Identification of Asthma Subtypes Using Clustering Methodologies

Adult patients with an exacerbation of asthma and a higher risk for pulmonary embolism: a cluster analysis

OBJECTIVE

Current literature acknowledges the complexity of exacerbation triggers in patients with asthma. We studied the clinical heterogeneity of patients with asthma exacerbation suspected of having pulmonary embolism using cluster analysis and compared the clusters regarding of the risks for pulmonary embolism.

METHODS

In a secondary analysis of a dataset from the University of Florida, USA, individuals who experienced asthma exacerbation between June 2011 and October 2018 were included. All patients had undergone pulmonary CT angiography. Overall, 18 variables consisting of demographic, clinical, comorbidity, and therapeutic characteristics were used to cluster patients. The clusters were then profiled and compared in the percentages of pulmonary embolism.

RESULTS

In total, 758 patients (226; 29.8% men) with an exacerbation of asthma were included in the analysis. The frequency of a confirmed pulmonary embolism was 145 (19.1%). Two distinct clusters were identified with a statistically significant difference in pulmonary embolism [p < 0.001, odds ratio (95%CI)=2.24 (1.55, 3.24)]. We developed a high-performance classifier to profile the low- and high-risk clusters (area under the curve = 0.923, positive likelihood ratio = 20.2). The three top important variables discriminating the two clusters were age, heart rate, and body mass index. Older age, lower heart rate, higher body mass index, black race, and positive medical history (including atrial fibrillation) were more frequent in the high-risk group. Despite the higher percentage of women in the high-risk group, the sex ratios were not significantly different between the clusters.

CONCLUSION

There are two clusters in patients with an exacerbation of asthma with different prognoses percentages of pulmonary embolism. The clusters can be well identified based on patient characteristics.

An interpretable cluster-based logistic regression model, with application to the characterization of response to therapy in severe eosinophilic asthma

Asthma is a disease characterized by chronic airway hyperresponsiveness and inflammation, with signs of variable airflow limitation and impaired lung function leading to respiratory symptoms such as shortness of breath, chest tightness and cough. Eosinophilic asthma is a distinct phenotype that affects more than half of patients diagnosed with severe asthma. It can be effectively treated with monoclonal antibodies targeting specific immunological signaling pathways that fuel the inflammation underlying the disease, particularly Interleukin-5 (IL-5), a cytokine that plays a crucial role in asthma. In this study, we propose a data analysis pipeline aimed at identifying subphenotypes of severe eosinophilic asthma in relation to response to therapy at follow-up, which could have great potential for use in routine clinical practice. Once an optimal partition of patients into subphenotypes has been determined, the labels indicating the group to which each patient has been assigned are used in a novel way. For each input variable in a specialized logistic regression model, a clusterwise effect on response to therapy is determined by an appropriate interaction term between the input variable under consideration and the cluster label. We show that the clusterwise odds ratios can be meaningfully interpreted conditional on the cluster label. In this way, we can define an effect measure for the response variable for each input variable in each of the groups identified by the clustering algorithm, which is not possible in standard logistic regression because the effect of the reference class is aliased with the overall intercept. The interpretability of the model is enforced by promoting sparsity, a goal achieved by learning interactions in a hierarchical manner using a special group-Lasso technique. In addition, valid expressions are provided for computing odds ratios in the unusual parameterization used by the sparsity-promoting algorithm. We show how to apply the proposed data analysis pipeline to the problem of sub-phenotyping asthma patients also in terms of quality of response to therapy with monoclonal antibodies.

Machine learning-derived phenotypic trajectories of asthma and allergy in children and adolescents: protocol for a systematic review

INTRODUCTION

Development of asthma and allergies in childhood/adolescence commonly follows a sequential progression termed the 'atopic march'. Recent reports indicate, however, that these diseases are composed of multiple distinct phenotypes, with possibly differential trajectories. We aim to synthesise the current literature in the field of machine learning-based trajectory studies of asthma/allergies in children and adolescents, summarising the frequency, characteristics and associated risk factors and outcomes of identified trajectories and indicating potential directions for subsequent research in replicability, pathophysiology, risk stratification and personalised management. Furthermore, methodological approaches and quality will be critically appraised, highlighting trends, limitations and future perspectives.

METHODS AND ANALYSES

10 databases (CAB Direct, CINAHL, Embase, Google Scholar, PsycInfo, PubMed, Scopus, Web of Science, WHO Global Index Medicus and WorldCat Dissertations and Theses) will be searched for observational studies (including conference abstracts and grey literature) from the last 10 years (2013-2023) without restriction by language. Screening, data extraction and assessment of quality and risk of bias (using a custom-developed tool) will be performed independently in pairs. The characteristics of the derived trajectories will be narratively synthesised, tabulated and visualised in figures. Risk factors and outcomes associated with the trajectories will be summarised and pooled estimates from comparable numerical data produced through random-effects meta-analysis. Methodological approaches will be narratively synthesised and presented in tabulated form and figure to visualise trends.

ETHICS AND DISSEMINATION

Ethical approval is not warranted as no patient-level data will be used. The findings will be published in an international peer-reviewed journal.

PROSPERO REGISTRATION NUMBER

CRD42023441691.

Cluster Analysis of Finnish Population-Based Adult-Onset Asthma Patients

BACKGROUND

Phenotypes of adult asthma have been identified in previous studies but rarely in population-based settings.

OBJECTIVE

To identify clusters of adult-onset asthma in a Finnish population-based study on subjects born before 1967.

METHODS

We used population-based data from 1350 asthmatics with adult-onset asthma (Adult Asthma in Finland) from Finnish national registers. Twenty-eight covariates were selected based on literature. The number of covariates was reduced by using factor analysis before cluster analysis.

RESULTS

Five clusters (CLU1-CLU5) were identified, 3 clusters with late-onset adult asthma (onset ≥40 years) and 2 clusters with onset at earlier adulthood (<40 years). Subjects in CLU1 (n = 666) had late-onset asthma and were nonobese, symptomatic, and predominantly female with few respiratory infections during childhood. CLU2 (n = 36) consisted of subjects who had earlier-onset asthma, were predominantly female, obese with allergic asthma, and had recurrent respiratory infections. Subjects in CLU3 (n = 75) were nonobese, older, and predominantly men with late-onset asthma, smoking history, comorbidities, severe asthma, least allergic diseases, low education, many siblings, and childhood in rural areas. CLU4 (n = 218) was a late-onset cluster consisting of obese females with comorbidities, asthma symptoms, and low education level. Subjects in CLU5 (n = 260) had earlier onset asthma, were nonobese, and predominantly allergic females.

CONCLUSIONS

Our population-based adult-onset asthma clusters take into account several critical factors such as obesity and smoking, and identified clusters that partially overlap with clusters identified in clinical settings. Results give us a more profound understanding of adult-onset asthma phenotypes and support personalized management.

Identification of cough-variant asthma phenotypes based on clinical and pathophysiologic data

BACKGROUND

Cough-variant asthma (CVA) may respond differently to antiasthmatic treatment. There are limited data on the heterogeneity of CVA.

OBJECTIVE

We aimed to classify patients with CVA using cluster analysis based on clinicophysiologic parameters and to unveil the underlying molecular pathways of these phenotypes with transcriptomic data of sputum cells.

METHODS

We applied k-mean clustering to 342 newly physician-diagnosed patients with CVA from a prospective multicenter observational cohort using 10 prespecified baseline clinical and pathophysiologic variables. The clusters were compared according to clinical features, treatment response, and sputum transcriptomic data.

RESULTS

Three stable CVA clusters were identified. Cluster 1 (n = 176) was characterized by female predominance, late onset, normal lung function, and a low proportion of complete resolution of cough (60.8%) after antiasthmatic treatment. Patients in cluster 2 (n = 105) presented with young, nocturnal cough, atopy, high type 2 inflammation, and a high proportion of complete resolution of cough (73.3%) with a highly upregulated coexpression gene network that related to type 2 immunity. Patients in cluster 3 (n = 61) had high body mass index, long disease duration, family history of asthma, low lung function, and low proportion of complete resolution of cough (54.1%). TH17 immunity and type 2 immunity coexpression gene networks were both upregulated in clusters 1 and 3.

CONCLUSION

Three clusters of CVA were identified with different clinical, pathophysiologic, and transcriptomic features and responses to antiasthmatics treatment, which may improve our understanding of pathogenesis and help clinicians develop individualized cough treatment in asthma.

