Cancer is the leading cause of death worldwide causing almost 10 million deaths in 2020. There is a need to see who develops cancer, if and how long they survive to help improve patient care. Usually, data from clinical trials is used to estimate survival from cancer however clinical trials are relatively short. Therefore there is some uncertainty when estimating how long someone lives after diagnosis. Real-world data such as medical records, with much longer follow up times, could be a suitable alternative to address this uncertainty. The aim of this project is the examine the incidence, prevalence, and survival of certain common, uncommon, and rarer cancers (breast, colorectal, lung, liver, prostate, head/neck, pancreatic, oesophagus, and stomach) and see how well mathematical models are able to predict cancer survival 1, 5, and 10 years after the first diagnosis using real-world data. We will carry out the analysis for each cancer but we will also split the analysis up by different age groups, gender, and other diseases to see the impact of these different groups on cancer survival. The results will be freely available as a user-friendly tool. This work will improve patient care by indirectly informing public health policy by evaluating the usefulness of real-world data in survival prediction for common cancers over the lifespan.
Cancer is the leading cause of mortality worldwide, and accounted for almost 10 million deaths in 2020. The continuous surveillance and monitoring of trends in cancer incidence and survival are needed for the development, implementation and evaluation of health policies aiming to reduce the burden of disease. Clincal trial data is used with extrapolation techniques to estimate long-term overall survival (OS). However, this extrapolation is often a key source of decision uncertainty, because they are forecasts of the future based only on observed data. Real-world evidence such as medical records can help to address this uncertainty. Therefore, the aim of this project is to examine the incidence, prevalence and survival of certain common and rare cancers (e.g., cancers of the breast, colon [colorectal], lung, liver, prostate, head/neck, pancreatic, oesophagus and stomach) using real world data. Using a retrospective cohort design we will estimate the age-sex incidence rates and prevalence and the overall observed survival of the studied cancers stratified by age, gender, region (GOLD only), and other comorbidities and will investigate how well standard survival functions predict the natural history of the studied cancers, 1, 5, or 10-years after diagnosis. From this study, we will be able to assess the survival of particular cancers. The evaluation of real-world data for cancer extrapolation is important due to the cost-effectiveness of extrapolation models, which typically assess costs and health outcomes over a lifetime horizon and are vitally important to understand cancer survival for patients, health care providers and regulatory authorities.
The primary outcome of this study is the all-cause mortality
Daniel Prieto-Alhambra - Chief Investigator - University of Oxford
Danielle Newby - Corresponding Applicant - University of Oxford
Antonella Delmestri - Collaborator - University of Oxford
Edward Burn - Collaborator - University of Oxford
Ian Koblbauer - Collaborator - University of Oxford
Jamie Elvidge - Collaborator - National Institute for Health and Clinical Excellence - NICE
Patricia Pedregal Pascual - Collaborator - University of Oxford
Ravinder Claire - Collaborator - National Institute for Health and Clinical Excellence - NICE
Wai Yi Man - Collaborator - University of Oxford
Rabia Khan - Collaborator - University of Oxford
Rabia Ali Khan - Collaborator - University of Oxford
ONS Death Registration Data