Comparison of algorithms that interpret genotypic HIV-1 drug resistance to determine the prevalence of transmitted drug resistance

Temporal Trends in HIV-1 Subtypes and Antiretroviral Drug Resistance Mutations in Istanbul, Türkiye (2021-2024): A Next-Generation Sequencing Study

HIV-1 genotyping and drug resistance tests are routinely performed in virology laboratories in some countries, aiding clinical management. In Istanbul, between January 2021 and March 2024, plasma samples from 1029 HIV-1-infected patients were analyzed using the NGS method, and mutation and drug resistance results were retrospectively evaluated alongside demographic data. Subtype B (54.4%) was most frequent in Turkish patients, while Subtype A1 (43.5%) was predominant among foreign nationals. The most common CRFs were CRF02_AG (3.8%) and CRF56_cpx (1.6%). According to the change in detection rates during the study period, Subtype B decreased, and Subtype A increased. The most frequent mutations detected were A62V (38.7%) and M184V (22.4%) for NRTIs; E138A (55.5%) and E138G (11.5%) for NNRTIs; M46I (33.3%) and M46L (25%) for PIs; and E92Q and G for INIs (total rate: 35.2%). Darunavir/ritonavir had the highest sensitivity rate, while resistance rates for NNRTIs and INIs increased over time. We anticipate that this study, in which we evaluate the routine use of an FDA-approved NGS kit alongside integrated bioinformatics data analysis and automated reporting software for the first time in Türkiye, will contribute to both national and international molecular epidemiological data and public health strategies by providing reliable results that align with international standarts.

Advances in computational approaches in identifying synergistic drug combinations

Accumulated empirical clinical experience, supported by animal or cell line models, has initiated efforts of predicting synergistic combinatorial drugs with more-than-additive effect compared with the sum of the individual agents. Aiming to construct better computational models, this review started from the latest updated data resources of combinatorial drugs, then summarized the reported mechanism of the known synergistic combinations from aspects of drug molecular and pharmacological patterns, target network properties and compound functional annotation. Based on above, we focused on the main in silico strategies recently published, covering methods of molecular modeling, mathematical simulation, optimization of combinatorial targets and pattern-based statistical/learning model. Future thoughts are also discussed related to the role of natural compounds, drug combination with immunotherapy and management of adverse effects. Overall, with particular emphasis on mechanism of action of drug synergy, this review may serve as a rapid reference to design improved models for combinational drugs.

Impact of Changes Over Time in the Stanford University Genotypic Resistance Interpretation Algorithm

INTRODUCTION

The Stanford HIV-1 genotypic resistance interpretation algorithm has changed substantially over its lifetime. In many studies, the algorithm version used is not specified. It is easy to assume that results across versions are comparable, but the effects of version changes on resistance calls are unknown. We evaluate these effects for 20 antiretroviral drugs.

METHODS

We calculated resistance interpretations for the same 5993 HIV-1 sequences, from participants in AIDS Clinical Trials Group studies, under 14 versions of the Stanford algorithm from 2002 to 2017. Trends over time were assessed using repeated-measures logistic regression. Changes in rule structure and scoring were examined.

RESULTS

For most drugs, the proportion of high-level resistance calls on the same sequences was greater using more recent algorithm versions; 16/20 drugs showed significant upward trends. Some drugs, especially tenofovir, had a substantial increase. Only darunavir had a decrease. Algorithm changes impacted calls for subtype C more than B. For intermediate and high-level resistance combined, effects were weaker and more varied. Over time, rules in the Stanford algorithm have become more complex and contain more subrules. The types of rule changes responsible for trends varied widely by drug.

DISCUSSION

Reporting the Stanford algorithm version used for resistance analysis is strongly recommended. Caution should be used when comparing results between studies, unless the same version of the algorithm was used. Comparisons using different Stanford versions may be valid for drugs with few changes over time, but for most comparisons, version matters, and for some drugs, the impact is large.

