AIDS 2010 Abstract - Outcomes of antiretroviral therapy (ART) among HIV-infected children at clínica de familia MIR (CFMIR), La Romana, Dominican Republic: predictors of switch to second-line ART, response to second-line therapy

Outcomes of antiretroviral therapy (ART) among HIV-infected children at clínica de familia MIR (CFMIR), La Romana, Dominican Republic: predictors of switch to second-line ART, response to second-line therapy

C.W. Elgarten¹, J.E. Myers^2,3, J.N. Mercedes⁴, J.O. Leonardo Guerrero⁴, C.T.B. Rodriguez Sirett⁴, N.P. Torres⁴, B.S. Taylor⁵, S. Hermosilla³, M.E. Sobieszczyk², J.A. Roman Pouriet⁴, F.J.C. Carazas^3,4, S.W. Nicholas^1,3

¹Columbia University, College of Physicians and Surgeons, New York, United States, ²Columbia University, Division of Infectious Diseases, Department of Medicine, College of Physicians and Surgeons, New York, United States, ³Columbia University Mailman School of Public Health, New York, United States, ⁴Clínica de Familia MIR, La Romana, Dominican Republic, ⁵University of Texas Health Science Center at San Antonio, Division of Infectious Diseases, Department of Medicine, San Antonio, United States

Background: Few studies have examined indications for and response to second-line ART among children in resource-limited settings with inconsistently available virologic monitoring.
Methods: We reviewed medical records of HIV-infected children initiating ART from 10/2004-12/2008 at CFMIR. Primary outcome was switch to second-line ART, defined as initiation of a PI-based regimen and concurrent NRTI change.
Results: 114 patients met study criteria; 100 records were reviewed. Median age was 6.1 years [inter-quartile range (IQR)=2.7-9.4], 49% were female, 66% were orphaned, and 18% resided in sugar plantation settlements of primarily Haitian migrants (bateys). 96% were infected by mother-to-child transmission (MTCT), 14% despite MTCT-prevention interventions. Median baseline CD4 was 426 cells/µL [IQR=190-846]. Initial regimen was NNRTI-based for 89% (PI-based for remainder). Six died: median time to death was 26 days [IQR=16.5-288]. No variable examined predicted mortality. Excluding patients who died, 13/94 switched to second-line ART (0.5 switches/10 person-years). Documented criteria for switch were the following: poor virologic response, 1; poor immunologic response, 5; clinical progression, 1; multiple criteria, 4; undocumented criteria, 2. Median time to switch was 18.7 months [IQR=13.4-26.1]. Second-line was didanosine+abacavir+lopinavir/ritonavir. Switch to second-line ART, compared with continued first-line, was associated with lower median baseline CD4 (118 versus 509 cells/µL, p=0.027), baseline CDC clinical category C (100% versus 62%, p=0.007) and lower median 12-month CD4 increase on first-line (145 versus 342 cells/µL, p=0.041). Anemia, malnutrition, batey residence and orphanhood did not predict switch. Of 14 starting second-line, one died 6 months post-switch; 13 survived (median follow-up: 18.6 months [IQR=11.6-26.3]). Median CD4 increases after 6 and 12 months were 166 and 331 cells/µL, respectively.
Conclusions: Regimen switch was associated with advanced clinical and immunological disease at ART initiation. These findings suggest that prompt diagnosis and timely treatment of HIV are associated with durability of first-line regimens, hence minimizing need for switch. Response to second-line ART was excellent.