Assessing the performance of longitudinal T-lymphocytes as biomarkers of immune recovery in HIV-infected children with or without TB co-infection.

In this paper, the receiver operating characteristic (ROC) curve was used to investigate the performance of longitudinal CD4+ T cell counts as a biomarker of disease recovery in HIV-TB co-infected children from sub-Saharan Africa. According to the standard of care for monitoring pediatric HIV infection, routine assessment of immunologic/virologic markers is required. However, due to the lack of well-equipped laboratories and trained personnel, monitoring these markers as regularly as recommended is not feasible in sub-Saharan Africa. Thus, in the absence of routine laboratory monitoring, methods enabling the prediction of patient outcomes from existing CD4+ T-cell data provide a cost-effective alternative to routine monitoring in the region. Simulation studies were carried out to evaluate the performance of the TB-specific ROC model. We found that post-treatment longitudinal CD4+ T cell count had a 79% (95% CI: 0.72-0.86) probability of correctly distinguishing a recovered from non-recovered TB patient compared to a probability of 70% (95% CI: 0.59-0.82) in TB negative patients. Our findings demonstrate that longitudinal CD4+ T-cell counts are reliable indicators of pediatric immune recovery and provide a cost-effective alternative for predicting patient outcomes in the absence of routine laboratory monitoring, particularly HIV RNA determination.