High levels of virological failure with major genotypic resistance mutations in HIV-1-infected children after 5 years of care according to WHO-recommended 1st-line and 2nd-line antiretroviral regimens in the Central African Republic: A cross-sectional study

A large cohort of 220 HIV-1-infected children (median [range] age: 12 [4-17] years) was cared and followed up in the Central African Republic, including 198 in 1st-line and 22 in 2nd-line antiretroviral regimens. Patients were monitored clinically and biologically for HIV-1 RNA load and drug resistance mutations (DRMs) genotyping. A total of 87 (40%) study children were virological responders and 133 (60%) nonresponders. In children with detectable viral load, the majority (129; 97%) represented a virological failure. In children receiving 1st-line regimens in virological failure for whom genotypic resistance test was available, 45% displayed viruses harboring at least 1 DRM to NNRTI or NRTI, and 26% showed at least 1 major DRM to NNRTI or NRTI; more than half of children in 1st-line regimens were resistant to 1st-generation NNRTI and 24% of the children in 1st-line regimens had a major DRMs to PI. Virological failure and selection of DRMs were both associated with poor adherence. These observations demonstrate high rate of virological failure after 3 to 5 years of 1st-line or 2nd-line antiretroviral treatment, which is generally associated with DRMs and therapeutic failure. Overall, more than half (55%) of children receiving 1st-line antiretroviral treatment for a median of 3.4 years showed virological failure and antiretroviral-resistance and thus eligible to 2nd-line treatment. Furthermore, two-third (64%) of children under 2nd-line therapy were eligible to 3rd-line regimen. Taken together, these observations point the necessity to monitor antiretroviral-treated children by plasma HIV-1 RNA load to diagnose as early as possible the therapeutic failure and operate switch to a new therapeutic line.

Reporting on the prevalence of antiretroviral drug resistance in a regional HIV population over 20 years: a word of caution

BACKGROUND

Failure to achieve complete viral suppression with antiretroviral drugs (ARV) may lead to uncontrolled HIV replication, ARV resistance and negative outcomes. Monitoring and reporting of HIV resistance trends is important but problematic. We examined prevalent resistance rates in an HIV population over 20 years and document how rates may appear to vary greatly based solely on which parameters are utilized.

METHODS

We determined the annual use of genotypic antiretroviral resistance testing (GART) from 1995 to 2014 for all patients receiving HIV care in southern Alberta, Canada, and the presence of resistance mutations in those tested. The impact on prevalent resistance rates of using cumulative or latest GART was also determined.

RESULTS

Between 1995 and 2014, the number of patients with GART increased from <1% to 71%. Prevalent resistance in patients with GART decreased from a high of 52% in 2003 to 25.8% in 2014. However, if prevalence rates were reported using all active patients as denominator, including those without GART, prevalence increased from 0.7% to 18.5%. Prevalence rates were 7% to 9% higher in any given year if cumulative GART rather than latest GART results were used.

CONCLUSIONS

While prevalence resistance rates are decreasing, the precise rates being reported may vary due to increasing number of patients tested annually, using either the entire population as denominator or only patients with GART, and using either last or cumulative GART. Defining these parameters is critical if prevalence is to be compared over time or between HIV populations.

Comparison of genotypic and virtual phenotypic drug resistance interpretations with laboratory-based phenotypes among CRF01_AE and subtype B HIV-infected individuals

