Predicting Disease Outcomes in Sarcoidosis Using PET Scanning
Research type
Research Study
Full title
Predicting Disease Outcomes in Sarcoidosis Using PET Scanning
IRAS ID
353393
Contact name
James Galloway
Contact email
Sponsor organisation
King’s College London
Duration of Study in the UK
1 years, 0 months, 0 days
Research summary
This study investigates whether disease activity captured via FDG PET imaging can predict progression in sarcoidosis using advanced machine learning techniques. Sarcoidosis is a complex, multisystem inflammatory condition with a highly variable course, ranging from mild, self-limiting disease to chronic, progressive organ dysfunction. Current tools to predict disease outcomes are limited, leading to potential under- or overtreatment.
This is a retrospective observational cohort study of 150 adults diagnosed with sarcoidosis who underwent FDG PET scans as part of routine NHS care within the past five years. By combining PET imaging data with clinical outcomes, the study aims to test whether organ involvement and metabolic activity, as assessed through machine learning, can predict disease progression. The primary focus is on progression risk, with secondary objectives exploring associations between PET imaging and steroid tapering, disease flares, treatment escalation, and progression of lung, cardiac, and liver disease.
PET imaging data will be analyzed using advanced segmentation software (TotalSegmentator) to quantify organ-specific metabolic activity, which will then be correlated with clinical outcomes through supervised machine learning algorithms. Statistical analyses, including logistic regression and Cox proportional hazards models, will assess both cross-sectional and longitudinal associations.
Given the minimal risks involved, anonymised patient data will be used without individual consent, adhering to UK GDPR regulations. Data will be securely stored, de-identified, and handled only by the clinical research team.
This study addresses an unmet clinical need, potentially offering a non-invasive tool to stratify high-risk patients, optimize treatment strategies, and prevent irreversible organ damage. Findings may pave the way for larger, multi-centre trials and enhance personalised care for sarcoidosis patients.
REC name
London - Surrey Borders Research Ethics Committee
REC reference
25/PR/0734
Date of REC Opinion
20 Jun 2025
REC opinion
Favourable Opinion