Cluster Analyses From the Real-World NOVELTY Study: Six Clusters Across the Asthma-COPD Spectrum

BACKGROUND

Asthma and chronic obstructive pulmonary disease (COPD) are complex diseases, the definitions of which overlap.

OBJECTIVE

To investigate clustering of clinical/physiological features and readily available biomarkers in patients with physician-assigned diagnoses of asthma and/or COPD in the NOVEL observational longiTudinal studY (NOVELTY; NCT02760329).

METHODS

Two approaches were taken to variable selection using baseline data: approach A was data-driven, hypothesis-free and used the Pearson dissimilarity matrix; approach B used an unsupervised Random Forest guided by clinical input. Cluster analyses were conducted across 100 random resamples using partitioning around medoids, followed by consensus clustering.

RESULTS

Approach A included 3796 individuals (mean age, 59.5 years; 54% female); approach B included 2934 patients (mean age, 60.7 years; 53% female). Each identified 6 mathematically stable clusters, which had overlapping characteristics. Overall, 67% to 75% of patients with asthma were in 3 clusters, and approximately 90% of patients with COPD were in 3 clusters. Although traditional features such as allergies and current/ex-smoking (respectively) were higher in these clusters, there were differences between clusters and approaches in features such as sex, ethnicity, breathlessness, frequent productive cough, and blood cell counts. The strongest predictors of the approach A cluster membership were age, weight, childhood onset, prebronchodilator FEV1, duration of dust/fume exposure, and number of daily medications.

CONCLUSIONS

Cluster analyses in patients from NOVELTY with asthma and/or COPD yielded identifiable clusters, with several discriminatory features that differed from conventional diagnostic characteristics. The overlap between clusters suggests that they do not reflect discrete underlying mechanisms and points to the need for identification of molecular endotypes and potential treatment targets across asthma and/or COPD.

Whispers of change in preschool asthma phenotypes: Findings in the French ELFE cohort

RATIONALE

Early life asthma phenotyping remains an unmet need in pediatric asthma. In France, severe pediatric asthma phenotyping has been done extensively; however, phenotypes in the general population remain underexplored. Based on the course and severity of respiratory/allergic symptoms, we aimed to identify and characterize early life wheeze profiles and asthma phenotypes in the general population.

METHODS

ELFE is a general population based birth cohort; which recruited 18,329 newborns in 2011, from 320 maternity units nationwide. Data was collected using parental responses to modified versions of ISAAC questionnaire on eczema, rhinitis, food allergy, cough, wheezing, dyspnoea and sleep disturbance due to wheezing at 3 time points: post-natal (2 months), infancy (age 1) and pre-school (age 5). We built a supervised trajectory for wheeze profiles and an unsupervised approach was used for asthma phenotypes. Chi squared (χ2) test or fisher's exact test was used as appropriate (p < 0.05).

RESULTS

Wheeze profiles and asthma phenotypes were ascertained at age 5. Supervised wheeze trajectory of 9161 children resulted in 4 wheeze profiles: Persistent (0.8%), Transient (12.1%), Incident wheezers at age 5 (13.3%) and Non wheezers (73.9%). While 9517 children in unsupervised clusters displayed 4 distinct asthma phenotypes: Mildly symptomatic (70%), Post-natal bronchiolitis with persistent rhinitis (10.2%), Severe early asthma (16.9%) and Early persistent atopy with late onset severe wheeze (2.9%).

CONCLUSION

We successfully determined early life wheeze profiles and asthma phenotypes in the general population of France.

Multidimensional endotyping using nasal proteomics predicts molecular phenotypes in the asthmatic airways

BACKGROUND

Unsupervised clustering of biomarkers derived from noninvasive samples such as nasal fluid is less evaluated as a tool for describing asthma endotypes.

OBJECTIVE

We sought to evaluate whether protein expression in nasal fluid would identify distinct clusters of patients with asthma with specific lower airway molecular phenotypes.

METHODS

Unsupervised clustering of 168 nasal inflammatory and immune proteins and Shapley values was used to stratify 43 patients with severe asthma (endotype of noneosinophilic asthma) using a 2 "modeling blocks" machine learning approach. This algorithm was also applied to nasal brushings transcriptomics from U-BIOPRED (Unbiased Biomarkers for the Prediction of Respiratory Diseases Outcomes). Feature reduction and functional gene analysis were used to compare proteomic and transcriptomic clusters. Gene set variation analysis provided enrichment scores of the endotype of noneosinophilic asthma protein signature within U-BIOPRED sputum and blood.

RESULTS

The nasal protein machine learning model identified 2 severe asthma endotypes, which were replicated in U-BIOPRED nasal transcriptomics. Cluster 1 patients had significant airway obstruction, small airways disease, air trapping, decreased diffusing capacity, and increased oxidative stress, although only 4 of 18 were current smokers. Shapley identified 20 cluster-defining proteins. Forty-one proteins were significantly higher in cluster 1. Pathways associated with proteomic and transcriptomic clusters were linked to T_H1, T_H2, neutrophil, Janus kinase-signal transducer and activator of transcription, TLR, and infection activation. Gene set variation analysis of the nasal protein and gene signatures were enriched in subjects with sputum neutrophilic/mixed granulocytic asthma and in subjects with a molecular phenotype found in sputum neutrophil-high subjects.

CONCLUSIONS

Protein or gene analysis may indicate molecular phenotypes within the asthmatic lower airway and provide a simple, noninvasive test for non-type 2 immune response asthma that is currently unavailable.

Subtypes of Asthma and Cold Weather-Related Respiratory Symptoms

(1) Poor asthma control increases the occurrence of cold weather-related symptoms among adult asthmatics. We assessed whether the subtype of asthma, taking into account the severity of the asthma, plays a role in these symptoms. (2) We conducted a population-based cross-sectional study of 1995 adult asthmatics (response rate 40.4%) living in northern Finland using a questionnaire that asked about cold weather-related respiratory symptoms including (1) shortness of breath, (2) prolonged cough, (3) wheezing, (4) phlegm production, and (5) chest pain, as well as questions related to the subtype of asthma. For women, the subtypes identified using latent class analysis were: (1) Controlled, mild asthma, (2) Partly controlled, moderate asthma, (3) Uncontrolled, unknown severity, and (4) Uncontrolled, severe asthma, and for men: (1) Controlled, mild asthma, (2) Uncontrolled, unknown severity, and (3) Partly controlled, severe asthma. (3) According to the subtypes of asthma, more severe and more poorly controlled asthma were related to the increased prevalence of cold weather-related respiratory symptoms when compared with those with mild, controlled asthma. This trend was especially clear for wheezing and chest pain. For example, in men, the adjusted prevalence ratio of wheezing was 1.55 (95% CI 1.09–2.19) in uncontrolled asthma with unknown severity and 1.84 (95% CI 1.26–2.71) in partly controlled severe asthma compared with controlled, mild asthma. (4) Our study provides evidence for the influence of subtypes of asthma on experiencing cold weather-related respiratory symptoms. Both women and men reported more cold weather-related symptoms when their asthma was more severe and uncontrolled compared with those who had mild and well-controlled asthma.

Asthma clustering methods: a literature-informed application to the children's health study data

OBJECTIVE

The heterogeneity of asthma has inspired widespread application of statistical clustering algorithms to a variety of datasets for identification of potentially clinically meaningful phenotypes. There has not been a standardized data analysis approach for asthma clustering, which can affect reproducibility and clinical translation of results. Our objective was to identify common and effective data analysis practices in the asthma clustering literature and apply them to data from a Southern California population-based cohort of schoolchildren with asthma.

METHODS

As of January 1, 2020, we reviewed key statistical elements of 77 asthma clustering studies. Guided by the literature, we used 12 input variables and three clustering methods (hierarchical clustering, k-medoids, and latent class analysis) to identify clusters in 598 schoolchildren with asthma from the Southern California Children's Health Study (CHS).

RESULTS

Clusters of children identified by latent class analysis were characterized by exhaled nitric oxide, FEV1/FVC, FEV1 percent predicted, asthma control and allergy score; and were predictive of control at two year follow up. Clusters from the other two methods were less clinically remarkable, primarily differentiated by sex and race/ethnicity and less predictive of asthma control over time.

CONCLUSION

Upon review of the asthma phenotyping literature, common approaches of data clustering emerged. When applying these elements to the Children's Health Study data, latent class analysis clusters-represented by exhaled nitric oxide and spirometry measures-had clinical relevance over time.

