Antiretroviral drug resistance poses a challenge to therapy efficacy. Although Integrase Strand Transfer Inhibitors (INSTIs) have a high genetic barrier to resistance, few studies in Brazil have focused on INSTI resistance. This study aimed to assess HIV mutations associated with INSTI resistance in vertically infected children in Maranhão and Pernambuco, Brazil.
MethodsWe conducted a retrospective analysis of antiretroviral resistance profiles in paediatric patients with virological failure receiving care through the Brazilian Unified Health System. Data on viral load, CD4+/CD8+ T-cell counts, and medication dispensing were obtained from Ministry of Health databases (SISGENO, SISCEL, and SICLOM). HIV-1 integrase sequences were analysed. Phylogenetic analysis (IQ-TREE) was used to identify subtypes and recombinants. Resistance-associated mutations were determined using the Stanford HIV Drug Resistance Database.
ResultsThirty-one paediatric patients (median age = 12-years) were included. Mean CD4+ count was 999 cells/mm³, and median viral load was 19,235 copies/mL. Subtype B and INSTI resistance mutations were detected in 74.5 % and 64.5 % of patients, respectively. High resistance rates to raltegravir (70.9 %) were observed, highlighting the importance of genotypic monitoring.
ConclusionA high frequency of INSTI resistance-associated mutations was identified, indicating the need for continued surveillance in this population.
Despite significant advances in Human Immunodeficiency Virus (HIV-1) treatment, antiretroviral drug resistance remains a critical challenge to the efficacy of Antiretroviral Therapy (ART), as it drives the accumulation of resistance-associated mutations, virological failure, and increased morbidity.1–5 Integrase Strand Transfer Inhibitors (INSTIs) are a class of antiretrovirals with a high genetic barrier to resistance, potent viral suppression, favourable safety profiles, and fewer drug-drug interactions.6 In Brazil, INSTIs are now recommended as the first- and second-line ART regimens following virological failure.7,8 Since 2017, the Brazilian Ministry of Health has adopted the preferred initial regimen of tenofovir disoproxil fumarate, lamivudine (3TC), and Dolutegravir (DTG), a second-generation INSTI.7,8 Recent national guidelines advocate for optimised dual therapy (3TC + DTG), particularly for patients with HIV-associated comorbidities.9 By 2020, approximately 410,000 Brazilian patients were receiving INSTI-based regimens.10 Given Brazil’s continental-scale implementation of these therapies, the widespread use of INSTIs may accelerate the emergence of resistance-associated mutations, potentially compromising their long-term effectiveness.
Tao et al. (2023) identified 99 sequences with INSTI resistance-associated mutations by analysing antiretroviral resistance profiles.11 Scutari et al. (2020) reported an INSTI resistance frequency of 36.4 %, among treatment-experienced individuals and those with long HIV infection duration.12 However, other studies have detected INSTI resistance mutations even in treatment-naïve patients.13,14 The identification of antiretroviral resistance mutations is widely recognised as a cost-effective strategy for optimising antiretrovirals and preventing the initiation or continuation of regimens that are likely to result in virological failure, which is a critical measure for HIV epidemiological control. The World Health Organization recommends routine resistance testing in children at diagnosis and after virological failure.15 According to Joint United Nations Programme on HIV/AIDS (UNAIDS), approximately one-third of children with HIV infection worldwide experience virological failure within 2-years of ART initiation.16–18
In Brazil, Ventosa-Cubillo et al. (2023) analysed HIV-1 sequences from 67 children and adolescents and identified INSTI-associated Drug Resistance Mutations (DRMs) in eight patients.19 Andrade et al. (2017) studied 117 children with HIV infection (mean age = 3.7-years) in Northern Brazil and reported high DRM frequencies, even among those who were treatment-naïve, particularly for non-nucleoside reverse transcriptase inhibitor-associated mutations.20 A retrospective Brazilian study on individuals with virological failure detected INSTI resistance mutations in 22.1 % of participants.10 Notably, a large multicentre study of mother-infant pairs in South Africa, Brazil, and Argentina demonstrated a strong association between maternal DRMs and their transmission to infants.21
Brazilian studies evaluating INSTI resistance have reported high frequencies of polymorphic variants associated with an increased risk of DRMs.22–24 However, data on INSTI resistance profiles, specifically in paediatric populations, remain scarce. Given Brazil's early adoption of INSTI-based regimens as first-line ART, understanding the resistance patterns in children is critical. Surveillance of DRM dissemination represents a fundamental strategy for HIV control, particularly in regions such as Brazil, where INSTIs were rapidly integrated into national treatment guidelines. This study aimed to characterise HIV-1 resistance mutations to INSTIs among children who were vertically infected and received care in the Maranhão and Pernambuco states of Northeast Brazil.
