PDF
Citation
Print
INTRODUCTION
Worldwide, thyroid cancer is the ninth most common cancer overall and the most common malignancy of the endocrine system over the past decades [1]. In 2020, approximately 586,000 new cases were estimated, accounting for 3.0% of the total cancer burden [1]. In Korea, thyroid cancer ranks sixth in males but second in females, with age-standardized incidence rates of 22.1 per 100,000 and 66.3 per 100,000, respectively [2]. In addition to the increased detection of thyroid cancer, exposure to ionizing radiation during childhood and adolescence is the only well-established risk factor for different types of thyroid carcinoma [3,4]. In terms of other factors that contribute to thyroid cancer, iodine is a vital micronutrient that is required for the synthesis of thyroid-stimulating hormones [5]. Insufficient intake of iodine therefore adversely affects thyroid function and may result in an increased risk of thyroid cancer development.
Approximately 65.6% of the iodine intake among Koreans is estimated to come from seaweeds [6,7]. In the traditional Korean diet, seaweed is commonly provided as dried or water-containing forms, including gim (Porphyra sp. laver), miyeok (Undaria pinnatifida sea mustard), and dashima (Laminaria sp. sea tangle) [8,9]. In addition to iodine, other bioactive molecules in seaweed, such as polysaccharides, proteins, polyphenols, carotenoids, and n-3 polyunsaturated fatty acids from seaweed are a beneficial source for human health [10]. Furthermore, in vitro and in vivo models have been suggested the anticarcinogenic effects of these components, which include regulating tumor metabolism, apoptosis, and cell proliferation [9,11,12]. However, epidemiological studies regarding the effect of dietary seaweed and iodine on thyroid cancer in the Korean population have been limited.
Furthermore, the sodium/iodine symporter (encoded by the SLC5A5 gene) is an intrinsic membrane protein that is involved in the metabolism of iodine from the bloodstream into the follicular cells of the thyroid using the sodium gradient generated by Na+/K+-ATPase [13-15]. SLC5A5 expression is lower in thyroid carcinomas or adenomas than in normal adjacent thyroid tissue [14]. Additionally, SLC5A5 expression is suggested to differentiate follicular adenomas from follicular carcinomas in cumulative studies [13,16]. Given the important role of SLC5A5 in the iodine uptake pathway and its effect on thyroid cancer prognosis, this study aimed to examine the associations of dietary seaweed and iodine intakes in relation to the risk of thyroid cancer in Korean adults to determine whether this association was modified by SLC5A5 polymorphisms.
METHODS
Study population
This hospital-based case-control study recruited a total of 41,109 participants who underwent health screening examinations, including thyroid ultrasonography, at the National Cancer Center (NCC), South Korea, between 2002 and 2014 (Fig. 1). Details of this project have been described elsewhere [17]. At baseline, all subjects provided the written consent form and completed a general questionnaire and a semiquantitative food frequency questionnaire (SQFFQ) about their demographics, lifestyles, and dietary consumption. The cases were defined as having a diagnosis of thyroid cancer, with the International Classification of Diseases, Tenth Revision (ICD-10) code of C73 according to data linkage to the Korea Central Cancer Registry. Among the 1,104 subjects diagnosed with thyroid cancer, we excluded 345 subjects who had missing blood samples (n=317) or did not complete the general questionnaire (n=28). Additionally, among 37,236 potential controls, we excluded those who did not complete the general questionnaire or with missing biomarker data or a history of thyroid-related disease, surgery, or medication use. A total of 759 cases and 759 ageand sex-matched controls who were free of other cancer types were selected for genotyping. General characteristics of total genotyping population were described elsewhere [18]. Of these, 117 cases and 173 controls with available information on the SQFFQ were included in the final analysis.
Data collection
Demographic information including age (years), sex (male/female), family history of thyroid cancer (yes/no), marital status (married/other), educational level (≤middle school, high school, and ≥college), occupation (housewives, professional, and office worker; sales and service; and agriculture, laborer, unemployed, and others), monthly income (<$1,667, $1,667–$3,333, and >$3,333), smoking status (ever/never), alcohol consumption (ever/never), and regular exercise (yes/no) were collected using a generally interviewer-administered questionnaire. Height and weight were measured using an Inbody version 3.0 (Biospace, Seoul, Korea) body composition analyzer or an X-SCAN PLUS II Body Composition Analyzer (Jawon Medical, Gyeongsan, Korea). Body mass index (BMI) was calculated as the ratio of weight (kg) to height squared (m2) and classified into normal (<23 kg/m2), overweight (23–24.9 kg/m2), and obese (≥25 kg/m2) groups.
