Graphical abstract
Keywords: Cardio-oncology, Cardiovascular disease death, Non-metastatic cancers, SEER
Highlights
-
•
The risk of CVD death in patients with non-metastatic cancer varies by cancer stage, cancer site, and survival time.
-
•
The risk of CVD death gradually exceeded that of primary neoplasm with longer follow-up from cancer diagnosis among patients with localized cancer of 14 of 21 individual cancers (high competing risk group).
-
•
Patients with non-metastatic cancer had a higher CVD death risk than the general US population.
Abstract
Introduction
Previous studies on cardiovascular disease (CVD) death risk in cancer patients mostly focused on overall cancer, age subgroups and single cancers.
Objectives
To assess the CVD death risk in non-metastatic cancer patients at 21 cancer sites.
Methods
A total of 1,672,561 non-metastatic cancer patients from Surveillance, Epidemiology, and End Results (SEER) datebase (1975–2018) were included in this population-based study, with a median follow-up of 12·7 years. The risk of CVD deaths was assessed using proportions, competing-risk regression, absolute excess risks (AERs), and standardized mortality ratios (SMRs).
Results
In patients with localized cancers, the proportion of CVD death and cumulative mortality from CVD in the high-competing risk group (14 of 21 unique cancers) surpassed that of primary neoplasm after cancer diagnosis. The SMRs and AERs of CVD were found higher in patients with non-metastatic cancer than the general US population (SMR 1·96 [95 %CI, 1·95-1·97]–19·85[95 %CI, 16·69-23·44]; AER 5·77–210·48), heart disease (SMR 1·94[95 %CI, 1·93-1·95]–19·25[95 %CI, 15·76-23·29]; AER 4·36–159·10) and cerebrovascular disease (SMR 2·05[95 %CI, 2·02-2·08]–24·71[95 %CI, 16·28-35·96]; AER 1·01–37·44) deaths. In the high-competing risk group, CVD-related SMR in patients with localized stage cancer increased with survival time but followed a reverse-dipper pattern in the low-competing risk group (7 of 21 cancers). The high-competing risk group had higher CVD-related death risks than the low-competing risk group.
Conclusion
The CVD death risk in patients with non-metastatic cancer varied by cancer stage, site and survival time. The risk of CVD mortality is higher in 14 out of 21 localized cancers (high-competing cancers). Targeted strategies for CVD management in non-metastatic cancer patients are needed.
Introduction
The number of non-metastatic cancer survivors is steadily growing owing to rapid technological advances in early cancer diagnosis and treatment. Non-metastatic cancer patients constitute about 50 % of cancer patients in the United States (US) [1], [2], [3]. Compared to metastatic cancer patients, those with non-metastatic cancers have better five-year survival rates of over 85 % [1], [2], [3]. However, cancer patients with better survival are more likely to die from noncancer-related death [4].
Cardiovascular disease (CVD) is the primary cause of death worldwide [5]. With improved prognosis, more patients with non-metastatic cancer have greater exposure to cardiovascular risks, such as cardiotoxicities of anticancer treatment and shared risk factors and mechanisms directly related to cancer biology [6], [7], [8], [9], [10], [11]. There are significant research concerns about the growing risk of CVD death among non-metastatic cancer patients.
The latest European Society of Cardiology (ESC) guidelines for cardio-oncology raise a scientific question regarding whether there are differences in CVD risk for patients at different stages of cancer [11]. Previous studies have mainly focused on the risk of CVD mortality of overall cancer survivors by age [4], [12], [13], [14], [15], [16] and adolescent and young adult cancers [17], [18], [19], but neglected that cardiovascular outcomes might differ significantly by cancer stage. In addition, studies on the risk of CVD death among non-metastatic cancer patients mostly focused on a single cancer type, especially breast cancer [20], [21]. However, the risk of CVD death among patients with non-metastatic cancers is likely to vary across cancer types due to cancer heterogeneity, warranting further investigation [19], [22]. In addition, uncertainties persist about the risk of CVD death among non-metastatic cancer patients by specific subgroups such as cancer stage,sex, and race.
Thus, we conducted a comprehensive evaluation to investigate the risk of CVD death among non-metastatic cancer survivors of 21 cancer sites. These results can provide a scientific basis from an epidemiological perspective for individualized managing non-metastatic cancer patients, and help healthcare systems to mitigate the burden of CVD death in patients with non-metastatic cancer.
Methods
Data source
The data for this study was obtained from the Surveillance, Epidemiology, and End Results (SEER) Program. The SEER program is a publicly available, authoritative database in US, and the data is quality assured due to its systematic, standardized, and periodic data collection procedure [23] (Supplementary Methods). Ethical approval was not required because of the publicly available data[12].
Study population
We identified patients diagnosed with non-metastatic cancer according to the International Classification of Diseases-10 (ICD-10) from SEER-9 (1975–2018). The 21 cancer types where the primary tumour was found are listed in the Online supplementary material Table S1. Non-metastatic cancer included any neoplasm that has not spread to distant organs from the primary tumour site [24], [25]. Patients from SEER-9 (1975–2018) were categorized using the SEER stage which the non-metastatic stage included localized and regional [25]. A localized stage was defined as an invasive neoplasm limited entirely to the primary organ [25]. In contrast, a regional stage involves the progression of the neoplasm into the surrounding tissues, organs or regional lymph nodes [25]. Given the long time span of records included in this study, we separately analyzed data from a more recent period between 2004 and 2015 to validate our results obtained from records from 1975 to 2018. This cohort of records was obtained from the SEER-18 program and contained some overlapping records with the SEER-9 cohort (but not overlapping use of data, see Supplementary Methods). SEER-18 categorized neoplasms according to the sixth edition of the American Joint Committee on Cancer (AJCC)[24] (specific classification criteria are reported in Online supplementary material Table S1). We therefore were able to compare data obtained using SEER stage (SEER-9:1975–2018) and AJCC stage (SEER-18: 2004–2015). The inclusion criteria for the study were: (1) cancer types defined with ICD-10 codes; (2) histological diagnosis; (3) active follow-up with survival time; (4) only one primary cancer; and (5) a clear cause of death. We excluded metastatic cancers, records with unknown race [26], and where follow-up time was less than two months [12]. (Online supplementary material Figure S1). The information of patients with metastatic cancer was shown in Supplementary Methods.
