Improved prediction of response to antiretroviral combination therapy using the genetic barrier to drug resistance

Detection of COVID-19-related biomarkers by electrochemical biosensors and potential for diagnosis, prognosis, and prediction of the course of the disease in the context of personalized medicine

As a more efficient and effective way to address disease diagnosis and intervention, cutting-edge technologies, devices, therapeutic approaches, and practices have emerged within the personalized medicine concept depending on the particular patient's biology and the molecular basis of the disease. Personalized medicine is expected to play a pivotal role in assessing disease risk or predicting response to treatment, understanding a person's health status, and, therefore, health care decision-making. This work discusses electrochemical biosensors for monitoring multiparametric biomarkers at different molecular levels and their potential to elucidate the health status of an individual in a personalized manner. In particular, and as an illustration, we discuss several aspects of the infection produced by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) as a current health care concern worldwide. This includes SARS-CoV-2 structure, mechanism of infection, biomarkers, and electrochemical biosensors most commonly explored for diagnostics, prognostics, and potentially assessing the risk of complications in patients in the context of personalized medicine. Finally, some concluding remarks and perspectives hint at the use of electrochemical biosensors in the frame of other cutting-edge converging/emerging technologies toward the inauguration of a new paradigm of personalized medicine.

Establishment and evaluation of prediction model for multiple disease classification based on gut microbial data

Diseases prediction has been performed by machine learning approaches with various biological data. One of the representative data is the gut microbial community, which interacts with the host's immune system. The abundance of a few microorganisms has been used as markers to predict diverse diseases. In this study, we hypothesized that multi-classification using machine learning approach could distinguish the gut microbiome from following six diseases: multiple sclerosis, juvenile idiopathic arthritis, myalgic encephalomyelitis/chronic fatigue syndrome, acquired immune deficiency syndrome, stroke and colorectal cancer. We used the abundance of microorganisms at five taxonomy levels as features in 696 samples collected from different studies to establish the best prediction model. We built classification models based on four multi-class classifiers and two feature selection methods including a forward selection and a backward elimination. As a result, we found that the performance of classification is improved as we use the lower taxonomy levels of features; the highest performance was observed at the genus level. Among four classifiers, LogitBoost-based prediction model outperformed other classifiers. Also, we suggested the optimal feature subsets at the genus-level obtained by backward elimination. We believe the selected feature subsets could be used as markers to distinguish various diseases simultaneously. The finding in this study suggests the potential use of selected features for the diagnosis of several diseases.

HIV evolution and diversity in ART-treated patients

Characterizing HIV genetic diversity and evolution during antiretroviral therapy (ART) provides insights into the mechanisms that maintain the viral reservoir during ART. This review describes common methods used to obtain and analyze intra-patient HIV sequence data, the accumulation of diversity prior to ART and how it is affected by suppressive ART, the debate on viral replication and evolution in the presence of ART, HIV compartmentalization across various tissues, and mechanisms for the emergence of drug resistance. It also describes how CD4+ T cells that were likely infected with latent proviruses prior to initiating treatment can proliferate before and during ART, providing a renewable source of infected cells despite therapy. Some expanded cell clones carry intact and replication-competent proviruses with a small fraction of the clonal siblings being transcriptionally active and a source for residual viremia on ART. Such cells may also be the source for viral rebound after interrupting ART. The identical viral sequences observed for many years in both the plasma and infected cells of patients on long-term ART are likely due to the proliferation of infected cells both prior to and during treatment. Studies on HIV diversity may reveal targets that can be exploited in efforts to eradicate or control the infection without ART.