Regarding dietary intake, participants were asked about their average consumption frequency for each of 106 food items in the SQFFQ during the previous year from nine frequency categories (never or rarely, once a month, 2–3 times per month, once or twice a week, 3–4 times per week, 5–6 times per week, once a day, twice a day, and 3 times per day) and were then asked about usual portion sizes using three portion size categories (small, medium, and large) [19]. Of the 106 food items, two seaweed items, including gim and miyeok/dashima, could be chosen in the SQFFQ. Seaweed consumption (g/day) of these two items and iodine intake (μg/day) from the whole diet were calculated using the Computer-Aided Nutritional Analysis Program (CAN-Pro) version 4.0 (Korean Nutrition Information Center, Seoul, Korea). In nutritional research, consumption of a diet component will result in a greater total energy intake and quantity of foods and thereby affect several important factors, such as body size and composition [20]. To account for the effect of a single food that is independent of total energy intake, consumption was estimated using the residual energy adjustment method and categorized into low and high intake based on the median value [21]. In particular, we first obtained residuals from the regression model with total energy as the independent variable and absolute intake of gim, miyeok/dashima, and iodine separately. The adjusted intake of these components was then calculated as the predicted values from the model of average energy intake as the independent variable and the residual and coefficient values above as intercepts [21].
Genotyping and imputation
The study participants were genotyped using the Infinium OncoArray-500K BeadChip (Illumina Inc., San Diego, CA, USA) with 499,170 single nucleotide polymorphisms (SNPs) (275,691 genome-wide tag SNPs and 223,479 cancer-specific SNPs). After quality control in PLINK version 1.07 for monomorphic variants (minor allele frequency [MAF]=0), MAF <0.01, call rate <95%, and deviation from Hardy-Weinberg equilibrium (P<1×10−6), a total of 345,675 SNPs remained. The data were then imputed using SHAPEIT (v2 r837) [22] and IMPUTE2 (2.3.2) [23], and 5,087,097 SNPs were available for further analysis. In this study, we selected rs77277498 (C>G) of the SLC5A5 gene for the final analysis.
Statistical analysis
The comparison between thyroid cancer cases and controls was analyzed using Student’s t test for continuous variables and the chi-square test (or Fisher’s exact test for expected cell frequencies <5) for categorical variables. A logistic regression analysis was implemented to assess the risk of thyroid cancer and calculate odds ratios (ORs) and 95% confidence intervals (95% CIs) for the measurement of associations. The multivariable model was adjusted for age, sex, family history of thyroid cancer, smoking status, alcohol consumption, and BMI. Dietary seaweed (including gim and miyeok/dashima) and iodine intakes on thyroid cancer risk were compared by quartile groups of high (greater than median amount) and low (less than the median amount) consumption. The association of each SNP with thyroid cancer and its interaction effect on dietary consumption and thyroid cancer association were examined in a dominant model. All statistical analyses for the present study were performed using R version 3.6.0 (R Core Team, R Foundation for Statistical Computing, Vienna, Austria).
RESULTS
Our study included 290 subjects comprising 117 thyroid cancer cases and 173 controls. Individuals with thyroid cancer were more likely to be females (P=0.001) and nonsmokers (P=0.016) (Table 1), whereas age, family history of thyroid cancer, marital status, educational level, occupation, monthly income, smoking status, alcohol consumption, regular exercise, and BMI were equally distributed between the patients and controls (P>0.05). Compared to those without dietary information, the current study population differed in terms of age, sex, educational level, monthly income, smoking status, and alcohol consumption (P<0.05) (Supplemental Table S1). Of these, marital status was not significantly different among cases (P=0.587), whereas family history was additionally different among controls with and without SQFFQ (P=0.022). In addition, although free thyroxine and thyroid stimulating hormone were measured at the health examination date for many participants who were genotyped, those for most of the participants in our final analysis were missing (Supplemental Table S2).
The characteristics of diet consumption and SLC5A5 variants in thyroid cancer cases and controls are presented in Table 2. The daily amount of gim intake was greater in the healthy subjects than in the cancer subjects (P=0.017), whereas the consumption of miyeok/dashima (P=0.811), total seaweed (P=0.338), and iodine (P=0.231) was comparable between the two groups. Additionally, the genotypes of the rs77277498 (P=0.027) variant were unequally distributed between the two groups.
