Distinguishing Asthma Phenotypes Using Machine Learning Approaches

The role of wheezing subtypes in the development of early childhood asthma

BACKGROUND

Early childhood wheezing is associated with asthma risk at later ages, emphasizing the need for understanding wheezing patterns and their implications for asthma development.

METHODS

Children in the F2-generation (n = 603) of the Isle of Wight Birth Cohort (IOWBC) were followed-up at 3, 6, 12, 24, 36, and 72 months. Prevalence of wheeze and wheeze type (general, infectious, and non-infectious) were recorded. Group-based trajectory models covering ages 3 to 36 months were used to identify early childhood wheezing trajectories for each type of wheeze. These trajectories were examined for their association with asthma status and lung function at 6 years and later.

RESULTS

Distinct trajectories for general ("Persistent", "Transient", "Progressive", and "Infrequent/Never"), infectious ("Persistent", "Transient", and "Infrequent/Never"), and non-infectious ("Progressive", "Early Occurrence", and "Infrequent/Never") wheezing were identified. Compared to the "Infrequent/Never" trajectories, four trajectories were associated with an increased risk of asthma, namely "Progressive" non-infectious, "Early Occurrence" non-infectious, "Persistent" infectious, and "Persistent" general wheeze trajectories.

CONCLUSIONS

The identification of wheeze trajectories across different etiologies as significant risk factors for asthma may aid in understanding the complex, multifactorial nature of asthma onset. The findings suggest that early identification of specific wheeze patterns, not just occurrence of wheezing, can inform clinical interventions and potentially mitigate the risk of developing asthma.

Computational Phenotyping of Obstructive Airway Diseases: A Systematic Review

Introduction

Computational sciences have significantly contributed to characterizing airway disease phenotypes, complementing medical expertise. However, comparing studies that derive phenotypes is challenging due to varying decisions made during phenotyping. We conducted a systematic review to describe studies that utilized unsupervised computational approaches for phenotyping obstructive airway diseases in children and adults.

Methods

We searched for relevant papers published between 2010 and 2020 in PubMed, EMBASE, Scopus, Web of Science, and Google Scholar. Additional sources included conference proceedings, reference lists, and expert recommendations. Two reviewers independently screened studies for eligibility, extracted data, and assessed study quality. Disagreements were resolved by a third reviewer. An in-house quality appraisal tool was used. Evidence was synthesized, focusing on populations, variables, and computational approaches used for deriving phenotypes.

Results

Of 120 studies included in the review, 60 focused on asthma, 19 on severe asthma, 28 on COPD, 4 on asthma-COPD overlap (ACO), and 9 on rhinitis. Among asthma studies, 31 focused on adults and 9 on children, with phenotypes related to atopy, age at onset, and disease severity. Severe asthma phenotypes were characterized by symptomatology, atopy, and age at onset. COPD phenotypes involved lung function, emphysematous changes, smoking, comorbidities, and daily life impairment. ACO and rhinitis phenotypes were mostly defined by symptoms, lung function, and sensitization, respectively. Most studies used hierarchical clustering, with some employing latent class modeling, mixture models, and factor analysis. The comprehensiveness of variable reporting was the best quality indicator, while reproducibility measures were often lacking.

Conclusion

Variations in phenotyping methods, study settings, participant profiles, and variables contribute to significant differences in characterizing asthma, severe asthma, COPD, ACO, and rhinitis phenotypes across studies. Lack of reproducibility measures limits the evaluation of computational phenotyping in airway diseases, underscoring the need for consistent approaches to defining outcomes and selecting variables to ensure reliable phenotyping.

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.

Unspecified asthma, childhood-onset, and adult-onset asthma have different causal genes: a Mendelian randomization analysis

Introduction

Asthma is a heterogeneous condition that is characterized by reversible airway obstruction. Childhood-onset asthma (COA) and adult-onset asthma (AOA) are two prominent asthma subtypes, each with unique etiological factors and prognosis, which suggests the existence of both shared and distinct risk factors.

Methods

Here, we employed a two-sample Mendelian randomization analysis to elucidate the causal association between genes within lung and whole-blood-specific gene regulatory networks (GRNs) and the development of unspecified asthma, COA, and AOA using the Wald ratio method. Lung and whole blood-specific GRNs, encompassing spatial eQTLs (instrumental variables) and their target genes (exposures), were utilized as exposure data. Genome-wide association studies for unspecified asthma, COA, and AOA were used as outcome data in this investigation.

Results

We identified 101 genes that were causally linked to unspecified asthma, 39 genes causally associated with COA, and ten genes causally associated with AOA. Among the identified genes, 29 were shared across some, or all of the asthma subtypes. Of the identified causal genes, ORMDL3 had the strongest causal association with both unspecified asthma (OR: 1.49; 95% CI:1.42-1.57; p=7.30x10-51) and COA (OR: 3.37; 95% CI: 3.02-3.76; p=1.95x10-102), whereas PEBP1P3 had the strongest causal association with AOA (OR: 1.28; 95% CI: 1.16-1.41; p=0.007).

Discussion

This study identified shared and unique genetic factors causally associated with different asthma subtypes. In so doing, our study emphasizes the need to move beyond perceiving asthma as a singular condition to enable the development of therapeutic interventions that target sub-type specific causal genes.

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.

Understanding the heterogeneity of childhood allergic sensitization and its relationship with asthma

PURPOSE OF REVIEW

To review the current state of knowledge on the relationship between allergic sensitization and asthma; to lay out a roadmap for the development of IgE biomarkers that differentiate, in individual sensitized patients, whether their sensitization is important for current or future asthma symptoms, or has little or no relevance to the disease.

RECENT FINDINGS

The evidence on the relationship between sensitization and asthma suggests that some subtypes of allergic sensitization are not associated with asthma symptoms, whilst others are pathologic. Interaction patterns between IgE antibodies to individual allergenic molecules on component-resolved diagnostics (CRD) multiplex arrays might be hallmarks by which different sensitization subtypes relevant to asthma can be distinguished. These different subtypes of sensitization are associated amongst sensitized individuals at all ages, with different clinical presentations (no disease, asthma as a single disease, and allergic multimorbidity); amongst sensitized preschool children with and without lower airway symptoms, with different risk of subsequent asthma development; and amongst sensitized patients with asthma, with differing levels of asthma severity.

SUMMARY

The use of machine learning-based methodologies on complex CRD data can help us to design better diagnostic tools to help practising physicians differentiate between benign and clinically important sensitization.

Novel Artificial Intelligence-Based Technology to Diagnose Asthma Using Methacholine Challenge Tests

PURPOSE

The methacholine challenge test (MCT) has high sensitivity but relatively low specificity for asthma diagnosis. This study aimed to develop and validate machine learning (ML) models to improve the diagnostic performance of MCT for asthma.

METHODS

Data from 1,501 patients with asthma symptoms who underwent MCT between 2015 and 2020 were analyzed. The patients were grouped as either the training (80%, n = 1,265) and test sets (20%, n = 236) depending on the time of referral. The conventional model (provocative concentration that causes a 20% decrease in forced expiratory volume in one second [FEV1]; PC20 ≤ 16 mg/mL) was compared with the prediction models derived from five ML methods: logistic regression, support vector machine, random forest, extreme gradient boosting, and artificial neural network. The area under the receiver operator characteristic curves (AUROC) and area under the precision-recall curves (AUPRC) of each model were compared. The prediction models were further analyzed using different input combinations of FEV1, forced vital capacity (FVC), and forced expiratory flow at 25%-75% of forced vital capacity (FEF25%-75%) values obtained during MCT.