Current Trends in Multidrug Optimization: An Alley of Future Successful Treatment of Complex Disorders

The identification of effective and long-lasting cancer therapies still remains elusive, partially due to patient and tumor heterogeneity, acquired drug resistance, and single-drug dose-limiting toxicities. The use of drug combinations may help to overcome some limitations of current cancer therapies by challenging the robustness and redundancy of biological processes. However, effective drug combination optimization requires the careful consideration of numerous parameters. The complexity of this optimization problem is clearly nontrivial and likely requires the assistance of advanced heuristic optimization techniques. In the current review, we discuss the application of optimization techniques for the identification of optimal drug combinations. More specifically, we focus on the application of phenotype-based screening approaches in the field of cancer therapy. These methods are divided into three categories: (1) modeling methods, (2) model-free approaches based on biological search algorithms, and (3) merged approaches, particularly phenotypically driven network biology methods and computation network models relying on phenotypic data. In addition to a brief description of each approach, we include a critical discussion of the advantages and disadvantages of each method, with a strong focus on the limitations and considerations needed to successfully apply such methods in biological research.

Current Trends in Multidrug Optimization

The identification of effective and long-lasting cancer therapies still remains elusive, partially due to patient and tumor heterogeneity, acquired drug resistance, and single-drug dose-limiting toxicities. The use of drug combinations may help to overcome some limitations of current cancer therapies by challenging the robustness and redundancy of biological processes. However, effective drug combination optimization requires the careful consideration of numerous parameters. The complexity of this optimization problem is clearly nontrivial and likely requires the assistance of advanced heuristic optimization techniques. In the current review, we discuss the application of optimization techniques for the identification of optimal drug combinations. More specifically, we focus on the application of phenotype-based screening approaches in the field of cancer therapy. These methods are divided into three categories: (1) modeling methods, (2) model-free approaches based on biological search algorithms, and (3) merged approaches, particularly phenotypically driven network biology methods and computation network models relying on phenotypic data. In addition to a brief description of each approach, we include a critical discussion of the advantages and disadvantages of each method, with a strong focus on the limitations and considerations needed to successfully apply such methods in biological research.

Highly Accurate Structure-Based Prediction of HIV-1 Coreceptor Usage Suggests Intermolecular Interactions Driving Tropism

HIV-1 entry into host cells is mediated by interactions between the V3-loop of viral glycoprotein gp120 and chemokine receptor CCR5 or CXCR4, collectively known as HIV-1 coreceptors. Accurate genotypic prediction of coreceptor usage is of significant clinical interest and determination of the factors driving tropism has been the focus of extensive study. We have developed a method based on nonlinear support vector machines to elucidate the interacting residue pairs driving coreceptor usage and provide highly accurate coreceptor usage predictions. Our models utilize centroid-centroid interaction energies from computationally derived structures of the V3-loop:coreceptor complexes as primary features, while additional features based on established rules regarding V3-loop sequences are also investigated. We tested our method on 2455 V3-loop sequences of various lengths and subtypes, and produce a median area under the receiver operator curve of 0.977 based on 500 runs of 10-fold cross validation. Our study is the first to elucidate a small set of specific interacting residue pairs between the V3-loop and coreceptors capable of predicting coreceptor usage with high accuracy across major HIV-1 subtypes. The developed method has been implemented as a web tool named CRUSH, CoReceptor USage prediction for HIV-1, which is available at http://ares.tamu.edu/CRUSH/.

Investigation of expert rule bases, logistic regression, and non-linear machine learning techniques for predicting response to antiretroviral treatment

Background

The extreme flexibility of the HIV type-1 (HIV-1) genome makes it challenging to build the ideal antiretroviral treatment regimen. Interpretation of HIV-1 genotypic drug resistance is evolving from rule-based systems guided by expert opinion to data-driven engines developed through machine learning methods.

Methods

The aim of the study was to investigate linear and non-linear statistical learning models for classifying short-term virological outcome of antiretroviral treatment. To optimize the model, different feature selection methods were considered. Robust extra-sample error estimation and different loss functions were used to assess model performance. The results were compared with widely used rule-based genotypic interpretation systems (Stanford HIVdb, Rega and ANRS).

