Thèse Modèle Multi Compartimentale Bronchique et Alvéolaire sur Puce pour une Médecine Personnalisée dans la Bronchopneumopathie Chronique Obstructive Bpco H/F - 33

Établissement : Université de Bordeaux
École doctorale : Sciences de la Vie et de la Santé
Laboratoire de recherche : Centre de recherche Cardio-Thoracique de Bordeaux
Direction de la thèse : Maéva ZYSMAN ORCID 0000000314592409
Début de la thèse : 2026-12-01
Date limite de candidature : 2026-05-20T23:59:59

La bronchopneumopathie chronique obstructive (BPCO) est la troisième cause de mortalité dans le monde. Elle se caractérise par une limitation progressive des débits d'air et des exacerbations fréquentes qui aggravent le pronostic des patients. Bien que les biothérapies aient révolutionné le traitement de l'asthme, leur application dans la BPCO reste difficile, en raison de l'hétérogénéité des patients et du faible taux de réussite des essais cliniques. L'approbation récente du dupilumab par l'agence américaine des produits alimentaires et des médicaments (Food and Drug Administration, FDA) pour les patients atteints de BPCO de type éosinophilique représente une avancée majeure, mais ses bénéfices restent limités à un sous-groupe de patients.
Les thérapies émergentes ciblant les alarmines d'origine épithéliale, telles que TSLP et IL-33, offrent une voie prometteuse pour le traitement de la BPCO, avec plusieurs essais cliniques de phase II et III en cours. Toutefois, l'identification des patients répondeurs aux biothérapies reste un défi majeur, en l'absence de biomarqueur universellement validé.
Les modèles innovants comme les organoïdes et les technologies de type 'organes-sur-puce' (organ-on-a-chip) offrent des données pertinentes sur la physiologie humaine et montrent un fort potentiel en tant qu'outils prédictifs, tant pour de nouveaux traitements que pour des médicaments déjà existants.
Notre projet vise à développer une plateforme de type multi-organes-sur-puce, capable de prédire la réponse des patients atteints de BPCO aux biothérapies, et de sélectionner les traitements les plus efficaces.
Cette étude répond à cet enjeu à travers trois objectifs (et les « work packages » qui en découlent) interconnectés :
WP1 : Génération et cryoconservation de cellules progénitrices (pulmonaires et musculaires) afin de créer des modèles in vitro représentatifs des patients.
WP2 : Développement d'une plateforme multi-organes-sur-puce intégrant les cellules pulmonaires (bronchiques et alvéolaires) pour reproduire la physiopathologie de la BPCO.
WP3 : Prédiction de la réponse des patients aux biothérapies et sélection des traitements les plus efficaces pour une biothérapie personnalisée de la BPCO.
En s'appuyant sur des modèles in vitro avancés, plus représentatifs de la physiologie humaine, cette approche pourrait ouvrir la voie à une biothérapie personnalisée dans la BPCO, améliorant ainsi les taux de succès thérapeutique.

Chronic obstructive pulmonary disease (COPD) is a major public health disease, ranked as the third cause of death worldwide. It is caused by significant exposure to noxious particles or gases, and characterized by progressive airflow limitation and alveolar destruction (emphysema). COPD accounts for 55% of chronic respiratory illnesses, affecting over 300 million people worldwide, and causing 3.3 million deaths a year. Exacerbations - acute episodes of worsening respiratory symptoms - are key events that can accelerate disease progression and the main costs of this disease through unscheduled hospitalisations.
Current pharmacological treatments for COPD, such as bronchodilators and corticosteroids, are mainly symptomatic. While biologics have transformed asthma care, their application to COPD remains difficult due to high patient heterogeneity. Some biologics have shown promise in small subgroups of COPD patients with frequent exacerbations, but overall success in COPD clinical trials is low (13%). Our recent real-world data also indicate that a limited proportion of COPD patients qualify for biologics. The 2024 FDA approval of Dupilumab, targeting IL-4 and IL-13, marks a milestone as the first antibody therapy for eosinophilic COPD, though it benefits only about 20% of patients. Emerging data on other biologics, such as mepolizumab (targeting IL-5), further support this targeted approach, with very recent positive results in hyper-eosinophilic COPD.
Biologics targeting alarmins show promise as COPD therapies, with encouraging results in phase 2 trials. Alarmins, epithelial-derived cytokines like TSLP and IL-33, are key upstream regulators of type 2 inflammation and involved in COPD pathobiology. Epithelial progenitor cells are the main source of excess IL-33 in COPD. Several phase II/III trials are evaluating biologics such as Tezepelumab (anti-TSLP), Astegolimab (anti-ST2), and IL-33 blockers like Tozorakimab and Itepekimab. Notably, Itepekimab reduced exacerbations rates in former but not current smokers.
The factors driving patient response to biologics in COPD remain poorly understood. Biomarkers like blood eosinophils, fractional exhaled nitric oxide (FeNO), and blood C-reactive protein (CRP) may help identify responders, but none have shown consistent predictive value or universal clinical validation. Therefore, there is an urgent need for reliable tools to guide biologic selection and identify responders. The absence of relevant models is a major barrier to personalized biologic response profiling in COPD. Microphysiological systems like organoids and organ-on-chip (OoC) platforms show promise for modelling disease complexity and predicting treatment responses, as seen in cystic fibrosis and immuno-oncology.

The project aims to develop a organ-on-chip platform to predict which COPD patients will respond to biologics and to select the most effective biologics, with a particular focus on exacerbation responses. The project has three key objectives:
1.Generate and biobank patient-derived lung, using bronchoalveolar lavages (BAL), and evaluate their potential to serve as biological avatars.
2.Develop a MOoC system, integrating human bronchial and alveolar epithelial cells, and establish a dynamic microfluidic environment to mimic in vivo interactions.
3.Assess the ability of the MOoC system to predict biologic responses, by exposing patient-derived cells to exacerbation-like stimuli and evaluating the impact of different biologic treatments in a patient-specific manner.

WP1: Generation and cryopreservation of progenitor cells from lung to create patient-relevant in vitro models
this WP aims at (1) generating airway and alveolar organoids from bronchoalveolar lavage (BAL) obtained from individuals with COPD, (2) characterizing lung organoids using singlecell RNA sequencing (scRNA seq) (3) establishing a biobank of airway and alveolar organoids
WP2 : Development of a multi-organ-on-chip platform that integrates lung (airway and alveolar) to mimic COPD pathophysiology
Aim: This WP aims to develop a patient-derived multi-organ-on-chip system to monitor inflammation and enable metabolomic analyses. To model epithelial interactions and assess responses to biologics, we will follow a stepwise approach using a robust and modular design. Initial experiments will be conducted with primary bronchial and alveolar epithelial cells from non-COPD lung tissue (TissUs Bronchiques et PulmonairEs (TUBE). Cells from COPD patients will be later integrated in the model.

WP3. Prediction of patient responsiveness to biologics and selection of the most effective biologics for personalized biotherapy in COPD
Aim: This WP aims at generating patient-derived multi-organ-on-chip models, treating them with biologics, and precisely measuring drug response under exacerbation-like conditions, effectively conducting a 'clinical trial in a chip. The objective is to evaluate the performance of this in vitro drug response in identifying clinically distinct subgroups and supporting the selection of the most effective biologics for each patient. When possible, we will assess the relationships between in vitro drug response and clinical outcome parameters. T