RESULTS

In total, 545 patients (36.3%) were diagnosed with asthma. The AUROC of the conventional model was 0.856 (95% confidence interval [CI], 0.852-0.861), and the AUPRC was 0.759 (95% CI, 0.751-0.766). All the five ML prediction models had higher AUROC and AUPRC values than those of the conventional model, and random forest showed both highest AUROC (0.950; 95% CI, 0.948-0.952) and AUROC (0.909; 95% CI, 0.905-0.914) when FEV1, FVC, and FEF25%-75% were included as inputs.

CONCLUSIONS

Artificial intelligence-based models showed excellent performance in asthma prediction compared to using PC20 ≤ 16 mg/mL. The novel technology could be used to enhance the clinical diagnosis of asthma.

Characteristics of Allergic, Eosinophilic, and Overlapping Asthma Phenotypes Among Pediatric Patients with Current Asthma: A Cross-Sectional Study from Saudi Arabia

Purpose

Asthma is one of the most common chronic diseases affecting 10%-30% of children in Saudi Arabia. Although data exist on adult asthma phenotyping and endotyping in Saudi Arabia, little is known about asthma phenotypes in Saudi children.

Patients and Methods

This cross-sectional study enrolled pediatric patients diagnosed with bronchial asthma and followed in the pediatric pulmonology clinic of the Abha Maternity and Children Hospital between August 2021 and May 2023.

Results

A total of 321 children (aged 5-14 years) were analyzed. The population was classified into allergic [169 (52.6%)], eosinophilic [144 (44.9%)], and overlapping allergic and eosinophilic asthma [97 (30.2%)] phenotypes. Regarding asthma severity, 35.5%, 50.2%, and 14.3% were classified as mild, moderate, and severe, respectively. Of the 321 patients in the study, 124 (38.6%) had at least one asthma exacerbation that required hospitalization. The number of reported missed school days in the previous year was 1571 days [190 (59.2%) patients reported at least one missed school day]. The factors associated with the likelihood of uncontrolled asthma for all study participants included: emergency room (ER) visit last year (OR = 3.7, 95% CI:0.6-15.9]), overlapping eosinophilic and allergic (OR = 3.2, 95% CI = 1.8-5.9), and allergic phenotype (OR = 2.7, 95% CI = 1.3-5.4). The level of asthma control differed significantly among the three asthma phenotypes (p = 0.037).

Conclusion

Allergic asthma is the most prevalent asthma phenotype in this study, followed by the eosinophilic phenotype. The research has also shown that several factors predict uncontrolled asthma, including a family history of asthma, previous admission to the PICU, and previous hospitalization ever. There is, therefore, a definite need for multicenter cohort studies to better understand the phenotypes and endotypes of childhood asthma, as it could offer therapeutic and prognostic relevance.

Understanding progression from pre-school wheezing to school-age asthma: Can modern data approaches help?

Preschool wheezing and childhood asthma create a heavy disease burden which is only exacerbated by the complexity of the conditions. Preschool wheezing exhibits both "curricular" and "aetiological" heterogeneity: that is, heterogeneity across patients both in the time-course of its development and in its underpinning pathological mechanisms. Since these are not fully understood, but clinical presentations across patients may nonetheless be similar, current diagnostic labels are imprecise-not mapping cleanly onto underlying disease mechanisms-and prognoses uncertain. These uncertainties also make a identifying new targets for therapeutic intervention difficult. In the past few decades, carefully designed birth cohort studies have collected "big data" on a large scale, incorporating not only a wealth of longitudinal clinical data, but also detailed information from modalities as varied as imaging, multiomics, and blood biomarkers. The profusion of big data has seen the proliferation of what we term "modern data approaches" (MDAs)-grouping together machine learning, artificial intelligence, and data science-to make sense and make use of this data. In this review, we survey applications of MDAs (with an emphasis on machine learning) in childhood wheeze and asthma, highlighting the extent of their successes in providing tools for prognosis, unpicking the curricular heterogeneity of these conditions, clarifying the limitations of current diagnostic criteria, and indicating directions of research for uncovering the etiology of the diseases underlying these conditions. Specifically, we focus on the trajectories of childhood wheeze phenotypes. Further, we provide an explainer of the nature and potential use of MDAs and emphasize the scope of what we can hope to achieve with them.

Automated detection of airflow obstructive diseases: A systematic review of the last decade (2013-2022)

BACKGROUND AND OBJECTIVE

Obstructive airway diseases, including asthma and Chronic Obstructive Pulmonary Disease (COPD), are two of the most common chronic respiratory health problems. Both of these conditions require health professional expertise in making a diagnosis. Hence, this process is time intensive for healthcare providers and the diagnostic quality is subject to intra- and inter- operator variability. In this study we investigate the role of automated detection of obstructive airway diseases to reduce cost and improve diagnostic quality.

METHODS

We investigated the existing body of evidence and applied Preferred Reporting Items for Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to search records in IEEE, Google scholar, and PubMed databases. We identified 65 papers that were published from 2013 to 2022 and these papers cover 67 different studies. The review process was structured according to the medical data that was used for disease detection. We identified six main categories, namely air flow, genetic, imaging, signals, and miscellaneous. For each of these categories, we report both disease detection methods and their performance.

RESULTS

We found that medical imaging was used in 14 of the reviewed studies as data for automated obstructive airway disease detection. Genetics and physiological signals were used in 13 studies. Medical records and air flow were used in 9 and 7 studies, respectively. Most papers were published in 2020 and we found three times more work on Machine Learning (ML) when compared to Deep Learning (DL). Statistical analysis shows that DL techniques achieve higher Accuracy (ACC) when compared to ML. Convolutional Neural Network (CNN) is the most common DL classifier and Support Vector Machine (SVM) is the most widely used ML classifier. During our review, we discovered only two publicly available asthma and COPD datasets. Most studies used private clinical datasets, so data size and data composition are inconsistent.

CONCLUSIONS

Our review results indicate that Artificial Intelligence (AI) can improve both decision quality and efficiency of health professionals during COPD and asthma diagnosis. However, we found several limitations in this review, such as a lack of dataset consistency, a limited dataset and remote monitoring was not sufficiently explored. We appeal to society to accept and trust computer aided airflow obstructive diseases diagnosis and we encourage health professionals to work closely with AI scientists to promote automated detection in clinical practice and hospital settings.

Current state and prospects of artificial intelligence in allergy

The field of medicine is witnessing an exponential growth of interest in artificial intelligence (AI), which enables new research questions and the analysis of larger and new types of data. Nevertheless, applications that go beyond proof of concepts and deliver clinical value remain rare, especially in the field of allergy. This narrative review provides a fundamental understanding of the core concepts of AI and critically discusses its limitations and open challenges, such as data availability and bias, along with potential directions to surmount them. We provide a conceptual framework to structure AI applications within this field and discuss forefront case examples. Most of these applications of AI and machine learning in allergy concern supervised learning and unsupervised clustering, with a strong emphasis on diagnosis and subtyping. A perspective is shared on guidelines for good AI practice to guide readers in applying it effectively and safely, along with prospects of field advancement and initiatives to increase clinical impact. We anticipate that AI can further deepen our knowledge of disease mechanisms and contribute to precision medicine in allergy.