HIV drug resistance assessments and interpretations can be obtained from genotyping (GT), virtual phenotyping (VP) and laboratory-based phenotyping (PT). We compared resistance calls obtained from GT and VP with those from PT (GT-PT and VP-PT) among CRF01_AE and subtype B HIV-1 infected patients. GT predictions were obtained from the Stanford HIV database. VP and PT were obtained from Janssen Diagnostics BVBA's vircoType(TM) HIV-1 and Antivirogram®, respectively. With PT assumed as the "gold standard," the area under the curve (AUC) and the Bland-Altman plot were used to assess the level of agreement in resistance interpretations. A total of 80 CRF01_AE samples from Asia and 100 subtype B from Janssen Diagnostics BVBA's database were analysed. CRF01_AE showed discordances ranging from 3 to 27 samples for GT-PT and 1 to 20 samples for VP-PT. The GT-PT and VP-PT AUCs were 0.76-0.97 and 0.81-0.99, respectively. Subtype B showed 3-61 discordances for GT-PT and 2-75 discordances for VP-PT. The AUCs ranged from 0.55 to 0.95 for GT-PT and 0.55 to 0.97 for VP-PT. Didanosine had the highest proportion of discordances and/or AUC in all comparisons. The patient with the largest didanosine FC difference in each subtype harboured Q151M mutation. Overall, GT and VP predictions for CRF01_AE performed significantly better than subtype B for three NRTIs. Although discrepancies exist, GT and VP resistance interpretations in HIV-1 CRF01_AE strains were highly robust in comparison with the gold-standard PT.

Transmitted Drug Resistance Mutations in Antiretroviral-Naïve Injection Drug Users with Chronic HIV-1 Infection in Iran

The growing incidence and transmission of drug resistant HIV-1 strains due to widespread use of antiretroviral therapy (ART) can jeopardize the success of first-line ART. While there is a known moderate prevalence of transmitted drug resistance (TDR) among newly infected Iranians, no data exist about the rate of these primary resistance mutations among the ART-naïve, chronically infected individuals who are, in fact, the main candidates for ART initiation. To address this issue, we collected blood samples from 40 ART-naïve injection drug-users (IDUs) with chronic HIV-1 infection (seroconversion time ranging from 2 to 9 years) living in Sanandaj, Iran, followed by sequencing of the protease and reverse-transcriptase regions from their HIV-1 genome. Phylogenetic analyses of the sequenced regions revealed that all samples were CRF35_AD. Transmitted resistance mutations were interpreted as surveillance drug-resistant mutations (SDRMs) based on the world health organization (WHO) algorithm. The frequency of SDRMs to any class of antiretroviral drugs was 15%, which included mutations to nucleoside reverse transcriptase inhibitors (NRTIs, 10%), with M41L and M184V as the most common (5%), and non-nucleoside reverse transcriptase inhibitors (NNRTIs, 5%), with K103N as the only detected mutation (5%). Although not in the WHO SDRMs list, several minor protease inhibitor resistant mutations listed in the International Antiviral Society-USA panel were identified, of which M36I, H69K, L89M/V/I (each one 100%) and K20R/T (92.5%) can be considered as polymorphic signatures for CRF35_AD.The relatively high rate of TDR mutations in our study raises concerns about the risk of treatment failure in chronically infected IDUs of Sanandaj city. These results suggest that routine resistance testing should be considered before the therapy initiation in this area. Additional surveillance studies are required to generalize this deduction to other cities of Iran.

Overcoming drug resistance through in silico prediction

Prediction tools are commonly used in pre-clinical research to assist target selection, to optimize drug potency or to predict the pharmacological profile of drug candidates. In silico prediction and overcoming drug resistance is a new opportunity that creates a high interest in pharmaceutical research. This review presents two main in silico strategies to meet this challenge: a structure-based approach to study the influence of mutations on the drug-target interaction and a system-biology approach to identify resistance pathways for a given drug. In silico screening of synergies between therapeutic and resistant pathways through biological network analysis is an example of technique to escape drug resistance. Structure-based drug design and in silico system biology are complementary approaches to reach few objectives at once: increase efficiency, reduce toxicity and overcoming drug resistance.

