Pancreatic cancer remains a very deadly disease, despite improvements in treatment and care. This is because patients often already have advanced disease when they are diagnosed, when there are fewer treatment options available. Patients with pancreatic cancer diagnosed at earlier stages are more likely to survive longer.
Unfortunately, pancreatic cancer is challenging to detect early as there are very few or only vague symptoms in early stages. However, many patients who develop pancreatic cancer are diagnosed with diabetes in the months/years before they receive their pancreatic cancer diagnosis. A new diagnosis of diabetes is therefore a potential indicator of pancreatic cancer which can be used for early detection.
We are really pleased to have received this award from Pancreatic Cancer UK and look forward to starting work on this project that aims to improve early detection of pancreatic cancer in patients with a recent diagnosis of diabetes. - Dr Pui San Tan, lead researcher.
However, as diabetes is a relatively common condition in the population, it remains impracticable to refer everyone with new onset diabetes for pancreatic cancer scans. Hence, GPs will need a more accurate tool to better identify patients with new onset diabetes who will be more likely to have pancreatic cancer for further tests.
Working closely with the Oxford Centre for Early Cancer Detection (OXCODE), Dr Pui San Tan and Professor Julia Hippisley-Cox (Nuffield Department of Primary Care Health Sciences) have successfully applied for Pancreatic Cancer UK’s Research Innovation Fund, which supports high-risk high-reward projects. With their ~£100,000 award, this team will be using machine learning to understand if it’s effective in detecting pancreatic cancer in patients who are diagnosed with new onset diabetes.
By using the QResearch database containing anonymised GP records of patients in England linked to hospital records, cancer registry and death records, they will search for characteristics within new-onset diabetes patients that are associated with a later diagnosis of pancreatic cancer. They will use this information to help develop a risk prediction equation. If effective, it could be used to develop a vital clinical decision tool to help GPs refer patients for the critical scans they need to diagnose pancreatic cancer earlier.