Introduction

The prevalence of thyroid cancer has been steadily increasing over the last three decades [1]. Thyroid cancer is more prevalent in females, with approximately three times more female patients than male [2]. Papillary thyroid carcinoma (PTC) accounts for the majority of thyroid cancer. Of late, however, there is strong evidence that overdiagnosis of papillary thyroid microcarcinoma (PTMC) may account for the increase in the incidence of reported thyroid cancer cases [1, 3, 4]. Nevertheless, the prevalence of large cancers and those diagnosed at later stages, where lymph node metastasis occurs, are still on the rise [5]. The effects of metabolic and environmental factors, such as radiation exposure, diabetes mellitus, obesity, and iodine deficiency or excess, on the development of PTC remain unclear [1].

Non-alcoholic fatty liver disease (NAFLD) is a metabolic disease characterized by hepatic macrovesicular [6]. It has had a global prevalence of around 25% in adults over the last four decades, making it the most prevalent chronic liver condition in adults [7]. The presence of NAFLD has been associated with the incidence of chronic kidney disease [8], hypothyroidism [9], thyroid autoimmunity [10], and various types of extrahepatic cancers, such as gastrointestinal tract and uterine cancer [11], colon cancer [12], bladder cancer [13], and thyroid and lung cancer in males [14], as well as metabolic syndrome (MetS) and insulin resistance (IR) [15, 16].

MetS was found associated with the aggressiveness of PTC [17]. And NAFLD is also a reversible metabolic disease like MetS. Although males with NAFLD have a high risk of developing thyroid cancer [14], to the best of our knowledge, there is no data on the correlation between NAFLD occurrence and the severity of PTC. Therefore, we undertook this study to explore the relationship between PTC and NAFLD.

Methods

Study protocol

This study used data from a group of patients with PTC who underwent first-time surgery for tumor removal between January 2020 and February 2022 at the First Affiliated Hospital of Wenzhou Medical University. Patients were excluded if they had a history of excessive alcohol intake and/or viral hepatitis, or were diagnosed with other types of thyroid cancers, such as follicular, medullary, or anaplastic cancers. Information on patients’ demographic characteristics, preoperative laboratory tests, imaging, and histological data were obtained from electronic medical records. The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Wenzhou Medical University and informed consent was obtained from all patients.

Imaging data

Two sonographers or radiologists performed an ultrasound and/or computed tomography (CT) scans on all patients [a chest CT was necessary for all patients due to the ongoing coronavirus disease-19 (COVID-19) pandemic during that time]. The patients were diagnosed with NAFLD based on specific ultrasonographic features, such as diffusely increased echogenicity of liver parenchyma, hepatomegaly, and vascular blunting [18] and if they had a liver-to-spleen attenuation ratio (L/S) < 1 through images obtained via non-contrast chest CT scan. NAFLD occurrence in the patients included in this study excluded viral hepatitis, liver cirrhosis, liver malfunction or cancer, and excessive alcohol intake as causal agents.

Data collection

Basic information and clinical data of all patients included in this study were collected from electronic medical records. Pre-operative data included details of age, gender, weight and height, history of alcohol consumption, history of type 2 diabetes mellitus (DM), history of hypertension, imaging and laboratory findings just before surgery, and clinicopathologic variables, including levels of thyroglobulin antibody (TgAb), thyroid peroxidase antibody (TPOAb), thyroid-stimulating hormone (TSH), high-density lipoprotein (HDL), low-density lipoprotein (LDL), and triglycerides (TG). Post-operative data included details of surgical outcomes and clinicopathologic variables mentioned above (values obtained after surgery). Body mass index (BMI in kg/m2) was calculated by dividing the weight in kilograms (kg) by the square of the height in meters (m).