IDEPI: rapid prediction of HIV-1 antibody epitopes and other phenotypic features from sequence data using a flexible machine learning platform

Since its identification in 1983, HIV-1 has been the focus of a research effort unprecedented in scope and difficulty, whose ultimate goals--a cure and a vaccine--remain elusive. One of the fundamental challenges in accomplishing these goals is the tremendous genetic variability of the virus, with some genes differing at as many as 40% of nucleotide positions among circulating strains. Because of this, the genetic bases of many viral phenotypes, most notably the susceptibility to neutralization by a particular antibody, are difficult to identify computationally. Drawing upon open-source general-purpose machine learning algorithms and libraries, we have developed a software package IDEPI (IDentify EPItopes) for learning genotype-to-phenotype predictive models from sequences with known phenotypes. IDEPI can apply learned models to classify sequences of unknown phenotypes, and also identify specific sequence features which contribute to a particular phenotype. We demonstrate that IDEPI achieves performance similar to or better than that of previously published approaches on four well-studied problems: finding the epitopes of broadly neutralizing antibodies (bNab), determining coreceptor tropism of the virus, identifying compartment-specific genetic signatures of the virus, and deducing drug-resistance associated mutations. The cross-platform Python source code (released under the GPL 3.0 license), documentation, issue tracking, and a pre-configured virtual machine for IDEPI can be found at https://github.com/veg/idepi.

Comparison of predicted susceptibility between genotype and virtual phenotype HIV drug resistance interpretation systems among treatment-naive HIV-infected patients in Asia: TASER-M cohort analysis

Background

Accurate interpretation of HIV drug resistance (HIVDR) testing is challenging, yet important for patient care. We compared genotyping interpretation, based on the Stanford University HIV Drug Resistance Database (Stanford HIVdb), and virtual phenotyping, based on the Janssen Diagnostics BVBA’s vircoTYPE™ HIV-1, and investigated their level of agreement in antiretroviral (ARV) naive patients in Asia, where non-B subtypes predominate.

Methods

Sequences from 1301 ARV-naive patients enrolled in the TREAT Asia Studies to Evaluate Resistance – Monitoring Study (TASER-M) were analysed by both interpreting systems. Interpretations from both Stanford HIVdb and vircoTYPE™ HIV-1 were initially grouped into 2 levels: susceptible and non-susceptible. Discrepancy was defined as a discordant result between the susceptible and non-susceptible interpretations from the two systems for the same ARV. Further analysis was performed when interpretations from both systems were categorised into 3 levels: susceptible, intermediate and resistant; whereby discrepancies could be categorised as major discrepancies and minor discrepancies. Major discrepancy was defined as having a susceptible result from one system and resistant from the other. Minor discrepancy corresponded to having an intermediate interpretation in one system, with a susceptible or resistant result in the other. The level of agreement was analysed using the prevalence adjusted bias adjusted kappa (PABAK).

Results

Overall, the agreement was high, with each ARV being in “almost perfect agreement”, using Landis and Koch’s categorisation. Highest discordance was observed for efavirenz (75/1301, 5.8%), all arising from susceptible Stanford HIVdb versus non-susceptible vircoTYPE™ HIV-1 predictions. Protease Inhibitors had highest level of concordance with PABAKs all above 0.99, followed by Nucleoside Reverse Transcriptase Inhibitors with PABAKs above 0.97 and non-NRTIs with the lowest PABAK of 0.88. The 68/75 patients with discordant efavirenz results harboured the V179D/E mutations compared to 7/1226 with no efavirenz discrepancy (p-value <0.001). In the 3-level comparison, all but one of the discrepancies was minor.

Conclusions

The two systems agreed well with lowest concordance observed for efavirenz. When interpreting HIVDR, especially in non-B subtypes, clinical correlation is crucial, in particular when efavirenz resistance is interpreted based on V179D/E.

Baseline CD4(+) T-cell counts and weighted background susceptibility scores strongly predict response to maraviroc regimens in treatment-experienced patients

BACKGROUND

Maraviroc-containing regimens are known to achieve virological suppression in many treatment-experienced patients. This study aimed to evaluate a more rigorous methodological approach to resistance-response analysis in large clinical studies and to better establish which subpopulations of patients were most likely to benefit from maraviroc by refining and extending previous subgroup analyses from the MOTIVATE studies.

METHODS

Individual weighted optimized background therapy (OBT) susceptibility scores were calculated by combining genotypic or phenotypic resistance testing with prior drug use information. Virological response (HIV-1 RNA<50 copies/ml at week 48) using each of these methods was compared with a commonly used method of counting active drugs. Baseline predictors of virological response, including weighted or unweighted scoring, maraviroc use, baseline CD4(+) T-cell count, HIV-1 plasma viral load and tropism, were assessed by logistic regression modelling.