A meta-analysis of genome-wide association studies of childhood wheezing phenotypes identifies ANXA1 as a susceptibility locus for persistent wheezing

Background

Many genes associated with asthma explain only a fraction of its heritability. Most genome-wide association studies (GWASs) used a broad definition of 'doctor-diagnosed asthma', thereby diluting genetic signals by not considering asthma heterogeneity. The objective of our study was to identify genetic associates of childhood wheezing phenotypes.

Methods

We conducted a novel multivariate GWAS meta-analysis of wheezing phenotypes jointly derived using unbiased analysis of data collected from birth to 18 years in 9568 individuals from five UK birth cohorts.

Results

Forty-four independent SNPs were associated with early-onset persistent, 25 with pre-school remitting, 33 with mid-childhood remitting, and 32 with late-onset wheeze. We identified a novel locus on chr9q21.13 (close to annexin 1 [ANXA1], p<6.7 × 10-9), associated exclusively with early-onset persistent wheeze. We identified rs75260654 as the most likely causative single nucleotide polymorphism (SNP) using Promoter Capture Hi-C loops, and then showed that the risk allele (T) confers a reduction in ANXA1 expression. Finally, in a murine model of house dust mite (HDM)-induced allergic airway disease, we demonstrated that anxa1 protein expression increased and anxa1 mRNA was significantly induced in lung tissue following HDM exposure. Using anxa1-/- deficient mice, we showed that loss of anxa1 results in heightened airway hyperreactivity and Th2 inflammation upon allergen challenge.

Conclusions

Targeting this pathway in persistent disease may represent an exciting therapeutic prospect.

Funding

UK Medical Research Council Programme Grant MR/S025340/1 and the Wellcome Trust Strategic Award (108818/15/Z) provided most of the funding for this study.

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.

Delineating asthma according to inflammation phenotypes with a focus on paucigranulocytic asthma

ABSTRACT

Asthma is characterized by chronic airway inflammation and airway hyper-responsiveness. However, the differences in pathophysiology and phenotypic symptomology make a diagnosis of "asthma" too broad hindering individualized treatment. Four asthmatic inflammatory phenotypes have been identified based on inflammatory cell profiles in sputum: eosinophilic, neutrophilic, paucigranulocytic, and mixed-granulocytic. Paucigranulocytic asthma may be one of the most common phenotypes in stable asthmatic patients, yet it remains much less studied than the other inflammatory phenotypes. Understanding of paucigranulocytic asthma in terms of phenotypic discrimination, distribution, stability, surrogate biomarkers, underlying pathophysiology, clinical characteristics, and current therapies is fragmented, which impedes clinical management of patients. This review brings together existing knowledge and ongoing research about asthma phenotypes, with a focus on paucigranulocytic asthma, in order to present a comprehensive picture that may clarify specific inflammatory phenotypes and thus improve clinical diagnoses and disease management.

Early childhood wheezing phenotypes and determinants in a South African birth cohort: longitudinal analysis of the Drakenstein Child Health Study

BACKGROUND

Developmental trajectories of childhood wheezing in low-income and middle-income countries (LMICs) have not been well described. We aimed to derive longitudinal wheeze phenotypes from birth to 5 years in a South African birth cohort and compare those with phenotypes derived from a UK cohort.

METHODS

We used data from the Drakenstein Child Health Study (DCHS), a longitudinal birth cohort study in a peri-urban area outside Cape Town, South Africa. Pregnant women (aged ≥18 years) were enrolled during their second trimester at two public health clinics. We followed up children from birth to 5 years to derive six multidimensional indicators of wheezing (including duration, temporal sequencing, persistence, and recurrence) and applied Partition Around Medoids clustering to derive wheeze phenotypes. We compared phenotypes with a UK cohort (the Avon Longitudinal Study of Parents and Children [ALSPAC]). We investigated associations of phenotypes with early-life exposures, including all-cause lower respiratory tract infection (LRTI) and virus-specific LRTI (respiratory syncytial virus, rhinovirus, adenovirus, influenza, and parainfluenza virus) up to age 5 years. We investigated the association of phenotypes with lung function at 6 weeks and 5 years.

FINDINGS

Between March 5, 2012, and March 31, 2015, we enrolled 1137 mothers and there were 1143 livebirths. Four wheeze phenotypes were identified among 950 children with complete data: never (480 children [50%]), early transient (215 children [23%]), late onset (104 children [11%]), and recurrent (151 children [16%]). Multivariate adjusted analysis indicated that LRTI and respiratory syncytial virus-LRTI, but not other respiratory viruses, were associated with increased risk of recurrent wheeze (odds ratio [OR] 2·79 [95% CI 2·05-3·81] for all LTRIs; OR 2·59 [1·30-5·15] for respiratory syncytial virus-LRTIs). Maternal smoking (1·88 [1·12-3·02]), higher socioeconomic status (2·46 [1·23-4·91]), intimate partner violence (2·01 [1·23-3·29]), and male sex (2·47 [1·50-4·04]) were also associated with recurrent wheeze. LRTI and respiratory syncytial virus-LRTI were also associated with early transient and late onset clusters. Wheezing illness architecture differed between DCHS and ALSPAC; children included in ALSPAC in the early transient cluster wheezed for a longer period before remission and late-onset wheezing started at an older age, and no persistent phenotype was identified in DCHS. At 5 years, airway resistance was higher in children with early or recurrent wheeze compared with children who had never wheezed. Airway resistance increased from 6 weeks to 5 years among children with recurrent wheeze.

INTERPRETATION

Effective strategies to reduce maternal smoking and psychosocial stressors and new preventive interventions for respiratory syncytial virus are urgently needed to optimise child health in LMICs.

FUNDING

UK Medical Research Council; The Bill & Melinda Gates Foundation; National Institutes of Health Human Heredity and Health in Africa; South African Medical Research Council; Wellcome Trust.

How AI is advancing asthma management? Insights into economic and clinical aspects

Asthma, an increasingly prevalent chronic respiratory condition, incurs significant economic costs worldwide. Artificial Intelligence (AI), particularly Machine Learning (ML), has been widely recognized as transformative when applied to asthma care. This commentary investigates how AI and ML may improve clinical outcomes while alleviating some of the costs associated with asthma care. AI's powerful analytical abilities could usher in an unprecedented era of preventive measures, particularly by identifying at-risk populations and anticipating environmental triggers. ML shows promise for enhancing real-time monitoring, early detection, and tailored treatment strategies in paediatric asthma, potentially reducing hospitalizations and emergency care costs. Emerging AI-powered wearable technologies are catalysing a revolutionary shift in patient monitoring, providing proactive interventions. Although optimistic, this commentary highlights a gap in empirical studies evaluating the cost-effectiveness of AI in asthma care and stresses the need for larger datasets to accurately represent the economic benefits of AI solutions. Additionally, this paper emphasizes the ethical considerations surrounding data privacy and algorithmic bias, which are vital for the successful and equitable integration of AI into healthcare settings. This editorial underscores the urgent necessity of conducting thorough analyses to assess all economic implications, facilitate optimized resource allocation, and foster a nuanced understanding of AI/ML technologies in asthma management that may reduce costs to healthcare systems.

