The Importance of Screening for Thyroid Cancer in Patients with Metabolic Syndrome or its Components: A Nationwide Population-Based Cohort Study.

Background The rapidly increasing coincidence of thyroid cancer and metabolic syndrome (MS) in recent decades suggests an association between the two disorders. To investigate this association, we conducted a nationwide study of a large-scale patient cohort. Methods Between 2009 and 2011, data were collected by the Korean National Health Insurance Service for 4,658,473 persons aged 40–70 years without thyroid cancer. During the 6-year follow-up period, participants were monitored for the development of thyroid cancer. The relative risks and incidences of thyroid cancer were calculated using multivariate Cox proportional hazards regression analyses after adjusting for age and body mass index.


Introduction
The global incidence of thyroid cancer has increased steeply over the past several decades. (1,2) Although improved detection of small subclinical thyroid cancer has contributed to this rise in incidence, (3) the effect of other modi able risk factors cannot be ignored. (4) The development of thyroid cancer is multifaceted and may be associated with other diseases or syndromes; however, the exact pathophysiologic mechanisms are not well established.
Over the same period, the prevalence of obesity and unhealthy metabolic conditions, such as metabolic syndrome (MS), has markedly increased and may be associated with the rising incidence of thyroid cancer. Indeed, MS and insulin resistance (IR) are known to increase the incidence of other cancers, including endometrial, lung, pancreatic, hepatocellular, prostate, and colorectal cancers. (5)(6)(7)(8)(9)(10) Additionally, IR, obesity, and type 2 diabetes are more prevalent in patients with differentiated thyroid carcinoma than in those without. (11) Previous studies investigating the association between various metabolic conditions and the development of thyroid cancer focused on body mass index (BMI) and IR, (12) but the results were limited and did not establish a clear connection between metabolic conditions and thyroid cancer.
Here, we report a large-scale, nationwide, population-based, cohort study investigating the association between MS or its components and the risk of thyroid cancer.

Study population
The present study utilised the South Korean population database provided by the National Health Insurance Service (NHIS), which is the public medical insurance system that covers up to 97% of the South Korean population and provides all insured persons with standardized health screening tests every two years. The data from the NHIS encompasses demographics, anthropometric and laboratory data, medical treatments, procedures, including surgeries, and diagnostic codes based on the International Classi cation of Disease-10 th revision-Clinical Modi cation (ICD-10-CM).
The complete NHIS cohort has been previously described. (13) From 2009 to 2011, 8,922,940 people aged 20 or older underwent a health examination provided by NHIS. Participants meeting the following criteria were excluded from this study: history of cardiovascular events before 2011 (n = 441,337), participation in less than two health examinations between 2009 and 2011 (n = 1,134,965), aged < 40 years or > 70 years (n = 2,636,776), prior diagnosis of thyroid cancer before 2011 (n=41,764), and those with any missing or inappropriate data (n = 9,625). In total, 4,658,473 people (2,468,996 men and 2,189,477 women) were enrolled in the study and followed until the date of diagnosis with thyroid cancer or last follow-up by December 31, 2017 ( Figure 1).
This study was conducted based on the ethical principles outlined in the Declaration of Helsinki. The study protocol was approved by the Institutional Review Board of Yonsei University Wonju College of Medicine (IRB number: CR 318349). The requirement for written informed consent was waived because anonymous and de-identi ed information was used for the analysis, and there was no intervention.

Health examination
The health examinations provided by NHIS measured participant height, weight, BMI (calculated by dividing weight [kg] by height [m] squared), waist circumference, systolic and diastolic blood pressure (measured in a seated position after resting for a minimum of ve minutes), and laboratory tests such as blood tests and urinalysis. Blood tests were performed after overnight fasting to measure serum levels of glucose, total cholesterol, triglyceride, high-density lipoprotein (HDL)-cholesterol, low-density lipoprotein (LDL)-cholesterol, and creatinine. Co-morbidities, including hypertension, diabetes mellitus, and dyslipidaemia, were indicated by ICD-10-CM codes or by claims for the agents prescribed for comorbidities.
De nitions of MS and the development of thyroid cancer MS was de ned according to guidelines of the American Heart Association and the National Heart, Lung, and Blood Institute, and the Asian-speci c waist circumference was adopted to de ne abdominal obesity. (14) Brie y, individuals with three or more of the following components were diagnosed with MS: abdominal obesity (waist circumference ≥ 90 cm for males and ≥ 85 cm for females); elevated blood pressure (systolic ≥ 130 mmHg and/or diastolic ≥ 85 mmHg or treatment with antihypertensive medication); hyperglycaemia (fasting plasma glucose ≥ 100 mg/dL or treatment with antidiabetic agents or insulin); low HDL-cholesterol levels (< 40 mg/dL for males and < 50 mg/dL for females), and hypertriglyceridemia (triglyceride ≥ 150 mg/dL or treatment with lipid-lowering medication).
The development of thyroid cancer was identi ed in hospitalization records with the ICD-10-CM code C73.
Individuals who were hospitalized for thyroid cancer between 2009 and 2011 were excluded from the study.

Statistical analyses
Continuous variables are presented as the mean ± standard deviation, and categorical variables are presented as percentages and absolute numbers. Baseline characteristics were compared using the chisquare test for categorical variables and the independent t-test for continuous variables. The incidence of thyroid cancer was calculated by dividing the number of thyroid cancer cases by 10,000 person-years. Multivariate Cox proportional hazard regression analyses were used to investigate the association of MS and its components with the risk of thyroid cancer by sex and to assess the impact of the number of MS components on the risk of thyroid cancer. Relative risks (RRs) with 95% con dence intervals (CIs) were assessed. These models were adjusted for age and BMI. Two-sided pvalues less than 0.05 were considered statistically signi cant. All statistical analyses were performed using STATA version 14 (StataCorp LP, College Station, TX, USA).

Baseline characteristics of the study population
The baseline characteristics of the study participants are summarized in Table 1. Within the study population, 1,012,782 (21.7%) participants had MS. During the 6-year follow-up period, 47,325 (1.0%) participants were newly diagnosed with thyroid cancer. Those who developed thyroid cancer were signi cantly younger (51.1 ± 7.5 years vs. 51.5 ± 8.1 years, p < 0.001) and signi cantly more were female (76.4% vs. 23.6%, p < 0.001) compared to those without thyroid cancer. Additionally, the mean BMI and the proportion of individuals with BMI greater than 25 was signi cantly higher in the thyroid cancer group than in the non-thyroid cancer group (Table 1). Although the prevalence of MS did not differ between those with thyroid cancer and those without (p = 0.616), the majority of MS components did vary between the groups. Speci cally, abdominal obesity and low HDL-cholesterol were signi cantly higher in thyroid cancer patients, whereas hyperglycaemia and hypertriglyceridemia were signi cantly lower in thyroid cancer patients (Table 1). In fact, the only MS component that did not vary signi cantly between the two groups was elevated blood pressure (p = 0.324) ( Table 1).
Association of MS and its components with the risk of thyroid cancer by sex The incidence of thyroid cancer per 10,000 person-years and the RRs of thyroid cancer were calculated after adjusting for age and BMI. MS was signi cantly associated with an increased risk of thyroid cancer in both men (p < 0.001, Table 2) and women (p < 0.001, Table 3). The incidence of thyroid cancer per 10,000 person-years was 6.2 in men with MS and 21.3 in women with MS, which was signi cantly higher than the incidence in men (5.2) and women (19.6) without MS (Tables 2, 3). Interestingly, the association of each MS component and the risk of thyroid cancer differed between males and females. All MS components were signi cantly associated with the risk of developing thyroid cancer in women, whereas hyperglycaemia (p = 0.723) and hypertriglyceridemia (p = 0.211) were not signi cantly associated with thyroid cancer risk in men (Tables 2, 3 followed by elevated blood pressure, and then abdominal obesity. In general, the incidence of thyroid cancer per 10,000 person-years in women with MS or with individual MS components was higher than in men (Tables 2, 3).   Table 4). The incidence of thyroid cancer per 10,000 person-years, which was adjusted for age and BMI, also gradually increased with the number of MS components and was much higher in women than in men (Table 4). whereas those with dyslipidaemia did not. (12) In the same review, the authors assigned quality levels to the evidence provided for each factor, ranging from 'very low' to 'very high'. Importantly, the evidence quality levels were 'low' or 'very low' for every metabolic parameter except IR and BMI, which were graded 'moderate'. (12) Thus, these results are not su cient to establish an association between metabolic status and the risk of thyroid cancer. Similarly, there was limited evidence regarding the impact of MS on the risk of developing thyroid cancer. To our knowledge, only one other study (15) has comprehensively evaluated the association between MS or its components and the risk of developing thyroid cancer in a large-scale patient cohort. In that study, 9, (17) The present study addresses this concern by adjusting for age and BMI prior to evaluation of risk.
Unlike earlier studies, the present study enrolled only subjects aged 40-70 years. This range was selected because thyroid cancer and MS are more prevalent at this age, and atypical events tend to occur in extremely young or old individuals. Although the study discussed above compared baseline characteristics between MS and non-MS groups, our study compared thyroid cancer and non-thyroid cancer groups, because thyroid cancer development was the target outcome in our study. All RRs and incidences of thyroid cancer were calculated after adjusting for age and BMI to avoid any statistical inappropriateness and to enhance the applicability of our results. Interestingly, the RRs and incidences of thyroid cancer gradually increased with the number of MS components in both men and women as well as in individuals with only one or two MS components who did not meet the de nition for MS. Altogether, these results further support the association between MS and the risk of thyroid cancer.
Several pathophysiological mechanisms have been proposed to explain how thyroid carcinogenesis is associated with MS or its components. For example, thyroid carcinogenesis may be driven by increased proliferation, angiogenesis, cellular mobility, and DNA damage due to IR and hyperglycaemia. IR may also overstimulate insulin-like growth factor-1, -2, and insulin receptors, which contribute to thyroid carcinogenesis. Additionally, obesity is associated with the release of cytokines, bioactive molecules, and oestrogen from visceral fat, immune cells, and adipocytes, which are also factors in thyroid cancer development. (18-23) These proposed mechanisms all revolve around IR and obesity, which do not fully encompass MS. Although IR and obesity are key factors in the pathophysiology of MS, the impact of other MS components on thyroid carcinogenesis remains unclear. Based on our results, abdominal obesity, low HDL-cholesterol, and elevated blood pressure impose greater RRs for developing thyroid cancer than hyperglycaemia and hypertriglyceridemia, which were not signi cant in men. Nevertheless, the exact pathophysiological mechanisms remain unclear. As correlation does not equal causation, it is important to identify whether the association of MS or its components and the development of thyroid cancer is due to the incidental overlapping of two prevalent disorders or due to connected pathophysiological mechanisms. To make this distinction, additional large-scale, multicentre-based studies are needed. If these two disorders are indeed mechanistically connected, then MS and its components would not only serve as important risk factors for thyroid cancer development but would also provide a clinically actionable prevention strategy via modi cation of lifestyle behaviours, such as physical exercise and diet.
In this study, we selected patients aged 40-70 years and excluded data that were not mathematizable or subjective, such as smoking status, alcohol consumption status, and physical activity, from the analyses.
We focused instead on concise analyses of basic demographics (age and sex), anthropometric and laboratory measurements, and the resultant statuses of MS and its components. Of note, these decisions may have introduced a selection bias in the study population due to the non-obligatory nature of the health examinations provided by the NHIS. We also acknowledge a lack of clinicopathological data speci c to thyroid cancer, including histologic subtype, primary tumour size, extrathyroidal extension, and lymph node or distant metastasis, which re ect the biological aggressiveness of the cancer. Furthermore, our study could not include the duration of MS or its components in the analyses.
Based on the results of the present study, MS and its components, especially abdominal obesity, low HDL-cholesterol, and elevated blood pressure, are likely to contribute to the development of thyroid cancer, with a higher incidence in women than in men. Although the exact mechanisms that connect the two disorders remain unclear, MS and its components are predictive of thyroid cancer risk and may be useful in screening efforts to diagnose thyroid cancer at early stages. Therefore, MS patients and patients with even a single MS component should be routinely screened for thyroid cancer.
Declarations Figure 1 The complete NHIS cohort has been previously described.