Results

A set of 3,143 treatment change episodes were extracted from the EuResist database. The dataset included patient demographics, treatment history and viral genotypes. A logistic regression model using high order interaction variables performed better than rule-based genotypic interpretation systems (accuracy 75.63% versus 71.74–73.89%, area under the receiver operating characteristic curve [AUC] 0.76 versus 0.68–0.70) and was equivalent to a random forest model (accuracy 76.16%, AUC 0.77). However, when rule-based genotypic interpretation systems were coupled with additional patient attributes, and the combination was provided as input to the logistic regression model, the performance increased significantly, becoming comparable to the fully data-driven methods.

Conclusions

Patient-derived supplementary features significantly improved the accuracy of the prediction of response to treatment, both with rule-based and data-driven interpretation systems. Fully data-driven models derived from large-scale data sources show promise as antiretroviral treatment decision support tools.

Advantages of predicted phenotypes and statistical learning models in inferring virological response to antiretroviral therapy from HIV genotype

Background

Inferring response to antiretroviral therapy from the viral genotype alone is challenging. The utility of an intermediate step of predicting in vitro drug susceptibility is currently controversial. Here, we provide a retrospective comparison of approaches using either genotype or predicted phenotypes alone, or in combination.

Methods

Treatment change episodes were extracted from two large databases from the USA (Stanford-California) and Europe (EuResistDB) comprising data from 6,706 and 13,811 patients, respectively. Response to antiretroviral treatment was dichotomized according to two definitions. Using the viral sequence and the treatment regimen as input, three expert algorithms (ANRS, Rega and HIVdb) were used to generate genotype-based encodings and VircoTYPE™ 4.0 (Virco BVBA, Mechelen, Belgium) was used to generate a predicted phenotype-based encoding. Single drug classifications were combined into a treatment score via simple summation and statistical learning using random forests. Classification performance was studied on Stanford- California data using cross-validation and, in addition, on the independent EuResistDB data.

Results

In all experiments, predicted phenotype was among the most sensitive approaches. Combining single drug classifications by statistical learning was significantly superior to unweighted summation ( P<2.2x10-16). Classification performance could be increased further by combining predicted phenotypes and expert encodings but not by combinations of expert encodings alone. These results were confirmed on an independent test set comprising data solely from EuResistDB.

Conclusions

This study demonstrates consistent performance advantages in utilizing predicted phenotype in most scenarios over methods based on genotype alone in inferring virological response. Moreover, all approaches under study benefit significantly from statistical learning for merging single drug classifications into treatment scores.

Modelled in vivo HIV Fitness under drug Selective Pressure and Estimated Genetic Barrier Towards Resistance are Predictive for Virological Response

Background

A method has been developed to estimate a fitness landscape experienced by HIV-1 under treatment selective pressure as a function of the genotypic sequence thereby also estimating the genetic barrier to resistance.

Methods

We evaluated the performance of two estimated fitness landscapes (nelfinavir [NFV] and zidovudine [AZT] plus lamivudine [3TC]) to predict week 12 viral load (VL) change for 176 treatment change episodes (TCEs) and probability of week 48 virological failure for 90 TCEs, in treatment experienced patients starting these drugs in combination.

Results

A higher genetic barrier for AZT plus 3TC, (quantified per additional mutation required to develop resistance against these drugs) was associated with a 0.54 (95% confidence interval [CI] 0.30–0.77) larger log10 VL reduction at 12 weeks ( P<0.0001) and a 0.39 (95% CI 0.23–0.66) lower odds of virological failure at 48 weeks ( P=0.0005), in analyses adjusting for the pre-TCE VL and the exact time-lag between the TCE and the date of determining response VL. The strength of these associations was comparable with those seen with expert interpretation systems (Rega, ANRS and HIVDB). A higher genetic barrier to NFV resistance was the only genotypic predictor that tended to be associated with a 0.19 (95% CI 0–0.39) higher log10 VL reduction at 12 weeks ( P=0.05) and a 0.63 (95% CI 0.36–1.09) lower odds of virological failure at 48 weeks ( P=0.10) per additional mutation.

Conclusions

These results suggest that an estimated genetic barrier derived from fitness landscapes may contribute to an improvement of predicted treatment outcome for NFV and this approach should be explored for other drugs.