Association of Depression, Poor Mental Health Status and Asthma Control Patterns in US Adults Using a Data-Reductive Latent Class Method

Objectives To explore the association between depression, poor mental health status, and asthma control patterns among US adults using a latent class analysis (LCA) approach. Methods We used data from 10,337 adults aged 18 years and above from the 2016 Behavioral Risk Factor Surveillance System (BRFSS) Asthma Call-back Survey. Data-reductive LCA was used to derive asthma control patterns in the population using class variables indicative of asthma control. Besides univariate analysis, adjusted and unadjusted logistic regression models were used to examine the association of depression and poor mental health on the derived asthma control patterns.  Results About 27.8% of adults aged <55 reported depression, while 27.3% aged ≥55 years were depressed. The latent class prevalence of asthma control patterns was 42.8%, 31.1%, and 26.1%, corresponding to good, fair, and poor asthma control patterns, respectively. In adults aged <55 years, odds of depression (OR=1.52, 95% CI=1.27-1.82) and poor mental health (OR=1.58, 95% CI=1.27-1.96) were higher in the poor asthma control group compared to the good asthma control group. Odds for depression (OR=1.28, 95% CI=1.06-1.53) were also higher in the moderate asthma control group compared to the good asthma control group. Among those aged ≥55 years, depression odds (OR=1.57, 95% CI=1.31-1.87) were higher in only the poor asthma control group. Conclusions These findings may have public health implications. Detecting, screening, and treating depression and mental health disorders may help improve asthma control in people with asthma.

Enigma variations: The multi-faceted problems of pre-school wheeze

Numerous publications on wheezing disorders in children younger than 6 years have appeared in the medical literature over the last decades with the aim of shedding light on the mechanistic pathways (endotypes) and treatment. Nevertheless, there is yet no consensus as to the appropriate way to manage preschool wheeze mainly because of the lack of a clear definition of "preschool asthma" and the paucity of scientific evidence concerning its underlying endotypes. A symptom-based approach is inadequate since the human airway can respond to external stimuli with a limited range of symptoms and signs, including cough and wheeze, and these manifestations represent the final expression of many clinical entities with potentially different pathophysiologies requiring different individualized treatments. Hence, new studies challenge the symptom-based approach and promote the importance of managing the wheezy child based on the "airway phenotype." This will enable the clinician to identify not only the child with a serious underlying pathology (e.g., a structural airway disorder or immunodeficiency) who is in need of prompt and specific treatment but also increase the specificity of treatment for the child with symptoms suggestive of an "asthma" syndrome. In the latter case, focus should be given to the identification of treatable traits. This review summarizes the current understanding in management of preschool wheezing and highlights the unmet need for further research.

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.

Uncovering PFAS and Other Xenobiotics in the Dark Metabolome Using Ion Mobility Spectrometry, Mass Defect Analysis, and Machine Learning

The identification of xenobiotics in nontargeted metabolomic analyses is a vital step in understanding human exposure. Xenobiotic metabolism, transformation, excretion, and coexistence with other endogenous molecules, however, greatly complicate the interpretation of features detected in nontargeted studies. While mass spectrometry (MS)-based platforms are commonly used in metabolomic measurements, deconvoluting endogenous metabolites from xenobiotics is also often challenged by the lack of xenobiotic parent and metabolite standards as well as the numerous isomers possible for each small molecule m/z feature. Here, we evaluate a xenobiotic structural annotation workflow using ion mobility spectrometry coupled with MS (IMS-MS), mass defect filtering, and machine learning to uncover potential xenobiotic classes and species in large metabolomic feature lists. Xenobiotic classes examined included those of known high toxicities, including per- and polyfluoroalkyl substances (PFAS), polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs), and pesticides. Specifically, when the workflow was applied to identify PFAS in the NIST SRM 1957 and 909c human serum samples, it greatly reduced the hundreds of detected liquid chromatography (LC)-IMS-MS features by utilizing both mass defect filtering and m/z versus IMS collision cross sections relationships. These potential PFAS features were then compared to the EPA CompTox entries, and while some matched within specific m/z tolerances, there were still many unknowns illustrating the importance of nontargeted studies for detecting new molecules with known chemical characteristics. Additionally, this workflow can also be utilized to evaluate other xenobiotics and enable more confident annotations from nontargeted studies.

Associations between dental caries and ponderal growth in children: A Cambodian study

Background

The evidence around the relationship between Early Childhood Caries (ECC) and undernutrition is sparse and mostly reported from cross-sectional data sets. This paper aimed to test the relationship between ECC and linear and ponderal growth trajectories.

Methods

This project involves secondary data analysis from the Cambodia Longitudinal Health and Nutrition Study. The analytical sample included a 2y-cohort of 894 children who were younger than 2 years of age at the time of first height and weight measurement. Statistical analysis used both logistic regression modelling and Latent Class Analysis to examine the effect of exposure to dental caries in the first 1000 days on weight for height Z-score (WHZ) and height for age Z-score (HAZ) trajectory class groups. The presence of any cavity and pulp involvement were examined using multinomial regression adjusting for gender, socioeconomic status, maternal age and education.

Findings

Within each class groupings (HAZ and WHZ groupings), there was a trend whereby those with one or more cavities had lower Z-scores across the three follow-up time points of observation. There was an association between exposure to caries and WHZ class membership whereby children with caries exposure were more likely belong to WHZ class groups with lower Z-scores over time.

Conclusions

The study offers evidence that ECC is correlated with less favourable ponderal growth categorized by WHZ trajectory class groups.

Modeling Wheezing Spells Identifies Phenotypes with Different Outcomes and Genetic Associates

Rationale: Longitudinal modeling of current wheezing identified similar phenotypes, but their characteristics often differ between studies. Objectives: We propose that a more comprehensive description of wheeze may better describe trajectories than binary information on the presence/absence of wheezing. Methods: We derived six multidimensional variables of wheezing spells from birth to adolescence (including duration, temporal sequencing, and the extent of persistence/recurrence). We applied partition-around-medoids clustering on these variables to derive phenotypes in five birth cohorts. We investigated within- and between-phenotype differences compared with binary latent class analysis models and ascertained associations of these phenotypes with asthma and lung function and with polymorphisms in asthma loci 17q12-21 and CDHR3 (cadherin-related family member 3). Measurements and Main Results: Analysis among 7,719 participants with complete data identified five spell-based wheeze phenotypes with a high degree of certainty: never (54.1%), early-transient (ETW) (23.7%), late-onset (LOW) (6.9%), persistent (PEW) (8.3%), and a novel phenotype, intermittent wheeze (INT) (6.9%). FEV₁/FVC was lower in PEW and INT compared with ETW and LOW and declined from age 8 years to adulthood in INT. 17q12-21 and CDHR3 polymorphisms were associated with higher odds of PEW and INT, but not ETW or LOW. Latent class analysis- and spell-based phenotypes appeared similar, but within-phenotype individual trajectories and phenotype allocation differed substantially. The spell-based approach was much more robust in dealing with missing data, and the derived clusters were more stable and internally homogeneous. Conclusions: Modeling of spell variables identified a novel intermittent wheeze phenotype associated with lung function decline to early adulthood. Using multidimensional spell variables may better capture wheeze development and provide a more robust input for phenotype derivation.

