This study aims to understand how cancer and heart disease (cardiovascular disease "CVD") are related and how they can affect patients at the same time. Because survival rates after a cancer diagnosis have improved, individuals with cancer face the risk of developing CVD. Cancer and CVD share common risk factors like being overweight, smoking, or having diabetes. Additionally, certain cancer treatments may increase the chances of developing CVD in cancer patients.
The study will use data from the UK's primary care system to look at how commonly overall and specific CVDs rates have changed over time in different groups of patients in relation to different types of cancer and other factors. These obtained historical data will be used to try to predict how common CVD will be in cancer patients over the next 10 years. There are models that can tell how likely someone is to have a heart attack or stroke in future based on information, such as their age, cholesterol level, height, and weight. However, it is unclear if these models work well when applied to patients who are diagnosed with cancer. A CVD risk prediction model called "QRISK" (derived from the “QResearch” database and "risk") will be tested to see if it can tell individuals with cancer their future cardiovascular risk.
In conclusion, this project will explore important disease patterns and predictions about CVD in cancer patients, which can help in understanding the burden of the condition on the healthcare system and improving patients’ overall health outcomes.
Due to improvements in cancer survival, cardiovascular disease (CVD) and cancer can co-exist and clinically overlap. It is believed that cancer and CVD share common risk factors such as obesity, diabetes mellitus, and smoking. At the same time, some anticancer therapies have been linked with an increased risk of cardiotoxicity in patients with cancer, which may enhance the development of CVD. The consequence of having CVD in patients with cancer can increase the risk of both mortality and morbidity. This in turn affects patients' prognosis and places a significant burden on the healthcare system by increasing hospitalization and admission rates.
We will analyse data from the UK's primary care setting linked with secondary data from Hospital Episode Statistics (HES) and caner registries to investigate temporal changes of CVDs over time and the development of various CV risk factors in relation to various cancer sites, stratified by age, sex, ethnicity, and deprivation status. Furthermore, we will estimate the 10-year projection of CVD and its risk factors in cancer patients by using time-trend analysis. Finally, the performance of the QRISK score in the cancer cohort will be tested statistically based on the given data to determine its performance in predicting future CVD in cancer patients, with the aim of developing new risk models tailored to patients diagnosed with cancer.
The main model through all these analyses will be a Cox proportional hazards regression. This model will be extended to incorporate competing risk event (i.e., dying from cancer), interactions (e.g., CVD risk with time), and effect modification according to different cancer sites. Considering all these aspects can result in a flexible parametric regression model that can predict absolute excess CVD risk over time.
During all phases of the project, the primary outcome will be cardiovascular disease (CVD) including acute myocardial infarction, heart failure, stroke, diabetes, atrial fibrillation, valvular heart disease, and peripheral vascular disease.
All outcomes will be determined during the follow-up period by matching with the appropriate Read/SNOMED codes found in electronic health records. The identification of hospitalisation cases resulting from CVDs is accomplished by establishing a linkage to the Hospital Episode Statistics Admitted Patient Care (HES APC) database, utilising the 10th version of the International Classification of Diseases (ICD-10). The mortality data from the Office for National Statistics (ONS) will be utilised in order to determine the precise reasons of death, such as cancer-related, cardiovascular, or all-cause mortality, based on ICD-10 codes.
Evangelos Kontopantelis - Chief Investigator - University of Manchester
Ali Alshahrani - Corresponding Applicant - University of Manchester
catharine Morgan - Collaborator - University of Manchester
Glen Martin - Collaborator - University of Manchester
Mamas Mamas - Collaborator - Keele University
Rathi Ravindrarajah - Collaborator - University of Manchester
HES Admitted Patient Care;NCRAS Cancer Registration Data;NCRAS National Radiotherapy Dataset (RTDS) data;NCRAS Systemic Anti-Cancer Treatment (SACT) data;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation