Impact of Health Behaviors on Cardiovascular Risk in Thyroid Cancer Survivors according to Metastasis: A Nationwide Cohort Study
Article information
Abstract
Background
Patients with thyroid cancer receive thyroid-stimulating hormone suppressive therapy and radioactive iodine therapy, which can alter metabolism and affect the association between lifestyle factors and cardiovascular disease (CVD) outcomes. However, it remains unclear how changes in lifestyle relate to CVD risk, particularly according to metastatic status. We investigated the impact of baseline and changing lifestyle behaviors on CVD risk among thyroid cancer patients by metastasis status.
Methods
This study was a nationwide cohort study using data from the Korean National Health Insurance Service claims database with the Cancer Clinical Library Database from the Korea Clinical Data Utilization Network for Research Excellence project. Over a mean follow-up of 5.7 years, 99,742 thyroid cancer patients were analyzed.
Results
Current smoking increased CVD risk in both no metastasis (NM) group (hazard ratio [HR], 1.77; 95% confidence interval [CI], 1.33 to 2.35) and the local metastasis (LM) group (HR, 1.30; 95% CI, 1.00 to 1.68), but not in those with distant metastasis. Regular activity reduced the risk in the NM group (HR, 0.84; 95% CI, 0.79 to 0.96). Alcohol consumption was not significantly associated with the CVD risk. Sustained smoking showed an increased CVD risk in both NM (HR, 1.95; 95% CI, 1.31 to 2.90) and LM (HR, 1.88; 95% CI, 1.28 to 2.77). Sustained regular physical activity was associated with a lower risk of CVD only in NM (HR, 0.68; 95% CI, 0.53 to 0.87).
Conclusion
Sustained smoking increased CVD risk, while regular physical activity was associated with a lower risk of CVD, particularly in those without distant metastasis.
INTRODUCTION
The global incidence of thyroid cancer has remained high in recent decades [1]. Despite this rise in diagnoses, disease-specific mortality has remained stable [2]. Thus, the number of thyroid cancer survivors has substantially increased worldwide [3]. Consequently, clinical attention has shifted from focusing solely on cancer control to long-term survivorship care, including the prevention and management of non-cancer comorbidities [4,5].
Although disease-specific mortality remains low, cardiovascular disease (CVD) has emerged as a leading cause of long-term morbidity and mortality among thyroid cancer patients [6]. Most patients with differentiated thyroid cancer undergo curative treatment, including thyroidectomy, often followed by radioactive iodine therapy (RAI) and levothyroxine replacement with supraphysiologic thyrotropin (TSH) suppression [7–9]. Prolonged TSH suppression leads to iatrogenic subclinical hyperthyroidism, which increases risk of atrial fibrillation, heart failure, and other CVD outcomes [10,11]. In addition, while RAI remains an essential component of adjuvant treatment of thyroid cancer, previous studies have reported a potential association between high cumulative doses and an increased CVD risk [12,13]. Tyrosine kinase inhibitors (TKIs), used in patients with RAI-refractory or advanced disease, can further contribute to cardiotoxicity [14].
Importantly, metastatic burden in thyroid cancer reflects not only disease severity but also the intensity and duration of cardiotoxic exposures. Patients with locally or distantly metastatic disease are more likely to receive repeated or higher-dose RAI, prolonged TSH suppression, and in case of advanced disease, received TKIs. These interventions lead to chronic inflammation, metabolic disturbances, and functional decline with an increase cardiovascular (CV) vulnerability in thyroid cancer [15]. While the overall CVD risk is elevated in thyroid cancer, metastatic status may act as a critical determinant of this risk by dictating the cumulative intensity of cancer-specific exposures. In contrast, large population-based studies in other malignancies have demonstrated that metastatic disease increases the risk of CVD events compared to non-metastatic cancer [16,17]. These findings suggest that disease burden and treatment modality can modulate CVD outcomes in cancer patients.
The association between lifestyle behaviors and CVD risk is well established [18]. This relationship can be modified by the cancer-specific exposures that vary systematically with metastatic status. Patients with more advanced thyroid cancer may change their health behaviors following a cancer diagnosis, resulting in different risk profiles compared to the general population. However, previous studies have assessed lifestyle factors cross-sectionally but have not accounted for behavioral changes or variation in disease burden. It remains to be determined whether lifestyle modifications can effectively reduce CV events in patients exposed to the high-intensity treatments associated with metastasis.
Therefore, we aimed to evaluate the association between both baseline lifestyle behaviors and changes over time and the risk of CVD among thyroid cancer patients according to metastatic status as a surrogate for treatment-related CV stress.
METHODS
Participants and data collection
This retrospective cohort study was based on data from the Korean Clinical Data Utilization Network for Research Excellence (K-CURE), a nationwide cancer research platform supported by the Ministry of Health and Welfare. The K-CURE integrates clinical, administrative, and registry data from 15 tertiary hospitals across South Korea, providing a standardized infrastructure for cancer epidemiology research [19]. It incorporates standardized cancer incidence and staging data from the Korea Central Cancer Registry, health screening records from the National Health Information Database maintained by the National Health Insurance Service (NHIS), mortality data from Statistics Korea, and nationwide insurance claims data from the Health Insurance Review and Assessment Service. The NHIS database covers the entire Korean population under a universal single-payer health system, including detailed information on demographics, medical service utilization, comorbidities, and health behavior questionnaires collected during biennial national health examinations [20]. The K-CURE includes comprehensive demographic and clinical data on cancer patients, such as age, sex, region of residence, diagnosis date, and cancer classification according to the 10th revision of the International Classification of Diseases (ICD-10), tumor characteristics including Surveillance, Epidemiology, and End Results Program (SEER) summary stage, histological subtype, and primary treatment modalities [19,21].
Of 178,087 participants aged 30 years or older who were diagnosed with thyroid cancer (ICD-10 code C73) between 2012 and 2017, 113,498 patients had undergone at least one general health screening within 1 year prior to the study. After excluding individuals with unknown SEER summary stage (n=6,482), prior CVD (n=3,437), missing covariate data (n=3,718), or death within 1 year of diagnosis (n=119), a total of 99,742 thyroid cancer survivors were analyzed and followed up through 2020 (Fig. 1).
Flowchart of participant selection. SEER, Surveillance, Epidemiology, and End Results Program.
This study complied with the ethical standards of the Declaration of Helsinki, and the protocol using secondary data received approval from the Eunpyeong St. Mary’s Hospital Institutional Review Board (IRB) of Catholic Medical Center, The Catholic University of Korea (IRB approval No. PC25ZISI0138). The requirement for written informed consent was waived by the IRB due to the use of previously collected, anonymized data. Access to the nationwide cancer research platform was approved by the National Cancer Data Center under the authorized registration number K-CURE2025100002.
Measurements and definitions of covariates
Baseline data on lifestyle factors were collected using standardized self-reported questionnaires administered during the national health screening examinations. Smoking status was classified as never, former, or current. Alcohol consumption was categorized based on daily average intake into non-drinker (0 g/day), moderate drinker (men <30 g/day; women <20 g/day), or heavy drinker (≥30 g/day for both sexes). Physical activity was assessed using two consecutive questionnaires derived from the World Health Organization (WHO)-developed and Korean-validated International Physical Activity Questionnaire, which assessed the frequency and intensity over the past 12 months [22]. Regular physical activity was defined as engaging in moderate-intensity exercise ≥1 day/week or vigorous-intensity exercise ≥1 day/week. Moderate-to-vigorous physical activity was defined as engaging in moderate exercise ≥5 days/week or vigorous exercise ≥3 days/week. Moderate-intensity activities included carrying light items, biking at a normal speed, and playing tennis. Vigorous-intensity activities included carrying heavy objects, running, aerobics, and fast biking.
To evaluate dynamic behavioral changes, lifestyle factors were assessed at two time points: baseline, defined as the last health screening within 1 year before thyroid cancer diagnosis, and follow-up. The pre-diagnosis assessment (T0) was defined as the last health screening conducted within 0–1 year before thyroid cancer diagnosis. The post-diagnosis assessment (T1) was defined as the first screening conducted after diagnosis. Behavioral change patterns were classified into four categories according to the transition from T0 to T1: (1) consistent non-engagers, who did not participate in the behavior at either timepoint; (2) consistent engagers, who maintained the behavior at both timepoints; (3) initiators, who started the behavior after baseline; and (4) discontinuers, who engaged in the behavior at the baseline but stopped by the follow-up.
Anthropometric and laboratory data were collected during the screening. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m2), and obesity was defined as BMI ≥25 kg/m2 based on WHO criteria for Asian populations [23]. Abdominal obesity was defined as waist circumference ≥ 90 cm in men and ≥85 cm in women [24]. Blood samples were collected after overnight fasting. Comorbidities were identified using examination results and diagnostic codes: diabetes mellitus was defined as fasting glucose ≥126 mg/dL or ≥1 claim per year for antidiabetic medication with ICD-10 codes E11–E14; hypertension as systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg or antihypertensive use with ICD-10 codes I10–I13, I15; and dyslipidemia as total cholesterol ≥240 mg/dL or use of lipid-lowering medication with ICD-10 code E78, as previous studies [25,26]. Thyroid cancer was defined by ICD-10 code C73. Metastasis at diagnosis was defined using SEER summary stage as no metastasis (NM), local metastasis (LM), or distant metastasis (DM). Information on thyroidectomy type (lobectomy or total), RAI, and levothyroxine use was also collected.
Definition of study outcomes
The primary outcome was incident CVD, defined as the first occurrence of myocardial infarction (MI) or stroke. MI was defined as hospitalization with a primary diagnosis of ICD-10 code I21 or I22. Ischemic stroke was identified by inpatient claims containing ICD-10 codes I63 or I64, accompanied by a claim for brain imaging procedures (computed tomography or magnetic resonance imaging).
Statistical analysis
Statistical analyses were conducted using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Baseline characteristics were summarized according to metastatic status and compared using one-way analysis of variance for continuous variables and chi-square tests for categorical variables. Incidence rates of CVD, MI, and stroke were calculated per 1,000 person-years. The follow-up duration was defined as the period from the date of cancer diagnosis to the date of the first CVD event and was reported as median (interquartile range [IQR]). Cox proportional hazards regression models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for the association between lifestyle factors and the risk of CVD, MI, and stroke. To account for varying follow-up durations across metastatic groups, time-to-event analysis with right-censoring was applied. Model 1 was unadjusted, providing crude estimates of the HR for CVD risk. Model 2 was adjusted for age and sex. Model 3 was additionally adjusted for BMI, blood pressure, lipid profile, income level, and other lifestyle factors. Model 4 was adjusted for all the covariates from model 3 and further accounted for thyroid cancer treatment modality (type of thyroidectomy, RAI, and levothyroxine therapy). Time-dependent variables for lifestyle changes were incorporated to assess their dynamic effects on CVD risk. Subgroup analyses were performed by metastatic status (NM, LM, and DM) to evaluate effect modification, and interaction terms were tested using likelihood ratio tests. Missing data were excluded using complete case analysis. A two-sided P value of <0.05 was considered statistically significant.
RESULTS
Baseline characteristics and CVD risk factors according to metastatic burden
The clinical and demographic characteristics are summarized in Table 1. Among the 99,742 thyroid cancer survivors, 77,098 (77.3%) were women, and the mean age was 50.0±10.2 years. At the time of enrollment, 44,324 participants (44.4%) had NM, 54,922 (55.1%) had LM, and 496 (0.5%) had DM.
Baseline Characteristics according to Status of Metastasis in Thyroid Cancer Patients
Patients with DM were older and had a less favorable cardio-metabolic profile, including higher BMI, higher blood pressure, fasting glucose, triglyceride levels, and lower high-density lipoprotein cholesterol levels compared to the NM and LM groups. The prevalence of obesity, central obesity, diabetes mellitus, and hypertension was also higher in the DM group. The prevalence of baseline CVD, including MI and stroke, was more than three times higher in the DM group than in those with NM or LM.
Regarding lifestyle behaviors, current smoking was more prevalent in the LM and DM groups than in the NM group. Non-drinking was most frequent in the DM group, while moderate alcohol consumption was highest in the LM group. The proportion of heavy drinkers remained relatively low across all groups (4.0%–4.6%). Regular physical activity was less common among patients with DM than among patients with NM or LM.
The median follow-up duration was 6.0 years (IQR, 4.0 to 7.0) in the NM and LM groups and 5.0 years (IQR, 3.0 to 7.0) in the DM group. Timing of baseline lifestyle assessment was similar across metastatic groups (same year as their diagnosis: 55.3% in the NM, 54.5% in the LM, and 55.4% in the DM group; 1 year prior: 44.7% in the NM, 45.5% in the LM, and 44.6% in the DM group).