Diagnoses and definitions

All patients underwent surgical resection according to the guidelines of the 2015 American Thyroid Association for the treatment of differentiated thyroid cancer in adults. The dissection of the central lymph nodes was routinely performed. Preoperative imaging, fine-needle aspiration, and intraoperative exploration were used to determine whether the lateral lymph nodes needed to be dissected. Patients were classified as having either multiple foci (having > 1 tumor in the thyroid tissue) or a solitary focus (only one tumor in the thyroid). The PTC tumor stage was defined based on the guidelines provided by the 8th American Joint Committee on Cancer (AJCC) edition. Information on tumor location, TNM staging, histologic differentiation of tumors, extra-thyroidal extension (ETE), and lymph node metastasis were collected from the pathological analysis reports.

A patient was diagnosed with NAFLD if steatosis was found in >5% of the hepatocytes. The diagnosis was linked to metabolic risk factors as none of the patients had a history of excessive alcohol consumption (≥30 g/day for male and ≥20 g/day for female) or any other chronic liver disease [19]. In our study, we diagnosed NAFLD mainly by examining ultrasound and/or CT images.

Hashimoto’s thyroiditis (HT) was diagnosed based on pathological results in our study. A BMI of >25 kg/m2 was considered to be high. Low HDL-C was defined as levels of HDL-C < 1.16 mmol/L (for male) and <1.29 mmol/L (for female). High LDL-C was defined as levels of LDL-C > 3.10 mmol/L. High TG was defined as levels of TG > 1.70 mmol/L.

Statistical analyses

The software SPSS (version 19.0; SPSS, Chicago, IL, USA) and StataSE (version 16.0; Standard Edition) were used to conduct statistical analyses. All categorical variables were expressed as percentages while continuous variables were represented as mean ± SD. To compare the clinical features of patients with and without NAFLD, we used the chi-square test for categorical variables and the rank-sum test or student’s t-test for continuous variables. Both, multivariate and simple binary logistic models were used to examine the relationship between NAFLD and PTC. Variables with P-values < 0.05 in the univariate logistic analyses were selected for multivariate analysis when screening for confounding factors. In addition, we investigated the association between NAFLD and clinicopathological characteristics of PTC within subgroups of HT, hypertension, high TG, DM, low HDL-C, high BMI, standard (non-DM & non-hypertension & non-HT & BMI ≤ 25 kg/m2), female and male. A P-value < 0.05 was considered statistically significant for all two-tailed statistical tests.

Results

Baseline characteristics

This study used data from 3468 patients with PTC, of which 2561 (73.8%) were female and 907 (26.2%) were male (Fig. 1). In all, 594 (17.1%) patients were diagnosed with NAFLD. The clinical data, including metabolic and pathological features, of the patients included in this study are listed in Table 1.

Fig. 1
figure 1

Flowchart of screened patients. PTC papillary thyroid carcinoma, TC thyroid carcinoma

Full size image
Table 1 Clinicopathologic characteristics for patients with and without NAFLD
Full size table

As shown in Table 1, the patients diagnosed with NAFLD (NAFLD group) were older (p = 0.000) and had higher BMI indices (p = 0.000), higher incidence of HT (p = 0.003), higher incidence of nodal metastasis (p = 0.007), higher incidence of multifocal tumors (p = 0.037), higher incidence of hypertension (p = 0.000) and DM (p = 0.000), higher incidence of BRAF V600E mutation (p = 0.002), more advanced tumors at diagnosis (later TNM stages) (p = 0.000), lower levels of HDL-C (p = 0.000), higher levels of LDL-C (p = 0.000), and higher levels of TG (p = 0.000) than patients in the non-NAFLD group. In addition, there were more males in the NAFLD group than in the non-NAFLD group (p = 0.000). There were no differences between the NAFLD and non-NAFLD groups in the occurrence of tumor subtype (classic versus follicular) (p = 0.762). Patients in both groups had similar T stages (p = 0.870), the incidence of ETE (p = 0.132), and thyroid-stimulating hormone (TSH) levels (p = 0.764).