Asthma and Allergy: Unravelling a Tangled Relationship with a Focus on New Biomarkers and Treatment

Asthma is a major driver of health care costs across ages. Despite widely disseminated asthma-treatment guidelines and a growing variety of effective therapeutic options, most patients still experience symptoms and/or refractoriness to standard of care treatments. As a result, most patients undergo a further intensification of therapy to optimize symptom control with a subsequent increased risk of side effects. Raising awareness about the relevance of evaluating aeroallergen sensitizations in asthmatic patients is a key step in better informing clinical practice while new molecular tools, such as the component resolved diagnosis, may be of help in refining the relationship between sensitization and therapeutic recommendations. In addition, patient care should benefit from reliable, easy-to-measure and clinically accessible biomarkers that are able to predict outcome and disease monitoring. To attain a personalized asthma management and to guide adequate treatment decisions, it is of paramount importance to expand clinicians' knowledge about the tangled relationship between asthma and allergy from a molecular perspective. Our review explores the relevance of allergen testing along the asthma patient's journey, with a special focus on recurrent wheezing children. Here, we also discuss the unresolved issues regarding currently available biomarkers and summarize the evidence supporting the eosinophil-derived neurotoxin as promising biomarker.

Establishing subclasses of childhood eczema, their risk factors and prognosis

Background

The heterogeneity of development and progression of eczema suggests multiple underlying subclasses for which aetiology and prognosis may vary. A better understanding may provide a comprehensive overview of eczema development and progression in childhood. Thus, we aimed to determine longitudinal eczema subclasses based on assessments and identify their associations with risk factors and allergic outcomes.

Methods

A total of 619 participants with a family history of allergic disease were assessed at 24 time‐points from birth to 12 years. At each time, eczema was defined as the report of current rash treated with topical steroid‐based preparations. Longitudinal latent class analysis was used to determine eczema subclasses. Subsequent analyses using regression models assessed the associations between eczema subclasses and potential risk factors and allergic outcomes at 18‐ and 25‐year follow‐ups (eczema, allergic rhinitis, asthma and allergic sensitization).

Results

We identified five eczema subclasses ‘early‐onset persistent’, ‘early‐onset resolving’, ‘mid‐onset persistent’, ‘mid‐onset resolving’ and ‘minimal eczema’. Filaggrin null mutations were associated with the early‐onset persistent (OR = 2.58 [1.09–6.08]) and mid‐onset persistent class (OR = 2.58 [1.32–5.06]). Compared with ‘minimal eczema’, participants from early‐onset persistent class had higher odds of eczema (OR = 11.8 [5.20–26.6]) and allergic rhinitis (OR = 3.13 [1.43–6.85]) at 18 and at 25 years eczema (OR = 9.37 [3.17–27.65]), allergic rhinitis (OR = 3.26 [1.07–9.93]) and asthma (OR = 2.91 [1.14–7.43]). Likewise, mid‐onset persistent class had higher odds of eczema (OR = 2.59 [1.31–5.14]), allergic rhinitis (OR = 1.70 [1.00–2.89]) and asthma (OR = 2.00 [1.10–3.63]) at 18 and at 25 years eczema (OR = 6.75 [3.11–14–65]), allergic rhinitis (OR = 2.74 [1.28–5.88]) and asthma (OR = 2.50 [1.25–5.00]). Allergic and food sensitization in early life was more common in those in the persistent eczema subclasses.

Conclusion

We identified five distinct eczema subclasses. These classes were differentially associated with risk factors, suggesting differences in aetiology, and also with the development of allergic outcomes, highlighting their potential to identify high‐risk groups for close monitoring and intervention.

Predicting ICU Mortality in Rheumatic Heart Disease: Comparison of XGBoost and Logistic Regression

Background

Rheumatic heart disease (RHD) accounts for a large proportion of Intensive Care Unit (ICU) deaths. Early prediction of RHD can help with timely and appropriate treatment to improve survival outcomes, and the XGBoost machine learning technology can be used to identify predictive factors; however, its use has been limited in the past. We compared the performance of logistic regression and XGBoost in predicting hospital mortality among patients with RHD from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database.

Methods

The patients with RHD in the MIMIC-IV database were divided into two groups retrospectively according to the availability of data and its clinical significance based on whether they survived or died. Backward stepwise regression was used to analyze the independent factors influencing patients with RHD, and to compare the differences between the two groups. The XGBoost algorithm and logistic regression were used to establish two prediction models, and the areas under the receiver operating characteristic curves (AUCs) and decision-curve analysis (DCA) were used to test and compare the models. Finally, DCA and the clinical impact curve (CIC) were used to validate the model.

Results

Data on 1,634 patients with RHD were analyzed, comprising 207 who died during hospitalization and 1,427 survived. According to estimated results for the two models using AUCs [0.838 (95% confidence interval = 0.786–0.891) and 0.815 (95% confidence interval = 0.765–0.865)] and DCA, the logistic regression model performed better. DCA and CIC verified that the logistic regression model had convincing predictive value.

Conclusions

We used logistic regression analysis to establish a more meaningful prediction model for the final outcome of patients with RHD. This model might be clinically useful for patients with RHD and help clinicians to provide detailed treatments and precise management.

Wheeze trajectories: Determinants and outcomes in the CHILD Cohort Study

BACKGROUND

Wheezing in early life is associated with asthma in adulthood; however, the determinants of wheezing trajectories and their associations with asthma and lung function in childhood remain poorly understood.

OBJECTIVE

In the CHILD Cohort Study, we aimed to identify wheezing trajectories and examine the associations between these trajectories, risk factors, and clinical outcomes at age 5 years.

METHODS

Wheeze data were collected at 8 time points from 3 months to 5 years of age. We used group-based trajectory models to derive wheeze trajectories among 3154 children. Associations with risk factors and clinical outcomes were analyzed by weighted regression models.

RESULTS

We identified 4 trajectories: a never/infrequent trajectory, transient wheeze, intermediate-onset (preschool) wheeze, and persistent wheeze. Higher body mass index was a common risk factor for all wheeze trajectories compared with that in the never/infrequent group. The unique predictors for specific wheeze trajectories included male sex, lower respiratory tract infections, and day care attendance for transient wheeze; paternal history of asthma, atopic sensitization, and child genetic risk score of asthma for intermediate wheeze; and maternal asthma for persistent wheeze. Blood eosinophil counts were higher in children with the intermediate wheeze trajectory than in those children with the other trajectories at the ages of 1 and 5 years. All wheeze trajectories were associated with decreased lung function and increased risk of asthma at age 5 years.

CONCLUSIONS

We identified 4 distinct trajectories in children from 3 months to 5 years of age, reflecting different phenotypes of early childhood wheeze. These trajectories were characterized by different biologic and physiologic traits and risk factors.

Childhood CCL18, CXCL10 and CXCL11 levels differentially relate to and predict allergy development

BACKGROUND

Chemokines are important mediators in immune cell recruitment, contributing to allergy development. However, extensive studies of chemokines in the circulation in relation to the presence and development of allergic diseases remain scarce. Our aim was to investigate associations of circulating allergy-related chemokines with the development of asthma and sensitization cross-sectionally and longitudinally in a population-based cohort.