Association between metastatic status, treatment factors, and CVD risk
To evaluate the independent impact of metastatic burden on CVD outcomes, we compared the risk of incident CVD across metastatic stages (Table 2). Compared to the NM group, the DM group exhibited a significantly higher risk of CVD in the fully adjusted model (model 4: HR, 2.29; 95% CI, 1.58 to 3.33). The risk in the LM group was comparable to that in the NM group (HR, 0.99; 95% CI, 0.89 to 1.09).
Association between Metastatic Status, Treatment, and CVD Risk
Regarding treatment-specific factors, levothyroxine (HR, 0.84; 95% CI, 0.66 to 1.07), total thyroidectomy (HR, 0.88; 95% CI, 0.76 to 1.01), and RAI therapy (HR, 1.11; 95% CI, 0.96 to 1.29) did not show significant independent associations with increased CVD risk in the fully adjusted model. TKIs were associated with a markedly elevated risk of CVD (HR, 5.33; 95% CI, 1.29 to 22.08). However, a limited number of patients was included in this group (n=17).
Association between lifestyle factors and CVD risk
In multivariable-adjusted models using never smokers as the reference group, current smokers had a significantly higher CVD risk (model 4: HR, 1.50; 95% CI, 1.24 to 1.81) (Table 3). Former smokers showed a non-significant trend toward lower CVD risk. Alcohol consumption was not significantly associated with CVD risk. Regular physical activity was significantly associated with a lower risk of CVD compared to inactivity (HR, 0.88; 95% CI, 0.80 to 0.97). Moderate-to-vigorous physical activity was not significantly associated with a reduced CVD risk (HR, 0.92; 95% CI, 0.81 to 1.04). Current smoking was consistently associated with the risk of MI and stroke (Supplemental Tables S1, S2) [27]. Regular physical activity showed a risk reduction in MI but not in stroke.
Association between Smoking, Alcohol Consumption, and Physical Activity and Cardiovascular Risk among Thyroid Cancer Patients
Association between lifestyle factors and CVD risk according to metastatic burden
When stratified by metastasis status, the association between lifestyle factors and CVD risk varied (Table 4). In the NM group, current smokers showed a HR of 1.77 (95% CI, 1.33 to 2.35), while former smokers did not have a significant association with CVD risk (HR, 1.00; 95% CI, 0.72 to 1.40). Regardless of the amount, alcohol consumption was not related to CVD risk. Regular physical activity decreased the risk of CVD with a HR of 0.84 (95% CI, 0.72 to 0.96), while moderate-to-vigorous physical activity was not significantly associated with CVD risk.
Association between Smoking, Alcohol Consumption, and Physical Activity and Cardiovascular Risk in Thyroid Cancer Patients according to Metastasis
In the LM group, current smokers had a significantly higher CVD risk (HR, 1.30; 95% CI, 1.00 to 1.68), while no significant association was observed in former smokers (HR, 0.87; 95% CI, 0.66 to 1.14). On the other hand, alcohol consumption, regular physical activity, and moderate-to-vigorous physical activity were not significantly associated with CVD risk.
Among patients with DM, the number of CVD events was small, but notable associations were observed. Former smokers had an elevated risk of CVD (HR, 5.37; 95% CI, 1.25 to 23.09), while current smokers showed a non-significant trend toward increased risk (HR, 2.62; 95% CI, 0.52 to 13.15). Although not statistically significant, moderate and heavy alcohol intake tended to increase the risk of CVD, physical activity showed a trend toward risk reduction.
Association between lifestyle changes and CVD risk
We further assessed the impact of changes in lifestyle behaviors on the risk of CVD using consistent non-engagers as the reference group (Table 5). Consistent smoking was associated with a significantly increased risk of CVD (HR, 1.90; 95% CI, 1.42 to 2.54), while smoking initiation and cessation were not significantly associated with CVD risk. Changes in alcohol consumption were not significantly associated with CVD risk. Consistent regular physical activity was associated with a significantly lower risk of CVD (HR, 0.77; 95% CI, 0.65 to 0.92), while initiating or discontinuing physical activity did not show significant associations. Consistent engagement in moderate-to-vigorous activity also showed a reduced CVD risk (HR, 0.78; 95% CI, 0.61 to 0.99).