Study variables and outcome
We extracted data on the SEER stage (localized and regional), age at diagnosis (0–34, 35–64, or 65 + years)[18], [27], race (White, Black, or others) [12], sex (female or male), tumour grade (high, low, other, or unknown) [28], year of diagnosis (1975–1994, or 1995–2018), year after diagnosis (<1, 1–3, 3–5, 5–10, 10–15 or 15 + years; the time interval include the lower limit), surgery (yes, no, or unknown), chemotherapy (yes or no evidence) and radiotherapy (yes or no evidence)[29], [30].
The cause of death in the SEER database was classified using ICD-10 codes and was derived from the National Center for Health Statistics [31], [32]. Death from any cause was the primary outcome and was categorized into death from primary neoplasm, other neoplasms, CVD[12], and other non-neoplasms, among which CVD including heart diseases, cerebrovascular diseases, hypertension without heart disease, atherosclerosis, aortic aneurysm and dissection, and other disease of arteries, arterioles, capillaries[12].(Online supplementary material Table S2).
Participants were followed up from the date of cancer diagnosis to death between 1 January 1975 and 31 December 2018 for the patients coded in SEER-9 and between 1 January 2004 and 31 December 2016 for those in SEER-18. Individuals who were lost to follow-up or survived the last follow-up were considered as censored observations.
Statistical analysis
The Chi-square test was applied to assess the association between categorical variables at baseline [33]. Proportion of deaths and cumulative mortality were calculated in comparisons among different causes of death. The proportion of deaths was used to describe the most common cause of death, which was defined as the mortality of cause-specific deaths to the general deaths [34]. We estimated cumulative mortality using competing-risk regression to investigate the competing relationship between CVD and primary neoplasm deaths [35]. Higher competing risks are defined as whether cumulative mortality from CVD exceeds that from primary neoplasm. Multivariate Fine-Gray competing-risk regressions were built with CVD death risk among non-metastatic pancreatic cancer as the reference (due to the lowest CVD death risk in pancreatic cancer) and adjusted for age at diagnosis, sex, race, grade, surgery, year of diagnosis, treatment with chemotherapy and radiotherapy (Supplementary Methods) [35]. The proportional hazards assumption was assessed graphically by plotting the cumulative mortality curves and was satisfied [36]. As reported previously, absolute excess risks (AERs) and standardized mortality ratios (SMRs) were calculated to compare CVD mortality in non-metastatic cancer patients with that in the general US population [12], [32], [37]. More details on the statistical methods are shown in Supplementary Methods.The R programming software version 3.4.4 was applied for statistical analysis. All p values < 0·05 were considered significant.
Results
Patient characteristics
A total of 1,672,561 patients with non-metastatic cancer from 21 cancer sites were identified from 1975 to 2018 (SEER-9) and contributed 14,157,822·3 person-years at risk. Most of the patients were 65 + years old (48·6%), female (54·4%), white (83·1%), and had localized (65·7%) and low-grade (46·8%) cancer (Table 1). The median follow-up period was 12·7 years [interquartile range (IQR) 6·3–20·6]. 1,793,619 non-metastatic cancer patients were identified from 2004 to 2015 (SEER-18), 36·5% were at AJCC I, 41·8% were at AJCC II, and 18·6% were at AJCC III, contributing a total of 9,264,742·5 person-years at risk (Online supplementary material Table S3). The median follow-up period was 6·3 years [interquartile range (IQR) 3·5–9·4]. The baseline of metastatic cancer patients was shown in supplementary material Table S4.
Table 1.
Baseline characteristics of non-metastatic cancer patients from 1975 to 2018.
| Characteristic | Total |
SEER stage (n/%) |
P value | |
|---|---|---|---|---|
| Localized | Regional | |||
| Overall | 1 672 561 | 1 098 466 (65.7) | 574 095 (34.3) | |
| Age at diagnosis | ||||
| 0–34 years | 54 148 (3.2) | 38 556 (3.5) | 15 592 (2.7) | < 0.001 |
| 35–64 years | 805 081 (48.1) | 519 636 (47.3) | 285 445 (49.7) | |
| 65 + years | 813 332 (48.6) | 540 274 (49.2) | 273 058 (47.6) | |
| Sex | ||||
| Male | 762 630 (45.6) | 519 675 (47.3) | 242 955 (42.3) | < 0.001 |
| Female | 909 931 (54.4) | 578 791 (52.7) | 331 140 (57.7) | |
| Race | ||||
| White | 1389 187 (83.1) | 920 686 (83.8) | 468 501 (81.6) | < 0.001 |
| Black | 157 615 (9.4) | 98 876 (9.0) | 58 739 (10.2) | |
| Other* | 125 759 (7.5) | 78 904 (7.2) | 46 855 (8.2) | |
| Grade | ||||
| Low | 783 077 (46.8) | 553 944 (50.4) | 229 133 (39.9) | < 0.001 |
| High | 440 884 (26.4) | 236 031 (21.5) | 204 853 (35.7) | |
| Other# | 4309 (0.3) | 2976 (0.3) | 1333 (0.2) | |
| Unknown | 444 291 (26.6) | 305 515 (27.8) | 138 776 (24.2) | |
| Year of diagnosis | ||||
| 1975–1994 | 454 929(27.2) | 257 222(23.4) | 197 707(34.4) | < 0.001 |
| 1995–2018 | 1 217 632(72.8) | 841 244(76.6) | 376 388(65.6) | |
| Year after diagnosis& | ||||
| < 1 year | 194 174 (11.6) | 86 093 (7.8) | 108 081 (18.8) | < 0.001 |
| 1–3 years | 314 511 (18.8) | 166 961 (15.2) | 147 550 (25.7) | |
| 3–5 years | 214 368 (12.8) | 138 956 (12.7) | 75 412 (13.1) | |
| 5–10 years | 381 120 (22.8) | 275 310 (25.1) | 105 810 (18.4) | |
| 10–15 years | 253 638 (15.2) | 191 998 (17.5) | 61 640 (10.7) | |
| ≥ 15 years | 314 750 (18.8) | 239 148 (21.8) | 75 602 (13.2) | |
| Surgery | ||||
| Yes | 1 341 816 (80.2) | 873 787 (79.5) | 468 029 (81.5) | < 0.001 |
| No | 319 984 (19.1) | 218 054 (19.9) | 101 930 (17.8) | |
| Unknown | 10 761 (0.6) | 6625 (0.6) | 4136 (0.7) | |
| Chemotherapy | ||||
| Yes | 1,316,722 (78.7) | 983,222 (89.5) | 333,500 (58.1) | < 0.001 |
| No evidence | 355,839 (21.3) | 115,244 (10.5) | 240,595 (41.9) | |
| Radiotherapy | ||||
| Yes | 1,146,432 (68.5) | 802,454 (73.1) | 343,978 (59.9) | < 0.001 |
| No evidence | 526,129 (31.5) | 296,012 (26.9) | 230,117 (40.1) | |
| Cancer site | ||||
| Head and neck | 87 743 (5.2) | 39 771 (3.6) | 47 972 (8.4) | < 0.001 |
| Lung | 105 355 (6.3) | 42 682 (3.9) | 62 673 (10.9) | |
| Breast (female) | 441 328 (26.4) | 287 154 (26.1) | 154 174 (26.9) | |
| Esophagus | 16 363 (1.0) | 7088 (0.6) | 9275 (1.6) | |
| Stomach | 32 654 (2.0) | 13 991 (1.3) | 18 663 (3.3) | |
| Liver & intrahepatic bile duct | 15 298 (0.9) | 9477 (0.9) | 5821 (1.0) | |
| Gallbladder | 4644 (0.3) | 2380 (0.2) | 2264 (0.4) | |
| Pancreas | 20 958 (1.3) | 4743 (0.4) | 16 215 (2.8) | |
| Colorectum | 248 489 (14.9) | 123 199 (11.2) | 125 290 (21.8) | |
| Anus | 7821 (0.5) | 4527 (0.4) | 3294 (0.6) | |
| Kidney and renal pelvis | 59 263 (3.5) | 43 483 (4.0) | 15 780 (2.7) | |
| Ureter | 1872 (0.1) | 749 (0.1) | 1123 (0.2) | |
| Bladder | 106 705 (6.4) | 84 289 (7.7) | 22 416 (3.9) | |
| Ovary | 15 662 (0.9) | 11 224 (1.0) | 4438 (0.8) | |
| Endometrium | 66 137 (4.0) | 53 305 (4.9) | 12 832 (2.2) | |
| Cervix uteri | 31 570 (1.9) | 19 202 (1.7) | 12 368 (2.2) | |
| Vulva | 7113 (0.4) | 4953 (0.5) | 2160 (0.4) | |
| Prostate gland | 285 302 (17.1) | 243 372 (22.2) | 41 930 (7.3) | |
| Penis | 2100 (0.1) | 1397 (0.1) | 703 (0.1) | |
| Bone | 6246 (0.4) | 3430 (0.3) | 2816 (0.5) | |
| Melanoma of the skin | 109 938 (6.6) | 98 050 (8.9) | 11 888 (2.1) | |
*Other includes American Indian/Alaska Native and Asian/Pacific Islander.
# Other includes pre-B, B-precursor and B-cell.
& The time interval does not include the upper limit.
Proportion of deaths
Patients with localized cancers were more likely to die from non-neoplasms than primary neoplasms, and CVD was the major cause of death (34·3%). However, this trend was not observed in patients with regional neoplasms (Fig. 1A). Heart disease (75·5%) and cerebrovascular diseases (17·5%) were the most common causes of CVD death, and this trend was consistent throughout the follow-up period, irrespective of the cancer stage (Fig. 1B), and further studies (SMR and HR) were additionally conducted for these two causes. Focusing on the proportion of deaths not caused by the primary or other neoplasms, CVDs were the primary cause of non-neoplasm deaths (50·4%) in non-metastatic cancer patients (Fig. 1C). The proportion of CVD deaths was only 3·6 % in patients with metastatic cancer (Online supplementary material Figure S2).
Fig. 1.
The proportion of death in patients with non-metastatic cancer. A. All causes of deaths; B. Causes of CVD deaths; C.Causes of non-neoplasms deaths. Abbreviations: CVD, cardiovascular disease.
In subgroup analysis of clinical features such as age at diagnosis, race, sex, tumour grade and year of diagnosis, a higher proportion of CVD deaths only occurred in patients with localized cancers. CVD deaths were more frequent than death from primary neoplasm in patients who are aged 65 + years, in white ethnicity, in both male and female, and in patients diagnosed at 1975–1994 and 1995–2018 (Online supplementary material Figure S3).
Online supplementary material Figure S4 depicts the proportions of deaths by different primary cancer sites. The proportion of primary neoplasm deaths declined, while that of CVD increased with survival time. For localized cancers, the proportion of CVD death surpassed that of primary neoplasm with survival time, especially in cancer of the head and neck, breast, colorectum, anus, kidney and renal pelvis, ureter, bladder, ovary, endometrium, cervix uteri, vulva, penis, prostate gland, and melanoma of the skin. The proportion of deaths across other cancer sites, including the lung, oesophagus, stomach, liver and intrahepatic bile duct, gallbladder, pancreas, and bone are presented in the Online supplementary material Figure S5.
