HIV-1 Disease Progression and Drug Resistance Mutations among Children on First-Line Antiretroviral Therapy in Ethiopia

Sub-Saharan Africa bears a disproportionate HIV burden, with Ethiopia among the most affected countries. HAART is central to preventing progression to AIDS, and CD4 count and viral load are key prognostic markers. Despite HAART scale-up in Ethiopia, pediatric coverage has lagged (23% in 2018), and emerging HIV drug resistance—particularly with regimens using NNRTIs with low genetic barriers and prior nevirapine exposure for PMTCT—threatens treatment outcomes. Viral replication under suboptimal drug pressure leads to resistance mutations, cross-resistance within drug classes, reduced future options, and poorer virologic and immunologic outcomes—often worse in children than adults. There is limited empirical evidence in Ethiopian children on immunologic outcomes and drug resistance. This study aimed to describe HIV-1 disease progression and drug resistance and evaluate the effect of drug resistance on disease progression among children on first-line ART in Ethiopia.

The study contextualizes prior evidence showing high HIV disease progression and HIVDR in adults in SSA, WHO guidance on using CD4 and viral load to monitor progression, and suboptimal pediatric HAART coverage in Ethiopia. Literature highlights that NNRTI-based regimens have low genetic barriers to resistance; single-dose nevirapine for PMTCT contributes to pediatric resistance; and resistance mutations can confer cross-resistance. Prior studies report poorer immune recovery with prolonged HAART exposure, associations of virologic failure and resistance with disease progression, and links between vitamin D deficiency, inflammation (hsCRP), and adverse HIV outcomes. Pediatric resistance patterns often include NNRTI mutations (e.g., K103N, Y181C) and NRTI mutations (e.g., M184V, K65R), while PI resistance is less frequent due to limited PI use.

Design and setting: A longitudinal study (2017–2019) of HIV-infected children (<15 years) on first-line ART at 40 ART sites across Ethiopia, nested within a national cohort at 63 facilities. ART in Ethiopia has been available free nationwide since 2005. First-line regimens historically included d4T+3TC+NVP, d4T+3TC+EFV, ZDV+3TC+EFV, and ZDV+3TC+NVP; second-line options included (ddI or TDF)+ABC+(LPV/r or SQV/r or NFV or IND/r). Sampling and population: Inclusion criteria were children on first-line ART for at least 6 months. Of 63 facilities and 13,649 patients in the parent study, 40 facilities had pediatric ART; all children at these facilities were included (n=554; analyzed n=551 with available data). Outcomes: Two primary outcomes were (1) disease progression measured as immunosuppression (CD4 <200 cells/mm3) over a retrospective 12-year period (2007–2019), and (2) HIV drug resistance (HIVDR). Data collection: Guardians were interviewed for sociodemographics and individual factors; medical records were abstracted for clinical and laboratory data (hemoglobin, CD4 counts, ART regimen history). Children with baseline viral load ≥1000 copies/mL were followed for 6 months with enhanced adherence counseling and re-tested. Laboratory methods: Whole blood collected in EDTA tubes; plasma separated by centrifugation (2000 rpm). HIV-1 viral load measured at regional labs using COBAS AmpliPrep/COBAS TaqMan or Abbott RealTime HIV-1. CD4 T-cell counts measured using FACScount. Hemoglobin measured with CELLDYN analyzer. hsCRP and vitamin D measured at national lab via Elecsys 2010 analyzer. For children with virologic failure (two VLs ≥1000 copies/mL 6 months apart with adherence counseling), HIV-1 pol gene amplified and sequenced using an in-house assay validated by the Chinese CDC on ABI-3730; sequences edited via ReCall; resistance interpreted with Stanford HIVDB v9.0. Definitions: Immunosuppression: CD4 <200 cells/mm3 after 6 months of HAART. Disease progression rate: change over time in immunosuppression status. Virologic failure: two consecutive VLs ≥1000 copies/mL within 6 months with adherence intervention. Drug resistance: per Stanford HIVDB; children with VL <1000 were considered susceptible in overall HIVDR analysis. Vitamin D deficiency: <20 ng/mL. Inflammation: hsCRP >3 mg/L. Statistical analysis: Descriptive stats for demographics/clinical variables. Disease progression illustrated with Kaplan–Meier; predictors assessed via Cox proportional hazards regression, reporting adjusted hazard ratios (AHRs) with 95% CIs (p<0.05). HIVDR computed via Stanford HIVDB; determinants assessed with logistic regression—variables with p<0.2 in bivariate entered multivariable model; adjusted odds ratios (AORs) with 95% CIs reported (p<0.05). Analyses in STATA 16.