Table 3 shows the ORs and 95% CIs of thyroid cancer risk for dietary seaweed (including gim and miyeok/dashima) and iodine intakes and SLC5A5 variants. The higher gim and iodine intake quartiles were inversely associated with thyroid cancer risk compared with the lower quartiles, with adjusted ORs of 0.50 (95% CI, 0.30 to 0.83) and 0.57 (95% CI, 0.33 to 0.95), respectively. In contrast, the intake amount of miyeok/dashima was not significantly associated with thyroid cancer risk (OR, 0.74; 95% CI, 0.44 to 1.23), whereas total seaweed consumption was borderline associated with thyroid cancer (OR, 0.61; 95% CI, 0.36 to 1.02). In addition, gim and miyeok/dashima intake frequency did not differ between cancer and noncancer subjects. Besides, a significant association was observed for rs77277498 (Table 3). Those carrying a G allele (G/C+G/G) had a decreased risk of thyroid cancer compared with those carrying the C/C genotype in the dominant model (OR, 0.52; 95% CI, 0.28 to 0.93).
Table 4 presents the results for the combined effects of dietary seaweed and iodine and a genetic variant of rs77277498 on thyroid cancer risk. The inverse association between iodine-based dietary intake and thyroid cancer seems to be stronger among G allele carriers than among C/C heterogenous carriers. In the multivariable model, those carrying the G allele with a high consumption of total seaweed, as well as its different forms, including gim and miyeok/dashima, had a significantly lower thyroid cancer risk than those not carrying the minor allele with a low intake, with ORs of 0.26 (95% CI, 0.10 to 0.62), 0.25 (95% CI, 0.10 to 0.56), and 0.31 (95% CI, 0.12 to 0.77), respectively. Regarding intake frequency, those carrying the G allele had a lower risk of thyroid cancer than those who did not (OR, 0.48; 95% CI, 0.23 to 0.99) among subjects who consumed fewer than three times of gim per month. Furthermore, compared with the G allele noncarrier subgroup with a low iodine intake, other subgroups were associated with a lower risk of thyroid cancer, with ORs of 0.42 (95% CI, 0.18 to 0.93) for G/G+G/C carriers with a low intake, 0.51 (95% CI, 0.27 to 0.93) for C/C carriers with a high intake, and 0.30 (95% CI, 0.12 to 0.73) for G/G+G/C carriers with a high intake. The interaction between gim intake frequency and the rs77277498 genotype was significant (P=0.02).
DISCUSSION
In this study, we found that high intakes of gim and iodine were associated with a decreased risk of thyroid cancer. In addition, our study results supported a putative role of the rs77277498 polymorphism in the SLC5A5 gene in the reduction in thyroid cancer risk. The G/G+G/C carriers with a high intake of these dietary factors had a lower risk of thyroid cancer than those carrying C/C genotype with a low intake. Additionally, we observed significantly interactive effects between gim intake frequency and rs77277498 in relation to thyroid cancer risk.
Previous studies have investigated the effect of seaweed consumption on the development of thyroid cancer in Japanese women. Data from the Japan Public Health Center-based Prospective Study revealed a positive association between frequent seaweed intake (almost daily vs. ≤2 days/week) and papillary carcinoma (hazard ratio [HR], 1.71; 95% CI, 1.01 to 2.90) but not total thyroid cancer (HR, 1.41; 95% CI, 0.86 to 2.32). Despite the higher overall incidence rate of thyroid cancer in the Japan Collaborative Cohort Study than in the Japan Public Health Center-based Prospective Study (20.9 vs. 17.5 per 100,000 person-years), seaweed consumption was still not significantly associated with either papillary carcinoma (HR, 1.15; 95% CI, 0.69 to 1.89) or total thyroid cancer (HR, 0.94; 95% CI, 0.51 to 1.74). In the current study, we further assessed total seaweed consumption in terms of both frequency and quantitative measurement of grams per day. However, we did not find any associations between total seaweed intake and the risk of thyroid cancer.
Regarding the major types of seaweed intake in Koreans, the chemopreventive effect of gim has been proposed to be associated with the anticancer effects of the polysaccharides, phospholipids, sterol, and peptides found in Porphyra sp. on apoptosis and cell proliferation inhibition [24]. This effect was evaluated in different human cell lines of hepatic carcinoma, cervical cancer, breast carcinoma, and gastric cancer [24]. In this study, miyeok/dashima consumption was not associated with a reduced risk of thyroid cancer. This might be due to the low-frequency intake in this population, with approximately 60% of the study participants consuming miyeok/dashima no more than three times per month. However, fucoidan from U. pinnatifida (miyeok) or Laminaria sp. (dashima) has been proven to exert cytotoxic activity against various cancer types including breast, colon, hepatocellular, lung, and prostate cancer cells, by inhibiting proliferation and inducing apoptosis via both mitochondrial-dependent and death receptor rough extrinsic-related pathways [25-30]. In addition, similar anticancer effects have been reported for bioactive compounds extracted from Laminaria japonica, such as glycoproteins, polysaccharides, laminarin, fucoxanthin, and phlorotannin [31-35].
Furthermore, iodine is an essential component for the synthesis of thyroid hormones [36]. Iodine deficiency has been hypothesized to induce thyroid hyperplasia and increase aneuploidy, thereby eliciting chromosomal changes in the thyroid [36]. Additionally, iodine deficiency together with epidermal growth factor and insulin-like growth factor I is suggested to cause chronic thyroid-stimulating hormone overstimulation, resulting in thyroid tumors [36]. Pooled estimates from a meta-analysis showed that a high intake of iodine was associated with a lower risk of thyroid cancer (OR, 0.74; 95% CI, 0.60 to 0.92), whereas iodine deficiency may not be significantly related to an increased risk (OR, 1.22; 95% CI, 0.94 to 1.58) [37]. Despite the limited number of individual studies in the meta-analysis and the different amounts of iodine intake, the findings appeared to be consistent with our results. In contrast, several studies have indicated the potentially harmful effects of excessive iodine intake regarding hypothyroidism, as excessive intake triggers the Wolff–Chaikoff effect [38-41]. Therefore, excessive intake of seaweed and iodine or not requires further recommendations.
It has been reported that the expression level of SLC5A5 has been reported to be lower in thyroid tumor tissues than that in normal tissues in Korean and Italian subjects [42,43]. Although SLC5A5 plays a crucial role in iodine metabolism and thyroid regulation, the relationship between SLC5A5 and thyroid cancer is complex and poorly understood [15,44]. Russo and collaborators found that mutations as well as other genetic alterations of SLC5A5 may not be the main cause of reduced iodine uptake in differentiated thyroid carcinomas [44,45]. However, in the present study, our findings suggested that SLC5A5 rs77277498 polymorphism may enhance the potential protective effect of dietary iodine-based intake on thyroid cancer. Nevertheless, the results regarding the methylation of the SLC5A5 promoter in thyroid cancer are still controversial, and thus, further studies on the role of SLC5A5 in thyroid cancer are warranted [44].
To the authors’ knowledge, no study has yet elucidated the effect modification of SLC5A5 on the associations between dietary seaweed and iodine intakes and thyroid cancer. In this study, a comprehensive and validated FFQ was administered by trained interviewers to acquire dietary information from the subjects [19,46-48]. Moreover, iodine and iodine-containing food consumption was estimated by residual energy adjustment, which has been suggested to be more powerful and robust to residual confounding than the standard method [21] and to diminish the effect of misreporting of energy and micronutrient intake [49]. Furthermore, by using the largely imputed data of approximately 5 million tags, this study not only determined the association between dietary iodine consumption and thyroid cancer risk but also identified how this association varies according to different genotypes of SLC5A5 polymorphisms.
Nevertheless, some limitations need to be considered when interpreting the present results. First, the presence of selection and recall biases related to a hospital-based case-control study should be mentioned. In addition, the associations between dietary intake and thyroid cancer could occur without a causal relationship because patients who are aware of their disease status might focus on healthy dietary behaviors [50,51], or because of the increased detection of early cases due to the thyroid cancer screening program available in Korea [52,53]. Second, the sample size was relatively small for the subgroup analyses by tumor subtype and tumor aggressiveness. However, our previous studies reported that most of the cases (97%) in our study population were papillary thyroid cancer [54]. Additionally, according to the Korea National Cancer Incidence Database, papillary carcinoma was reported to be the most frequent histologic type (94.2%) among incident thyroid cancer cases in Korea during 1997 and 2011 [55,56]. The results therefore provide supportive evidence for papillary thyroid cancer but not for other subtypes, such as differentiated or follicular thyroid cancer. The relatively small sample size might also not represent the consumption by Korea adults. However, by using dietary data from the total Cancer Screenee Cohort of 10,810 participants [50], our estimates of iodine and iodine-based food consumption in the present study did not differ from those that were not included in the final analysis (Supplemental Table S3). Third, due to the data unavailability, we were unable to examine or adjust for the effect of previous radiation exposure, which is a strong risk factor for thyroid cancer. Finally, because of the lack of relevant items on the SQFFQ, we could not investigate the effect of all items of edible seaweed and their bioactive components, only the three most commonly consumed types of seaweed in Korea [57,58]. Thus, the SQFFQ may therefore underestimate the actual amount of iodine intake in the Korean population.
In summary, our findings provide evidence to support the putative effects of dietary gim and iodine intakes against thyroid cancer risk, and the strength of this association can be modified by rs77277498 genotypes.
XML Download