METHODS

The chemokines CCL17, CCL22, CXCL10, CXCL11 and CCL18 were measured in plasma samples from children in the Manchester Asthma and Allergy Study. Samples were available from cord blood at birth (n = 376), age 1 (n = 195) and age 8 (n = 334). Cross-sectional and longitudinal association analyses were performed in relation to asthma and allergic sensitization, as well as allergic phenotype clusters previously derived using machine learning in the same study population.

RESULTS

In children with asthma and/or allergic sensitization, CCL18 levels were consistently elevated at 1 and/or 8 years of ages. In a longitudinal model including information on asthma from 4 time points (5, 8, 11 and 16 years of ages), we observed a significant association between increasing CCL18 levels at age 1 and a higher risk of asthma from early school age to adolescence (OR = 2.9, 95% CI 1.1-7.6, p = .028). We observed similar associations in longitudinal models for allergic sensitization. Asthma later in life was preceded by increased CXCL10 levels after birth and decreased CXCL11 levels at birth.

CONCLUSION

Elevated CCL18 levels throughout childhood precede the development of asthma and allergic sensitization. The Th1-associated chemokines CXCL10 and CXCL11 also associated with the development of both outcomes, with differential temporal effects.

Seeing the Forest for the Trees: Evaluating Population Data in Allergy-Immunology

A population-level study is essential for understanding treatment effects, epidemiologic phenomena, and health care best practices. Evaluating large populations and associated data requires an analytic framework, which is commonly used by statisticians, epidemiologists, and data scientists. This document will serve to provide an overview of these commonly employed methods in allergy and immunology research. We will draw upon recent examples from the allergy-immunology literature to contextualize discrete principles of relevance to population-level analysis that include statistical features of a study population, elements of statistical inference, regression analysis, and an overview of machine learning practices. Our intent is to guide the reader through a practical description of this important quantitative discipline and facilitate greater understanding about data and result display in the medical literature.

Association between occupational exposure to irritant agents and a distinct asthma endotype in adults

AIM

The biological mechanisms of work-related asthma induced by irritants remain unclear. We investigated the associations between occupational exposure to irritants and respiratory endotypes previously identified among never asthmatics (NA) and current asthmatics (CA) integrating clinical characteristics and biomarkers related to oxidative stress and inflammation.

METHODS

We used cross-sectional data from 999 adults (mean 45 years old, 46% men) from the case-control and familial Epidemiological study on the Genetics and Environments of Asthma (EGEA) study. Five respiratory endotypes have been identified using a cluster-based approach: NA1 (n=463) asymptomatic, NA2 (n=169) with respiratory symptoms, CA1 (n=50) with active treated adult-onset asthma, poor lung function, high blood neutrophil counts and high fluorescent oxidation products level, CA2 (n=203) with mild middle-age asthma, rhinitis and low immunoglobulin E level, and CA3 (n=114) with inactive/mild untreated allergic childhood-onset asthma. Occupational exposure to irritants during the current or last held job was assessed by the updated occupational asthma-specific job-exposure matrix (levels of exposure: no/medium/high). Associations between irritants and each respiratory endotype (NA1 asymptomatic as reference) were studied using logistic regressions adjusted for age, sex and smoking status.

RESULTS

Prevalence of high occupational exposure to irritants was 7% in NA1, 6% in NA2, 16% in CA1, 7% in CA2 and 10% in CA3. High exposure to irritants was associated with CA1 (adjusted OR aOR, (95% CI) 2.7 (1.0 to 7.3)). Exposure to irritants was not significantly associated with other endotypes (aOR range: 0.8 to 1.5).

CONCLUSION

Occupational exposure to irritants was associated with a distinct respiratory endotype suggesting oxidative stress and neutrophilic inflammation as potential associated biological mechanisms.

Phenotypic Variations of Mild-to-Moderate Obstructive Pulmonary Diseases According to Airway Inflammation and Clinical Features

Purpose

Asthma and chronic obstructive pulmonary disease (COPD) are complex and heterogeneous inflammatory diseases. We sought to investigate distinct disease profiles based on clinical, cellular and molecular data from patients with mild-to-moderate obstructive pulmonary diseases.

Patients and Methods

Patients with mild-to-moderate allergic asthma (n=30) and COPD (n=30) were prospectively recruited. Clinical characteristics and induced sputum were collected. In total, 35 mediators were assessed in induced sputum. Logistic regression analysis was conducted to identify the optimal factors that were able to discriminate between asthma and COPD. Further, the data were explored using hierarchical clustering in order to discover and compare clusters of combined samples of asthma and COPD patients. Clinical parameters, cellular composition, and sputum mediators of asthma and COPD were assessed between and within obtained clusters.

Results

We found five clinical and biochemical variables, namely IL-6, IL-8, CCL4, FEV₁/VC ratio pre-bronchodilator (%), and sputum neutrophils (%) that differentiated asthma and COPD and were suitable for discrimination purposes. A combination of those variables yielded high sensitivity and specificity in the differentiation between asthma and COPD, although only FEV₁/VC ratio pre-bronchodilator (%) proven significant in the combined model. In cluster analysis, two main clusters were identified: cluster 1, asthma predominant with evidence of eosinophilic airway inflammation and low level of Th1 and Th2 cytokines; and cluster 2, COPD predominant with elevated levels of Th1 and Th2 mediators.

Conclusion

The inflammatory profile of sputum samples from patients with stable mild-to-moderate asthma and COPD is not disease specific, varies within the disease and might be similar between these diseases. This study highlights the need for phenotyping the mild-to-moderate stages according to their clinical and molecular features.

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.

Recurrent Severe Preschool Wheeze: From Pre-Specified Diagnostic Labels to Underlying Endotypes

Rationale: Preschool wheezing is heterogeneous, but the underlying mechanisms are poorly understood.

Objectives: To investigate lower airway inflammation and infection in preschool children with different clinical diagnoses undergoing elective bronchoscopy and BAL.

Methods: We recruited 136 children aged 1–5 years (105 with recurrent severe wheeze [RSW]; 31 with nonwheezing respiratory disease [NWRD]). Children with RSW were assigned as having episodic viral wheeze (EVW) or multiple-trigger wheeze (MTW). We compared lower airway inflammation and infection in different clinical diagnoses and undertook data-driven analyses to determine clusters of pathophysiological features, and we investigated their relationships with prespecified diagnostic labels.

Measurements and Main Results: Blood eosinophil counts and percentages and allergic sensitization were significantly higher in children with RSW than in children with a NWRD. Blood neutrophil counts and percentages, BAL eosinophil and neutrophil percentages, and positive bacterial culture and virus detection rates were similar between groups. However, pathogen distribution differed significantly, with higher detection of rhinovirus in children with RSW and higher detection of Moraxella in sensitized children with RSW. Children with EVW and children with MTW did not differ in terms of blood or BAL-sample inflammation, or bacteria or virus detection. The Partition around Medoids algorithm revealed four clusters of pathophysiological features: 1) atopic (17.9%), 2) nonatopic with a low infection rate and high use of inhaled corticosteroids (31.3%), 3) nonatopic with a high infection rate (23.1%), and 4) nonatopic with a low infection rate and no use of inhaled corticosteroids (27.6%). Cluster allocation differed significantly between the RSW and NWRD groups (RSW was evenly distributed across clusters, and 60% of the NWRD group was assigned to cluster 4; P < 0.001). There was no difference in cluster membership between the EVW and MTW groups. Cluster 1 was dominated by Moraxella detection (P = 0.04), and cluster 3 was dominated by Haemophilus or Staphylococcus or Streptococcus detection (P = 0.02).