Hazard Ratio with 95% CI of Cardiovascular Disease after Behavior Change in Thyroid Cancer Patients
The risk of MI or stroke was elevated in consistent smokers (MI: HR, 2.03; 95% CI, 1.35 to 3.06; stroke: HR, 1.78; 95% CI, 1.19 to 2.66), while initiation and cessation of smoking were not significantly associated. Consistent engagement in regular physical activity was associated with a reduced risk of MI (HR, 0.71; 95% CI, 0.56 to 0.91). Neither changes in nor sustained patterns of alcohol consumption and physical activity were significantly associated with stroke risk.
Association between lifestyle changes and CVD risk according to metastasis status
Among patients with NM, those who continued smoking showed a significantly higher CVD risk, with an adjusted HR (aHR) of 1.95 (95% CI, 1.31 to 2.90) (Table 6, Fig. 2). A similar trend was observed in the LM group, where continuous smokers had an aHR of 1.88 (95% CI, 1.28 to 2.77). However, no significant association was observed in the DM group. Changes in alcohol consumption were not significantly associated with CVD risk in any group (P for interaction=0.4322). Consistent engagement in regular physical activity and moderate-to-vigorous activity was associated with lower CVD risk (HR, 0.68; 95% CI, 0.53 to 0.87; and HR, 0.68; 95% CI, 0.46 to 0.98, respectively). However, no significant association was observed in the LM (HR, 0.89; 95% CI, 0.70 to 1.13) or DM group.
Hazard Ratio with 95% CI of Cardiovascular Disease after Behavior Change in Thyroid Cancer Survivor according to Metastasis
Adjusted hazard ratios (HRs) for cardiovascular disease by lifestyle change pattern among thyroid cancer patients, stratified by metastatic status. (A) No metastasis and (B) lymph node metastasis.
Consistent smoking was significantly associated with increased risk of MI in the LM groups (HR, 2.60; 95% CI, 1.59 to 4.22). However, there was no significant association between changes in alcohol consumption or physical activity and with the risk of MI across metastatic status group (Supplemental Table S3) [27]. Regarding stroke risk, consistent smoking was significantly associated with an increased risk of stroke without metastasis (HR, 2.39; 95% CI, 1.47 to 3.90). In contrast, no significant associations were observed between changes in alcohol consumption or physical activity and stroke risk across any metastatic status group (Supplemental Table S4) [27].
DISCUSSION
In this large nationwide cohort study of thyroid cancer patients, we observed that metastatic burden was independently associated with CVD risk. DM group showed a 2.29-fold higher risk compared with the NM group. With this difference in CV vulnerability, we investigated how baseline lifestyle behaviors and their subsequent changes were associated with CVD risk across metastatic status. We found that both baseline unhealthy lifestyle behaviors and sustained adverse lifestyle patterns over time were independently associated with increased risk of CVD. Among lifestyle factors, smoking and physical inactivity had significant associations with CVD risk, while alcohol consumption did not. These associations remained consistent across different metastatic burden.
Our findings are consistent with a previous study demonstrating that healthy lifestyle behaviors are associated with a lower risk of CVD among cancer survivors. A recent meta-analysis reported that adherence to favorable lifestyle factors was associated with an approximately 47% reduction in CVD incidence [28]. Our study extended these findings by demonstrating similar cardioprotective associations in thyroid cancer, despite its relatively favorable prognosis.
