Machine learning to classify the focus score and Sjögren's disease using digitalised salivary gland biopsies: a retrospective cohort study

Julien Duquesne¹, Louis Basseto¹, Charlotte Claye¹, Michael Barnes², Elena Pontarini³, Amaya Gallagher-Syed², Michele Bombardieri³, Benjamin A Fisher⁴, Saba Nayar⁴, Rachel Brown⁵, Athanasios Tzioufas⁶, Andreas Goules⁶, Loukas Chatzis⁶, Kyle Thompson⁷, Joe Berry⁸, Wan-Fai Ng⁹, Matilde Bandeira¹⁰, Vasco C Romão¹⁰, Maria Dolores López-Presa¹¹, Gaetane Nocturne¹², Wassila Ouerdane¹³, Thierry Molina¹⁴, Thierry Lazure¹⁵, Clovis Adam¹⁶, Xavier Mariette¹², Vincent Bouget¹, Samuel Bitoun¹⁷

Affiliations

¹ Scienta Lab, Paris, France.
² Centre for Translational Bioinformatics, William Harvey Research Institute, London, UK.
³ Experimental Medicine and Rheumatology, William Harvey Research Institute, London, UK.
⁴ School of Infection, Inflammation and Immunology, University of Birmingham, Birmingham, UK; National Institute of Health and Care Research Birmingham Biomedical Research Centre, University Hospitals Birmingham NHS Foundation Trust and University of Birmingham, Birmingham, UK.
⁵ Cellular Pathology, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK.
⁶ Department of Pathophysiology, Joint Academic Rheumatology Program, School of Medicine, National and Kapodistrian University of Athens, Laiko General Hospital, Athens, Greece; Research Institute for Systemic Autoimmune Diseases, Athens, Greece; Laboratory of Immunobiology, Center for Clinical, Experimental Surgery and Translational Research, Biomedical Research Foundation of the Academy of Athens, Athens, Greece.
⁷ Wellcome Centre for Mitochondrial Research, Institute of Neuroscience, Newcastle University, Newcastle upon Tyne, UK.
⁸ Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, UK.
⁹ Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, UK; HRB Clinical Research Facility, University College Cork, Cork, Ireland.
¹⁰ Rheumatology Department, Hospital de Santa Maria, Centro Hospitalar Universitário Lisboa Norte and Rheumatology Research Unit, Instituto de Medicina Molecular, Faculdade de Medicina, Universidade de Lisboa, Lisbon Academic Medical Centre, Lisbon, Portugal.
¹¹ Department of Pathology, Hospital de Santa Maria, ULS Santa Maria, Lisbon, Portugal.
¹² Rheumatology Department, Université Paris-Saclay, Assistance Publique-Hôpitaux de Paris (AP-HP), Hôpital Bicêtre, INSERM, CEA, UMR 1184, FHU CARE, Paris, France.
¹³ Université Paris-Saclay, Centrale Supélec, MICS, Gif-sur-Yvette, France.
¹⁴ Anatomie et Cytologie Pathologiques, AP-HP Université Paris Cité, Hôpital Necker, Paris, France.
¹⁵ Anatomie et Cytologie Pathologiques, AP-HP Université Paris-Saclay, Hôpital Bicêtre, Paris, France.
¹⁶ Laboratoire de Neuropathologie, AP-HP Université Paris-Saclay, Hôpital Bicêtre, Paris, France.
¹⁷ Rheumatology Department, Université Paris-Saclay, Assistance Publique-Hôpitaux de Paris (AP-HP), Hôpital Bicêtre, INSERM, CEA, UMR 1184, FHU CARE, Paris, France. Electronic address: samuel.bitoun@aphp.fr.

PMID: 41038216
DOI: 10.1016/S2665-9913(25)00181-X

Machine learning to classify the focus score and Sjögren's disease using digitalised salivary gland biopsies: a retrospective cohort study

Julien Duquesne et al. Lancet Rheumatol. 2025.

. 2025 Sep 29:S2665-9913(25)00181-X.

doi: 10.1016/S2665-9913(25)00181-X. Online ahead of print.

Authors

Affiliations

¹ Scienta Lab, Paris, France.
² Centre for Translational Bioinformatics, William Harvey Research Institute, London, UK.
³ Experimental Medicine and Rheumatology, William Harvey Research Institute, London, UK.
⁴ School of Infection, Inflammation and Immunology, University of Birmingham, Birmingham, UK; National Institute of Health and Care Research Birmingham Biomedical Research Centre, University Hospitals Birmingham NHS Foundation Trust and University of Birmingham, Birmingham, UK.
⁵ Cellular Pathology, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK.
⁶ Department of Pathophysiology, Joint Academic Rheumatology Program, School of Medicine, National and Kapodistrian University of Athens, Laiko General Hospital, Athens, Greece; Research Institute for Systemic Autoimmune Diseases, Athens, Greece; Laboratory of Immunobiology, Center for Clinical, Experimental Surgery and Translational Research, Biomedical Research Foundation of the Academy of Athens, Athens, Greece.
⁷ Wellcome Centre for Mitochondrial Research, Institute of Neuroscience, Newcastle University, Newcastle upon Tyne, UK.
⁸ Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, UK.
⁹ Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, UK; HRB Clinical Research Facility, University College Cork, Cork, Ireland.
¹⁰ Rheumatology Department, Hospital de Santa Maria, Centro Hospitalar Universitário Lisboa Norte and Rheumatology Research Unit, Instituto de Medicina Molecular, Faculdade de Medicina, Universidade de Lisboa, Lisbon Academic Medical Centre, Lisbon, Portugal.
¹¹ Department of Pathology, Hospital de Santa Maria, ULS Santa Maria, Lisbon, Portugal.
¹² Rheumatology Department, Université Paris-Saclay, Assistance Publique-Hôpitaux de Paris (AP-HP), Hôpital Bicêtre, INSERM, CEA, UMR 1184, FHU CARE, Paris, France.
¹³ Université Paris-Saclay, Centrale Supélec, MICS, Gif-sur-Yvette, France.
¹⁴ Anatomie et Cytologie Pathologiques, AP-HP Université Paris Cité, Hôpital Necker, Paris, France.
¹⁵ Anatomie et Cytologie Pathologiques, AP-HP Université Paris-Saclay, Hôpital Bicêtre, Paris, France.
¹⁶ Laboratoire de Neuropathologie, AP-HP Université Paris-Saclay, Hôpital Bicêtre, Paris, France.
¹⁷ Rheumatology Department, Université Paris-Saclay, Assistance Publique-Hôpitaux de Paris (AP-HP), Hôpital Bicêtre, INSERM, CEA, UMR 1184, FHU CARE, Paris, France. Electronic address: samuel.bitoun@aphp.fr.

