Shahbazi Khamas, Shahriyar and Brinkman, Paul and Neerincx, Anne H. and Vijverberg, Susanne J. H. and Hashimoto, Simone and Blankestijn, Jelle M. and Duitman, Jan Willem and Dekker, Tamara and Smids, Barbara S. and Terheggen-Lagro, Suzanne W. J. and Lutter, Rene and Metwally, Nariman K. A. and Sondaal, Fleur and Haarman, Eric G. and Sterk, Peter J. and Adcock, Ian M. and Auffray, Charles and Bang, Corinna and Bansal, Aruna T. and Buntrock-Döpke, Heike and Bonnelykke, Klaus and Bush, Andrew and Chawes, Bo Lund and Chung, Kian Fan and Corcuera-Elosegui, Paula and Dahlen, Sven-Erik and Djukanovic, Ratko and Fleming, Louise J. and Fowler, Stephen J. and Franke, Andre and Frey, Urs and Gorenjak, Mario and Brandstetter, Susanne and Harner, Susanne and Hedlin, Gunilla and Kabesch, Michael and Zounemat-Kermani, Nazanin and Kheiroddin, Parastoo and Kiefer, Alexander and Konradsen, Jon R. and Kraneveld, Aletta D. and Lopez-Fernandez, Leyre and Murray, Clare S. and Nordlund, Björn and Pino-Yanes, Maria and Potocnik, Uros and Roberts, Graham and Stokholm, Jakob and Sorensen, Soren Johannes and Sardon-Prado, Olaia and Shaw, Dominick E. and Singer, Florian and Sousa, Ana R. and Thorsen, Jonathan and Toncheva, Antoaneta A. and Vissing, Nadja H. and Wolff, Christine and Abdel-Aziz, Mahmoud I. and Maitland-van der Zee, Anke H. (2026) Complementary Predictors for Asthma Attack Prediction in Children: Salivary Microbiome, Serum Inflammatory Mediators, and Past Attack History. ALLERGY, 81 (2). pp. 413-426. ISSN 0105-4538, 1398-9995
Full text not available from this repository. (Request a copy)Abstract
Background Early identification of children at risk of asthma attacks is important for optimizing treatment strategies. We aimed to integrate salivary microbiome and serum inflammatory mediator profiles with asthma attacks history to develop a comprehensive predictive model for future attacks. Methods This study contained a discovery (SysPharmPediA) and a replication phase (U-BIOPRED). School-aged children with asthma were classified into at risk and no-risk groups, based on the presence or absence of one or more severe attacks during one-year follow-up. Prediction models were developed using random forest on the training set (70%) with data on past asthma attacks, microbiome composition, serum inflammatory mediator levels, and their combinations and then tested on the rest of the population (30%). Outcomes were replicated in a subset of children with severe asthma from U-BIOPRED. Results Complete data were available for 154 children (SysPharmPediA = 121, U-BIOPRED = 33). In discovery, the model based on past attacks resulted in an area under the receiving characteristic curve (AUROCC) similar to 0.7. Models including six salivary bacteria or six inflammatory mediators achieved similar results. The combined model incorporating seven features, past asthma attacks, Capnocytophaga, Corynebacterium, and Cardiobacterium, TIMP-4, VEGF, and MIP-3 beta achieved the highest accuracy with AUROCC similar to 0.87. The combined model in the U-BIOPRED limited to available inflammatory mediators (VEGF), and incorporating past asthma attacks, Capnocytophaga, Corynebacterium, and Cardiobacterium, resulted in an AUROCC of 0.84. Conclusion Serum inflammatory mediators and salivary microbiome complement asthma attacks history for predicting future attacks. These results highlight the imperative for continued investigation into oral microbiota and its interaction with the immune system.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | MATRIX METALLOPROTEINASES; EXACERBATIONS; DISEASE; 16S rRNA; asthma; biomarker; exacerbations; precision medicine; saliva |
| Subjects: | 600 Technology > 610 Medical sciences Medicine |
| Divisions: | Medicine > Lehrstuhl für Kinder- und Jugendmedizin |
| Depositing User: | Dr. Gernot Deinzer |
| Date Deposited: | 13 Aug 2026 08:55 |
| Last Modified: | 13 Aug 2026 08:55 |
| URI: | https://pred.uni-regensburg.de/id/eprint/67472 |
Actions (login required)
 |
View Item |