The association between current smoking and increased CVD risk observed in our study aligns with well-established biological mechanisms. Smoking induces endothelial dysfunction, promotes atherogenesis, and increases arterial stiffness [29]. These effects are further exacerbated in cancer survivors due to cancer-related metabolic alterations, treatment-induced factors, and inflammatory changes [30]. Among thyroid cancer survivors, CV damage can be exacerbated by cancer treatment, including RAI, TSH suppression, and TKIs therapy, resulting in metabolic dysregulation [6,31]. In the NM and LM groups, current smokers showed the highest risk of CVD, with 1.78- and 1.30-fold increases. Interestingly, former smokers with DM exhibited a 5.37-fold higher CVD, suggesting residual vascular damage from prior long-term exposure [32]. This residual risk is associated with baseline cardiometabolic comorbidities such as hypertension, diabetes, dyslipidemia, and obesity that were more prevalent in the DM group. In our analysis, TKI therapy was associated with an increased risk of CVD (HR, 5.33), although this finding should be interpreted with caution given the limited sample size (n=17). Nevertheless, it is clinically plausible that TKI exposure contributes to this risk. The first-line TKIs used for progressive RAI-refractory differentiated thyroid cancer primarily inhibit the vascular endothelial growth factor–vascular endothelial growth factor receptor signaling pathway, which plays a pivotal role in maintaining vascular homeostasis [33]. Inhibition of this pathway is known to cause vascular complications such as hypertension and arterial thromboembolism [34]. Furthermore, patients with advanced metastatic disease often have a substantially reduced CV reserve, making them particularly susceptible to the synergistic vascular toxicity of potent pharmacologic agents and persistent smoking. This unexpected pattern reflects reverse causality or survivor bias, where a current smoker with advanced disease may die before developing overt CVD events. Moreover, given RAI therapy and long-term levothyroxine suppression in the DM group, it is plausible that cancer treatment-related CVD stress interacted with previous smoking-induced endothelial injury to increase susceptibility to CVD. However, the CV effects of RAI therapy remain controversial. While some studies have suggested possible radiation-related vascular changes, including increased carotid intima media thickness [12,35], others have failed to show consistent associations [13,36].
Beyond baseline smoking status, changes in smoking behavior were associated with CVD outcomes in our cohort. Thyroid cancer survivors who continued smoking after diagnosis exhibited the highest risk of CVD, with a 1.9-fold increase, compared to sustained never smokers. This finding underscores the persistent vascular toxicity of tobacco exposure, even in the post-treatment period. The observed pattern of CV risk across different smoking behavioral groups in our study is generally consistent with findings from a nationwide Korean cancer survivor cohort study [37]. In that study, the highest CVD risk was observed in continuing smokers (HR, 1.86; 95% CI, 1.74 to 1.98), followed by starters (HR, 1.51; 95% CI, 1.33 to 1.71), and quitters (HR, 1.20; 95% CI, 1.12 to 1.28), relative to sustained nonsmokers. This pattern likely reflects the delayed nature of CV risk reduction following smoking cessation, as vascular recovery often requires several years to become evident [38], especially in individuals with underlying endothelial injury or additional CV stress resulting from cancer treatment. Although our models were adjusted for levothyroxine use and RAI therapy, the unique pathophysiology of thyroid cancer could still influence CV outcomes in ways not fully captured by conventional covariate adjustment.
Alcohol consumption was not significantly associated with the risk of CVD across any metastasis status. These findings suggest that alcohol is not a major modifiable determinant of CV health in this specific cancer population. Previous meta-analyses have reported a J-shaped relationship between alcohol intake and CVD in the general population, with low-to-moderate intake potentially offering protective effects [39]. However, such benefits may not translate to cancer survivors, particularly those undergoing TSH suppression, or those with altered metabolic function. Furthermore, in the analysis of behavioral changes, individuals in the DM group reported no alcohol consumption after cancer diagnosis. This limited the statistical power to evaluate the impact of alcohol consumption on CV risk in this subgroup, particularly within drinking categories.
Regarding physical activity, our results demonstrated that regular engagement in physical activity was associated with a 12% reduction in the risk of CVD. Sustained engagement in regular physical activity throughout the follow-up period further reduced CV risk, with a HR of 0.77. This association was statistically significant only in the NM group for sustained physical activity, indicating a clear cardioprotective benefit in thyroid cancer survivors without metastasis who maintained long-term regular physical activity. In contrast to regular physical activity, moderate-to-vigorous physical activity was not significantly associated with CVD risk among thyroid cancer survivors. This finding persisted after stratification by metastatic status. One plausible explanation is that the beneficial effects of physical activity on CVD outcomes are more closely linked to total volume and consistency of activity rather than intensity alone [40]. While previous studies in the general population reported additive benefits from higher-intensity activity [41,42], the distinct physiological and treatment-related characteristics of thyroid cancer survivors may modulate the CV response to moderate-to-vigorous physical activity. Additionally, thyroid cancer survivors with advanced disease often experience fatigue, reduced cardiorespiratory fitness, and musculoskeletal limitations that restrict their ability to engage in or sustain high-intensity exercise [43,44]. As a result, they may not reach the threshold intensity or duration required to elicit the same physiological benefits observed in the general population. These findings underscore the need for cancer-specific physical activity guidelines that account not only for intensity but also for treatment history, functional capacity, and comorbidity burden.