Material and methodsStudy population and settingThis retrospective study evaluated the antiretroviral resistance profiles of paediatric patients diagnosed with HIV who experienced virological failure and were undergoing antiretroviral treatment. The participants were followed by the Unified Health System in the states of Maranhão and Pernambuco in Northeast Brazil. Virological failure was defined as two consecutive detectable viral loads with a minimum interval of 4-weeks (https://www.gov.br/aids/pt-br/central-de-conteudo/pcdts/pcdt_hiv_modulo_1_2024.pdf). The study included children and adolescents (aged < 18-years) with a confirmed HIV infection who had undergone genotyping and sequencing of the HIV-1 integrase region. We obtained 12 sequences from the Maranhão State and 19 from the Pernambuco State, collected between 2018 and 2022.
Data were collected from the Maranhão Central Public Health Laboratory, Maranhão Specialised Healthcare Secretariat, Oswaldo Cruz University Hospital of the University of Pernambuco, and the Brazilian Ministry of Health. Databases containing information on viral load results (SISGENO), CD4+ and CD8+ T-lymphocyte counts (SISCEL), and antiretroviral medication dispensing data (SICLOM) were accessed. Data were collected from January to June 2024 using the medical records of patients who were diagnosed with HIV and followed up in outpatient clinics. Information was extracted from the available records of the study sites in the Medical Archives Services. A specific questionnaire was used to collect sociodemographic, laboratory, and clinical variables, which were subsequently analysed using Stata 13.0® software (United States).
Phylogenetic analysisHIV-1 query sequences (n = 31) were initially analysed using the REGA HIV-1 subtyping tool (version 3.0; http://dbpartners.stanford.edu:8080/RegaSubtyping/stanford-hiv/typingtool/) and COMET HIV-1 for the preliminary viral subtyping classification.25 Reference sequences were obtained from the HIV sequence database at the Los Alamos National Laboratory (http://www.hiv.lanl.gov/components/sequence/HIV/search/search.html). Sequence alignments of the study and reference sequences were performed using AliView and manually edited using BioEdit.26,27 Nucleotide substitution models were inferred based on the maximum likelihood and Bayesian information criterion using MEGA software version 11.28 A general time-reversible model with gamma distribution and invariant sites (GTR+G + I) was selected for the alignment analysis. Phylogenetic inferences were made using the maximum likelihood method in IQ-TREE online, with statistical support assessed using 1000 bootstrap replicates.29 The resulting phylogenetic trees were visualised using FigTree software version 1.4.3 (http://tree.bio.ed.ac.uk/software/figtree/).
Antiretroviral resistance analysisHIV-1 drug resistance mutation analysis was performed by submitting the sequences to the Stanford University HIV Drug Resistance Database (http://hivdb.stanford.edu), which utilises a standardised list of major HIV-1 DRMs from the Stanford HIV Database (https://cms.hivdb.org/prod/downloads/resistance-mutation-handout/resistance-mutation-handout.pdf). Antiretroviral resistance was classified according to the HIV Drug Resistance Database output as susceptible, intermediate (including both low- and intermediate-level resistance), and resistant (high-level resistance).