Cumulative mortality from CVD versus primary neoplasm
We analyzed cumulative mortality against years after primary neoplasm diagnosis and found that in patients with localized cancer, the highest cumulative mortality was not the primary neoplasm or other neoplasms, but mostly due to CVD and other non-neoplastic causes (Online supplementary material Figure S6). CVDs in these patients become a primary cause of death within 10–20 years after cancer diagnosis regardless of race, sex, grade, and year of diagnosis. Higher CVD mortality with increasing years after diagnosis was also evident in other age groups (subgroup aged 35–64 and 65 + but not 0–34) (Online supplementary material Figure S7). In patients with regional cancer and metastatic cancer, primary neoplasm remained the highest cumulative mortality risk (Online supplementary material Figure S8 and Figure S9).
Subgroup analyses by sites of primary neoplasms are presented in Fig. 2. Among patients with localized cancers, the cumulative mortality from CVD gradually exceeded that from primary neoplasm in 14 individual cancers after cancer diagnosis. These 14 cancers included cancer of the head and neck, breast (female), colorectum, anus, kidney and renal pelvis, ureter, bladder, endometrium, vulva, prostate gland, penis, melanoma of the skin, ovary, and cervix uteri (Fig. 2). In contrast, the cumulative mortality from CVD did not surpass that from primary neoplasm in the seven cancer sites, including lung, oesophagus, stomach, liver and intrahepatic bile duct, gallbladder, pancreas, and bone cancers (Online supplementary material Figure S10). Therefore, of the 21 cancers, 14 and seven were divided into high- and low-competing risk groups, respectively. In patients with regional cancers, primary neoplasm conferred the highest cumulative mortality among the 21 cancers (Online supplementary material Figure S11).
Fig. 2.
The cumulative mortality in patients with localized cancer based on cancer sites (high competing risk group). Abbreviations: CVD, cardiovascular disease.
Risk of CVD death compared to the general population
The SMRs and AERs of CVD were found higher in patients with non-metastatic cancer than the general US population (SMR 1·96 [95 %CI, 1·95-1·97]–19·85[95 %CI, 16·69-23·44]; AER 5·77–210·48), heart disease (SMR 1·94[95 %CI, 1·93-1·95]–19·25[95 %CI, 15·76-23·29]; AER 4·36–159·10) and cerebrovascular disease (SMR 2·05[95 %CI, 2·02-2·08]–24·71[95 %CI, 16·28-35·96]; AER 1·01–37·44) regardless of stage, sex, age, race, grade and year of diagnosis (Online supplementary material Table S5). Individual analysis of SMR from 1975 to 2018 for CVD, heart disease and cerebrovascular disease revealed that the SMR of CVD, heart disease and cereborovascular disease for patients with localized cancers all increased after 3 years post-diagnosis (<1 to 15 + years after diagnosis 4·43-7·64, 4·46-7·36, 4·33-8·56 respectively) in the high competing risk group (Fig. 3A). In the low-competing risk group, the SMR by CVD, heart disease and cerebrovascular disease compared to the general US population is 9·24, 9·36 and 8·40 < 1 year after diagnosis and followed a reverse-dipper pattern, with a prominent drop, a gradual rise and finally, a slight fall (Fig. 3B).
Fig. 3.
The risk of cardiovascular death by competing risk and years after diagnosis in patients with non-metastatic cancer (localized stage and AJCC I). The resulting SMR from the two staging systems follows the same trend: (A and B) SEER stage from 1975 to 2018 (SEER-9); (C and D) AJCC stage from 2004 to 2015 (SEER-18). The data from 1975 to 2018 contributed to observing long-term outcomes (A and B), while the data from 2004 to 2015 reflected modern medical practices (C and D).The time interval does not include the upper limit.
Similar to the SEER-9 cohort, the risk of CVD, heart disease and cerebrovascular death in AJCC I stage cancer patients (SEER-18) varied by survival time and competing risk group. In the high-competing risk group, the trend of CVD, heart disease and cerebrovascular SMR increased 3 years after diagnosis (Fig. 3C). In the low-competing risk group, SMR related to CVD, heart disease, and cerebrovascular disease did not increase with follow up time (Fig. 3D). Subgroup analysis revealed that there is an increase in SMRs and AERs of CVD, heart disease and cerebrovascular disease regardless of sex, age at diagnosis, race and tumour grade in patients with non-metastatic cancer and patients with tumour of AJCC stages I-IV (SEER-18) (Online supplementary material Table S6).The trend of CVD-related SMR in individual cancer sites was generally similar to that of all sites combined (Online supplementary material Table S7-S12).
Cause-specific hazard ratios among 21 cancer sites
For localized cancers,the risk of CVD, heart disease and cerebrovascular disease death in the high-competing risk group about 3·53–5·79 times, 3·19–5·23 times, and 3·79–6·00 times higher, while those in low-competing risk group were 1·50–3·43 times, 1·41–3·10 times and 1·79–3·50 times higher, respectively. The risk of CVD death varied by cancer site for the anus (HR 5·79, [95 %CI, 4·88-6·87]), ureter (HR 5·50, [95 %CI, 4·50-6·73]), bladder (HR 5·42, [95 %CI, 4·66-6·31]), penis (HR 5·41, [95 %CI, 4·49-6·52]), head and neck (HR 5·31, [95 %CI, 4·56-6·18]), vulva (HR 5·22, [95 %CI, 4·44-6·14]), colorectum (HR 5·00, [95 %CI, 4·30-5·82]), breast (HR 4·62, [95 %CI, 3·97-5·38]), kidney and renal pelvis (HR 4·50, [95 %CI, 3·86-5·25]), endometrium (HR 4·14, [95 %CI, 3·55-4·83]), ovary (HR 4·06, [95 %CI, 3·46-4·77]), cervix uteri (HR 3·80, [95 %CI, 3·24-4·46]), melanoma of the skin (HR 3·55, [95 %CI, 3·05-4·14]), prostate gland (HR 3·53, [95 %CI, 3·02-4·12]), stomach (HR 3·43, [95 %CI, 2·94-4·01]), gallbladder (HR 3·19, [95 %CI, 2·66-3·82]), bone (HR 3·18, [95 %CI, 2·55-3·96]), lung (HR 3·03, [95 %CI, 2·60-3·52]), esophagus (HR 2·25, [95 %CI, 1·90-2·66]), liver and intrahepatic bile duct (HR 1·50, [95 %CI, 1·25-1·80])(Fig. 4).
Fig. 4.
Cause-specific hazard ratios by 21 cancer sites with localized stage. Abbreviations: CVD, cardiovascular disease
Discussion
This is the first comprehensive evaluation of CVD death risk among patients with non-metastatic cancer from 21 cancer sites, using population-based data with over 40 years of follow-up. Our study found that CVD death was the most clinically significant competing risk in most non-metastatic cancers. In 14 of the 21 localized cancers (cancer sites in head and neck, breast(female), colorectum, anus, kidney and renal pelvis, ureter, bladder, ovary, endometrium, cervix, vulva, prostate gland, penis, and melanoma of the skin), CVD gradually surpassed primary neoplasm as the primary cause of death after cancer diagnosis and mortality due to CVD increased as survival time increased. In addition, the risk of CVD mortality was higher in patients with non-metastatic cancer than in the general US population. It was found for the first time that the risk of CVD mortality varied by cancer stage, cancer site and survival time (Graphical abstract).
Our data showed that with longer follow-up, the risk of CVD death exceeded that of primary neoplasm among 14 of the 21 localized cancers (high competing risk group). Our results are similar to previous data showing that 7 of 14 cancers (bladder, prostate, endometrial, colorectum, melanoma, kidney, oral cavity and pharynx) had a higher proportion of heart disease death [15]. Another population-based study reported that 7 of 28 cancers had a higher proportion of CVD death than primary cancer [12]. We found that localized but not regional cancers had a higher proportion of CVD death compared to death caused by primary neoplasms in 14 out of 21 primary neoplasm sites, demonstrating that the risk of CVD is particularly high in patients localized neoplasms. However, this trend was not observed in the competing risk analysis [12]. This might be because previous studies conducted analyses on the overall stage cancer population, thereby concealing essential subgroup features [4], [12], [15] such as patients with localized cancer. In addition, our results were consistent with previous studies using competing risk mode, suggesting that the risk of CVD death surpassed primary neoplasm death in early-stage breast cancer [20], [21] and early-stage endometrial cancer[34].
For the first time, we reported the varying trends in CVD-related SMR by competing risk groups, and survival time. Patients in the low-competing risk group (consisting of cancers with poor prognoses) are more exposed to acute cardiotoxicity after anticancer treatments[38], [39] compared to patients in the high-competing risk group. Psychological burden also plays an important role in patient well-being[40].These factors contribute to the high SMR for CVD mortality within the first year of diagnosis. Unsurprisingly, the SMR drops after the initial high risk caused by acute drug toxicity. As survival time increased, patients in the low-competing risk group are at cumulative risk of cardiac events [41], [42] and CVD death in the long term[43]. For example, the risk of cardiac-specific mortality in patients with limited-stage small cell lung cancer was higher at 5 years after diagnosis[43]. Conversely, since the diagnosis of cancer, patients in the high-competing risk group (consisting of cancers with good prognosis) are exposed to shared cardiovascular risk factors and chronic cardiotoxicity from anticancer therapy, resulting in the gradual accumulation of risk of CVD death [6], [44]. A population-based study with data on 28 cancer sites reported that the highest risk for CVD death was observed in the first year of cancer diagnosis [12]. However, by disaggregating our analysis by cancer site and stage, we demonstrated that analysis of overall cancers is not applicable to all cancer sites [12].
In patients with localized cancer, the risks of death from CVD, heart disease, and cerebrovascular disease varied by cancer sites, with anus, ureter, bladder, penis, head and neck, vulva, and colorectum cancers conferring the higher risk, and cancers of the pancreas, liver and intrahepatic bile duct and esophagus conferring the lower risk. Interestingly, early-stage breast cancer patients which were considered one of cardio-oncology research hotspots [20], [45], do not have the highest HR for CVD death among the 21 cancer sites. Our results highlighted that the scope of cancer sites in cardio-oncology should be expanded, particularly the risk of CVD death in urinary, vulva, head and neck and colorectum cancers.
Increased risk of CVD death in patients with localized cancer may be multifactorial. Shared risk factors (e.g., smoking, hypertension and diabetes) for CVD and cancer promote the risk of CVD death [6]. Cancer patients with localized tumours with longer survival times are more likely to be exposed to these risk factors and thus have higher risks of CVD death, especially the high-competing risk cancers who has longer survival time. Additionally, the cardiotoxicity of cancer treatment increases the risk of CVD death [10], especially the low-competing risk cancers who requires extra treatment. Finally, increasing evidence suggests that cancer damages systemic vasculature and the heart. Cancer-induced inflammation, called neutrophil extracellular trap (NET), accumulates in heart and vasculature, resulting in cardiovascular dysfunction [46], [47]. The heterogeneity in the risk of CVD death by cancer types may also be multifactorial: (1) the prognostic heterogeneity in different cancer; (2) the heterogeneity in cardiotoxic anticancer of different cancer; (3) the CVD risk heterogeneity in different cancers[48].”
The primary strength of this study was the large population and breadth that allowed us to explore the association between the risk of CVD death and different cancer types, with sufficient statistical power to conduct comprehensive subgroup analyses in different cancer stages. Our results were verified in the SEER stage (SEER-9: 1975–2018) and AJCC stage (SEER-18: 2004–2015) to minimize the bias from different stage systems, follow-up time and year of diagnosis. The resulting SMR follows the same trend, indicating that our results are applicable to both staging systems. The long-term follow-up from 1975 to 2018 contributed to observing long-term outcomes [4], [29], [48], [49], while the cohort from 2004 to 2015 most likely reflected modern medical practices.