Conclusions: We identified four clusters of severe preschool wheeze, which were distinguished by using sensitization, peripheral eosinophilia, lower airway neutrophilia, and bacteriology.

A systematic review of asthma case definitions in 67 birth cohort studies

BACKGROUND

Birth cohort studies are a valuable source of information about potential risk factors for childhood asthma. To better understand similarities and variations in findings between birth cohort studies, the methodologies used to measure asthma require consideration.

OBJECTIVE

To review and appraise the definitions of "asthma" used in birth cohort studies.

METHODS

A literature search, conducted in December 2017 in the MEDLINE database and birth cohort repositories, identified 1721 citations published since 1990. Information extracted included: study name, year of publication, sample size, sample age, prevalence of asthma (%), study region, source of information about asthma, measured outcome, and asthma case definition. A meta-analysis evaluated whether asthma prevalence in cohorts from Europe and North America varied by the studies' definition of asthma and by their data sources.

RESULTS

The final review included 67 birth cohorts, of which 48 (72%) were from Europe, 14 (21%) from North America, 3 (5%) from Oceania, 1 (1%) from Asia and 1 (1%) from South America. We identified three measured outcomes: "asthma ever", "current asthma", and "asthma" without further specification. Definitions of "asthma ever" were primarily based upon an affirmative parental response to the question whether the child had ever been diagnosed with asthma by a physician. The most frequently used definition of "current asthma" was "asthma ever" and either asthma symptoms or asthma medications in the last 12 months. This definition of "current asthma" was used in 16 cohorts. There was no statistically significant difference in the pooled asthma prevalence in European and North American cohorts that used questionnaire alone versus other data sources to classify asthma.

CONCLUSION

There is substantial heterogeneity in childhood asthma definitions in birth cohort studies. Standardisation of asthma case definitions will improve the comparability and utility of future cohort studies and enable meta-analyses.

Novel Machine Learning Can Predict Acute Asthma Exacerbation

BACKGROUND

Asthma exacerbations result in significant health and economic burden, but are difficult to predict.

RESEARCH QUESTION

Can machine learning (ML) models with large-scale outpatient data predict asthma exacerbations?

STUDY DESIGN AND METHODS

We analyzed data extracted from electronic health records (EHRs) of asthma patients treated at the Cleveland Clinic from 2010 through 2018. Demographic information, comorbidities, laboratory values, and asthma medications were included as covariates. Three different models were built with logistic regression, random forests, and a gradient boosting decision tree to predict: (1) nonsevere asthma exacerbation requiring oral glucocorticoid burst, (2) ED visits, and (3) hospitalizations.

RESULTS

Of 60,302 patients, 19,772 (32.8%) had at least one nonsevere exacerbation requiring oral glucocorticoid burst, 1,748 (2.9%) requiring and ED visit and 902 (1.5%) requiring hospitalization. Nonsevere exacerbation, ED visit, and hospitalization were predicted best by light gradient boosting machine, an algorithm used in ML to fit predictive analytic models, and had an area under the receiver operating characteristic curve of 0.71 (95% CI, 0.70-0.72), 0.88 (95% CI, 0.86-0.89), and 0.85 (95% CI, 0.82-0.88), respectively. Risk factors for all three outcomes included age, long-acting β agonist, high-dose inhaled glucocorticoid, or chronic oral glucocorticoid therapy. In subgroup analysis of 9,448 patients with spirometry data, low FEV₁ and FEV₁ to FVC ratio were identified as top risk factors for asthma exacerbation, ED visits, and hospitalization. However, adding pulmonary function tests did not improve models' prediction performance.

INTERPRETATION

Models built with an ML algorithm from real-world outpatient EHR data accurately predicted asthma exacerbation and can be incorporated into clinical decision tools to enhance outpatient care and to prevent adverse outcomes.

Artificial Intelligence and Machine Learning in Chronic Airway Diseases: Focus on Asthma and Chronic Obstructive Pulmonary Disease

Chronic airway diseases are characterized by airway inflammation, obstruction, and remodeling and show high prevalence, especially in developing countries. Among them, asthma and chronic obstructive pulmonary disease (COPD) show the highest morbidity and socioeconomic burden worldwide. Although there are extensive guidelines for the prevention, early diagnosis, and rational treatment of these lifelong diseases, their value in precision medicine is very limited. Artificial intelligence (AI) and machine learning (ML) techniques have emerged as effective methods for mining and integrating large-scale, heterogeneous medical data for clinical practice, and several AI and ML methods have recently been applied to asthma and COPD. However, very few methods have significantly contributed to clinical practice. Here, we review four aspects of AI and ML implementation in asthma and COPD to summarize existing knowledge and indicate future steps required for the safe and effective application of AI and ML tools by clinicians.

Trajectories of early-onset rhinitis in the Singapore GUSTO mother-offspring cohort

BACKGROUND

The natural history of childhood rhinitis is not well described.

OBJECTIVE

This study aimed to identify different rhinitis trajectories in early childhood and their predictors and allergic associations.

METHODS

Rhinitis symptoms were ascertained prospectively from birth until 6 years using standardized questionnaires in 772 participants. Rhinitis was defined as one or more episodes of sneezing, runny and/or blocked nose >2 weeks duration. Latent trajectories were identified using group-based modelling, and their predictive risk factors and allergic associations were examined.

RESULTS

Three rhinitis trajectory groups were identified: 7.6% (n = 59) were termed early transient rhinitis, 8.6% (n = 66) late transient rhinitis, and 6.6% (n = 51) persistent rhinitis. The remaining 77.2% (n = 596) were classified as non-rhinitis/reference group. Early transient rhinitis subjects were more likely of Indian ethnicity, had siblings, reported childcare attendance, early wheezing and eczema in the first 3 years of life. Late transient rhinitis was associated with antenatal exposure to smoking, higher maternal education levels, and wheezing at age 36-72 months. Persistent rhinitis was associated with male gender, paternal and maternal history of atopy, eczema, and house dust mite sensitization.

CONCLUSIONS & CLINICAL RELEVANCE

Risk factors for early transient rhinitis involve a combination of genetic and early environmental exposures, whereas late transient rhinitis may relate to maternal factors and early respiratory infections independent of atopy. In contrast, persistent rhinitis is strongly associated with atopic risk and likely represents the typical trajectory associated with allergic disorders. Allergic rhinitis symptoms may commence as early as the first year of life and may inform development of early interventive strategies.

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.

The role of artificial intelligence in identifying asthma in pediatric inpatient setting

Background

The incidence of asthma in Chinese children has rapidly increased as a result of inadequate management. This is mainly due to the failure of many primary-level pediatricians to distinguish asthma from common respiratory diseases, such as bronchitis and pneumonia. Such misdiagnoses often lead to the abuse of antibiotics and systemic glucocorticoids. Additionally, if asthma is not diagnosed early, chronic airway inflammation results in lesions that not only hamper children’s athletic abilities, but serve as the primary cause for adult chronic airway diseases, such as chronic obstructive pulmonary disease (COPD).

