Unsupervised Machine-Learning-Based Endotype Discovery Using Iterative Resampling in Dupilumab-Treated Patients

Moreno-Jiménez, Emma; Morgado, Natalia; García-Sánchez, Asunción; Triviño, Juan Carlos; Estravís, Miguel; Gómez-García, Manuel; Gil-Melcón, María; Lázaro-Sastre, Milagros; Sanz, Catalina; Isidoro-García, María; Dávila, Ignacio

doi:10.3390/ijms27125266

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Unsupervised Machine-Learning-Based Endotype Discovery Using Iterative Resampling in Dupilumab-Treated Patients

by

Emma Moreno-Jiménez

^1,2

,

Natalia Morgado

^1,2

,

Asunción García-Sánchez

^2,3,4

,

Juan Carlos Triviño

⁵

,

Miguel Estravís

^1,2,3

,

Manuel Gómez-García

^2,3,6,

María Gil-Melcón

⁷,

Milagros Lázaro-Sastre

^2,8,

Catalina Sanz

^1,2,3,*

,

María Isidoro-García

^2,3,6,9,† and

Ignacio Dávila

^2,3,4,8,†

¹

Departamento de Microbiología y Genetica, Universidad de Salamanca, 37007 Salamanca, Spain

²

Instituto de Investigación Biomédica de Salamanca, 37007 Salamanca, Spain

³

Instituto de Salud Carlos III, Red de Enfermedades Inflamatorias—RICORS, 28029 Madrid, Spain

⁴

Departamento de Biomedicina y Ciencias del Diagnostico, Universidad de Salamanca, 37007 Salamanca, Spain

⁵

Synlab Group, Sistemas Genómicos SL, 46980 Paterna, Spain

⁶

Servicio de Bioquímica Clínica, Hospital Universitario de Salamanca, 37007 Salamanca, Spain

⁷

Servicio de Otorhinolaringología y Cirugía de cabeza y cuello, Hospital Universitario de Salamanca, 37007 Salamanca, Spain

⁸

Servicio de Aergología Department, Hospital Universitario de Salamanca, 37007 Salamanca, Spain

⁹

Departamento de Medicina, Universidad de Salamanca, 37007 Salamanca, Spain

^*

Author to whom correspondence should be addressed.

^†

These authors shared senior authorship.

Int. J. Mol. Sci. 2026, 27(12), 5266; https://doi.org/10.3390/ijms27125266

Submission received: 25 May 2026 / Revised: 5 June 2026 / Accepted: 6 June 2026 / Published: 10 June 2026

(This article belongs to the Special Issue Molecular Crosstalk in Allergy, Barrier Dysfunction, and Asthma)

Download Versions Notes

Abstract

Asthma is a heterogeneous inflammatory disorder involving multiple immune pathways, frequently presenting alongside comorbidities such as chronic rhinosinusitis with nasal polyps (CRSwNP). Although biologic therapies such as dupilumab have shown clinical efficacy, the molecular mechanisms underlying variable treatment responses remain poorly understood. This study aimed to characterize transcriptomic patterns that distinguish asthmatic patients from healthy controls and to evaluate transcriptomic changes induced by dupilumab. Whole-blood RNA-seq was performed in 66 samples, 18 patients (G0) with severe asthma before and after 6 months of dupilumab treatment compared with 30 non-asthmatic controls. Differentially expressed genes (DEGs) were identified and validated by quantitative PCR (qPCR). Clinical responses were assessed using the FEV1, Exacerbations, Oral corticosteroids, Symptoms (FEOS) score and the Sino-Nasal Outcome Test-22 (SNOT-22). A total of 1124 DEGs were identified, distinguishing asthmatic patients from controls. Notably, ABCC1, CYP4F12, FBN1, IKZF2, and RAB44 were differentially expressed across all patients’ subgroups and are proposed as putative general disease biomarkers. Unsupervised machine learning analysis of pre- vs. post-dupilumab transcriptomic profiles identified two distinct patient subgroups within G0, here termed G1 and G2. When comparing baseline vs. post-treatment samples in the overall cohort (G0), only 12 DEGs were identified. In contrast, stratified analysis revealed 1288 DEGs in G1 and 354 DEGs in G2, suggesting divergent molecular response to treatment. Additionally, baseline expression of DIXDC1 was identified as a predictor of CRSwNP non-super-responders. By applying unsupervised machine learning to transcriptomic profiles, this exploratory study identifies two distinct endotypes with divergent molecular mechanisms of response to dupilumab, supporting a precision medicine approach to biologic therapy in severe asthma.

Keywords: severe asthma; CRSwNP; dupilumab; transcriptomics; endotypes; super-response; precision medicine

Share and Cite

MDPI and ACS Style

Moreno-Jiménez, E.; Morgado, N.; García-Sánchez, A.; Triviño, J.C.; Estravís, M.; Gómez-García, M.; Gil-Melcón, M.; Lázaro-Sastre, M.; Sanz, C.; Isidoro-García, M.; et al. Unsupervised Machine-Learning-Based Endotype Discovery Using Iterative Resampling in Dupilumab-Treated Patients. Int. J. Mol. Sci. 2026, 27, 5266. https://doi.org/10.3390/ijms27125266

AMA Style

Moreno-Jiménez E, Morgado N, García-Sánchez A, Triviño JC, Estravís M, Gómez-García M, Gil-Melcón M, Lázaro-Sastre M, Sanz C, Isidoro-García M, et al. Unsupervised Machine-Learning-Based Endotype Discovery Using Iterative Resampling in Dupilumab-Treated Patients. International Journal of Molecular Sciences. 2026; 27(12):5266. https://doi.org/10.3390/ijms27125266

Chicago/Turabian Style

Moreno-Jiménez, Emma, Natalia Morgado, Asunción García-Sánchez, Juan Carlos Triviño, Miguel Estravís, Manuel Gómez-García, María Gil-Melcón, Milagros Lázaro-Sastre, Catalina Sanz, María Isidoro-García, and et al. 2026. "Unsupervised Machine-Learning-Based Endotype Discovery Using Iterative Resampling in Dupilumab-Treated Patients" International Journal of Molecular Sciences 27, no. 12: 5266. https://doi.org/10.3390/ijms27125266

APA Style

Moreno-Jiménez, E., Morgado, N., García-Sánchez, A., Triviño, J. C., Estravís, M., Gómez-García, M., Gil-Melcón, M., Lázaro-Sastre, M., Sanz, C., Isidoro-García, M., & Dávila, I. (2026). Unsupervised Machine-Learning-Based Endotype Discovery Using Iterative Resampling in Dupilumab-Treated Patients. International Journal of Molecular Sciences, 27(12), 5266. https://doi.org/10.3390/ijms27125266

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Unsupervised Machine-Learning-Based Endotype Discovery Using Iterative Resampling in Dupilumab-Treated Patients

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI