AI-assisted diagnosis of ICPI pneumonitis
Research type
Research Study
Full title
AI-assisted diagnosis and prediction in Immune Checkpoint Inhibitor (ICPI)-mediated pneumonitis
IRAS ID
329056
Contact name
James O Jones
Contact email
Sponsor organisation
Cambridge University hospitals and the University of Cambridge
Duration of Study in the UK
5 years, 0 months, 0 days
Research summary
Immune checkpoint inhibitors (ICPIs) are a type of cancer treatment that work by activating a patient's own immune system to attack the cancer. ICPIs have been very successful in a number of cancer types. However, they can also cause the immune system to attack the body's healthy organs. One of the more common organs affected is the lungs, which become inflamed, causing the patient to feel breathless (this is called 'pneumonitis'). This complication can be very serious, some patients need hospital treatment and it is sometimes fatal. More research is needed to understand the risk factors for it happening and how to pick it up early so it can be treated quickly.
This research project aims to study this clinical problem. We will check the records of patients previously treated with ICPIs at Addenbrookes Hospital in Cambridge (about 1200 records) to find about 100 people who developed pneumonitis, and at least 100 similar patients who did not. This is a retrospective study, so we will be looking back at patients treated in the past, and there will be no changes to patients' treatment as part of the study.
De-identified data will be shared with our collaborators in the Department of Maths and Theoretical Physics at the University of Cambridge. They will use artificial intelligence (AI) based methods to look for patterns in previous health problems, test results and medical images (like X rays and CT scans), that are difficult for humans to detect. We hope that this will allow us to develop tools to pick up signs of pneumonitis at an early stage when it is more treatable.
REC name
East of England - Cambridge Central Research Ethics Committee
REC reference
23/EE/0197
Date of REC Opinion
19 Sep 2023
REC opinion
Favourable Opinion