Methods

A number of machine learning–based models including CatBoost, Logistic Regression, Naïve Bayes, and Support Vector Machines (SVM) have been developed to identify asthma via utilizing retrospective electronic medical records (EMRs) of patients. These models were evaluated independently using EMRs from both the Pulmonology Department and other departments of the Children’s Hospital, Zhejiang University School of Medicine, China.

Results

Two independent test sets were applied for performance evaluation. TestSet-1 consisted of 325 positive asthma cases and 428 negative cases from the Pulmonology Department. TestSet-2 was composed of 2,123 cases from non-pulmonology departments, and included 337 positive and 1,786 negative cases. Experimental results showed that the CatBoost model outperformed other models on both test sets with an accuracy of 84.7% and an area under the curve (AUC) of 90.9% on TestSet-1, and an accuracy of 96.7% and an AUC of 98.1% on TestSet-2.

Conclusions

The artificial intelligence (AI) model could rapidly and accurately identify asthma in general medical wards of children, and may aid primary pediatricians in the correct diagnoses of asthma. It possesses great clinical value and practical significance in improving the control rate of asthma in children, optimizing medical resources, and limiting the abuse of antibiotics and systemic glucocorticoids.

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.

Prediction of breakthrough pain during labour neuraxial analgesia: comparison of machine learning and multivariable regression approaches

INTRODUCTION

Risk-prediction models for breakthrough pain facilitate interventions to forestall inadequate labour analgesia, but limited work has used machine learning to identify predictive factors. We compared the performance of machine learning and regression techniques in identifying parturients at increased risk of breakthrough pain during labour epidural analgesia.

METHODS

A single-centre retrospective study involved parturients receiving patient-controlled epidural analgesia. The primary outcome was breakthrough pain. We randomly selected 80% of the cohort (training cohort) to develop three prediction models using random forest, XGBoost, and logistic regression, followed by validation against the remaining 20% of the cohort (validation cohort). Area-under-the-receiver operating characteristic curve (AUC), sensitivity, specificity, and positive and negative predictive values (PPV and NPV) were used to assess model performance.

RESULTS

Data from 20 716 parturients were analysed. The incidence of breakthrough pain was 14.2%. Of 31 candidate variables, random forest, XGBoost and logistic regression models included 30, 23, and 15 variables, respectively. Unintended venous puncture, post-neuraxial analgesia highest pain score, number of dinoprostone suppositories, neuraxial technique, number of neuraxial attempts, depth to epidural space, body mass index, pre-neuraxial analgesia oxytocin infusion rate, maternal age, pre-neuraxial analgesia cervical dilation, anaesthesiologist rank, and multiparity, were identified in all three models. All three models performed similarly, with AUC 0.763-0.772, sensitivity 67.0-69.4%, specificity 70.9-76.2%, PPV 28.3-31.8%, and NPV 93.3-93.5%.

CONCLUSIONS

Machine learning did not improve the prediction of breakthrough pain compared with multivariable regression. Larger population-wide studies are needed to improve predictive ability.

Research Techniques Made Simple: Latent Class Analysis

Latent class analysis (LCA) is a statistical technique that allows for identification, in a population characterized by a set of predefined features, of hidden clusters or classes, that is, subgroups that have a given probability of occurrence and are characterized by a specific and predictable combination of the analyzed features. Compared with other methods of so called data segmentation, such as hierarchical clustering, LCA derives clusters using a formal probabilistic approach and can be used in conjunction with multivariate methods to estimate parameters. The optimal number of classes is the one that minimizes the degree of relationship among cases belonging to different classes, and it is decided by relying on methods such as the Bayesian Information Criterion that capitalize on the value of the negative log-likelihood function, a well-established measure of the goodness of fit of a statistical model. LCA has not been extensively used in dermatology. The areas of application are manifold, from the phenotype classification to the analysis of behavior in relation with risk factors to the performance of diagnostic tests.

Distinguishing Wheezing Phenotypes from Infancy to Adolescence. A Pooled Analysis of Five Birth Cohorts

Rationale: Pooling data from multiple cohorts and extending the time frame across childhood should minimize study-specific effects, enabling better characterization of childhood wheezing.

Objectives: To analyze wheezing patterns from early childhood to adolescence using combined data from five birth cohorts.

Methods: We used latent class analysis to derive wheeze phenotypes among 7,719 participants from five birth cohorts with complete report of wheeze at five time periods. We tested the associations of derived phenotypes with late asthma outcomes and lung function, and investigated the uncertainty in phenotype assignment.

Results: We identified five phenotypes: never/infrequent wheeze (52.1%), early onset preschool remitting (23.9%), early onset midchildhood remitting (9%), persistent (7.9%), and late-onset wheeze (7.1%). Compared with the never/infrequent wheeze, all phenotypes had higher odds of asthma and lower forced expiratory volume in 1 second and forced expiratory volume in 1 second/forced vital capacity in adolescence. The association with asthma was strongest for persistent wheeze (adjusted odds ratio, 56.54; 95% confidence interval, 43.75–73.06). We observed considerable within-class heterogeneity at the individual level, with 913 (12%) children having low membership probability (<0.60) of any phenotype. Class membership certainty was highest in persistent and never/infrequent, and lowest in late-onset wheeze (with 51% of participants having membership probabilities <0.80). Individual wheezing patterns were particularly heterogeneous in late-onset wheeze, whereas many children assigned to early onset preschool remitting class reported wheezing at later time points.

Conclusions: All wheeze phenotypes had significantly diminished lung function in school-age children, suggesting that the notion that early life episodic wheeze has a benign prognosis may not be true for a proportion of transient wheezers. We observed considerable within-phenotype heterogeneity in individual wheezing patterns.

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.

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.

Childhood Asthma: Advances Using Machine Learning and Mechanistic Studies

A paradigm shift brought by the recognition that childhood asthma is an aggregated diagnosis that comprises several different endotypes underpinned by different pathophysiology, coupled with advances in understanding potentially important causal mechanisms, offers a real opportunity for a step change to reduce the burden of the disease on individual children, families, and society. Data-driven methodologies facilitate the discovery of "hidden" structures within "big healthcare data" to help generate new hypotheses. These findings can be translated into clinical practice by linking discovered "phenotypes" to specific mechanisms and clinical presentations. Epidemiological studies have provided important clues about mechanistic avenues that should be pursued to identify interventions to prevent the development or alter the natural history of asthma-related diseases. Findings from cohort studies followed by mechanistic studies in humans and in neonatal mouse models provided evidence that environments such as traditional farming may offer protection by modulating innate immune responses and that impaired innate immunity may increase susceptibility. The key question of which component of these exposures can be translated into interventions requires confirmation. Increasing mechanistic evidence is demonstrating that shaping the microbiome in early life may modulate immune function to confer protection. Iterative dialogue and continuous interaction between experts with different but complementary skill sets, including data scientists who generate information about the hidden structures within "big data" assets, and medical professionals, epidemiologists, basic scientists, and geneticists who provide critical clinical and mechanistic insights about the mechanisms underpinning the architecture of the heterogeneity, are keys to delivering mechanism-based stratified treatments and prevention.