Prevalence and risk factors of virologic failure and HIV-1 drug resistance among children and adolescents in Nairobi, Kenya

Abstract

Highly active antiretroviral therapy (HAART) in people infected with the Human Immunodeficiency Virus type-1 (HIV-1) is effective when backed up with adequate clinical, immunological, and virologic monitoring. However, a lack of viral suppression after at least 6 months of ART - known as virologic failure (VF) - may result in the emergence of HIV-1 drug resistance (HIV-DR). The twin phenomena of VF and HIV-DR can then lead to an increased incidence of opportunistic infections (OIs), the need for costlier second-line HAART drugs and a higher likelihood of death. To evaluate the prevalence and risk factors of VF and HIV-DR among children and adolescents in Kenya, a study was conducted in the Lea Toto Programme (LTP). The LTP is a multi-Centre community outreach Programme in Nairobi, Kenya providing care to 3,500 HIV-infected children and adolescents who live in resource-limited settings. This prospective cohort study involved 438 HIV-infected children and adolescents aged 2-17 years, who were recruited between December 2013 and April 2014: 210 (47.9%) females and 228 (52.1%) males. Medical and demographic data including, HIV-1 viral loads, CD4+ T-cell counts, information on adherence to HAART, HIV-1 drug resistance mutations (DRMs), malnutrition and OIs were collected over a period of four years. A threshold of 1,000 HIV-1 RNA copies/ml was used to determine treatment outcome. Treatment regimens with the Non-Nucleoside Reverse Transcriptase Inhibitor (NNRTI) drugs Nevirapine (NVP) and Efavirenz (EFV), were prescribed to 284 (64.8%) and 141 (32.2%) of the children, respectively. Treatment regimens with the protease inhibitor drugs Ritonavir-boosted Lopinavir (LPV/r) were prescribed to 13 (3%) individuals in the cohort. The median log10 viral load (VL) for the cohort at baseline was 4.82 (Interquartile range [(IQR]: 4.37-5.39). After 24 months of first-line ART, the log10 VL fell to 2.43 (IQR, 2.28-2.59). The prevalence of VF during first-line antiretroviral therapy (ART) was 32.2%, 23.9%, 24.2%, and 26.1% after 12, 24, 36, and 48 months of treatment, respectively. For the 438 individuals initiated on first-line ART, the outcomes were as follows: nine (2.1%) died; 49 (11.2%) transferred out to other Comprehensive Care Centres (CCCs); 20 (4.6%) were lost to follow-up; 114 (26.0%) failed first-line ART and were switched to second-line ART. Sanger sequencing determined that among these 114 individuals, there were a total of 16 HIV-1 subtypes: four (25%) pure HIV-1 subtypes, namely A1, B, C and D; and 12 (75%) possibly recombinant HIV-1 subtypes, including A1B, DG, A1J, CRF_01AE and CRF_35AD. The VF rate among the 114 second-line ART individuals was 19.3% (n=22) after a median of 17 months of treatment. Of these 22 individuals, 15 (68.2%) were switched to Salvage ART; two (1.8%) were lost to follow-up; two (1.8%) transferred out of the LTP; and three (13.6%) died. The median log10 VL for the second-line ART cohort fell from 4.86 (IQR, 4.42-5.20) at baseline to 2.06 (IQR, 1.81-2.31) HIV RNA copies/ml at the end of the study. At second-line ART baseline, 95.6% (n=109) of the cohort had at least 1 Nucleoside Reverse Transcriptase Inhibitor (NRTI) DRM and 92.1% (n=105) had at least 1 NNRTI DRM. Fifty percent (n=57) had 1 or more Thymidine Analogue mutations (TAMs), 50% (n=57) had 1 or more Type-I TAMs, while 45.7% (n=52) had 1 or more Type-II TAMs. No DRMs to the protease inhibitor drugs LPV/r were found. Among the 15 patients switched from second-line ART to salvage ART, 13 succeeded on treatment, one (6.7%) transferred out of the LTP and one (6.7%) died. The risk factors for VF and HIV-DR were determined using Cox Proportional Hazards Ratio (CPHR) analysis. Firstly, this analysis found that children and adolescents with suboptimal adherence to first-line ART were 37 times more likely to experience VF than those with optimal adherence (Hazard ratio [HR] = 36.99, 95% Confidence Interval [CI]=8.21-166.66, P<0.001). Secondly, those with severe acute malnutrition (SAM) were three times more likely to experience VF than the well-nourished (HR= 2.93, 95% CI=1.54-5.99, P<0.01). Thirdly, teenagers aged 14 to 17 years were three times more likely to have suboptimal adherence to first-line ART than children aged two to five years old (HR = 2.67, 95% CI=1.36-5.24, P<0.01). Fourthly, children and adolescents with suboptimal adherence to second-line ART were seven times more likely to experience VF than those with optimal adherence. Analysis of the salvage ART cohort data revealed a statistically significant increase in the number of NRTI DRMs (from 35 to 69) and OIs (from 22 to 34) among the 15 patients transitioned from second-line ART to salvage ART (P=6.12x10-10 and P=2.14x10-6, respectively). In light of these data, measures to reduce the rates of VF and HIV-1 among children and adolescents should be ramped up. These measures include increased child and adolescent HIV testing, counselling, linkage to care, and better adherence to ART. Moreover, teenagers should be encouraged to form peer support groups so that they can receive moral support and reduce self-stigma. On the HIV-1 treatment Programmatic level, the Government should revise their guidelines to recommend the use of protease inhibitors more frequently, as they have a higher barrier to HIV-DR. Given that HIV-1 infection is lifelong, reducing the rates of VF and HIV-DR among children and adolescents is crucial to ensuring that they will live long, healthy and productive lives.