Mortality In Cancer Patients: Cardio Mortality Meta-Analysis
Overview
Cardiovascular disease (CVD) is a significant cause of non-cancer-related death among cancer survivors, although the level of risk varies depending on the type of cancer.
This study aims to review and synthesize existing evidence regarding patterns and the extent of CVD mortality risk in cancer survivors.
A comprehensive search of Medline (OVID), CINAHL, and Scopus databases was conducted from January 1, 2000, to July 16, 2023. The review focused on studies that reported CVD mortality in cancer patients compared to reference populations, using standardised mortality ratios (SMRs). A meta-analysis was performed using a random-effects model to account for variability across studies.
A total of 136 studies from 16 countries were included, with sample sizes ranging from 157 to over 7.5 million individuals. The majority of the studies (72%) were conducted in the United States, followed by Europe (16%). The most commonly studied cancers were gastrointestinal, hematological, and breast cancers. The analysis extracted 876 SMRs for various CVD conditions. Of these, 61% indicated an elevated risk of CVD death (SMR >1), 12% indicated a lower risk (SMR <1), and 27% showed an equivalent risk compared to the general population. The overall meta-analysis revealed that cancer survivors had a 55% higher risk of CVD mortality (SMR = 1.55, 95% CI = 1.40–1.72) compared to the general population. The risk varied across different CVD conditions, with SMRs ranging from 1.36 for heart disease to 1.56 for cerebrovascular disease. CVD risk also differed by cancer type, with the highest risk observed in brain and central nervous system tumors (SMR = 2.82), while testicular/germ cell tumors had the lowest (SMR = 1.14).
Cancer survivors face an elevated risk of early death from CVD, though the risk is influenced by both cancer type and CVD condition. Further research is needed to explore the underlying mechanisms driving this increased risk to develop targeted interventions.
Introduction
Cardiovascular disease (CVD) and cancer are the leading causes of death and illness globally. Research increasingly shows that cancer survivors face a higher risk of developing CVD and have higher mortality rates from CVD compared to the general population. This elevated risk can be attributed to shared risk factors such as smoking, obesity, physical inactivity, unhealthy diets, and excessive alcohol consumption. Additionally, cancer treatments like chemotherapy, radiotherapy, immunotherapy, and targeted therapies have been linked to cardiovascular complications. Cancer survivors may also have lower adherence to medications targeting CVD risk factors.
Circulatory system diseases, which include both cardiovascular diseases (affecting the heart and blood vessels) and cerebrovascular diseases (affecting the blood vessels in the brain), are closely related due to common underlying pathologies. These conditions are often grouped under the umbrella term “CVDs.” While CVD is recognized as a significant cause of non-cancer deaths among long-term cancer survivors, the exact risk level for CVD mortality varies across studies. To date, no meta-analysis has comprehensively quantified this risk across different populations of cancer survivors or specific types of CVD. Therefore, this study aimed to conduct a systematic review and meta-analysis to compare the patterns and levels of CVD mortality between cancer survivors and the general population.
Method
This systematic review and meta-analysis followed the PRISMA guidelines to investigate the risk of cardiovascular disease (CVD) mortality among individuals with cancer compared to the general population. The review sought to determine whether this risk varied by cancer type.
Studies were included if they featured patients of all ages diagnosed with any type of cancer and reported CVD-related mortality data using standardized mortality ratios (SMRs). Eligible studies were peer-reviewed, in English, and published after 2000. Various study designs were considered, including cohort and case-control studies, while pediatric-focused studies were excluded. Additionally, studies involving individuals who developed secondary cancers after being treated for pediatric cancers were allowed. The databases Medline (OVID), CINAHL (EBSCOhost), and SCOPUS were searched from January 2000 to July 2023 using a broad range of cancer, mortality, and SMR-related terms. The reference lists of included studies were also examined. Screening of titles, abstracts, and full texts was done independently by two authors, with any discrepancies resolved by a third author. The study protocol was registered with PROSPERO (CRD42020209215).
The meta-analysis was conducted using Stata 17, and SMRs across all cancer and CVD types were pooled using random-effects models to account for heterogeneity. Overall pooled estimates and confidence intervals were reported for specific CVD types, including heart disease and stroke, across broad cancer categories. The summary effect sizes were calculated using the DerSimonian-Laird random-effects model.
Data analysis
The primary outcome of interest was the risk of non-cancer death from CVD, expressed as SMRs. SMRs compare observed CVD mortality in the cancer population to what would be expected in a similar reference population. An SMR greater than one indicates a higher-than-expected mortality rate. Data related to CVD were collected using the International Classification of Diseases (ICD) codes. Nine broad categories of CVDs were considered, including diseases of the heart, ischemic heart disease, heart failure, hypertension, atherosclerosis, and stroke.
Data were extracted independently by two authors, capturing information such as the first author’s surname, year of publication, country, study design, population demographics, cancer types, and SMR values. The results were categorized by 15 different cancer types, including breast, gastrointestinal, lung, and hematologic cancers. Study quality was assessed using the Newcastle-Ottawa Scale (NOS), which evaluates non-randomized studies based on selection, comparability, and outcome. Studies scored between 6 and 9 on the NOS, but no study was excluded based on quality alone.