Limitations
There are several limitations in this study. First, the wide time period of the data (1975–2018) may have led to a confounding effect from calendar year periods, as both oncological and cardiovascular treatments changed over time. Nonetheless, subgroup analyses according to the year of diagnosis, multivariate analyses adjusted by year of diagnosis, and data from 2004 to 2015 (SEER 18), which reflects modern medical practices, demonstrated that these confounding factors did not have an effect on the study outcome. Second, the lack of immunotherapy and specific information on chemotherapy and radiotherapy in SEER prevented further analysis of their impact on the risk of CVD death. Third, the SEER did not offer variables on cardiovascular comorbidities, cardioprotective treatment, and cardiometabolic risk factors, limiting further exploration [12], [18], [48]. Nevertheless, our results provide a description of the risk of CVD death instead of causal result. Fourth, the risk of more specific heart disease subtypes is still unknown in patients with non-metastatic cancer, due to the lack of detailed death classification of heart disease. Fifth, those patients with more severe disease are more likely to drop out of the study, and the event rate may be underestimated. Finally, the lack of information on recurrence and metastasis during follow-up in the SEER limited further exploration. Further sensitivity analyses were needed in participants without metastasis and recurrence during follow-up.
Conclusion
The risk of CVD death in patients with non-metastatic cancer varies by cancer stage, cancer site, and survival time. In particular, the risk of CVD death gradually exceeded that of primary neoplasm with longer follow-up from cancer diagnosis among patients with localized cancer of 14 of 21 individual cancers (high competing risk group). In addition, patients with non-metastatic cancer had a higher CVD death risk than the general US population. The cornerstone of cardio-oncology entails managing CVD risk and the interaction of CVD and cancer. Our study consistently elucidates the interplay between CVD and cancer, underscoring the critical significance of considering competing risks and enhancing the comprehension of cardio-oncology. Strategies to prevent, manage, and minimize the risk of CVD death are needed for the growing population of patients with non-metastatic cancer which contributes to improving their survival time and quality of life.
Data sharing
The datasets in our study are publicly available from the SEER database (https://seer.cancer.gov).
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We thank the research staff from SEER and the National Cancer Institute in US.
Funding
National Natural Science Foundation of China (Nos. 81871504, 32171355 and 82172103), China Postdoctoral Science Foundation (No.2023M741567), National key specialist funding cultivation fund (No. Z202304), Guangdong Basic and Applied Basic Research Foundation (No. 2023A1515110724) and Fundo para o Desenvolvimento das Ciências e da Tecnologia (FDCT 0055/2022/A1).
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.jare.2024.03.017.
Contributor Information
Kang Zhang, Email: kang.zhang@gmail.com.
Caiwen Ou, Email: oucaiwen@smu.edu.cn.
Appendix A. Supplementary material
The following are the Supplementary data to this article:
References
- 1.Miller K.D., Nogueira L., Mariotto A.B., Rowland J.H., Yabroff K.R., Alfano C.M., et al. Cancer treatment and survivorship statistics, 2019. CA Cancer J Clin. 2019;69(5):363–385. doi: 10.3322/caac.21565. [DOI] [PubMed] [Google Scholar]
- 2.Siegel R.L., Miller K.D., Fuchs H.E., Jemal A. Cancer statistics, 2021. CA Cancer J Clin. 2021;71(1):7–33. doi: 10.3322/caac.21654. [DOI] [PubMed] [Google Scholar]
- 3.Siegel R.L., Miller K.D., Jemal A. Cancer statistics, 2020. CA Cancer J Clin. 2020;70(1):7–30. doi: 10.3322/caac.21590. [DOI] [PubMed] [Google Scholar]
- 4.Zaorsky N.G., Churilla T.M., Egleston B.L., Fisher S.G., Ridge J.A., Horwitz E.M., et al. Causes of death among cancer patients. Ann Oncol. 2017;28(2):400–407. doi: 10.1093/annonc/mdw604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Roth G.A., Forouzanfar M.H., Moran A.E., Barber R., Nguyen G., Feigin V.L., et al. Demographic and epidemiologic drivers of global cardiovascular mortality. N Engl J Med. 2015;372(14):1333–1341. doi: 10.1056/NEJMoa1406656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Koene R.J., Prizment A.E., Blaes A., Konety S.H. Shared risk factors in cardiovascular disease and cancer. Circulation. 2016;133(11):1104–1114. doi: 10.1161/CIRCULATIONAHA.115.020406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Curigliano G., Lenihan D., Fradley M., Ganatra S., Barac A., Blaes A., et al. Management of cardiac disease in cancer patients throughout oncological treatment: ESMO consensus recommendations. Ann Oncol. 2020;31(2):171–190. doi: 10.1016/j.annonc.2019.10.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gilchrist S.C., Barac A., Ades P.A., Alfano C.M., Franklin B.A., Jones L.W., et al. Cardio-oncology rehabilitation to manage cardiovascular outcomes in cancer patients and survivors: A scientific statement from the American Heart Association. Circulation. 2019;139(21):e997–e1012. doi: 10.1161/CIR.0000000000000679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Campia U., Moslehi J.J., Amiri-Kordestani L., Barac A., Beckman J.A., Chism D.D., et al. Cardio-oncology: Vascular and metabolic perspectives: A scientific statement from the American Heart Association. Circulation. 2019;139(13):e579–e602. doi: 10.1161/CIR.0000000000000641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Herrmann J., Lenihan D., Armenian S., Barac A., Blaes A., Cardinale D., et al. Defining cardiovascular toxicities of cancer therapies: An International Cardio-Oncology Society (IC-OS) consensus statement. Eur Heart J. 2022;43(4):280–299. doi: 10.1093/eurheartj/ehab674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lyon A.R., López-Fernández T., Couch L.S., Asteggiano R., Aznar M.C., Bergler-Klein J., et al. 2022 ESC guidelines on cardio-oncology developed in collaboration with the European Hematology Association (EHA), the European Society for Therapeutic Radiology and Oncology (ESTRO) and the International Cardio-Oncology Society (IC-OS) Eur Heart J. 2022 doi: 10.1093/eurheartj/ehac244. [DOI] [PubMed] [Google Scholar]