Identifying clinical phenotypes in extremely low birth weight infants-an unsupervised machine learning approach

There is increasing evidence that patient heterogeneity significantly hinders advancement in clinical trials and individualized care. This study aimed to identify distinct phenotypes in extremely low birth weight infants. We performed an agglomerative hierarchical clustering on principal components. Cluster validation was performed by cluster stability assessment with bootstrapping method. A total of 215 newborns (median gestational age 27 (26-29) weeks) were included in the final analysis. Six clusters with different clinical and laboratory characteristics were identified: the "Mature" (Cluster 1; n = 60, 27.9%), the mechanically ventilated with "adequate ventilation" (Cluster 2; n = 40, 18.6%), the mechanically ventilated with "poor ventilation" (Cluster 3; n = 39, 18.1%), the "extremely immature" (Cluster 4; n = 39, 18.1%%), the neonates requiring "Intensive Resuscitation" in the delivery room (Cluster 5; n = 20, 9.3%), and the "Early septic" group (Cluster 6; n = 17, 7.9%). In-hospital mortality rates were 11.7%, 25%, 56.4%, 61.5%, 45%, and 52.9%, while severe intraventricular hemorrhage rates were 1.7%, 5.3%, 29.7%, 47.2%, 44.4%, and 28.6% in clusters 1, 2, 3, 4, 5, and 6, respectively (p < 0.001).Conclusion: Our cluster analysis in extremely preterm infants was able to characterize six distinct phenotypes. Future research should explore how better phenotypic characterization of neonates might improve care and prognosis. What is Known: • Patient heterogeneity is becoming more acknowledged as a cause of clinical trial failure. • Machine learning algorithms can find patterns within a heterogeneous group. What is New: • We identified six different phenotypes of extremely preterm infants who exhibited distinct clinical and laboratorial characteristics.

Air pollution-related asthma profiles among children/adolescents: A multi-group latent class analysis

BACKGROUND

This study aimed to investigate the asthma profile among children/adolescents and the relationship of the prevalence of air pollution profiles using latent class analysis (LCA).

OBJECTIVES

In this cross-sectional study, a case rural community was selected in an industrial area, and two rural control communities were selected in unexposed areas.

METHODS

Hourly concentrations of PM₁₀, SO₂, NO_2, and total volatile organic compounds were obtained from the records of a fixed air quality monitoring station, and the concentrations of benzene, toluene, xylenes styrene were measured during six campaigns. Asthma data was collected using the International Study of Asthma and Allergies in Childhood in 7-18 years old children/adolescents. The modeling was conducted using LCA.

RESULTS

A higher amount of air pollution indices were observed in the case than both control communities. LCA divided the participants into three clusters; "healthy" (92.8%), "moderate" (2.8%), and "severe" (4.4%). A higher probability of severe asthma (6.8%) was observed in the case than control communities (2.6% and 1.8%). Additionally, after adjusting for possible confounders, the odds of asthma was lower in the control communities than the case in both moderate and sever classes (Odds Ratios in the range of 0.135-0.697).

CONCLUSIONS

This study indicates asthma profiles of children/adolescents and the higher prevalence of severe class in the area, explaining the possible effect of air pollution.

Clustering of Unhealthy Behaviors: Protocol for a Multiple Behavior Analysis of Data From the Canadian Longitudinal Study on Aging

BACKGROUND

Health behaviors such as physical inactivity, unhealthy eating, smoking tobacco, and alcohol use are leading risk factors for noncommunicable chronic diseases and play a central role in limiting health and life satisfaction. To date, however, health behaviors tend to be considered separately from one another, resulting in guidelines and interventions for healthy aging siloed by specific behaviors and often focused only on a given health behavior without considering the co-occurrence of family, social, work, and other behaviors of everyday life.

OBJECTIVE

The aim of this study is to understand how behaviors cluster and how such clusters are associated with physical and mental health, life satisfaction, and health care utilization may provide opportunities to leverage this co-occurrence to develop and evaluate interventions to promote multiple health behavior changes.

METHODS

Using cross-sectional baseline data from the Canadian Longitudinal Study on Aging, we will perform a predefined set of exploratory and hypothesis-generating analyses to examine the co-occurrence of health and everyday life behaviors. We will use agglomerative hierarchical cluster analysis to cluster individuals based on their behavioral tendencies. Multinomial logistic regression will then be used to model the relationships between clusters and demographic indicators, health care utilization, and general health and life satisfaction, and assess whether sex and age moderate these relationships. In addition, we will conduct network community detection analysis using the clique percolation algorithm to detect overlapping communities of behaviors based on the strength of relationships between variables.

RESULTS

Baseline data for the Canadian Longitudinal Study on Aging were collected from 51,338 participants aged between 45 and 85 years. Data were collected between 2010 and 2015. Secondary data analysis for this project was approved by the Ottawa Health Science Network Research Ethics Board (protocol ID #20190506-01H).

CONCLUSIONS

This study will help to inform the development of interventions tailored to subpopulations of adults (eg, physically inactive smokers) defined by the multiple behaviors that describe their everyday life experiences.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/24887.

Genetics of Asthma and Allergic Diseases

Asthma genes have been identified through a range of approaches, from candidate gene association studies and family-based genome-wide linkage analyses to genome-wide association studies (GWAS). The first GWAS of asthma, reported in 2007, identified multiple markers on chromosome 17q21 as associates of the childhood-onset asthma. This remains the best replicated asthma locus to date. However, notwithstanding undeniable successes, genetic studies have produced relatively heterogeneous results with limited replication, and despite considerable promise, genetics of asthma and allergy has, so far, had limited impact on patient care, our understanding of disease mechanisms, and development of novel therapeutic targets. The paucity of precise replication in genetic studies of asthma is partly explained by the existence of numerous gene-environment interactions. Another important issue which is often overlooked is that of time of the assessment of the primary outcome(s) and the relevant environmental exposures. Most large GWASs use the broadest possible definition of asthma to increase the sample size, but the unwanted consequence of this is increased phenotypic heterogeneity, which dilutes effect sizes. One way of addressing this is to precisely define disease subtypes (e.g. by applying novel mathematical approaches to rich phenotypic data) and use these latent subtypes in genetic studies.

Heterogeneity of Childhood Asthma in Korea: Cluster Analysis of the Korean Childhood Asthma Study Cohort

PURPOSE

Asthma is a heterogeneous airway disease occurring in children, and it has various clinical phenotypes. A clear differentiation of the clinical phenotypes can provide better asthma management and prediction of asthma prognosis. Little is currently known about asthma phenotypes in Korean children. This study was designed to identify asthma phenotypes in school-aged Korean children.

METHODS

This study enrolled 674 children with physician-diagnosed asthma from the Korean childhood Asthma Study (KAS) cohort. The physicians verified the relevant histories of asthma and comorbid diseases, as well as airway lability and hyper-responsiveness from the results of pulmonary function tests and bronchial provocation tests. Questionnaires regarding the participants' baseline characteristics, their environment and self-rating of asthma control were collected at the time of enrollment. Laboratory tests were performed to assess allergy and airway inflammation. Children with asthma were classified by hierarchical cluster analysis.

RESULTS

Of the 674 patients enrolled from the KAS cohort, 447 were included in the cluster analysis. Cluster analysis of these 447 children revealed 4 asthma phenotypes: cluster 1 (n = 216, 48.3%) which was characterized by male-dominant atopic asthma; cluster 2 (n = 79, 17.7%) which was characterized by early-onset atopic asthma with atopic dermatitis; cluster 3 (n = 47, 10.5%) which was characterized by puberty-onset, female-dominant atopic asthma with the low lung function; and cluster 4 (n = 105, 23.5%) which was characterized by early-onset, non-atopic dominant asthma.

CONCLUSIONS

The asthma phenotypes among Korean children can be classified into 4 distinct clusters. Long-term follow-up with these phenotypes will be needed to define their prognosis and response to treatment.

Comparison Analysis of Prediction Model for Respiratory Diseases

Millions of people around the world have one or many respiratory-related illnesses. Many chronic respiratory diseases like asthma, COPD, pneumonia, respiratory distress, etc. are considered to be a significant public health burden. To reduce the mortality rate, it is better to perform early prediction of respiratory disorders and treat them accordingly. To build an efficient prediction model for various types of respiratory diseases, machine learning approaches are used. The proposed methodology builds classifier model using supervised learning algorithms like random forest, decision tree, and multi-layer perceptron neural network (MLP-NN) for the detection of different respiratory diseases of ICU admitted patients. It achieves accuracy of nearly 99% by various machine learning approaches.