Result
The initial database search was conducted on July 19, 2021, with an update on July 16, 2023. A total of 4501 records were identified, and after removing duplicates, 1725 records remained. Following full-text screening, 136 studies met the inclusion criteria. These studies spanned 16 countries across four continents, with the majority (98 studies, 72%) conducted in the United States. Europe contributed 22 studies (16%), while Australia and South Korea had four each, and Japan had three. Other studies were from China, Canada, and cross-national settings. No studies originated from Africa or South America.
The studies were published between 2000 and 2023, with over three-quarters (105 studies) published in the past five years. Most studies (134 out of 136) used a retrospective cohort design, with one prospective cohort study and one secondary data analysis from a randomized screening trial. The most common data source was the SEER program, used by 95 studies (70%), while cancer registries were used in 29 studies (21%).
Study periods ranged from 1943 to 2020, and sample sizes varied from 157 to over 7.5 million individuals. Approximately half of the study populations were female, and five studies focused on adolescents or young adults aged 15 to 39 years. The most frequently studied cancer types were gastrointestinal (34 studies), hematological (31 studies), breast (29 studies), urological (25 studies), gynecological (23 studies), and prostate cancer (22 studies).
The studies reported 876 standardized mortality ratios (SMRs) for cardiovascular diseases (CVDs) across 15 cancer sites. Of these, 61% (535 SMRs) showed an elevated risk of CVD mortality (SMR >1), 12% (109 SMRs) indicated a reduced risk (SMR <1), and 27% (232 SMRs) were not statistically significant. A meta-analysis revealed an increased risk of CVD mortality in cancer patients compared to the general population (SMR = 1.55, 95% CI = 1.40–1.72). This increased risk was consistent for heart disease (SMR = 1.36) and stroke (SMR = 1.56).
For specific cancers, the risk of CVD mortality ranged from 1.14 for testicular cancer to 2.82 for brain and CNS tumors, with non-significant results for thyroid and prostate cancer. Further analysis showed varying risks for different types of CVDs across cancer types. For heart disease, SMRs ranged from 1.32 for all cancers to 2.46 for hematological cancers, while for stroke, SMRs ranged from 1.25 for urological cancers to 5.95 for brain tumors.
Most studies (86%) used reference populations controlled for age and sex, while 11% controlled for either age or sex, and 3% provided no relevant information.
Conclusion
This systematic review and meta-analysis represents a comprehensive synthesis of the burden of cardiovascular disease (CVD) mortality among cancer populations compared to the general population, integrating findings from 136 studies across 16 countries over the past 23 years. The analysis revealed a notable 55% increase in CVD mortality among cancer patients relative to the general population, with significant variability depending on cancer type and CVD.
The highest relative risk was observed in individuals with brain and central nervous system (CNS) tumors (Standardized Mortality Ratio [SMR]: 2.82), likely linked to an elevated incidence of cerebrovascular disease and thrombosis within this group. This heightened risk underscores the critical concern of “double jeopardy,” given the already high cancer-specific mortality rates. Hematological cancers also exhibited a pronounced risk of mortality from ischemic heart disease, possibly due to the cardiotoxic effects of cancer treatments.
Interestingly, thyroid and prostate cancers showed no significant differences in CVD mortality risks compared to the general population. The absence of cardiotoxic treatments in thyroid cancer, primarily managed through surgery, may explain this finding. Similarly, early-stage prostate cancer, which is often treated with surgery or radiation, may not carry the same risks as advanced stages subjected to cardiotoxic therapies like androgen deprivation therapy.
Variability in mortality risks among different cancers may reflect differing distributions of cardiotoxic treatments and background CVD risk factors. Certain anticancer drugs, such as anthracyclines and taxanes, have been associated with adverse cardiovascular events, contributing to the increased risk in populations with hematological malignancies.
The study also highlighted that CVD mortality risk was elevated across all cancer types, with a 36% increase for heart disease and a 56% increase for cerebrovascular diseases. The risk was particularly high among patients with brain and CNS tumors (SMR: 5.95), gastrointestinal cancers (SMR: 2.31), and head and neck cancers (SMR: 1.81). This may be attributed to the direct impacts of tumors on cerebral vasculature or cancer-associated hypercoagulability.
Despite advancements in cardio-oncology primarily focusing on cancers with better prognoses, the findings suggest a need for increased attention on high-risk populations such as those with brain and CNS tumors. Interventions aimed at managing cardiovascular risk factors through lifestyle modifications may help mitigate the elevated risk of CVD mortality.
The study acknowledged limitations, including challenges in interpreting CVD data due to variability in reporting and study design. Most studies were conducted in the United States, raising questions about the generalizability of the findings across different populations and healthcare contexts.
In summary, this research highlights the elevated risk of CVD mortality among cancer survivors, emphasizing the need for further investigation into the mechanisms contributing to this increased risk and the development of targeted interventions. Enhanced data reporting and standardization in research practices will be vital for improving understanding and management of CVD in cancer populations. Future research should delve deeper into the underlying factors driving this risk to inform effective preventive strategies.