- 12.Sturgeon K.M., Deng L., Bluethmann S.M., Zhou S., Trifiletti D.M., Jiang C., et al. A population-based study of cardiovascular disease mortality risk in US cancer patients. Eur Heart J. 2019;40(48):3889–3897. doi: 10.1093/eurheartj/ehz766. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Strongman H., Gadd S., Matthews A.A., Mansfield K.E., Stanway S., Lyon A.R., et al. Does Cardiovascular mortality overtake cancer mortality during cancer survivorship?: An english retrospective cohort study. JACC Cardio Oncol. 2022;4(1):113–123. doi: 10.1016/j.jaccao.2022.01.102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fang F., Fall K., Mittleman M.A., Sparén P., Ye W., Adami H.O., et al. Suicide and cardiovascular death after a cancer diagnosis. N Engl J Med. 2012;366(14):1310–1318. doi: 10.1056/NEJMoa1110307. [DOI] [PubMed] [Google Scholar]
- 15.Stoltzfus K.C., Zhang Y., Sturgeon K., Sinoway L.I., Trifiletti D.M., Chinchilli V.M., et al. Fatal heart disease among cancer patients. Nat Commun. 2020;11(1):2011. doi: 10.1038/s41467-020-15639-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Koczwara B., Meng R., Miller M.D., Clark R.A., Kaambwa B., Marin T., et al. Late mortality in people with cancer: a population-based Australian study. Med J Aust. 2021;214(7):318–323. doi: 10.5694/mja2.50879. [DOI] [PubMed] [Google Scholar]
- 17.Henson K.E., Reulen R.C., Winter D.L., Bright C.J., Fidler M.M., Frobisher C., et al. Cardiac mortality among 200 000 five-year survivors of cancer diagnosed at 15 to 39 years of age: The teenage and young adult cancer survivor study. Circulation. 2016;134(20):1519–1531. doi: 10.1161/CIRCULATIONAHA.116.022514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wang L., Wang F., Chen L., Geng Y., Yu S., Chen Z. Long-term cardiovascular disease mortality among 160 834 5-year survivors of adolescent and young adult cancer: An American population-based cohort study. Eur Heart J. 2021;42(1):101–109. doi: 10.1093/eurheartj/ehaa779. [DOI] [PubMed] [Google Scholar]
- 19.Murphy C.C., Lupo P.J., Roth M.E., Winick N.J., Pruitt S.L. Disparities in cancer survival among adolescents and young adults: A population-based study of 88 000 patients. J Natl Cancer Inst. 2021;113(8):1074–1083. doi: 10.1093/jnci/djab006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Abdel-Qadir H., Austin P.C., Lee D.S., Amir E., Tu J.V., Thavendiranathan P., et al. A population-based study of cardiovascular mortality following early-stage breast cancer. JAMA Cardiol. 2017;2(1):88–93. doi: 10.1001/jamacardio.2016.3841. [DOI] [PubMed] [Google Scholar]
- 21.Patnaik J.L., Byers T., DiGuiseppi C., Dabelea D., Denberg T.D. Cardiovascular disease competes with breast cancer as the leading cause of death for older females diagnosed with breast cancer: a retrospective cohort study. Breast Cancer Res : BCR. 2011;13(3):R64. doi: 10.1186/bcr2901. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Meacham C.E., Morrison S.J. Tumour heterogeneity and cancer cell plasticity. Nature. 2013;501(7467):328–337. doi: 10.1038/nature12624. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.National Cancer Institute. About the SEER program [Available from: https://seer.cancer.gov/about/.
- 24.Greene FL. American Joint Committee on Cancer, American Cancer Society. AJCC Cancer Staging Manual. 6th ed. New York: Springer-Verlag; 2002.
- 25.National Cancer Institute. SEER: Surveillance, Epidemiology, and End Results Program [Available from: https://seer.cancer.gov/.
- 26.Pearson-Stuttard J., Guzman-Castillo M., Penalvo J.L., Rehm C.D., Afshin A., Danaei G., et al. Modeling future cardiovascular disease mortality in the United States: National Trends and racial and ethnic disparities. Circulation. 2016;133(10):967–978. doi: 10.1161/CIRCULATIONAHA.115.019904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Guan T., Jiang Y., Luo Z., Liang Y., Feng M., Lu Z., et al. Long-term risks of cardiovascular death in a population-based cohort of 1,141,675 older patients with cancer. Age Ageing. 2023;52(5) doi: 10.1093/ageing/afad068. [DOI] [PubMed] [Google Scholar]
- 28.Thomas A, Rhoads A, Pinkerton E, Schroeder MC, Conway KM, Hundley WG, et al. Incidence and Survival Among Young Women With Stage I-III Breast Cancer: SEER 2000-2015. JNCI cancer spectrum. 2019;3(3):pkz040. [DOI] [PMC free article] [PubMed]
- 29.Fung C., Fossa S.D., Milano M.T., Sahasrabudhe D.M., Peterson D.R., Travis L.B. Cardiovascular disease mortality after chemotherapy or surgery for testicular nonseminoma: A population-based study. J Clin Oncol. 2015;33(28):3105–3115. doi: 10.1200/JCO.2014.60.3654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Dong F., Shen Y., Gao F., Shi X., Xu T., Wang X., et al. Nomograms to predict individual prognosis of patients with primary small cell carcinoma of the bladder. J Cancer. 2018;9(7):1152–1164. doi: 10.7150/jca.23344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Centers for Disease Control and Prevention: National Center for Health Statistics [Available from: https://www.cdc.gov/nchs.