RESULTS

Genotypic or phenotypic weighted methods were similarly predictive of virological response and better than counting active drugs. Weighted scoring and baseline CD4(+) T-cell count were the strongest predictors of virological response (P<0.0001): ≈70% of maraviroc patients with a weighted score ≥2 had a virological response, rising to ≈80% when the baseline CD4(+) T-cell count was ≥50 cells/mm(3).

CONCLUSIONS

Approximately 80% of patients with a CD4(+) T-cell count ≥50 cells/mm(3) receiving maraviroc with the equivalent of at least two fully active agents achieved HIV-1 RNA<50 copies/ml at week 48 in the MOTIVATE studies. Genotypic and phenotypic weighted scores were similarly predictive of virological response.

Impact of HIV-1 group O genetic diversity on genotypic resistance interpretation by algorithms designed for HIV-1 group M

BACKGROUND

HIV-1 group O (HIV-O) is characterized by a high genetic divergence from HIV-1 group M viruses. Little is known about the therapeutic impact of this diversity. The aim of this study was to assess in a large series of samples (1) the genotypic impact of natural polymorphism of the HIV-O reverse transcriptase and protease genes; and (2) the predictive value of resistance interpretation algorithms developed for HIV-1 group M when used for highly mutated HIV-O viruses.

METHODS

Sixty-eight antiretroviral-naive and 9 highly antiretroviral-experienced HIV-O-infected patients were included. The viruses were sequenced and resistance-associated mutations were identified using 3 different algorithms (Agence Nationale de Recherches sur le SIDA et les hépatites virales, Rega, Stanford).

RESULTS

All HIV-O samples naturally exhibited the A98G and V179E resistance mutations in the reverse transcriptase region; 54% of samples presented the Y181C mutation, conferring resistance to nonnucleoside reverse transcriptase inhibitors. Twelve minor resistance mutations, present in more than 75% of the protease sequences, led to the different algorithms giving discrepant results for nelfinavir and saquinavir susceptibility. A marked virological response was observed in 8 of the 9 antiretroviral-experienced patients, despite the prediction of limited activity of the combination for 5 to 8 patients according to the algorithm used.

CONCLUSIONS

The high level of natural polymorphism in HIV-O genes, and the important discrepancies between genotypic resistance interpretation and the virological response, emphasize the need for resistance algorithm rules better adapted to HIV-O.

The relatedness of HIV epidemics in the United States-Mexico border region

Phylogeography can improve the understanding of local and worldwide HIV epidemics, including the migration of subepidemics across national borders. We analyzed HIV-1 sequences sampled from Mexico and San Diego, California to determine the relatedness of these epidemics. We sampled the HIV epidemics in (1) Mexico by downloading all publicly available HIV-1 pol sequences from antiretroviral-naive individuals in GenBank (n = 100) and generating similar sequences from cohorts of injection drug users and female sex workers in Tijuana, Mexico (n = 27) and (2) in San Diego, California by pol sequencing well-characterized primary (n = 395) and chronic (n = 267) HIV infection cohorts. Estimates of population structure (F(ST)), genetic distance cluster analysis, and a cladistic measure of migration events (Slatkin-Maddison test) were used to assess the relatedness of the epidemics. Both a test of population differentiation (F(ST) = 0.06; p < 0.01) and a cladistic estimate of migration events (84 migrations, p < 0.01) indicated that the Tijuana and San Diego epidemics were not freely mixing. A conservative cluster analysis identified 72 clusters (two or more sequences), with two clusters containing both Mexican and San Diego sequences (permutation p < 0.01). Analysis of this very large dataset of HIV-1 sequences suggested that the HIV-1 epidemics in San Diego, California and Tijuana, Mexico are distinct. Larger epidemiological studies are needed to quantify the magnitude and associations of cross-border mixing.