Endotypes identified by cluster analysis in asthmatics and non-asthmatics and their clinical characteristics at follow-up: the case-control EGEA study

Background

Identifying relevant asthma endotypes may be the first step towards improving asthma management. We aimed identifying respiratory endotypes in adults using a cluster analysis and to compare their clinical characteristics at follow-up.

Methods

The analysis was performed separately among current asthmatics (CA, n=402) and never asthmatics (NA, n=666) from the first follow-up of the French EGEA study (EGEA2). Cluster analysis jointly considered 4 demographic, 22 clinical/functional (respiratory symptoms, asthma treatments, lung function) and four blood biological (allergy-related, inflammation-related and oxidative stress-related biomarkers) characteristics at EGEA2. The clinical characteristics at follow-up (EGEA3) were compared according to the endotype identified at EGEA2.

Results

We identified five respiratory endotypes, three among CA and two among NA: CA1 (n=53) with active treated adult-onset asthma, poor lung function, chronic cough and phlegm and dyspnoea, high body mass index, and high blood neutrophil count and fluorescent oxidation products level; CA2 (n=219) with mild asthma and rhinitis; CA3 (n=130) with inactive/mild untreated allergic childhood-onset asthma, high frequency of current smokers and low frequency of attacks of breathlessness at rest, and high IgE level; NA1 (n=489) asymptomatic, and NA2 (n=177) with respiratory symptoms, high blood neutrophil and eosinophil counts. CA1 had poor asthma control and high leptin level, CA2 had hyper-responsiveness and high interleukin (IL)-1Ra, IL-5, IL-7, IL-8, IL-10, IL-13 and TNF-α levels, and NA2 had high leptin and C reactive protein levels. Ten years later, asthmatics in CA1 had worse clinical characteristics whereas those in CA3 had better respiratory outcomes than CA2; NA in NA2 had more respiratory symptoms and higher rate of incident asthma than those in NA1.

Conclusion

These results highlight the interest to jointly consider clinical and biological characteristics in cluster analyses to identify endotypes among adults with or without asthma.

Comprehensive Cluster Analysis for COPD Including Systemic and Airway Inflammatory Markers

Chronic obstructive pulmonary disease (COPD) is a complex, multidimensional and heterogeneous disease. The main purpose of the present study was to identify clinical phenotypes through cluster analysis in adults suffering from COPD. A retrospective study was conducted on 178 COPD patients in stable state recruited from ambulatory care at University hospital of Liege. All patients were above 40 years, had a smoking history of more than 20 pack years, post bronchodilator FEV1/FVC <70% and denied any history of asthma before 40 years. In this study, the patients were described by a total of 84 mixed sets of variables with some missing values. Hierarchical clustering on principal components (HCPC) was applied on multiple imputation. In the final step, patients were classified into homogeneous distinct groups by consensus clustering. Three different clusters, which shared similar smoking history were found. Cluster 1 included men with moderate airway obstruction (n = 67) while cluster 2 comprised men who were exacerbation-prone, with severe airflow limitation and intense granulocytic airway and neutrophilic systemic inflammation (n = 56). Cluster 3 essentially included women with moderate airway obstruction (n = 55). All clusters had a low rate of bacterial colonization (5%), a low median FeNO value (<20 ppb) and a very low sensitization rate toward common aeroallergens (0-5%). CAT score did not differ between clusters. Including markers of systemic airway inflammation and atopy and applying a comprehensive cluster analysis we provide here evidence for 3 clusters markedly shaped by sex, airway obstruction and neutrophilic inflammation but not by symptoms and T2 biomarkers.

Risks for cold frequency vary by sex: role of asthma, age, TLR7 and leukocyte subsets

Viral respiratory infections are usually benign but can trigger asthma exacerbations. The factors associated with upper respiratory tract infection (cold) frequency are not fully understood, nor is it clear whether such factors differ between women and men.To determine which immunological and clinical variables associate with the frequency of self-reported respiratory infections (colds), 150 asthma cases and 151 controls were recruited. Associations between antiviral immune response variables: toll-like receptor (TLR)7/8 gene expression, plasmacytoid dendritic cell (pDC) numbers and interferon-α, tumour necrosis factor and interleukin-12 production, and asthma were then examined that might explain cold frequency.People with asthma cases reported more colds per year (median 3 versus 2; p<0.001) and had lower baseline TLR7 gene expression (odds ratio 0.12; p=0.02) than controls. Associations between many variables and cold frequency differed between women and men. In women, high blood neutrophil counts (β=0.096, p=0.002), and younger age (β=-0.017, p<0.001), but not exposure to children, were independently associated with more frequent colds. In men, low TLR7 expression (β=-0.96, p=0.041) and high CLEC4C gene expression (a marker of pDC; β=0.88, p=0.008) were independently associated with more frequent colds. Poor asthma symptom control was independently associated with reduced TLR8 gene expression (β=-1.4, p=0.036) and high body mass index (β=0.041, p=0.004).Asthma, age and markers of inflammation and antiviral immunity in peripheral blood are associated with frequent colds. Interestingly, the variables associated with cold frequency differed between women and men.

New medical big data for P4 medicine on allergic conjunctivitis

Allergic conjunctivitis affects approximately 15-20% of the global population and can permanently deteriorate one's quality of life (QoL) and work productivity, leading to societal work force costs. Although not fully understood, allergic conjunctivitis is a multifactorial disease with a complex network of environmental, lifestyle, and host contributory risk factors. To effectively enhance the quality of treatment for patients with allergic conjunctivitis, as well as other allergic diseases, the field must first comprehend the pathology underlying various individualized subjective symptoms and stratify the disease according to risk factors and presentations. Such competent stratification and societal reconstruction that targets the alleviation of the damage due to allergic diseases would greatly help ramify personalized treatments and prevent the projected increase in societal costs imposed by allergic diseases. Owing to the rapid advancements in the information and technology sector, medical big data are greatly accessible and useful to decipher the pathophysiology of many diseases. Such data collected through multi-omics and mobile health have been effective for research on chronic diseases including allergic and immune-mediated diseases. Novel big data containing vast and continuous information on individuals with allergic conjunctivitis and other allergic symptoms are being used to search for causative genes of diseases, gain insights into new biomarkers, prevent disease progression, and, ultimately, improve QoL. The individualized and holistic data accrued from new angles using technological innovations are helping the field realize the principles of P4 medicine: predictive, preventive, personalized, and participatory medicine.

Atopic phenotypes and their implication in the atopic march

INTRODUCTION

Eczema, allergic rhinitis, and asthma are traditionally considered atopic (or allergic) diseases. They are complex, multifactorial, and are caused by a variety of different mechanisms, which result in multiple heterogeneous clinical phenotypes. Atopic march is usually interpreted as the sequential development of symptoms from eczema in infancy, to asthma, and then allergic rhinitis. Areas covered: The authors reviewed the evidence on the multimorbidity of eczema, asthma, and rhinitis, and the implication of results of data-driven analyses on the concept framework of atopic march. A literature search was conducted in the PubMed and Web of Science for peer-reviewed articles published until July 2020. Application of Bayesian machine learning framework to rich phenotypic data from birth cohorts demonstrated that the postulated linear progression of symptoms (atopic march) does not capture the heterogeneity of allergic phenotypes. Expert opinion: Eczema, wheeze, and rhinitis co-exist more often than would be expected by chance, but their relationship can be best understood in a multimorbidity framework, rather than through atopic march sequence. The observation of their co-occurrence does not imply any specific relationship between them, and certainly not a progressive or causal one. It is unlikely that a sngle mechanism such as allergic sensitization underpins different multimorbidity manifestations.

Asthma control factors in the Gulf Cooperation Council (GCC) countries and the effectiveness of ICS/LABA fixed dose combinations: a dual rapid literature review

Background

Asthma control is influenced by multiple factors. These factors must be considered when appraising asthma interventions and their effectiveness in the Gulf Cooperation Council (GCC) countries (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and United Arab Emirates [UAE]). Based on published studies, the most prevalent asthma treatment in these countries are fixed dose combinations (FDC) of inhaled corticosteroid and long-acting beta-agonist (ICS/LABA). This study is a rapid review of the literature on: (a) factors associated with asthma control in the GCC countries and (b) generalisability of ICS/LABA FDC effectiveness studies.