To the best of our knowledge, this is the first study to investigate the association between both baseline and dynamic changes in health behaviors and their impact on CVD outcomes among thyroid cancer survivors. The large sample size (>99,000 survivors) and extended follow-up period of 5.7 years enhanced the reliability of our results. By incorporating metastatic status, we were able to stratify survivors by metastatic burden. It allows a nuanced analysis of lifestyle-CVD associations across clinically relevant subgroups. Additionally, we adjusted for key treatment modalities, including RAI and levothyroxine replacement, thereby enhancing the internal validity of our findings.
However, several limitations should be considered when interpreting our findings. Due to the design of the study, a causal relationship cannot be established. Although we observed a significant association between behavioral trajectories and CVD outcomes, the direction of these relationships remains uncertain, and reverse causality is possible. Lifestyle behaviors were assessed through self-reported questionnaires, which could introduce recall bias or misclassification. Additionally, dynamic behavior changes were assessed using two or three time points corresponding to national health screenings. This may not capture transient or intermediate changes and limits the granularity with which behavior trajectories can be analyzed over time. Although the models were adjusted for demographic, clinical, and treatment-related confounders, the possibility of residual confounding remains. Important unmeasured variables, such as dietary patterns, psychosocial stress, or genetic predispositions, could influence the observed associations. Regarding the assessment timing, the DM group had the shorter median follow-up duration (5.0 years) compared to the NM/LM groups (6.0 years). However, Cox time-to-event modeling accounted for this difference, and significant associations observed despite the shorter follow-up suggest an early and clinically relevant impact of lifestyle factors in high-risk survivors. Furthermore, while we incorporated various treatment modalities into our analysis, some data granularity was limited. Because RAI therapy and levothyroxine use were obtained as binary indicators, dose-response analyses based on cumulative dose or biochemical intensity were not feasible. Another limitation is that our analyses were restricted to a thyroid cancer survivorship cohort without a direct comparison with the general population. While the CVD risk of cancer survivors compared with the general population has been documented, our study instead focuses on the prognostic value of longitudinal behavioral trajectories within the thyroid cancer survivorship period.
Despite these limitations, our findings provide important insights into the long-term CVD risk of thyroid cancer survivors. The consistent association between sustained smoking and increased CVD risk underscores the critical importance of smoking cessation as a core component of survivorship care. In contrast, while moderate-to-vigorous physical activity was not associated with reduced risk of CVD outcomes, sustained regular physical activity could attenuate CVD risk especially in survivors with NM. These findings suggest that even modest levels of physical activity can impact on CVD. Importantly, the lack of association between changes in alcohol consumption and CVD risk suggests that lifestyle recommendations for thyroid cancer survivors should be developed based on survivor-specific data rather than generalized population-based evidence.
In conclusion, our findings underscore the clinical relevance of monitoring and promoting healthy behavioral trajectories throughout cancer survivorship. Consideration of behavioral risk patterns could enhance the effectiveness of survivorship care in reducing CVD morbidity in this population.
Supplementary Material
Supplemental Table S1.
Association between Smoking, Alcohol Consumption, and Physical Activity and Myocardial Infarction Risk among Patients with Thyroid Cancer
Supplemental Table S2.
Association between Smoking, Alcohol Consumption, and Physical Activity and Stroke Risk among Patients with Thyroid Cancer
Supplemental Table S3.
Hazard Ratio with 95% CI of Myocardial Infarction after Behavior Change in Thyroid Cancer Survivor according to Metastasis
Supplemental Table S4.
Hazard Ratio with 95% CI of Stroke after Behavior Change in Thyroid Cancer Survivors according to Metastasis
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
ACKNOWLEDGMENTS
This study was performed using the database from the Korean Clinical Data Utilization Network for Research Excellence (K-CURE) and the National Health Insurance Service and the results do not necessarily represent the opinion of the National Health Insurance service.
AUTHOR CONTRIBUTIONS
Conception or design: K.D.H., D.J.L. Acquisition, analysis, or interpretation of data: J.L., J.L., J.P., K.K., J.S.B. Drafting the work or revising: J.L. (Jeongmin Lee). Final approval of the manuscript: J.L., J.L., J.P., K.K., J.S.B., K.D.H., D.J.L.