Biomarkers of Metabolic Outcomes of Second-Line Antiretroviral Regimens in Low and Middle-Income Countries

High rates of virological failure among people living with HIV (PLHIV) on second-line antiretroviral therapy (ART) in low and middle-income countries (LMICs) raise concerns regarding long-term outcomes and sustainability of the ART programs. Data from high-income countries show that PLHIV have a higher prevalence of metabolic complications associated with ART use, particularly nucleoside/nucleotide reverse transcriptase inhibitors (N(t)RTIs) and HIV-associated immune dysregulation. This thesis examined the role of biomarkers in predicting virological and metabolic outcomes in a gender-balanced and ethnically diverse cohort of PLHIV who had failed first-line ART regimens and who were randomised to starting either N(t)RTI or raltegravir plus ritonavir-boosted lopinavir (LPV/r) in the SECOND-LINE trial. The SECOND-LINE trial has provided a unique opportunity to examine the associations of plasma concentration of LPV/r and virological outcomes (chapter 3); bone turnover markers and mineral density (BMD) changes (chapter 4); immune activation and inflammatory biomarkers in relation to both body composition changes (chapter 5) and CD4 T-cell recovery (chapter 6). We have shown that the presence of an undetectable random concentration of LPV/r as an objective marker of adherence was predictive of virological failure of the second-line ART regimens. The N(t)RTI backbone (vs raltegravir) in the second-line regimen was associated with higher rates of clinically significant BMD loss at the proximal femur over 48 weeks. At week 12, procollagen type-1-N-terminal pro-peptide (P1NP), a biomarker of bone formation, increased among patients randomised to N(t)RTI (vs raltegravir) and predicted clinically significant BMD loss over 48 weeks. Furthermore, higher baseline levels of IL-6, neopterin, and D-dimer predicted a better CD4 T-cell trajectory and were associated with greater limb and trunk fat gain over 96 weeks regardless of ART regimen. In summary, our results show that, firstly, measuring random drug concentration may offer a practical tool for identifying non-adherence and inform interventions. Secondly, in LMICs with access to dual-energy X-ray absorptiometry limited by prohibitive costs, P1NP measured early after ART initiation may identify PLHIV at a high-risk of rapid bone loss and facilitate timely interventions. Lastly, biomarkers provide valuable insights into the pathogenesis of suboptimal CD4 T-cell recovery and fat gain independent of ART regimen