Advanced machine learning for predicting individual risk of flares in rheumatoid arthritis patients tapering biologic drugs.
Flare prediction
Machine learning
Rheumatoid arthritis
Journal
Arthritis research & therapy
ISSN: 1478-6362
Titre abrégé: Arthritis Res Ther
Pays: England
ID NLM: 101154438
Informations de publication
Date de publication:
27 02 2021
27 02 2021
Historique:
received:
13
04
2020
accepted:
10
02
2021
entrez:
28
2
2021
pubmed:
1
3
2021
medline:
22
6
2021
Statut:
epublish
Résumé
Biological disease-modifying anti-rheumatic drugs (bDMARDs) can be tapered in some rheumatoid arthritis (RA) patients in sustained remission. The purpose of this study was to assess the feasibility of building a model to estimate the individual flare probability in RA patients tapering bDMARDs using machine learning methods. Longitudinal clinical data of RA patients on bDMARDs from a randomized controlled trial of treatment withdrawal (RETRO) were used to build a predictive model to estimate the probability of a flare. Four basic machine learning models were trained, and their predictions were additionally combined to train an ensemble learning method, a stacking meta-classifier model to predict the individual flare probability within 14 weeks after each visit. Prediction performance was estimated using nested cross-validation as the area under the receiver operating curve (AUROC). Predictor importance was estimated using the permutation importance approach. Data of 135 visits from 41 patients were included. A model selection approach based on nested cross-validation was implemented to find the most suitable modeling formalism for the flare prediction task as well as the optimal model hyper-parameters. Moreover, an approach based on stacking different classifiers was successfully applied to create a powerful and flexible prediction model with the final measured AUROC of 0.81 (95%CI 0.73-0.89). The percent dose change of bDMARDs, clinical disease activity (DAS-28 ESR), disease duration, and inflammatory markers were the most important predictors of a flare. Machine learning methods were deemed feasible to predict flares after tapering bDMARDs in RA patients in sustained remission.
Sections du résumé
BACKGROUND
Biological disease-modifying anti-rheumatic drugs (bDMARDs) can be tapered in some rheumatoid arthritis (RA) patients in sustained remission. The purpose of this study was to assess the feasibility of building a model to estimate the individual flare probability in RA patients tapering bDMARDs using machine learning methods.
METHODS
Longitudinal clinical data of RA patients on bDMARDs from a randomized controlled trial of treatment withdrawal (RETRO) were used to build a predictive model to estimate the probability of a flare. Four basic machine learning models were trained, and their predictions were additionally combined to train an ensemble learning method, a stacking meta-classifier model to predict the individual flare probability within 14 weeks after each visit. Prediction performance was estimated using nested cross-validation as the area under the receiver operating curve (AUROC). Predictor importance was estimated using the permutation importance approach.
RESULTS
Data of 135 visits from 41 patients were included. A model selection approach based on nested cross-validation was implemented to find the most suitable modeling formalism for the flare prediction task as well as the optimal model hyper-parameters. Moreover, an approach based on stacking different classifiers was successfully applied to create a powerful and flexible prediction model with the final measured AUROC of 0.81 (95%CI 0.73-0.89). The percent dose change of bDMARDs, clinical disease activity (DAS-28 ESR), disease duration, and inflammatory markers were the most important predictors of a flare.
CONCLUSION
Machine learning methods were deemed feasible to predict flares after tapering bDMARDs in RA patients in sustained remission.
Identifiants
pubmed: 33640008
doi: 10.1186/s13075-021-02439-5
pii: 10.1186/s13075-021-02439-5
pmc: PMC7913400
doi:
Substances chimiques
Antirheumatic Agents
0
Biological Products
0
Biomarkers
0
Types de publication
Journal Article
Randomized Controlled Trial
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
67Références
Rheumatology (Oxford). 2015 Aug;54(8):1408-14
pubmed: 25731769
J Glob Health. 2018 Dec;8(2):020303
pubmed: 30405904
J Rheumatol. 2013 Oct;40(10):1650-7
pubmed: 23950185
N Engl J Med. 2016 Sep 29;375(13):1216-9
pubmed: 27682033
Ann Rheum Dis. 2016 Aug;75(8):1428-37
pubmed: 27261493
Ann Rheum Dis. 2017 Jun;76(6):960-977
pubmed: 28264816
Ann Rheum Dis. 2015 Feb;74(2):389-95
pubmed: 24288014
Arthritis Rheum. 2009 Aug;60(8):2262-71
pubmed: 19644846
Ann Rheum Dis. 2016 Jan;75(1):45-51
pubmed: 25660991
BMC Med. 2019 Dec 16;17(1):230
pubmed: 31842878
Ann Rheum Dis. 2016 Jan;75(1):52-8
pubmed: 25873634
Ann Rheum Dis. 2015 Feb;74(2):381-8
pubmed: 24285493
Ann Rheum Dis. 2016 Sep;75(9):1637-44
pubmed: 26483255