Stable and low prevalence of transmitted HIV type 1 drug resistance despite two decades of antiretroviral therapy in Hong Kong

Transmitted HIV resistance is of both clinical and public health importance. Baseline genotypic resistance testing was performed for HIV-1-infected treatment-naive patients who were newly diagnosed between 2003 and 2007 and attended the government HIV clinic in Hong Kong. International AIDS Society-USA mutation figures and the Stanford resistance interpretation algorithm were used to identify resistance mutations and drug susceptibility, respectively. The pattern and factors associated with resistance were examined. The presence of one or more IAS-USA resistance mutations was found in 26 (3.6%) of 731 patients over the 5-year study period. Overall, protease inhibitor (PI) resistance mutations were most common (16), followed by nucleoside reverse transcriptase inhibitors (NRTIs) (8) and nonnucleoside reverse transcriptase inhibitors (NNRTIs) (3). Resistance to drugs in one, two, and three classes was present in 25 (3.4%), 1 (0.1%), and 0, respectively. Seventy-eight (10.7%) had strains of reduced susceptibility, as predicted by the Stanford algorithm to display at least low-level resistance to one or more drugs of the three classes. Intermediate or high-level resistance was found in 1.6% overall, and in descending order for NRTIs, PIs, and NNRTIs. There was no temporal trend of increase in resistance. Sex between men, Chinese ethnicity, and lower baseline CD4 were associated with harboring resistant strains as elucidated by either method. We conclude that transmitted HIV-1 drug resistance is uncommon in up to two decades of antiretroviral therapy in Hong Kong. The situation has to be continually monitored for any change in significance.

A comparison of interpretation by three different HIV type 1 genotypic drug resistance algorithms using sequences from non-clade B HIV type 1 strains

The advent of affordable ART has benefited HIV-infected individuals. Prospective studies have shown that the availability of drug resistance reports for infected individuals has allowed more effective regimens to be prescribed as compared to a control group whose physicians had no access to drug resistance reports. There is a paucity of information on the performance of genotypic algorithms on non-clade B HIV-1 strains, especially clade C. In this study the results obtained on submission of HIV-1 RT and PR sequences of non-clade B strains to the Stanford University HIV drug resistance database (SHDB) were compared to the results obtained from Geno2Pheno (G2P) and DR_Seqan (DS). For the study, we took samples from 93 treatment-naive individuals and 21 samples from 19 infected individuals showing detectable viral load while on ART. There were discrepancies in the clade identification results obtained from the SHDB and G2P databases. This feature was not available in DS. The mean observed concordance between SHDB and G2P was 85.6% while between SHDB and DC it was 37%. When the level of concordance was determined based on exposure to ART, the G2P was found to have a better level of concordance (76.8%) to SHDB as compared to SHDB versus DS (36%). We do not have phenotypic data for the strains included in this study and hence we are not in a position to assign a particular algorithm as being superior. These results also show a possible need for a subtype-specific algorithm for interpretation of HIV-1 genotypic drug resistance.

Antiretroviral drug resistance testing in adult HIV-1 infection: 2008 recommendations of an International AIDS Society-USA panel

Resistance to antiretroviral drugs remains an important limitation to successful human immunodeficiency virus type 1 (HIV-1) therapy. Resistance testing can improve treatment outcomes for infected individuals. The availability of new drugs from various classes, standardization of resistance assays, and the development of viral tropism tests necessitate new guidelines for resistance testing. The International AIDS Society-USA convened a panel of physicians and scientists with expertise in drug-resistant HIV-1, drug management, and patient care to review recently published data and presentations at scientific conferences and to provide updated recommendations. Whenever possible, resistance testing is recommended at the time of HIV infection diagnosis as part of the initial comprehensive patient assessment, as well as in all cases of virologic failure. Tropism testing is recommended whenever the use of chemokine receptor 5 antagonists is contemplated. As the roll out of antiretroviral therapy continues in developing countries, drug resistance monitoring for both subtype B and non-subtype B strains of HIV will become increasingly important.