• Sample characteristics: Mean age 9.33 years (SD ±2.21); 58.3% <10 years; 89.7% urban. Most common first-line regimens: AZT+3TC+NVP (35.9%), d4T+3TC+NVP (30.3%), AZT+3TC+EFV (14.0). HAART substitution history in 47.5%.
• Virologic suppression improved from baseline to 6 months: VL >1000 copies/mL decreased from 37.8% to 21.4% after adherence counseling and follow-up.
• Inflammation and vitamin D: 61.3% had hsCRP >3 mg/L; 51.5% had vitamin D <20 ng/mL at measurement.
• Disease progression (immunosuppression): Overall immunosuppression over 12 years was 11.25% (95% CI: 7.5–15.1). Rate of disease progression: 6.3 per 100 person-years (95% CI: 4.2–8.1). Immunosuppression at mean 10.5 years on HAART was 38.2%, increasing to 67.8% at 12 years (p<0.01). Higher immunosuppression among those with OIs (22.64%), unsuppressed VL (26.27%), hsCRP >3 mg/L, vitamin D <20 ng/mL, and longer HAART duration (>133 months).
• HIV-1 drug resistance (HIVDR): Overall HIVDR prevalence 14.52% (95% CI: 10.21–12.76); multi-drug resistance 6.2%. Among children with VL >1000 copies/mL, HIVDR was 67.8%. Class-specific resistance: NNRTI 11.4%; NRTI 10.1%; PI high-level resistance rare (NFV 0.9%). High-level resistance to specific drugs among resistant isolates included: 3TC 60.9%, FTC 60.9%, DDI 43.6%, ABC 34%, TDF 22.7%, d4T 21.8%; NNRTIs: NVP 40.0%, EFV 39.1%, RPV 30.0%, ETR 12.7%.
• Common resistance mutations: NRTI—M184V (30.1%), K65R (12.1%), D67N (5.6%), K70R (5.6%), Y115F (5.0). NNRTI—K103N (14.8%), Y181C (11.8%), G190A (7.7%), V106M (5%).
• Predictors of disease progression (Cox model): History of OI AHR 3.38 (95% CI: 1.84–6.23); HIVDR AHR 2.21 (1.42–3.64); virologic failure AHR 2.82 (1.20–3.53); vitamin D <20 ng/mL AHR 4.53 (2.07–9.94); all p≤0.02.
• Determinants of HIVDR (logistic regression): Orphan status AOR 1.81 (95% CI: 1.15–3.08; p=0.02); history of HAART substitution AOR 4.76 (2.09–6.46; p=0.04); hemoglobin ≤12 g/dL AOR 1.24 (1.07–2.04; p=0.03).

The study demonstrates substantial HIV disease progression over time among Ethiopian children on first-line HAART, with immunosuppression rising markedly after about a decade on therapy. This progression is closely associated with virologic failure, HIV drug resistance, vitamin D deficiency, and history of opportunistic infections, aligning with prior evidence that sustained viremia and resistance undermine immune recovery. The high pediatric HIVDR prevalence (14.5%)—notably higher than reported in Ethiopian adults—likely reflects challenges unique to children (adherence, regimen changes, PMTCT exposure). The mutation spectrum (M184V, K65R, K103N, Y181C, G190A) is consistent with NNRTI- and NRTI-based first-line regimen pressures common in the setting; PI resistance was rare due to limited PI use. Orphan status, HAART substitution, and lower hemoglobin were independently associated with resistance, suggesting social vulnerability and advanced disease contribute to resistance development. These findings underscore the need for earlier detection of virologic failure, routine resistance monitoring, optimization of pediatric regimens away from low barrier NNRTIs, attention to nutritional/vitamin D status, and minimizing unnecessary regimen substitutions. Collectively, the results address the research question by quantifying progression and resistance burdens and identifying modifiable predictors that can inform programmatic interventions to improve pediatric HIV outcomes.

Immunosuppression increased substantially with longer HAART duration, and HIV drug resistance was common, particularly among children with virologic failure. Integrating HIVDR testing into routine pediatric HIV care, strengthening viral load monitoring, addressing vitamin D deficiency (e.g., supplementation), and minimizing regimen substitutions may improve immunologic outcomes. Future research should prospectively assess interventions to reduce virologic failure and resistance, evaluate transitions to higher barrier regimens in children, and investigate causal pathways linking vitamin D status, inflammation, and immune recovery.

The study included only children alive and on treatment; those lost to follow-up or deceased—potentially with poorer immunologic outcomes—were not captured, possibly underestimating disease progression and resistance. Additional limitations include retrospective retrieval of long-term CD4 data and potential confounding related to changes in ART regimens and programmatic practices over the 12-year period.