Statistical analysisData were analysed using the Stata 13.0® software. Medians and Interquartile Ranges (IQR) were determined for continuous variables, whereas categorical variables were expressed as counts and percentages. Statistical analyses involved Chi-Square tests for categorical variables and the Wilcoxon rank-sum test for non-parametric comparisons. The Mann-Whitney U test was used to assess the differences between medians. For all tests, the significance level was set at p < 0.05.
ResultsThis study analysed 31 paediatric patients diagnosed with HIV-1 infection who experienced virological failure, with a median age of 12-years (IQR: 9–16). Of them, 18 (58 %) were male patients and 13 (42 %) female patients. The median CD4+ T-cell count and viral load at the time of virological failure were 999 cells/mm3 (IQR: 707–1432.5) and 19,235 copies/mL (IQR: 1508.5–270,375), respectively. HIV-1 subtyping of the integrase region revealed subtype B (n = 23, 74.2 %), BF recombinants (n = 5, 16.1 %), and subtype F1 (n = 3, 9.7 %) (Fig. 1). Additionally, six patients (19.4 %) had previously used INSTIs (Table 1). Analysis of resistance to INSTIs (Bictegravir [BIC], Dolutegravir [DTG], Elvitegravir [EVG], Cabotegravir [CAB], and Raltegravir [RAL]) showed that 20 sequences (64.5 %) remained susceptible to BIC and DTG, representing the INSTIs with the lowest resistance levels. In contrast, the highest resistance rate was observed for RAL (n = 21, 67.8 %) (Fig. 2). Twenty-one HIV-1 sequences (67.8 %) exhibited major INSTI resistance mutations. The most frequent resistance mutations were G140S and Q148H (both n = 10, 32.2 %). The Q148K mutation in the integrase gene was observed in 11 patients (35.5 %). Importantly, the Q148H/K mutations conferred resistance to all INSTIs, indicating a high level of resistance in the paediatric population. The N155H mutation was detected in seven patients (22.6 %), of whom two have been receiving antiretroviral therapy that includes INSTIs, and conferred resistance to CAB, EVG, and RAL. Mutations at integrase position 143 (Y143H/R/G) were documented in four patients and conferred resistance to RAL. Among the current ART regimens administered to the patients, the most common were Zidovudine (AZT) + Lamivudine (3TC) + Lopinavir/ritonavir (LPV/r) (n = 10; 32.2 %), AZT + 3TC + Nevirapine (NVP) (n = 6; 19.4 %), and AZT + 3TC + efavirenz (EFV) (n = 5; 16.1 %). The duration of the current ART regimen ranged from 1- to 10-years, with a median of 5.5-years (Table 2).
Demographic and laboratory characteristics of paediatric patients living with HIV-1 under virological failure ‒ Northeast Brazil.
HIV-1 sequences with associated demographic, laboratory, viral subtype, and resistance mutation data.
3TC, Lamivudine; ABC, Abacavir; AZT, Zidovudine; DRV/r, Darunivir/ritonavir; EFV, Efavirenz; LPV/r, Lopinavir/Ritonavir; NVP, Nevirapine; RAL, Raltegravir.
The current study aimed to retrospectively analyse the antiretroviral resistance profiles of INSTIs in virologically failing patients under 16-years of age who were followed by Brazil's Unified Health System. Given the scarcity of research on ART resistance in paediatric populations, our study population selection is a key strength. We analysed 31 HIV-1 integrase region sequences from virologically failing children and adolescents in Northeast Brazil for antiretroviral resistance mutations. Six patients (19.4 %) received INSTIs before genotypic testing for virological failure. The highest resistance frequency was observed for RAL (n = 21; 67.8 %), indicating a high-level resistance to at least one INSTI.
