaemic children. Zidovudine appears to be well tolerated in young children living in nonmalarious areas, and, among African children, concerns about abacavir hypersensitivity have not been substantiated. Optimization of first-line ART regimens needs to take account of the toxicities in HIV-infected children, in particular as they will take ART much longer than adults and during the period of growth and development. The benefits of ART in pregnancy are clear, but long-term follow-up of ART-exposed infants in LMICs through integrated surveillance systems would be invaluable.