Methods

To review local factors associated with asthma control and, generalisability of published ICS/LABA FDC studies, two rapid reviews were conducted. Review 1 targeted literature pertaining to asthma control factors in GCC countries. Eligible studies were appraised, and clustering methodology used to summarise factors. Review 2 assessed ICS/LABA FDC studies in conditions close to actual clinical practice (i.e. effectiveness studies). Eligibility was determined by reviewing study characteristics. Evaluation of studies focused on randomised controlled trials (RCTs). In both reviews, initial (January 2018) and updated (November 2019) searches were conducted in EMBASE and PubMed databases. Eligible studies were appraised using the Critical Appraisal Skills Program (CASP) checklists.

Results

We identified 51 publications reporting factors associated with asthma control. These publications reported studies conducted in Saudi Arabia (35), Qatar (5), Kuwait (5), UAE (3), Oman (1) and multiple countries (2). The most common factors associated with asthma control were: asthma-related education (13 articles), demographics (11articles), comorbidities (11 articles) and environmental exposures (11 articles). Review 2 identified 61 articles reporting ICS/LABA FDC effectiveness studies from countries outside of the GCC. Of these, six RCTs were critically appraised. The adequacy of RCTs in informing clinical practice varied when appraised against previously published criteria.

Conclusions

Asthma-related education was the most recurring factor associated with asthma control in the GCC countries. Moreover, the generalisability of ICS/LABA FDC studies to this region is variable. Hence, asthma patients in the region, particularly those on ICS/LABA FDC, will continue to require physician review and oversight. While our findings provide evidence for local treatment guidelines, further research is required in GCC countries to establish the causal pathways through which asthma-related education influence asthma control for patients on ICS/LABA FDC therapy.

Cluster Analysis: A New Approach for Identification of Underlying Risk Factors and Demographic Features of First Trimester Pregnancy Women

Thyroid pathology is reported internationally in 5-10% of all pregnancies. The overall aim of this research was to determine the prevalence of hypothyroidism and risk factors during the first trimester screening in a Mexican patients sample. We included the records of 306 patients who attended a prenatal control consultation between January 2016 and December 2017 at the Women's Institute in Monterrey, Mexico. The studied sample had homogeneous demographic characteristics in terms of age, weight, height, BMI (body mass index) and number of pregnancies. The presence of at least one of the risk factors for thyroid disease was observed in 39.2% of the sample. Two and three clusters were identified, in which patients varied considerably among risk factors, symptoms and pregnancy complications. Compared to Cluster 0, one or more symptoms or signs of hypothyroidism occurred, while Cluster 1 was characterized by healthier patients. When three clusters were used, Cluster 2 had a higher TSH (thyroid stimulating hormone) value and pregnancy complications. There were no significant differences in perinatal variables. In addition, high TSH levels in first trimester pregnancy are characterized by pregnancy complications and decreased newborn weight. Our findings underline the high degree of disease heterogeneity with existing pregnant hypothyroid patients and the need to improve the phenotyping of the syndrome in the Mexican population.

Identification of Specific IgE Antibodies and Asthma Control Interaction and Association Using Cluster Analysis in a Bulgarian Asthmatic Children Cohort

(1) Background: Asthma is a complex heterogeneous disease that likely comprises several distinct disease phenotypes, where the clustering approach has been used to classify the heterogeneous asthma population into distinct phenotypes; (2) Methods: For a period of 1 year, we evaluated medical history data of 71 children with asthma aged 3 to 17 years, performing pulmonary function tests, drew blood for IgE antibodies against inhalation and food allergies detection, and Asthma Control Questionnaire (ACQ); (3) Results: Five distinct phenotypes were determined. Cluster 1 (n = 10): (non-atopic) the lowest IgE level, very low ACQ, and median age of diagnosis. Cluster 2 (n = 28): (mixed) the highest Body mass index (BMI) with the latest age of diagnosis and high ACQ and bronchodilator response (BDR) levels and median and IgE levels. Cluster 3 (n = 19) (atopic) early diagnosis, highest BDR, highest ACQ score, highest total, and high specific IgE levels among the clusters. Cluster 4 (n = 9): (atopic) the highest specific IgE result, relatively high BMI, and IgE with median ACQ score among clusters. Cluster 5 (n = 5): (non-atopic) the earliest age for diagnosis, with the lowest BMI, the lowest ACQ score, and specific IgE levels, with high BDR and the median level of IgE among clusters; (4) Conclusions: We identified asthma phenotypes in Bulgarian children according to IgE levels, ACQ score, BDR, and age of diagnosis.

The management of severe asthma in 2020

Asthma is a chronic inflammatory disease of the airways affecting more than 300 million patients worldwide. The disease can be of various severity ranging from very mild to severe. The severe form of the disease only affects about 5% of patients but is responsible for a large component of the overall disease burden and results in about half of direct asthma-related costs. This led to important investments in research, which allowed better understanding of its pathophysiology and the development of new therapeutic strategies. Despite those breakthroughs, recent guidelines still emphasize the need to distinguish uncontrolled or difficult-to-control asthma from severe asthma. Indeed, a significant number of patients referred to severe asthma clinics are non-severe uncontrolled patients. However, the basics of asthma management such as ensuring that the patient has the right diagnosis, recognition of contributing comorbidities, avoidance of exposure to sensitizing agents in allergic individuals, regular assessment of control, and patient education should not be forgotten. The major improvements in pathophysiology arose from the evidences that asthma is of heterogeneous nature. Such heterogeneity has been particularly studied in severe asthma, leading to the recognition of more homogeneous groups referred to as phenotypes. Appropriate phenotyping of individual patients allows enforcing more specific treatment plans for patients, which is a step toward precision medicine. The high morbidity and socioeconomic burden of severe asthma has led to the development of optimized therapeutic strategies in addition to the commercialization of new drugs. Many of these have targeted the eosinophilic component of inflammatory asthma phenotypes while there is still a need to develop new therapies for non-eosinophilic asthma. When asthma is not controlled by optimal therapy, including a high-dose of inhaled corticosteroid (ICS) and a long-acting beta-agonist (LABA), a long acting anticholinergic agent can be added and if insufficient, a variety of biologic agents is now available in many countries. When biologics are not an option, thermoplasty and macrolides have also become available. Despite many recent breakthroughs in severe asthma, much research needs to be done. Improvement in availability of targeted asthma medications and asthma prevention should be top priorities.

Challenges of Clustering Multimodal Clinical Data: Review of Applications in Asthma Subtyping

Background

In the current era of personalized medicine, there is increasing interest in understanding the heterogeneity in disease populations. Cluster analysis is a method commonly used to identify subtypes in heterogeneous disease populations. The clinical data used in such applications are typically multimodal, which can make the application of traditional cluster analysis methods challenging.

Objective

This study aimed to review the research literature on the application of clustering multimodal clinical data to identify asthma subtypes. We assessed common problems and shortcomings in the application of cluster analysis methods in determining asthma subtypes, such that they can be brought to the attention of the research community and avoided in future studies.

Methods

We searched PubMed and Scopus bibliographic databases with terms related to cluster analysis and asthma to identify studies that applied dissimilarity-based cluster analysis methods. We recorded the analytic methods used in each study at each step of the cluster analysis process.

Results

Our literature search identified 63 studies that applied cluster analysis to multimodal clinical data to identify asthma subtypes. The features fed into the cluster algorithms were of a mixed type in 47 (75%) studies and continuous in 12 (19%), and the feature type was unclear in the remaining 4 (6%) studies. A total of 23 (37%) studies used hierarchical clustering with Ward linkage, and 22 (35%) studies used k-means clustering. Of these 45 studies, 39 had mixed-type features, but only 5 specified dissimilarity measures that could handle mixed-type features. A further 9 (14%) studies used a preclustering step to create small clusters to feed on a hierarchical method. The original sample sizes in these 9 studies ranged from 84 to 349. The remaining studies used hierarchical clustering with other linkages (n=3), medoid-based methods (n=3), spectral clustering (n=1), and multiple kernel k-means clustering (n=1), and in 1 study, the methods were unclear. Of 63 studies, 54 (86%) explained the methods used to determine the number of clusters, 24 (38%) studies tested the quality of their cluster solution, and 11 (17%) studies tested the stability of their solution. Reporting of the cluster analysis was generally poor in terms of the methods employed and their justification.

Conclusions

This review highlights common issues in the application of cluster analysis to multimodal clinical data to identify asthma subtypes. Some of these issues were related to the multimodal nature of the data, but many were more general issues in the application of cluster analysis. Although cluster analysis may be a useful tool for investigating disease subtypes, we recommend that future studies carefully consider the implications of clustering multimodal data, the cluster analysis process itself, and the reporting of methods to facilitate replication and interpretation of findings.

Preschool Wheezing: Trajectories and Long-Term Treatment

Wheezing is very common in infancy affecting one in three children during the first 3 years of life. Several wheeze phenotypes have been identified and most rely on temporal pattern of symptoms. Assessing the risk of asthma development is difficult. Factors predisposing to onset and persistence of wheezing such as breastfeeding, atopy, indoor allergen exposure, environmental tobacco smoke and viral infections are analyzed. Inhaled corticosteroids are recommended as first choice of controller treatment in all preschool children irrespective of phenotype, but they are particularly beneficial in terms of fewer exacerbations in atopic children. Other therapeutic options include the addition of montelukast or the intermittent use of inhaled corticosteroids. Overuse of inhaled steroids must be avoided. Therefore, adherence to treatment and correct administration of the medications need to be checked at every visit.

The Future of Asthma Care: Personalized Asthma Treatment

Although once considered a single disease entity, asthma is now known to be a complex inflammatory disease engaging a range of causal pathways. The most frequent forms of asthma are identified by sputum/blood eosinophilia and activation of type 2 inflammatory pathways involving interleukins-3, -4, -5, and granulocyte-macrophage colony-stimulating factor. The use of diagnostics that identify T2 engagement linked to the selective use of highly targeted biologics has opened up a new way of managing severe disease. Novel technologies, such as wearables and intelligent inhalers, enable real-time remote monitoring of asthma, creating a unique opportunity for personalized health care.

Systems biology and big data in asthma and allergy: recent discoveries and emerging challenges

Asthma is a common condition caused by immune and respiratory dysfunction, and it is often linked to allergy. A systems perspective may prove helpful in unravelling the complexity of asthma and allergy. Our aim is to give an overview of systems biology approaches used in allergy and asthma research. Specifically, we describe recent "omic"-level findings, and examine how these findings have been systematically integrated to generate further insight.Current research suggests that allergy is driven by genetic and epigenetic factors, in concert with environmental factors such as microbiome and diet, leading to early-life disturbance in immunological development and disruption of balance within key immuno-inflammatory pathways. Variation in inherited susceptibility and exposures causes heterogeneity in manifestations of asthma and other allergic diseases. Machine learning approaches are being used to explore this heterogeneity, and to probe the pathophysiological patterns or "endotypes" that correlate with subphenotypes of asthma and allergy. Mathematical models are being built based on genomic, transcriptomic and proteomic data to predict or discriminate disease phenotypes, and to describe the biomolecular networks behind asthma.The use of systems biology in allergy and asthma research is rapidly growing, and has so far yielded fruitful results. However, the scale and multidisciplinary nature of this research means that it is accompanied by new challenges. Ultimately, it is hoped that systems medicine, with its integration of omics data into clinical practice, can pave the way to more precise, personalised and effective management of asthma.

A modern approach to identifying and characterizing child asthma and wheeze phenotypes based on clinical data

‘Asthma’ is a complex disease that encapsulates a heterogeneous group of phenotypes and endotypes. Research to understand these phenotypes has previously been based on longitudinal wheeze patterns or hypothesis-driven observational criteria. The aim of this study was to use data-driven machine learning to identify asthma and wheeze phenotypes in children based on symptom and symptom history data, and, to further characterize these phenotypes. The study population included an asthma-rich population of twins in Sweden aged 9–15 years (n = 752). Latent class analysis using current and historical clinical symptom data generated asthma and wheeze phenotypes. Characterization was then performed with regression analyses using diagnostic data: lung function and immunological biomarkers, parent-reported medication use and risk-factors. The latent class analysis identified four asthma/wheeze phenotypes: early transient wheeze (15%); current wheeze/asthma (5%); mild asthma (9%), moderate asthma (10%) and a healthy phenotype (61%). All wheeze and asthma phenotypes were associated with reduced lung function and risk of hayfever compared to healthy. Children with mild and moderate asthma phenotypes were also more likely to have eczema, allergic sensitization and a family history of asthma. Furthermore, those with moderate asthma phenotype had a higher eosinophil concentration (β 0.21, 95%CI 0.12, 0.30) compared to healthy and used short-term relievers at a higher rate than children with mild asthma phenotype (RR 2.4, 95%CI 1.2–4.9). In conclusion, using a data driven approach we identified four wheeze/asthma phenotypes which were validated with further characterization as unique from one another and which can be adapted for use by the clinician or researcher.

Distinct asthma phenotypes with low maximal attainment of lung function on cluster analysis

OBJECTIVE

Asthma is a heterogeneous disease with varying clinical presentations, severity and ability to achieve asthma control. The present study aimed to characterize clinical phenotypes of asthma in an Indian cohort of subjects using a cluster analysis approach.

METHODS

Patients with confirmed asthma (N = 100) and at least 6-months of follow-up data, identified by retrospective chart review, were included in this study. Demographics, age at disease onset, disease duration, body mass index, serial spirometry and allergen sensitization were assessed. Asthma control was assessed prospectively using Global Initiative for Asthma and Asthma Control Test. R version 3.4.3 was used for statistical analysis. Ward's minimum-variance hierarchical clustering method was performed using an agglomerative (bottom-up) approach. To compare differences between clusters, analysis of variance using Kruskal-Wallis test (continuous variables) and chi-square test (categorical variables) was used.

RESULTS

Cluster analysis of 100 treatment-naive patients with asthma identified four clusters. Cluster 1, (N = 40), childhood onset of disease, normal body weight, equal gender distribution and achieved normal lung function. Cluster 2 (N = 16) included adolescent disease-onset, obese, majority males and had poor attainment of maximum lung functions. Cluster 3 (N = 20) were older, late-onset of disease, obese, majority male and had poor attainment of maximum lung function. Cluster 4 (N = 24) had adult-onset of disease, obese, predominantly female and achieved normal lung function.

CONCLUSIONS

In an Indian cohort of well-characterized patients with asthma, cluster analysis identified four distinct clinical phenotypes of asthma, two of which had poor attainment of maximum lung functions.

The Genetics and Genomics of Asthma

Asthma is a common, clinically heterogeneous disease with strong evidence of heritability. Progress in defining the genetic underpinnings of asthma, however, has been slow and hampered by issues of inconsistency. Recent advances in the tools available for analysis-assaying transcription, sequence variation, and epigenetic marks on a genome-wide scale-have substantially altered this landscape. Applications of such approaches are consistent with heterogeneity at the level of causation and specify patterns of commonality with a wide range of alternative disease traits. Looking beyond the individual as the unit of study, advances in technology have also fostered comprehensive analysis of the human microbiome and its varied roles in health and disease. In this article, we consider the implications of these technological advances for our current understanding of the genetics and genomics of asthma.

Data-driven adult asthma phenotypes based on clinical characteristics are associated with asthma outcomes twenty years later

BACKGROUND

Research based on cluster analyses led to the identification of particular phenotypes confirming phenotypic heterogeneity of asthma. The long-term clinical course of asthma phenotypes defined by clustering analysis remains unknown, although it is a key aspect to underpin their clinical relevance. We aimed to estimate risk of poor asthma events between asthma clusters identified 20 years earlier.

METHODS

The study relied on two cohorts of adults with asthma with 20-year follow-up, ECRHS (European Community Respiratory Health Survey) and EGEA (Epidemiological study on Genetics and Environment of Asthma). Regression models were used to compare asthma characteristics (current asthma, asthma exacerbations, asthma control, quality of life, and FEV₁ ) at follow-up and the course of FEV₁  between seven cluster-based asthma phenotypes identified 20 years earlier.

RESULTS

The analysis included 1325 adults with ever asthma. For each asthma characteristic assessed at follow-up, the risk for adverse outcomes differed significantly between the seven asthma clusters identified at baseline. As compared with the mildest asthma phenotype, ORs (95% CI) for asthma exacerbations varied from 0.9 (0.4 to 2.0) to 4.0 (2.0 to 7.8) and the regression estimates (95% CI) for FEV₁ % predicted varied from 0.6 (-3.5 to 4.6) to -9.9 (-14.2 to -5.5) between clusters. Change in FEV₁ over time did not differ significantly across clusters.

CONCLUSION

Our findings show that the long-term risk for poor asthma outcomes differed between comprehensive adult asthma phenotypes identified 20 years earlier, and suggest a strong tracking of asthma activity and impaired lung function over time.

Discovering Pediatric Asthma Phenotypes on the Basis of Response to Controller Medication Using Machine Learning

RATIONALE

Pediatric asthma has variable underlying inflammation and symptom control. Approaches to addressing this heterogeneity, such as clustering methods to find phenotypes and predict outcomes, have been investigated. However, clustering based on the relationship between treatment and clinical outcome has not been performed, and machine learning approaches for long-term outcome prediction in pediatric asthma have not been studied in depth.