The predominant HIV-1 subtype was B (74.2 %), which is consistent with phylogenetic data from previous Brazilian studies that identified this as the most common subtype in both the Northeast and other Brazilian regions, except for Southern Brazil.29–31 Among the non-B subtypes, BF showed the highest incidence (19 %), aligning with the findings of a previous study conducted in the Maranhão State, Northeast Brazil. However, 2017 data identified HIV-F as the most common subtype in the Pernambuco State, Northeast Brazil,32 which contrasts with our findings. The predominance of subtype B may reflect its status as the first subtype to be isolated in the Western world, which facilitated its subsequent spread.
BIC and DTG showed the lowest resistance rates, with 64.5 % of sequences remaining susceptible. Notably, DTG, included in Brazil's first-line adult ART regimen as per the Ministry of Health guidelines, has recently been formulated as dispersible tablets for HIV treatment and prevention in children aged 2-months to 6-years,33 significantly improving paediatric treatment feasibility.
Regarding the most prevalent INSTI resistance mutations, in descending order of frequency, we identified G140S, Q148H, N155H, Q148K, Y143H, Y143HR, and Y143R. The high frequency of G140S and Q148H mutations (both n = 10; 32.2 %) raises significant concerns, as they confer resistance to all INSTIs. Mutations at positions 140, 143, 148, and 155 of the HIV-1 integrase gene induce resistance to multiple INSTIs.33 For instance, N155H (22.6 %, n = 7/31) conferred resistance to most INSTIs except DTG. These findings in the paediatric population may suggest consequences associated with prior INSTI exposure, either through the individual’s treatment history or potential maternal transmission. Mutations most associated with INSTI resistance (G140S, Q148H, and N155H) were observed in various combinations among patients infected with subtypes B and F1. Considering the population's median age of 12-years, approximately 35 % showed DTG resistance (only introduced in Brazil's public health system in 2016), and only 19.4 % (n = 6/31) had prior INSTI exposure. We hypothesised that INSTI resistance may be emerging rapidly, with treatment adherence likely playing a crucial preventive role.34 Mutation distribution analysis revealed that G140S and Q148H frequently co-occurred in patients infected with subtype B virus, whereas N155H appeared in patients infected with subtype B or F1 virus. Patients infected with recombinant BF virus showed INSTI-sensitive profiles, with only one patient exhibiting resistance mutations.
Despite the inclusion criterion adopted in the study, selecting only patients with virological failure who had genotyping results available may have introduced a selection bias towards individuals already suspected of having HIV drug resistance, continuous genotypic monitoring of these patients is essential for therapeutic adjustments to improve clinical outcomes, particularly in paediatric populations. This study highlights the critical need for the surveillance of INSTI resistance mutations, as emerging resistance may compromise treatment efficacy and safety. Systematic resistance monitoring ensures that INSTIs remain a viable therapeutic option, while advancing HIV research to develop more potent treatments and tailored prevention strategies for vulnerable populations. Through these efforts, healthcare systems can implement preventive measures, where early detection of viral resistance may curb the transmission of drug-resistant strains and guide policymakers in optimising antiretroviral selection.
Ethics approval statementThis study was approved by the Research Ethics Committee of Federal University of Pernambuco, under approval number 5.180.229 (December 20, 2021), in accordance with the guidelines of the National Health Council of Brazil (Resolution 466/12).
Patient consent statementThe ethics committee accepted the exemption from the patient consent statement, as only HIV-1 medical records and genomic data were obtained.
Permission to reproduce material from other sourcesPermission to reproduce material from other sources was obtained where necessary. All reproduced content is appropriately cited and used with authorization from the original copyright holders.
Declaration of generative AI and AI-assisted technologies in the writing processNone.
FundingThis research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Data availabilityThe data that support the findings of this study are available from the corresponding author upon reasonable request.
The authors declare no conflicts of interest.
We acknowledge the support from the Scientific Initiation Programme of the National Council for Scientific and Technological Development (PIBICCNPq) and Instituto de Medicina Integral Professor Fernando Figueira (IMIP).







