Long-term non-progression in children with HIV: estimates from international cohort data

Objectives: To estimate the probability of long-term nonprogression (LTNP) in the absence of antiretroviral treatment (ART) in children with perinatally acquired HIV, and the impact of LTNP definitions on these estimates. Design: Analysis of longitudinal routine care data (follow-up to 2016) collected through a collaboration of cohorts of children in routine HIV care across Europe and Thailand. Methods: LTNP was defined as reaching age 8 years without disease progression (defined as an AIDS diagnosis or immunosuppression based on WHO immunosuppression-for-age thresholds, age-adjusted CD4$^{+}$ z-scores or CD4$^{+}$ counts). ART initiation was treated as a competing risk (children initiating ART before age 8 were not considered to have LTNP). We included children born domestically in six national HIV cohorts (n = 2481). Additional analyses included domestic-born children enrolled in national cohorts in infancy (aged <12 months, n = 1144, six cohorts), or all domestic-born children in national and nonnational cohorts (n = 4542, 18 cohorts). Results were stratified by birth year. Results: Among children born domestically in national cohorts in 2004–2007, the probability [95% confidence interval (CI)] of LTNP at age 8 years was 10% (6–15%) based on WHO immunosuppression-for-age criteria. This was lower for children born earlier when ART use was less frequent. Results were similar using other immunosuppression thresholds. Estimates were lower when restricted to domestic-born children in national cohorts enrolled in infancy, and higher when including all domestic-born children. Conclusion: Up to 10% of children born during 2004–2007 had LTNP at age 8. Our findings may help identify participants with LTNP for research into posttreatment control and HIV cure.