Using Predictive Analytics to Improve Patient Outcomes in Neuroendocrine Tumours

Date of Approval: 
2018-04-27 00:00:00
Lay Summary: 
Neuroendocrine tumors (NETs) are a type of cancer that develops from cells that produce hormones (such as insulin). These cells are widely spread among the organs of the human body They are most commonly found in the gut (stomach, colon) but are also found in the lungs and pancreas. Once patients are diagnosed with a NET there are several different options for treatment including surgery, drug treatment, chemotherapy and radiation. There is little information about what the best treatment strategy is and if this differs for different patients. This study will look at patients who have been diagnosed with NETs using information from the National Cancer Registration and Analysis Service linked to the routinely collected health information in the Clinical Practice Research Datalink and the hospital treatment data available in the Health Episodes Statistics database. Our aim is to identify the most common treatments and determine if there are factors (such as age, weight, gender of the patient, where the tumour is located or how far advanced the cancer is) that can be used to predict how long the cancer patient will survive.
Technical Summary: 
The objective of this study is to identify predictors of survival in the five year follow-up period after the initial diagnosis of Neuroendocrine tumours (NETs). This is a retrospective cohort study utilizing data from CPRD linked to Hospital Episode statistics (HES), UK Cancer Registry databases and Office for National Statistics (ONS) mortality data. The study cohort will be followed for the 5 year period from initial diagnosis. Predictors of survival will be identified using machine learning algorithms based on patient's characteristics. Patient subgroups, treatment pathways and diagnosis pathways will be identified using clustering. Descriptive statistics will be produced on patient characteristics at baseline and treatment pathways in the post-index period.
Health Outcomes to be Measured: 
Primary: 5 year survival from NETs diagnosis
Application Number: 

Ruby Saharan - Chief Investigator - NOVARTIS
Paul McGettigan - Corresponding Applicant - Novartis Ireland Limited
Ailis Fagan - Collaborator - NOVARTIS
Jamie Weaver - Collaborator - The Christie NHS Foundation Trust
Martin McSharry - Collaborator - Optum
Spyros Roumpanis - Collaborator - F. Hoffmann - La Roche Ltd
Wasat Mansoor - Collaborator - The Christie NHS Foundation Trust

HES Admitted Patient Care;NCRAS Cancer Registration Data;NCRAS Systemic Anti-Cancer Treatment (SACT) data;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation