Prevalence of Cardiovascular Disease in Patients With Potentially Curable Malignancies

Background Although a common challenge for patients and clinicians, there is little population-level evidence on the prevalence of cardiovascular disease (CVD) in individuals diagnosed with potentially curable cancer. Objectives We investigated CVD rates in patients with common potentially curable malignancies and evaluated the associations between patient and disease characteristics and CVD prevalence. Methods The study included cancer registry patients diagnosed in England with stage I to III breast cancer, stage I to III colon or rectal cancer, stage I to III prostate cancer, stage I to IIIA non-small-cell lung cancer, stage I to IV diffuse large B-cell lymphoma, and stage I to IV Hodgkin lymphoma from 2013 to 2018. Linked hospital records and national CVD databases were used to identify CVD. The rates of CVD were investigated according to tumor type, and associations between patient and disease characteristics and CVD prevalence were determined. Results Among the 634,240 patients included, 102,834 (16.2%) had prior CVD. Men, older patients, and those living in deprived areas had higher CVD rates. Prevalence was highest for non-small-cell lung cancer (36.1%) and lowest for breast cancer (7.7%). After adjustment for age, sex, the income domain of the Index of Multiple Deprivation, and Charlson comorbidity index, CVD remained higher in other tumor types compared to breast cancer patients. Conclusions There is a significant overlap between cancer and CVD burden. It is essential to consider CVD when evaluating national and international treatment patterns and cancer outcomes.

C ancer is associated with significant morbidity and mortality in England. 1 Cancer and cardiovascular disease (CVD) survival is improving. 2,3 However, they share risk factors and pathophysiological processes 4 and may coexist. 5 Furthermore, cancer and its treatment may result in cardiac complications. 6 CVD may influence cancer management and contribute to disparities 7 in the UK, [8][9][10][11][12] in older adults, 13,14 and internationally. 2,15 Pre-existing CVD in individuals with potentially curable cancer has been described in various countries. [16][17][18][19] However, this has not been widely reported in England. The impact of cancer-related factors and social deprivation on cancer and CVD has also not been previously assessed. Investigating the intersection of cancer and CVD is central to understanding outcomes, informing cancer policy, and service provision.
We analyzed the prevalence of pre-existing CVD in a cohort of individuals with potentially curable tumors in England, as differences in cancer management due to comorbidities may affect survival. We also assessed the associations between CVD prevalence and patient and tumor characteristics.

As part of the Virtual Cardio-Oncology Research
Initiative program, 20 we linked Public Health England National Cancer Registration Dataset (NCRD), 21 Hospital Episode Statistics (HES), 22 and National Institute for Cardiovascular Outcomes Research (NICOR) 23 data to identify CVD recorded in hospital records and registry datasets. We linked English cancer registry data (NCRD) and 6 CVD-specific audits managed by NICOR (Supplemental Table 1). Four NICOR databases were included in this study: the Myocardial Ischaemia National Audit Project, 24 National Adult Cardiac Surgery Audit, 25  Robust quality assurance checks are in place for the NICOR and HES datasets. 28,29 Therefore, CVD prevalence was defined according to either presence of an inpatient hospitalization CVD diagnosis code and/or a NICOR CVD audit record. NCRD Table 2). We excluded patients with synchronous tumors diagnosed in the same site with similar prognostic features and those with synchronous tumors diagnosed in different sites.
We included patients 25 to 100 years of age at cancer diagnosis, those with residency in England, and those with complete data on vital status, sex, and National Health Service number (to allow linkage).
We restricted the analysis to potentially curable tumors (stage I-III breast cancer, stage I-III colon or   rectal cancer, stage I-III prostate cancer, stage I-IIIA NSCLC, stage I-IV DLBCL, and stage I-IV Hodgkin lymphoma) (Supplemental Table 3 Table 4).
We identified CVD comorbidities using the ICD-10 code list (Supplemental Table 5   1.6%-6.7%) (Supplemental Table 14). Prior CVD rates showed no laterality differences in the breast cancer and NSCLC cohorts.
We present the unadjusted logistic regression analysis (Supplemental Table 15) and the adjusted logistic regression analysis (  The highest absolute number of individuals with prior  The observed CVD prevalence across tumor groups is shown in Figure 3. Age-and sex-standardized CVD prevalence was much lower than the observed prevalence, as patients with cancer were older than the  Table 3). Patients receiving surgery, radiotherapy, or chemotherapy had lower odds of CVD compared with those not treated in most individual tumor cohorts (breast, colon, rectal, DLBCL, and Hodgkin lymphoma), but not in the prostate and NSCLC cohorts.

DISCUSSION
Our study was a large-scale, population-based analysis describing CVD prevalence in individuals with potentially curable cancers. Understanding the intersection between cancer and CVD is key to informing anticancer treatment decisions, interpreting outcomes, and planning health care provision. 37 We used linked national registry datasets of patients diagnosed with potentially curable malignancies over 6 years in England and found an overlap between cancer and CVD in 16.2% of individuals.

An analysis of English National Cancer Diagnosis
Audit data linked to primary care records showed that  Comorbidities are more common among lung cancer survivors and less frequent among breast and prostate cancer survivors. 39 One study documented that 43.6% of patients diagnosed with potentially curable NSCLC in England from 2012 to 2016 had CVD, 36 which affected resection and mortality rates. 21 In the general population, older age is associated with a higher prevalence of CVD, 40 and CVD contributes to an increasing burden of morbidity and disability in community-dwelling older individuals. Prospective trials and cancer registry analyses have documented higher risk for heart failure in patients with potentially curable malignancies and CVD and cardiovascular risk factors receiving cytotoxic or targeted therapies. [41][42][43][44][45] Similar concerns exist for patients potentially suitable for locoregional treatments. 36,46 Pre-existing CVD may represent a contraindication for pursuing specific anticancer treatment options or require adjustments, possibly hindering the chances of cure in individuals with potentially curable cancer.
In future analyses, we plan to examine the geographic variation of CVD rates and its impact on anticancer treatments.
CVD is also an increasingly prevalent exclusion criterion for studies investigating novel anticancer treatments. 47 This has substantial implications on limiting not only the access of patients with cancer to experimental treatments but also trial results applicability, 48 trial design, drug development, and drug labeling. 49 Our study confirms that men, older individuals, and those living in socioeconomically deprived areas had a higher CVD burden. These factors have important impact on the prevalence of CVD in patients with potentially curable malignancies ( Figure 4). Male sex is a risk factor for higher coronary artery disease rates and mortality. 50 Patients undergoing surgery, radiotherapy, or chemotherapy have lower odds of CVD compared with those not treated in the overall cohort and in most individual tumor cohorts. The burden of comorbidities increases with age 39,51 and may influence overall and non-cancer-related mortality 52-54 but also affect anticancer treatment tolerance. 55 For patients with breast cancer, CVD may also influence tumor-specific mortality. 56 We demonstrated an increasing prevalence of CVD associated with worse deprivation in all tumor cohorts except Hodgkin lymphoma. In this analysis, a higher score of the income domain of the Index of Multiple Deprivation, which corresponds to lower income and higher levels of deprivation, was also associated with more advanced tumor stage. Socioeconomic inequalities have a significant impact on cancer presentation, diagnosis, and treatment. 57 Despite efforts aiming to reduce them in England, their impact on cancer survival has not substantially changed. 58 An accurate review of care pathways for patients with cancer and comorbidities may mitigate their detrimental effect on outcomes. 59 Our analysis suggests that CVD can be ascertained

CENTRAL ILLUSTRATION Study Design and Key Findings
ACKNOWLEDGMENTS This project involves data that has been provided by, or derived from patients and collected by the NHS as part of their care and support.
The data is collated, maintained and quality assured by the National Cancer Registration and Analysis Service, which is part of Public Health England (PHE) and data has also been provided by the Healthcare Quality Improvement Partnership from the National Cardiac Audit Programme, part of the National Clinical Audit and Patient Outcomes Programme which they commission. Access to the data was facilitated by the PHE Office for Data Release. TRANSLATIONAL OUTLOOK: The overlap between cancer and CVD burden is substantial and may explain cancer treatment patterns and outcomes. Future work is focused on understanding the impact of prevalent CVD on cancer management and the relationships between geography and access to cardio-oncology resources. Understanding the intersection between cancer and CVD is key to informing anticancer treatment decisions, interpreting outcomes, and planning health care provision.