Relationship between NAFLD and clinicopathological features

As demonstrated in Table 2, univariate analysis found that NAFLD was associated with a higher incidence of nodal metastasis (OR = 1.312, 95% CI: 1.099–1.565), higher incidence of BRAF V600E mutation (OR = 1.518, 95% CI: 1.167–1.975), later tumor stages (OR = 3.409, 95% CI: 2.593–4.482), and higher incidence of multifocal diseases (OR = 1.232, 95% CI: 1.012–1.498) than non-NAFLD. However, tumor sizes were similar between the NAFLD and non-NAFLD groups (OR = 1.096, 95% CI: 0.890–1.348). In the multivariate analysis, after adjustment for risk factors in each univariate analysis, such as age, gender, BMI, DM, hypertension, HT, HDL-C, LDL-C, and TG, NAFLD still emerged as a risk factor for the development of nodal metastasis (OR = 1.285, 95% CI: 1.052–1.570) and was associated with a higher incidence of BRAF V600E mutation (OR = 1.504, 95% CI: 1.148–1.972) and later TNM stage (OR = 2.310, 95% CI: 1.700–3.139). However, NAFLD was not associated with the incidence of multifocal tumors (OR = 1.195, 95% CI: 0.972–1.469).

Table 2 Univariate and multivariate analyses of the risk factors for clinicopathological features in PTC
Full size table

Subgroup analyses associated with NAFLD

We further analyzed the available data to investigate the relationship between the clinicopathological characteristics of PTC and the occurrence of NAFLD. As demonstrated in Fig. 2, we selected nine subgroups, such as HT, hypertension, high TG, DM, low HDL-C, high BMI, standard, female and male. After adjusting for risk factors in each clinicopathological characteristic (Table 3), we discovered that in females and patients with HT, hypertension, high TG levels, low HDL-C levels, and high BMI, NAFLD was still a risk factor for the development of nodal metastasis, incidence of BRAF V600E mutation and later TNM stage. In contrast, NAFLD was not a risk factor for any of the above-mentioned clinicopathological characteristics in males with PTC. Additionally, the analyses revealed that NAFLD was not associated with tumor size in all subgroups. Furthermore, there were no obvious differences in the incidence of multifocal tumors between the non-NAFLD and NAFLD groups in any of the subgroups, except for those in the HT subgroup (OR = 1.672, 95% CI: 1.052–2.658). In addition, we found that NAFLD was only associated with later TNM stage (OR = 3.380, 95% CI: 1.656–6.899) in patients without hypertension, HT, DM or overweight (standard group).

Fig. 2
figure 2

Multivariate analysis for relationship between NAFLD and clinicopathological features of PTC in 9 subgroups. NAFLD non-alcoholic fatty liver disease, BMI body mass index, HT Hashimoto’s thyroiditis, TNM stage Tumor Node Metastasis stage, DM diabetes, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein Cholesterol, TG triglycerides. Standard* subgroup consist of participants without hypertension, HT, DM or overweight. Analyses after adjusted for risk factors in each clinicopathological characteristic (e.g., lymph node metastasis, BRAF V600E mutation, TNM stage, multifocality, tumor size), details of adjustments are shown in Table 3

Full size image
Table 3 Adjustments for the multivariate analysis in nine subgroups
Full size table

Discussion

Currently, only two studies have investigated the association between fatty liver disease and thyroid cancer. One study from China found that males with NAFLD had a higher risk of developing thyroid cancer and that blood levels of alanine transferase in this group were positively associated with the risk of developing thyroid cancer [14]. Another study from China involving 352,911 participants found that the presence of metabolic dysfunction-associated fatty liver disease (MAFLD) was a significant risk factor for the development of thyroid cancer [20]. However, no studies to date have explored the relationship between NAFLD and the clinicopathological characteristics of PTC.

Our study showed that NAFLD may be an independent risk factor associated with the development of lymph node metastasis, incidence of BRAF V600E mutation, and later TNM stage in females with PTC. Despite increased nodal metastasis and a later TNM stage, BRAF V600E mutation was also linked in certain reports to PTC aggressiveness and a worse prognosis [21, 22]. Nevertheless, this study found no significant relationship between NAFLD and tumor size in patients with PTC. The most recent WHO classification of thyroid neoplasms does not include papillary thyroid micro-carcinoma as a subtype of PTC because a subset of these tumors displayed aggressive pathologic characteristics and clinical behaviors, including regional and distant metastasis and structural recurrence after surgery [23,24,25]. Thus, a small tumor size does not necessarily mean a better prognosis.

Subgroup multivariate analysis indicated that NAFLD was an important risk factor for the development of lymph node metastasis, high incidence of BRAF V600E mutation, and development of advanced tumor stage at diagnosis even after accounting for the confounding effects of Hashimoto’s thyroiditis, hypertension, diabetes, high BMI, high TG, low HDL-C, and gender. However, further analysis showed that NAFLD was not associated with any aggressive clinicopathological characteristics in males, possibly due to the notably worse prognosis generally associated with male patients with PTC [26, 27].

In the subgroup of patients with HT, the NAFLD group had a significantly higher incidence of multifocal lesions than the non-NAFLD group. Several studies have verified that patients with HT, who develop PTC, have an increased risk of having multifocal tumors [28, 29]. Our results suggested that the occurrence of NAFLD and HT together increases the incidence of multifocal tumors; however, further studies are required to confirm this pattern.

The molecular mechanisms underlying the effects of NAFLD on PTC are not well understood; therefore, variables other than NAFLD may play important roles in the progression of PTC. For example, the pathophysiology of NAFLD is closely associated IR [30], which might promote thyroid cancer cell proliferation through insulin and insulin-like growth factor signaling pathways [31,32,33,34]. Furthermore, as NAFLD progresses, the release of leptin may cause pyroptotic cell death in macrophages and hepatocytes via CD8+ T lymphocytes [35], which may in turn support the proliferation and invasiveness of cancers such as thyroid cancer [36]. Patients with PTC have elevated levels of serum leptin and thyroid cancer cells with high levels of leptin receptors are associated with tumor aggressiveness [37, 38]. On the other hand, adiponectin inhibits angiogenic growth and its levels are inversely correlated with fat deposits; therefore, it is unsurprising that both, the development of NAFLD and thyroid cancer growth, are negatively correlated with adiponectin levels [39, 40]. Besides these factors, the production of TSH stimulates thyrocyte proliferation and is positively correlated with the expression of vascular endothelial growth factor. An elevated TSH level is a standalone risk factor for the development of NAFLD [41] and likely plays a role in the onset and progression of thyroid tumors in obese patients [42]. In our study, however, no significant differences in pre-surgical TSH levels between the NAFLD and non-NAFLD groups were apparent.

Inevitably, our study has several limitations. (1) We utilized abdominal ultrasonography and/or CT as non-invasive means for diagnosing NAFLD. However, liver biopsies are the gold standard for this diagnosis because conventional ultrasound and CT scans are not sensitive enough to diagnose mild steatosis, particularly in obese people, and the accuracy of ultrasound is hampered by reader variability [43, 44]. Since we did not diagnose NAFLD via liver biopsies, we may have missed diagnosing patients with early-stage NAFLD. (2) Despite the usually favorable prognosis of patients with PTC after surgical treatment, there are no ongoing follow-up evaluations to check for recurrence or distant metastases. (3) Some potential confounders, such as how long a patient has had NAFLD, smoking, and genetic factors were not considered in this study.

In summary, this study indicates that the presence of NAFLD may be positively associated with aggressiveness of PTC in female patients. However, additional research is required to understand the biochemical pathways that link PTC with NAFLD.