Blood biomarkers of neuronal injury in paediatric cerebral malaria and severe malarial anaemia

Abstract

Abstract Persistent neurodisability is a known complication in paediatric survivors of cerebral malaria and severe malarial anaemia. Tau, ubiquitin C-terminal hydrolase-L1, neurofilament-light chain, and glial fibrillary acidic protein have proven utility as biomarkers that predict adverse neurologic outcomes in adult and paediatric disorders. In paediatric severe malaria, elevated tau is associated with mortality and neurocognitive complications. We aimed to investigate whether a multi-analyte panel including ubiquitin C-terminal hydrolase-L1, neurofilament-light chain, and glial fibrillary acidic protein can serve as biomarkers of brain injury associated with mortality and neurodisability in cerebral malaria and severe malarial anaemia. In a prospective cohort study of Ugandan children, 18 months to 12 years of age with cerebral malaria (n = 182), severe malarial anaemia (n = 158), and asymptomatic community children (n = 118), we measured admission blood levels of ubiquitin C-terminal hydrolase-L1, neurofilament-light chain, and glial fibrillary acidic protein. We investigated differences in biomarker levels, associations with mortality, blood–brain barrier integrity, neurodeficits and cognitive Z-scores in survivors up to 24-month follow-up. Admission ubiquitin C-terminal hydrolase-L1 levels were elevated >95th percentile of community children in 71 and 51%, and neurofilament-light chain levels were elevated >95th percentile of community children in 40 and 37% of children with cerebral malaria and severe malarial anaemia, respectively. Glial fibrillary acidic protein was not elevated in disease groups compared with controls. In cerebral malaria, elevated neurofilament-light chain was observed in 16 children who died in hospital compared with 166 survivors (P = 0.01); elevations in ubiquitin C-terminal hydrolase-L1 levels were associated with degree of blood–brain barrier disruption (P = 0.01); and the % predictive value for neurodeficits over follow-up (discharge, 6-, 12-, and 24 months) increased for ubiquitin C-terminal hydrolase-L1 (60, 67, 72, and 83), but not neurofilament-light chain (65, 68, 60, and 67). In cerebral malaria, elevated ubiquitin C-terminal hydrolase-L1 was associated with worse memory scores in children <5 years at malaria episode who crossed to over 5 years old during follow-up cognitive testing [β −1.13 (95% confidence interval −2.05, −0.21), P = 0.02], and elevated neurofilament-light chain was associated with worse attention in children ≥5 years at malaria episode and cognitive testing [β −1.08 (95% confidence interval −2.05, −1.05), P = 0.03]. In severe malarial anaemia, elevated ubiquitin C-terminal hydrolase-L1 was associated with worse attention in children <5 years at malaria episode and cognitive testing [β −0.42 (95% confidence interval −0.76, −0.07), P = 0.02]. Ubiquitin C-terminal hydrolase-L1 and neurofilament-light chain levels are elevated in paediatric cerebral malaria and severe malarial anaemia. In cerebral malaria, elevated neurofilament-light chain is associated with mortality whereas elevated ubiquitin C-terminal hydrolase-L1 is associated with blood–brain barrier dysfunction and neurodeficits over follow-up. In cerebral malaria, both markers are associated with worse cognition, while in severe malarial anaemia, only ubiquitin C-terminal hydrolase-L1 is associated with worse cognition.