The pancreas is an organ in the body, which helps with food digestion and controlling blood sugar (glucose) levels. Patients with cancer of the pancreas generally have very poor outlook. Only 5 among 100 people with pancreatic cancer survive 5 years after it is found. In the UK, cancer of the pancreas is the tenth most common cancer. Each year, roughly 10,500 new cases occur and around 10,000 people die from it. National UK data on the trends of pancreatic cancer are nearly 10 years old and information on how it affects different groups of the population is limited. Unfortunately, there is no screening test for pancreatic cancer and tools for picking it up early aren’t very good.
We will use anonymous data from general practice records in England, between 2006-2022, to find the number of people who either get pancreatic cancer or die from it. We will look at their age, sex, ethnicity, and how deprived or wealthy they are. We will also check how well tools from other studies for predicting the condition work by combining these tools. Finally, we will find out what kinds of people are more likely to get pancreatic cancer in the future.
The results of this study will provide the most up-to-date knowledge on the number of people with pancreatic cancer in England. We will learn more about which groups of people are more or less likely to develop pancreatic cancer. The findings will also help GPs spot cancer of the pancreas earlier.
Background
Pancreatic cancer (PC) is one of the most lethal cancers in adults. Nearly 460,000 people with PC were diagnosed worldwide in 2018 with 430,000 deaths. Cancer incidence and mortality rates vary widely. The most recent estimates on the trends of PC at UK national level are nearly 10 years old. Additionally, national-level information on the impact of health inequalities and/or ethnicity on PC incidence is scarce. PC has very poor survival with only 5% of people surviving 5 years after diagnosis. This is often due to late-stage diagnosis. There is currently no screening test for PC and early detection tools are lacking.
Aims
Improve our understanding of the incidence and mortality of PC by socio-demographic factors in England and develop and validate new tools to predict the risk of PC in primary care.
Methods
Using population-based and case-control study designs, eligible population will be people aged ≥18years without a prior history of PC between 2006-2022 in England. Outcomes include annual incidence and mortality rates of PC. Poisson regression models will be used to estimate incidence and mortality rates overall and by age, sex, ethnicity, social deprivation and whether they live in rural/urban areas. Using information from an existing systematic review, we will validate existing risk prediction models in linked primary, secondary data and mortality records and perform model ensemble and stack regression to combine risk factors from these models. Subsequently, using machine learning approaches (random forest, deep learning, gradient boosting), we will analyse the trajectories of the risk factors leading to the disease to develop and validate a new risk prediction model for PC.
Expected impacts
Our study is expected to: a) update national statistics on PC epidemiology; b) provide insights in temporal trends, e.g. age of onset and ethnic differences in incidence; and c) improve the disease prediction.
Incident and death by pancreatic cancer in England (2006 -2022) across various strata of interest: age, sex, social deprivation, ethnicity, and geographic region in England and rural/urban areas.
Rosa Parisi - Chief Investigator - University of Manchester
Salwa Zghebi - Corresponding Applicant - University of Manchester
Alison Wright - Collaborator - University of Manchester
catharine Morgan - Collaborator - University of Manchester
Evangelos Kontopantelis - Collaborator - University of Manchester
Glen Martin - Collaborator - University of Manchester
Mairead Geraldine McNamara - Collaborator - University of Manchester
Martin Rutter - Collaborator - University of Manchester
Rathi Ravindrarajah - Collaborator - University of Manchester
Sam Merriel - Collaborator - University of Exeter
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;CPRD Aurum Ethnicity Record;CPRD GOLD Ethnicity Record;Practice Level Rural-Urban Classification