OBJECTIVES

Our objectives were to use our novel machine learning algorithm, predictor pursuit (PP), to discover pediatric asthma phenotypes on the basis of asthma control in response to controller medications, to predict longitudinal asthma control among children with asthma, and to identify features associated with asthma control within each discovered pediatric phenotype.

METHODS

We applied PP to the Childhood Asthma Management Program study data (n = 1,019) to discover phenotypes on the basis of asthma control between assigned controller therapy groups (budesonide vs. nedocromil). We confirmed PP's ability to discover phenotypes using the Asthma Clinical Research Network/Childhood Asthma Research and Education network data. We next predicted children's asthma control over time and compared PP's performance with that of traditional prediction methods. Last, we identified clinical features most correlated with asthma control in the discovered phenotypes.

RESULTS

Four phenotypes were discovered in both datasets: allergic not obese (A⁺/O^-), obese not allergic (A^-/O⁺), allergic and obese (A⁺/O⁺), and not allergic not obese (A^-/O^-). Of the children with well-controlled asthma in the Childhood Asthma Management Program dataset, we found more nonobese children treated with budesonide than with nedocromil (P = 0.015) and more obese children treated with nedocromil than with budesonide (P = 0.008). Within the obese group, more A⁺/O⁺ children's asthma was well controlled with nedocromil than with budesonide (P = 0.022) or with placebo (P = 0.011). The PP algorithm performed significantly better (P < 0.001) than traditional machine learning algorithms for both short- and long-term asthma control prediction. Asthma control and bronchodilator response were the features most predictive of short-term asthma control, regardless of type of controller medication or phenotype. Bronchodilator response and serum eosinophils were the most predictive features of asthma control, regardless of type of controller medication or phenotype.

CONCLUSIONS

Advanced statistical machine learning approaches can be powerful tools for discovery of phenotypes based on treatment response and can aid in asthma control prediction in complex medical conditions such as asthma.

Clustering asthma symptoms and cleaning and disinfecting activities and evaluating their associations among healthcare workers

Asthma is a heterogeneous disease with varying severity and subtypes. Recent reviews of epidemiologic studies have identified cleaning and disinfecting activities (CDAs) as important risk factors for asthma-related outcomes among healthcare workers. However, the complexity of CDAs in healthcare settings has rarely been examined. This study utilized a complex survey dataset and data reduction approaches to identify and group healthcare workers with similar patterns of asthma symptoms, and then explored their associations with groups of participants with similar patterns of CDAs. Self-reported information on asthma symptoms/care, CDAs, demographics, smoking status, allergic status, and other characteristics were collected from 2030 healthcare workers within nine selected occupations in New York City. Hierarchical clustering was conducted to systematically group participants based on similarity of patterns of the 27 asthma symptom/care variables, and 14 product applications during CDAs, separately. Word clouds were used to visualize the complex information on the resulting clusters. The associations of asthma health clusters (HCs) with exposure clusters (ECs) were evaluated using multinomial logistic regression. Five HCs were identified (HC-1 to HC-5), labelled based on predominant features as: "no symptoms", "winter cough/phlegm", "mild asthma symptoms", "undiagnosed/untreated asthma", and "asthma attacks/exacerbations". For CDAs, five ECs were identified (EC-1 to EC-5), labelled as: "no products", "housekeeping/chlorine", "patient care", "general cleaning/laboratory", and "disinfection products". Using HC-1 and EC-1 as the reference groups, EC-2 was associated with HC-4 (odds ratio (OR) = 3.11, 95% confidence interval (95% CI) = 1.46-6.63) and HC-5 (OR = 2.71, 95% CI = 1.25-5.86). EC-3 was associated with HC-5 (OR = 2.34, 95% CI = 1.16-4.72). EC-4 was associated with HC-5 (OR = 2.35, 95% CI = 1.07-5.13). EC-5 was associated with HC-3 (OR = 1.81, 95% CI = 1.09-2.99) and HC-4 (OR = 3.42, 95% CI = 1.24-9.39). Various combinations of product applications like using alcohols, bleach, high-level disinfectants, and enzymes to disinfect instruments and clean surfaces captured by the ECs were identified as risk factors for the different asthma symptoms clusters, indicating that prevention efforts may require targeting multiple products. The associations of HCs with EC can be used to better inform prevention strategies and treatment options to avoid disease progression. This study demonstrated hierarchical clustering and word clouds were useful techniques for analyzing and visualizing a complex dataset with a large number of potentially correlated variables to generate practical information that can inform prevention activities.

Korean childhood asthma study (KAS): a prospective, observational cohort of Korean asthmatic children

Background

Asthma is a syndrome composed of heterogeneous disease entities. Although it is agreed that proper asthma endo-typing and appropriate type-specific interventions are crucial in the management of asthma, little data are available regarding pediatric asthma.

Methods

We designed a cluster-based, prospective, observational cohort study of asthmatic children in Korea (Korean childhood Asthma Study [KAS]). A total of 1000 Korean asthmatic children, aged from 5 to 15 years, will be enrolled at the allergy clinics of the 19 regional tertiary hospitals from August 2016 to December 2018. Physicians will verify the relevant histories of asthma and comorbid diseases, as well as airway lability from the results of spirometry and bronchial provocation tests. Questionnaires regarding subjects’ baseline characteristics and their environment, self-rating of asthma control, and laboratory tests for allergy and airway inflammation will be collected at the time of enrollment. Follow-up data regarding asthma control, lung function, and environmental questionnaires will be collected at least every 6 months to assess outcome and exacerbation-related aggravating factors. In a subgroup of subjects, peak expiratory flow rate will be monitored by communication through a mobile application during the overall study period. Cluster analysis of the initial data will be used to classify Korean pediatric asthma patients into several clusters; the exacerbation and progression of asthma will be assessed and compared among these clusters. In a subgroup of patients, big data-based deep learning analysis will be applied to predict asthma exacerbation.

Discussion

Based on the assumption that asthma is heterogeneous and each subject exhibits a different subset of risk factors for asthma exacerbation, as well as a different disease progression, the KAS aims to identify several asthma clusters and their essential determinants, which are more suitable for Korean asthmatic children. Thereafter we may suggest cluster-specific strategies by focusing on subjects’ personalized aggravating factors during each exacerbation episode and by focusing on disease progression. The KAS will provide a good academic background with respect to each interventional strategy to achieve better asthma control and prognosis.

How the Smartphone Is Changing Allergy Diagnostics

PURPOSE OF REVIEW

Evidence-based clinical diagnosis of allergic disorders is increasingly challenging. Clinical decision support systems implemented in mobile applications (apps) are being developed to assist clinicians in diagnostic decisions at the point of care. We reviewed apps for allergic diseases general diagnosis, diagnostic refinement and diagnostic personalisation. Apps designed for specific medical devices are not addressed.

RECENT FINDINGS

Apps with potential usefulness in the initial diagnosis and diagnostic refinement of respiratory, food, skin and drug allergies are described. Apps to support diagnostic personalisation are not yet available. There is an urgent need to increase the scientific evidence on the real usefulness of these apps, as well as to develop new scientifically grounded apps designed and validated to support all allergic diseases and diagnostic levels. Apps have the potential to change the diagnosis of allergic diseases becoming part of the routine diagnostics toolset, but its usefulness needs to be established.

Trajectories of childhood immune development and respiratory health relevant to asthma and allergy

Events in early life contribute to subsequent risk of asthma; however, the causes and trajectories of childhood wheeze are heterogeneous and do not always result in asthma. Similarly, not all atopic individuals develop wheeze, and vice versa. The reasons for these differences are unclear. Using unsupervised model-based cluster analysis, we identified latent clusters within a prospective birth cohort with deep immunological and respiratory phenotyping. We characterised each cluster in terms of immunological profile and disease risk, and replicated our results in external cohorts from the UK and USA. We discovered three distinct trajectories, one of which is a high-risk ‘atopic’ cluster with increased propensity for allergic diseases throughout childhood. Atopy contributes varyingly to later wheeze depending on cluster membership. Our findings demonstrate the utility of unsupervised analysis in elucidating heterogeneity in asthma pathogenesis and provide a foundation for improving management and prevention of childhood asthma.

Asthma causes wheezy and troubled breathing, and can be life-threatening. Scientists and doctors understand that asthma begins in early childhood. Chest infections, exposure to bacteria, viruses, and allergies may cause or trigger asthma. One person with asthma may not have the same origins as another. But it is not yet clear how various triggers may interact to trigger or exacerbate asthma.