- 32.Sung H., Hyun N., Leach C.R., Yabroff K.R., Jemal A. Association of first primary cancer with risk of subsequent primary cancer among survivors of adult-onset cancers in the United States. JAMA. 2020;324(24):2521–2535. doi: 10.1001/jama.2020.23130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Feng W., Li Z., Guo W., Fan X., Zhou F., Zhang K., et al. Association between fasting glucose variability in young adulthood and the progression of coronary artery calcification in middle age. Diabetes Care. 2020;43(10):2574–2580. doi: 10.2337/dc20-0838. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Felix A.S., Bower J.K., Pfeiffer R.M., Raman S.V., Cohn D.E., Sherman M.E. High cardiovascular disease mortality after endometrial cancer diagnosis: Results from the surveillance, epidemiology, and end results (SEER) database. Int J Cancer. 2017;140(3):555–564. doi: 10.1002/ijc.30470. [DOI] [PubMed] [Google Scholar]
- 35.Austin P.C., Lee D.S., Fine J.P. Introduction to the analysis of survival data in the presence of competing risks. Circulation. 2016;133(6):601–609. doi: 10.1161/CIRCULATIONAHA.115.017719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Delgado J., Pereira A., Villamor N., López-Guillermo A., Rozman C. Survival analysis in hematologic malignancies: Recommendations for clinicians. Haematologica. 2014;99(9):1410–1420. doi: 10.3324/haematol.2013.100784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Dores G.M., Curtis R.E., Dalal N.H., Linet M.S., Morton L.M. Cause-specific mortality following initial chemotherapy in a population-based cohort of patients with classical hodgkin lymphoma, 2000–2016. J Clin Oncol. 2020;38(35):4149–4162. doi: 10.1200/JCO.20.00264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Søndergaard M.M.A., Nordsmark M., Nielsen K.M., Poulsen S.H. Cardiovascular burden and adverse events in patients with esophageal cancer treated with chemoradiation for curative intent. JACC Cardio Oncol. 2021;3(5):711–721. doi: 10.1016/j.jaccao.2021.10.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Wang X., Palaskas N.L., Yusuf S.W., Abe J.I., Lopez-Mattei J., Banchs J., et al. Incidence and onset of severe cardiac events after radiotherapy for esophageal cancer. J Thorac Oncol. 2020;15(10):1682–1690. doi: 10.1016/j.jtho.2020.06.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Papadopoulos FC. Suicide and cardiovascular death after a cancer diagnosis. N Engl J Med. 2012;367(3):276-7; author reply 7. [DOI] [PubMed]
- 41.Chan J.S.K., Tang P., Ng K., Dee E.C., Lee T.T.L., Chou O.H.I., et al. Cardiovascular risks of chemo-immunotherapy for lung cancer: A population-based cohort study. Lung Cancer (Amsterdam, Netherlands) 2022;174:67–70. doi: 10.1016/j.lungcan.2022.10.010. [DOI] [PubMed] [Google Scholar]
- 42.Chitturi K.R., Xu J., Araujo-Gutierrez R., Bhimaraj A., Guha A., Hussain I., et al. Immune checkpoint inhibitor-related adverse cardiovascular events in patients with lung cancer. JACC Cardio Oncol. 2019;1(2):182–192. doi: 10.1016/j.jaccao.2019.11.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Verma V., Fakhreddine M.H., Haque W., Butler E.B., Teh B.S., Simone C.B., 2nd Cardiac mortality in limited-stage small cell lung cancer. Radiother Oncol. 2018;128(3):492–497. doi: 10.1016/j.radonc.2018.06.011. [DOI] [PubMed] [Google Scholar]
- 44.Kenzik K.M., Balentine C., Richman J., Kilgore M., Bhatia S., Williams G.R. New-onset cardiovascular morbidity in older adults with stage I to III colorectal cancer. J Clin Oncol. 2018;36(6):609–616. doi: 10.1200/JCO.2017.74.9739. [DOI] [PubMed] [Google Scholar]
- 45.Weberpals J., Jansen L., Müller O.J., Brenner H. Long-term heart-specific mortality among 347 476 breast cancer patients treated with radiotherapy or chemotherapy: A registry-based cohort study. Eur Heart J. 2018;39(43):3896–3903. doi: 10.1093/eurheartj/ehy167. [DOI] [PubMed] [Google Scholar]
- 46.Cedervall J., Herre M., Dragomir A., Rabelo-Melo F., Svensson A., Thålin C., et al. Neutrophil extracellular traps promote cancer-associated inflammation and myocardial stress. Oncoimmunology. 2022;11(1):2049487. doi: 10.1080/2162402X.2022.2049487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Cedervall J., Zhang Y., Huang H., Zhang L., Femel J., Dimberg A., et al. Neutrophil extracellular traps accumulate in peripheral blood vessels and compromise organ function in tumor-bearing animals. Cancer Res. 2015;75(13):2653–2662. doi: 10.1158/0008-5472.CAN-14-3299. [DOI] [PubMed] [Google Scholar]
- 48.Strongman H., Gadd S., Matthews A., Mansfield K.E., Stanway S., Lyon A.R., et al. Medium and long-term risks of specific cardiovascular diseases in survivors of 20 adult cancers: a population-based cohort study using multiple linked UK electronic health records databases. Lancet (London, England) 2019;394(10203):1041–1054. doi: 10.1016/S0140-6736(19)31674-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Oliveira G.H., Al-Kindi S.G., Hoimes C., Park S.J. Characteristics and survival of malignant cardiac tumors: A 40-year analysis of >500 patients. Circulation. 2015;132(25):2395–2402. doi: 10.1161/CIRCULATIONAHA.115.016418. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.