Predictors of Treatment Failure in Cambodian Children With Human Immunodeficiency Virus Infection

Abstract

To the Editors: Jittamala et al1 in their article in the September issue of Pediatric Infectious Disease Journal on predictors of treatment failure documented that nevirapin but not human immunodeficiency virus-ribonucleic acid titer was predictive for virologic failure in a cohort of Thai children after 4.5 years of follow-up. We analyzed 102 children aged 4 to 16 years who were infected by human immunodeficiency virus in our orphanages in Phnom Penh, Cambodia, in 2003–2009 on genotypic resistance as a cause of first-line failure and for adherence to HAART, virologic, clinical, and immunologic response to HAART. Of 102 children, only 7 failed after 18 to 36 months on first-line treatment stavudine (d4T), lamivudine (3TC) plus nevirapine. Clinical checkup has been done daily by physicians. CD4 cells were investigated every 3 months and viral load 6 monthly. Genotypic/phenotypic resistance tests were performed after December 2004 in case of clinical or immunologic failure by National Pediatric Hospital laboratory affiliated to Institute Pasteur Cambodge, Phnom Penh. Mortality on acquired immunodeficiency syndrome and acquired immunodeficiency syndrome-related opportunistic infections in our group of 102 children was <3%, and only 3 children died in 2003–2009. Two children died even before the onset of therapy (late arrival to clinic), 1 at the beginning of HAART (first month). All had CD4 counts <10 cells/mL (<5%) at admission. The rest of the 98 children were treated with HAART from 18 to 72 months: 7 of 99 (7%) failed between 15 and 36 months. As a failure, we considered when CD4 count decreased (2 measurements within 6 months) and viral load showed increasing tendency (>1000 copies/mL). Only viral load increase (without immunologic correlate in 1 child) or only CD4 decrease (discordant failure in 2 children) was not considered as indicator for replacement of the first-line. Most of the children failed on D4T + 3TC + nevirapine, which was replaced by abacavir (ABC), didanosine (DDI), zidovudine (ZDV), or tenofovir (TDF) with protease inhibitors—lopinavir/ritonavir or efavirenz (available after 2007). Commonest mutations of the reverse transcription (responsible for both genotypic and phenotypic resistance) among nucleoside analog reverse transcriptase inhibitors (NRTIs) were M184V (5 children), V75M (3 children), and T215Y and among non-NRTIs (NNRTIs), commonest mutations were K103N (2 children) and K101E (2 children). Those who failed on first-line were treated by TDF, ABC or DDI, and ZDV. DDI was still left in the NRTI backbone with TDF or ZDV, or ABC plus protease inhibitors was added. All 7 children who were treated with second-line showed rapid decline of viral load (from 5.4 to <2.4 log) and increase of CD4 percentage (used in children and young adults indication response—virologic and immunologic). There are no studies on efficacy and pharmacokinetics of new form fusion, entry, and C-C chemokine receptor type 5 (CCR5) antagonist-class antiretrovirals in children. Therefore, optimal dosing especially in children with body weight <15 kg is unknown. Cambodia, in contrast to Europe and the United States, has few therapeutic options for pediatric HAART in developing countries. Only lopinavir/ritonavir and nelfinavir produced by Thai and Indian pharmaceutical industries were available in 2003–2007. In contrast to the results published by Jittamala et al1 reporting failure in 40 of 202 children (20%), failure rate in own count of 102 children within 6 years was 7%. Of 4 children who died, 3 had not received any treatment and 1 died after 1 month; all 4 came very late to assess the effect of potential therapy because they arrived in terminal stage (<10 CD4 cells). We have no explanation for low number of deaths (3%) and virologic/immunologic failures (7%) among our group of children in Cambodia within 6 years of follow-up. Probably, high proportion of those receiving efavirenz (90%) versus nevirapin (8%) because of concomitant anti-TB therapy for tuberculosis may explain our 7% failure rate (after 6 years of follow-up). Efavirenz and adherence were protective against failure in the Thai study1 and a review in several studies on pediatric ART.2 All our children were placed in 2 orphanages with good nutrition (5 × daily) on inpatient basis, with up to 100% adherence, which may also contribute to low failure rate. Irad Beldjebel, MD St. Maximilian Kolbe Tropical Clinic of Infectious Diseases, SEUC Sangkat Buoeung Tumpong, Khan Meanchey Phnom Penh, Kingdom of Cambodia Julia Vujcikova, MD Tropical Programs SEUC Sihanoukville, Kingdom of Cambodia Jaroslava Sokolova, MSc Department of Clinical Disciplines School of Health Care and Social Work Trnava University Trnava, Slovak Republic St. Elizabeth University College of Health and Social Sciences Bratislava, Slovak Republic Vladimir Krcmery, MD, DSc St. Elizabeth University College of Health and Social Sciences Bratislava, Slovak Republic