Treatment Failure in HIV-Infected Children on Second-line Protease Inhibitor-Based Antiretroviral Therapy.

Background: Human immunodeficiency virus (HIV)-infected children failing second-line antiretroviral therapy (ART) have no access to third-line antiretroviral drugs in many resource-limited settings. It is important to identify risk factors for second-line regimen failure.
Methods: HIV-infected children initiating protease inhibitor (PI)-containing second-line ART within the Program for HIV Prevention and Treatment observational cohort study in Thailand between 2002 and 2010 were included. Treatment failure was defined as confirmed HIV type 1 RNA load >400 copies/mL after at least 6 months on second-line regimen or death. Adherence was assessed by drug plasma levels and patient self-report. Cox proportional hazards regression analyses were used to identify risk factors for failure.
Results: A total of 111 children started a PI-based second-line regimen, including 59 girls (53%). Median first-line ART duration was 1.9 years (interquartile range [IQR], 1.4-3.3 years), and median age at second-line initiation was 10.7 years (IQR, 6.3-13.4 years). Fifty-four children (49%) experienced virologic failure, and 2 (2%) died. The risk of treatment failure 24 months after second-line initiation was 41%. In multivariate analyses, failure was independently associated with exposure to first-line ART for >2 years (adjusted hazard ratio [aHR], 1.8; P = .03), age >13 years (aHR, 2.9; P < .001), body mass index-for-age z score < -2 standard deviations at second-line initiation (aHR, 2.8; P = .03), and undetectable drug levels within 6 months following second-line initiation (aHR, 4.5; P < .001).
Conclusions: Children with longer exposure to first-line ART, entry to adolescence, underweight, and/or undetectable drug levels were at higher risk of failing second-line ART and thus should be closely monitored.
(© The Author 2015. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.)