PMID: 41038216
DOI: 10.1016/S2665-9913(25)00181-X

Abstract

Background: The classification of Sjögren's disease partly relies on focus score grading from a minor salivary gland biopsy. Expert regrading of the focus score leads to disease reclassification in half of cases. This study aimed to leverage machine learning to automatically classify the focus score and Sjögren's disease to identify new histological disease subtypes based on minor salivary gland biopsy.

Methods: This retrospective cohort study included minor salivary gland biopsy scanned haematoxylin and eosin slides from six expert centres (three centres in the UK and one each in Greece, Portugal, and France) of the European H2020 NECESSITY consortium. Participants with sicca but without Sjögren's disease and patients with Sjögren's disease and a focus score of either at least 1 or less than 1 where included. All patients with Sjögren's disease fulfilled the American College of Rheumatology-European League Against Rheumatism 2016 criteria. A deep learning model was trained on slides from five centres and validated on slides from the sixth centre. The primary outcome was the area under the receiver operator curve (AUROC) to classify the focus score and Sjögren's disease. Shapley values, an explainable machine learning technology, were computed to identify histological patterns driving the model's classification. People with lived experience of Sjögren's disease were involved in the decision to fund this research and in the dissemination of the findings.

Findings: The study was conducted between Oct 13, 2021, and Sept 5, 2024, and included 545 participants with a mean age of 54·2 (SD 13·5); 490 (90%) were female and 55 (10%) were male. After external validation, the model had an AUROC of 0·88 (95% CI 0·82-0·94) for the focus score classification task and an AUROC of 0·89 (0·82-0·94) for Sjögren's disease classification. The performance of Sjögren's disease classification for patients who were negative for anti-Sjögren's syndrome-related antigen A was 0·92 (0·87-1·00). Of histological patterns identified by the model, a new pattern of CD8⁺ T cells around acinar epithelial cells was associated with Sjögren's disease diagnosis.

Interpretation: This study showed that deep learning can reliably classify the focus score and Sjögren's disease using minor salivary gland biopsy exclusively. The study identified that CD8⁺ T-cell infiltration in acini was associated with Sjögren's disease. Further studies are needed to validate the models.

Funding: Société Française de Rhumatologie, European Alliance of Associations for Rheumatology.

PubMed Disclaimer

Conflict of interest statement

Declaration of interests BF declares grants or contracts from Janssen, Servier, Galapagos, Celgene, and Novartis; consulting fees from Novartis, Roche, Bristol Myers Squibb (BMS), Janssen, Servier, UCB, Sanofi, Galapagos, AstraZeneca, Kiniksa, Amgen, Otsuka, and Cullinan; and payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Servier and Novartis. GN declares grants or contracts from Société Française de Rhumatologie, Arthritis Foundation, Dreamer, FOREUM; consulting fees from AbbVie, Novartis, Galapagos, and Amgen; support for attending meetings or travel from Amgen and UCB; and participation on a data safety monitoring board or advisory board for Boehringer Ingelheim. MB declares grants or contracts from Janssen, GSK, and UCB; payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Novartis, UCB, GSK, and Amgen; and participation on a data safety monitoring board or advisory board for Ono Therapeutics, Horizons Therapeutic, Novartis, Janssen, and GSK. MB declares support for attending meetings from Viartris, Atral, Medac, and Accord. VB declares stock or stock options from Scienta Lab. W-FN declares royalties from Oxford University Press and inventors royalties from the Newcastle Sjögren's stratification tool and consulting fees from Amgen, BMS, Novartis, Johnson & Jonhson, Sanofi, Argenx, Iqvia, Quotients, Resolves Therapeutics, EQT, and Veloxis. XM declares consulting fees from BMS, GSK, Novartis, Otsuka, and Pfizer. SB declares grants from EULAR and SFR for this study. LB and CC are employed by Scienta Lab, and VB and JD are shareholders of Scienta Lab. All other authors declare no competing interests.

LinkOut - more resources

Full Text Sources
- Elsevier Science
Research Materials
- NCI CPTC Antibody Characterization Program

Save citation to file

Email citation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

Machine learning to classify the focus score and Sjögren's disease using digitalised salivary gland biopsies: a retrospective cohort study

Affiliations

Machine learning to classify the focus score and Sjögren's disease using digitalised salivary gland biopsies: a retrospective cohort study

Authors

Affiliations

Abstract

Conflict of interest statement

LinkOut - more resources

Full Text Sources

Research Materials