To disentangle how these factors contribute to asthma, experts have tried to group people with asthma into subgroups. Unfortunately, the groups often vary from expert to expert. Now, some scientists are using computers to sort patients with asthma. The scientists let the computers decide the best criteria for sorting patients. This way the machines may identify patterns that are not obvious to humans.

Using this computer-based approach, Tang et al. sorted Australian children with asthma into 3 groups based on their early life allergies and respiratory health. One group has high-risk asthma with frequent chest infections and strong allergic responses. The other two groups are low-risk, but they respond differently to allergy and infection. Common tests used by doctors to diagnose patients with allergy or asthma may not work the same with all three groups. The bacteria found in the nose influence the risk of asthma, even in patients who are well, and the way this occurs varies by group. Similar groups were also found among children with asthma in the United States and the United Kingdom.

Learning more about subgroups of patients with asthma may help other scientists and doctors design better ways to diagnose, treat, or prevent asthma. Working together with scientists around the world to determine how to best describe subgroups of people according to asthma type and risk is a critical step in the process. Tang et al. hope other scientist will test whether these three groups are also found in people from other parts of the world.

Informatics and machine learning to define the phenotype

INTRODUCTION

For the past decade, the focus of complex disease research has been the genotype. From technological advancements to the development of analysis methods, great progress has been made. However, advances in our definition of the phenotype have remained stagnant. Phenotype characterization has recently emerged as an exciting area of informatics and machine learning. The copious amounts of diverse biomedical data that have been collected may be leveraged with data-driven approaches to elucidate trait-related features and patterns. Areas covered: In this review, the authors discuss the phenotype in traditional genetic associations and the challenges this has imposed.Approaches for phenotype refinement that can aid in more accurate characterization of traits are also discussed. Further, the authors highlight promising machine learning approaches for establishing a phenotype and the challenges of electronic health record (EHR)-derived data. Expert commentary: The authors hypothesize that through unsupervised machine learning, data-driven approaches can be used to define phenotypes rather than relying on expert clinician knowledge. Through the use of machine learning and an unbiased set of features extracted from clinical repositories, researchers will have the potential to further understand complex traits and identify patient subgroups. This knowledge may lead to more preventative and precise clinical care.

Features of asthma which provide meaningful insights for understanding the disease heterogeneity

BACKGROUND

Data-driven methods such as hierarchical clustering (HC) and principal component analysis (PCA) have been used to identify asthma subtypes, with inconsistent results.

OBJECTIVE

To develop a framework for the discovery of stable and clinically meaningful asthma subtypes.

METHODS

We performed HC in a rich data set from 613 asthmatic children, using 45 clinical variables (Model 1), and after PCA dimensionality reduction (Model 2). Clinical experts then identified a set of asthma features/domains which informed clusters in the two analyses. In Model 3, we reclustered the data using these features to ascertain whether this improved the discovery process.

RESULTS

Cluster stability was poor in Models 1 and 2. Clinical experts highlighted four asthma features/domains which differentiated the clusters in two models: age of onset, allergic sensitization, severity, and recent exacerbations. In Model 3 (HC using these four features), cluster stability improved substantially. The cluster assignment changed, providing more clinically interpretable results. In a 5-cluster model, we labelled the clusters as: "Difficult asthma" (n = 132); "Early-onset mild atopic" (n = 210); "Early-onset mild non-atopic: (n = 153); "Late-onset" (n = 105); and "Exacerbation-prone asthma" (n = 13). Multinomial regression demonstrated that lung function was significantly diminished among children with "Difficult asthma"; blood eosinophilia was a significant feature of "Difficult," "Early-onset mild atopic," and "Late-onset asthma." Children with moderate-to-severe asthma were present in each cluster.

CONCLUSIONS AND CLINICAL RELEVANCE

An integrative approach of blending the data with clinical expert domain knowledge identified four features, which may be informative for ascertaining asthma endotypes. These findings suggest that variables which are key determinants of asthma presence, severity, or control may not be the most informative for determining asthma subtypes. Our results indicate that exacerbation-prone asthma may be a separate asthma endotype and that severe asthma is not a single entity, but an extreme end of the spectrum of several different asthma endotypes.

Altered fatty acid metabolism and reduced stearoyl-coenzyme a desaturase activity in asthma

BACKGROUND

Fatty acids and lipid mediator signaling play an important role in the pathogenesis of asthma, yet this area remains largely underexplored. The aims of this study were (i) to examine fatty acid levels and their metabolism in obese and nonobese asthma patients and (ii) to determine the functional effects of altered fatty acid metabolism in experimental models.

METHODS

Medium- and long-chain fatty acid levels were quantified in serum from 161 human volunteers by LC/MS. Changes in stearoyl-coenzyme A desaturase (SCD) expression and activity were evaluated in the ovalbumin (OVA) and house dust mite (HDM) murine models. Primary human bronchial epithelial cells from asthma patients and controls were evaluated for SCD expression and activity.

RESULTS

The serum desaturation index (an indirect measure of SCD) was significantly reduced in nonobese asthma patients and in the OVA murine model. SCD1 gene expression was significantly reduced within the lungs following OVA or HDM challenge. Inhibition of SCD in mice promoted airway hyper-responsiveness. SCD1 expression was suppressed in bronchial epithelial cells from asthma patients. IL-4 and IL-13 reduced epithelial cell SCD1 expression. Inhibition of SCD reduced surfactant protein C expression and suppressed rhinovirus-induced IP-10 secretion, which was associated with increased viral titers.

CONCLUSIONS

This is the first study to demonstrate decreased fatty acid desaturase activity in humans with asthma. Experimental models in mice and human epithelial cells suggest that inhibition of desaturase activity leads to airway hyper-responsiveness and reduced antiviral defense. SCD may represent a new target for therapeutic intervention in asthma patients.

The heterogeneity of viral bronchiolitis: A lack of universal consensus definitions

Viral bronchiolitis is one of the most common hospital presentations in infancy and as such represents a major healthcare burden worldwide. However despite this, there are currently no effective targeted therapies nor can those infants at highest risk for developing severe disease or subsequent respiratory morbidity be predicted on initial hospital presentation. Current definitions of bronchiolitis in the published literature vary significantly in terms of the age range at presentation, specific clinical symptoms, causative virus, and the inclusion or exclusion of infants with previous presentations and/or various comorbidities. In this review, we highlight how this heterogeneity among definitions contributes to a lack of clarity on this condition and its likely multiple endotypes. We argue that without a new universal consensus definition or sets of definitions, progress into bronchiolitis will continue to be stalled.

Disaggregating asthma: Big investigation versus big data

We are facing a major challenge in bridging the gap between identifying subtypes of asthma to understand causal mechanisms and translating this knowledge into personalized prevention and management strategies. In recent years, "big data" has been sold as a panacea for generating hypotheses and driving new frontiers of health care; the idea that the data must and will speak for themselves is fast becoming a new dogma. One of the dangers of ready accessibility of health care data and computational tools for data analysis is that the process of data mining can become uncoupled from the scientific process of clinical interpretation, understanding the provenance of the data, and external validation. Although advances in computational methods can be valuable for using unexpected structure in data to generate hypotheses, there remains a need for testing hypotheses and interpreting results with scientific rigor. We argue for combining data- and hypothesis-driven methods in a careful synergy, and the importance of carefully characterized birth and patient cohorts with genetic, phenotypic, biological, and molecular data in this process cannot be overemphasized. The main challenge on the road ahead is to harness bigger health care data in ways that produce meaningful clinical interpretation and to translate this into better diagnoses and properly personalized prevention and treatment plans. There is a pressing need for cross-disciplinary research with an integrative approach to data science, whereby basic scientists, clinicians, data analysts, and epidemiologists work together to understand the heterogeneity of asthma.

Asthma phenotypes in childhood

INTRODUCTION

Asthma is no longer thought of as a single disease, but rather a collection of varying symptoms expressing different disease patterns. One of the ongoing challenges is understanding the underlying pathophysiological mechanisms that may be responsible for the varying responses to treatment. Areas Covered: This review provides an overview of our current understanding of the asthma phenotype concept in childhood and describes key findings from both conventional and data-driven methods. Expert Commentary: With the vast amounts of data generated from cohorts, there is hope that we can elucidate distinct pathophysiological mechanisms, or endotypes. In return, this would lead to better patient stratification and disease management, thereby providing true personalised medicine.