Research aiming to help predict the progression of lung disease in patients with scleroderma has been awarded funding from Scleroderma & Raynaud's UK (SRUK), a charity dedicated to improving the lives of people with scleroderma and Raynaud's phenomenon.
The research, led by Dr Elizabeth Renzoni, respiratory consultant at Royal Brompton and Harefield hospitals, will look at machine learning tools and how these can be used to help patients with scleroderma associated interstitial lung disease (SSc-ILD). The research will be carried out in collaboration with the Royal Free London NHS Foundation Trust.
What is SSc-ILD?
Systemic sclerosis (SSc), also known as scleroderma, is a rare condition which causes hardening and thickening of the skin, internal organs and blood vessels. It is an autoimmune disease where the body’s own immune system attacks the connective tissue in the body, leading to scarring and thickening of the tissue.
Interstitial Lung Disease (ILD) is a term used to describe a group of conditions where there is thickening of the supportive tissues found between the air sacs of the lungs (lung scarring or fibrosis). The disease can be caused by a variety of things or have no identifiable cause.
The majority of patients who suffer from SSc also suffer from ILD (SSc-ILD) although the extent of the ILD and how it progresses over time varies from patient to patient. At least a third of patients with scleroderma will have increasing lung scarring, worsening breathlessness and reduced survival.
Uncertainty for patients
Patients diagnosed with SSc-ILD can experience uncertainty and anxiety about their future and how the disease will progress, severely affecting their quality of life. This uncertainty of how the disease will progress also makes treatment decision difficult for clinicians.
Uncertainty about the likelihood of progression of the lung disease makes treatment decisions difficult. Dr Renzoni and her team are keen to reduce this uncertainty and believe there is need for better indicators to accurately predict the likelihood of the lung fibrosis getting worse. This will allow for effective treatments to be started as soon as possible and prevent irreversible scarring, whilst also avoiding potentially toxic treatments in patients who do not need them.
Machine learning
The research funding will allow Dr Renzoni and her team aim to evaluate the chest CT scans of 500 patients with SSc-ILD, taken when they were first diagnosed. The team will look at the detailed follow up data of these patients and run a machine learning tool to see whether it can predict the progression of ILD in patients with SSc over time.
Dr Renzoni says; “High resolution CT plays a central role in the diagnosis and staging of SSc-ILD. Deep learning algorithms have the potential to provide CT signal that goes beyond the ability of human scoring.
“We plan to assess whether an existing algorithm developed in idiopathic ILD (SOFIA algorithm) can provide information about likelihood of progression in SSc-ILD patients. We also plan to do the groundwork to begin to develop an algorithm specific to SSc.”
The group hope to use the data as a basis to perform a larger study which will include other centres to determine if the results are applicable to all patient with SSc-ILD.
Dr Renzoni explains that the groups ultimate aim is to “develop an accessible algorithm specific to SSc-ILD patients that can predict progression and can be used across centres to guide management”.
To find out more about this project, or any of our other research, please contact us.