Dataset Information

Machine Learning Prediction of Treatment Response to Inhaled Corticosteroids in Asthma.

ABSTRACT:

Background

Although inhaled corticosteroids (ICS) are the first-line therapy for patients with persistent asthma, many patients continue to have exacerbations. We developed machine learning models to predict the ICS response in patients with asthma.

Methods

The subjects included asthma patients of European ancestry (n = 1371; 448 children; 916 adults). A genome-wide association study was performed to identify the SNPs associated with ICS response. Using the SNPs identified, two machine learning models were developed to predict ICS response: (1) least absolute shrinkage and selection operator (LASSO) regression and (2) random forest.

Results

The LASSO regression model achieved an AUC of 0.71 (95% CI 0.67-0.76; sensitivity: 0.57; specificity: 0.75) in an independent test cohort, and the random forest model achieved an AUC of 0.74 (95% CI 0.70-0.78; sensitivity: 0.70; specificity: 0.68). The genes contributing to the prediction of ICS response included those associated with ICS responses in asthma (TPSAB1, FBXL16), asthma symptoms and severity (ABCA7, CNN2, PTRN3, and BSG/CD147), airway remodeling (ELANE, FSTL3), mucin production (GAL3ST), leukotriene synthesis (GPX4), allergic asthma (ZFPM1, SBNO2), and others.

Conclusions

An accurate risk prediction of ICS response can be obtained using machine learning methods, with the potential to inform personalized treatment decisions. Further studies are needed to examine if the integration of richer phenotype data could improve risk prediction.

SUBMITTER: Ong MS

PROVIDER: S-EPMC10970828 | biostudies-literature | 2024 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine Learning Prediction of Treatment Response to Inhaled Corticosteroids in Asthma.

Ong Mei-Sing MS Sordillo Joanne E JE Dahlin Amber A McGeachie Michael M Tantisira Kelan K Wang Alberta L AL Lasky-Su Jessica J Brilliant Murray M Kitchner Terrie T Roden Dan M DM Weiss Scott T ST Wu Ann Chen AC

Journal of personalized medicine 20240225 3

<h4>Background</h4>Although inhaled corticosteroids (ICS) are the first-line therapy for patients with persistent asthma, many patients continue to have exacerbations. We developed machine learning models to predict the ICS response in patients with asthma.<h4>Methods</h4>The subjects included asthma patients of European ancestry (<i>n</i> = 1371; 448 children; 916 adults). A genome-wide association study was performed to identify the SNPs associated with ICS response. Using the SNPs identified, ...[more]

PMID: 38540988

Similar Datasets

Project description:BackgroundPrevious studies have demonstrated that glucocorticoid-induced transcript 1 gene (GLCCI1) rs37973 mutant genotype is associated with poor inhaled corticosteroid (ICS) response in asthmatics. As human airway relaxation is regulated by circulation epinephrine, which can be enhanced by corticosteroid. It is unknown whether or not GLCCI1 rs37973 is associated with circulation epinephrine and cortisol concentrations in asthma. The aim of this study is to evaluate these relationships.MethodsA total of 182 asthmatics and 180 healthy controls were recruited for the study. 30 mild-to-moderate asthmatics received fluticasone propionate (125 μg, bid) treatment for 12 weeks. GLCCI1 rs37973 genotyping was performed with the iPlex MassARRAY genotyping platform. The plasma concentrations of cortisol and epinephrine of each participant were detected by enzyme linked immunosorbent assay (ELISA) kits.ResultsGLCCI1 rs37973 homozygotes mutant genotype GG had a higher plasma epinephrine concentration (median concentration 27.032 pg/ml, nGG = 36; median concentration 23.149 pg/ml, nAA+AG = 146; P = 0.015) and cortisol concentration (median concentration 1.141 ng/ml, nGG = 36; median concentration 0.921 ng/ml, nAA+AG = 146; P = 0.013). Both epinephrine concentration and cortisol concentration in plasma were positively correlated with FEV1 (r = 0.889 and r = 0.821, respectively. nasthma = 182). For asthmatics treated with ICS, rs37973 was associated with change in plasma epinephrine and cortisol concentration in a recessive model (AA + AG vs GG), with GG had less improvement in epinephrine concentration [ΔEPIAA+AG = 6.843 (9.26) pg/ml, nAA+AG = 26; ΔEPIGG = -1.666 (6.52) pg/ml, nGG = 4; P = 0.018] and cortisol concentration [ΔCORAA+AG = 0.3040 (0.21) ng/ml, nAA+AG = 26; ΔCORGG = -0.066 (0.24) ng/ml, nGG = 4; P = 0.009].ConclusionsOur study suggested that the poor ICS response in GLCCI1 rs37973 mutant genotype might be related to the less increased amplitudes of plasma epinephrine and cortisol in asthmatic patients.Trial registrationChiCTR-RCC-13003634 www.chictr.org.cn. Active since September 27, 2013.

Project description:ObjectiveTo describe recovery of adrenal insufficiency in asthmatic children treated with inhaled corticosteroids (ICS) and cortisol replacement therapy.DesignRetrospective, observational study.PatientsA total of 113 patients, 74 male; age 10.4 (3.3-16.5) years; beclomethasone-equivalent ICS dose, 800 μg, (100-1,000), tested by low dose short Synacthen (tetracosactide) test (LDSST), were studied. Test results were classified by basal and peak cortisol concentration: "normal" (basal >100 nmol/L, peak >500 nmol/L), "suboptimal" (basal >100 nmol/L, peak 350-499 nmol/L), "abnormal" (basal <100 nmol/L and/or peak <350 nmol/L). Patients with suboptimal results received hydrocortisone during periods of stress only, and those with abnormal responses received daily hydrocortisone, increased during periods of stress. A total of 73 patients (68%) had ≥2 LDSSTs over 2.2 years (0.2-7.7).MeasurementsChange in cortisol response to repeat LDSST (movement between diagnostic groups, difference in basal and peak cortisol >15% [2× the inter-assay coefficient of variation]), change in BMI and height standard deviation score (SDS).ResultsBaseline test results were abnormal in 17 patients (15%) and all of them had repeat tests. In 13 patients (76%), test results improved (normal in six, suboptimal in seven) and four (24%) remained abnormal. Baseline tests results were suboptimal in 54 patients (48%), of whom 50 (93%) were retested. Repeat tests were normal in 36 patients (72%), remained suboptimal in 11 (22%), and were abnormal in three (6%). Baseline tests results were normal in 42 patients, of whom six patients (14%) were retested. Results remained normal in three (50%), were suboptimal in two (33%), and abnormal in one (17%). Basal and peak cortisol levels increased by >15% in 33/73 (45%) and 42/73 (57%) patients, respectively, and decreased by >15% in 14/73 (19%) and 7/73 (10%), respectively. There was no significant change in height or BMI SDS.ConclusionRecovery of adrenal function is common and occurs during continued ICS and cortisol replacement therapy.

Dataset Information

Machine Learning Prediction of Treatment Response to Inhaled Corticosteroids in Asthma.

Background

Methods

Results

Conclusions

Publications

Machine Learning Prediction of Treatment Response to Inhaled Corticosteroids in Asthma.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets