Dairy product consumption and thyroid cancer risk: a systematic review and meta-analysis of observational studies

Introduction

Thyroid cancer (TC) is the most common malignant tumor of the endocrine system and the most frequent head and neck malignancy, with an increasing incidence and a trend toward younger age of onset (1). According to the 2022 global cancer report from the International Agency for Research on Cancer (IARC), new cases of TC in China have exceeded 60,000, accounting for more than half of the global incidence (2). The risk of TC in women is about three times that in men (3). Moreover, a 17-year study indicated rising incidence and mortality rates of TC in China (4). Multiple factors are implicated in thyroid carcinogenesis, including genetic predisposition, history of ionizing radiation exposure, environmental endocrine disruptors, and dietary patterns. To date, ionizing radiation is a well-established risk factor for TC (5-8).

In recent years, the relationship between dietary factors and TC risk has been extensively investigated, with diet recognized as a modifiable risk factor. Greca et al. demonstrated that dietary modifications can influence the risk of disease persistence or recurrence in TC patients (9). Furthermore, research indicates that alterations in dietary habits can disrupt gut microbiota, promote autoimmunity, increase the number of pro-inflammatory cells, and potentially elevate TC risk (10). Barrea et al. identified associations between specific dietary patterns and the clinical severity and aggressiveness of TC (11).

Dairy products are rich in calcium, magnesium, vitamin D, and various other bioactive components that may affect cancer risk and progression (12). Dairy intake has been variably linked to TC occurrence. For instance, in women with a body mass index (BMI) ≥25 kg/m², dairy consumption appears to be negatively associated with TC risk (13), a finding consistent with the study by Fiore et al. (14). Conversely, Przybylik-Mazurek et al., in a case-controlled study of 345 participants, found no correlation between dairy product intake and TC risk (15). Some researchers have also pointed out that certain dairy products are rich in iodine, and excessive iodine intake may itself increase TC risk to some extent. Thus, the correlation between dairy consumption and TC remains controversial.

Given that dietary structure is a key modifiable risk factor for chronic diseases, its appropriate management throughout life, particularly for early risk reduction and late-stage benefits, is crucial. Therefore, in this meta-analysis, we summarized available observational studies up to September 1, 2025, to evaluate the potential relationship between dairy consumption and TC. This aims to contribute to chronic disease-nutrition balance and play a role in disease prevention and progression. We present this article in accordance with the PRISMA reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2643/rc).

Methods

Search strategy

Three electronic databases, PubMed, EMBASE, and Web of Science, were systematically searched for scholarly articles published between January 1, 1996 and September 1, 2025. The search strategy followed the PICOS (participants, interventions, comparisons, outcomes, and study design) principle. Search terms combined keywords related to dairy products (“dairy” OR “dairy products” OR “milk” OR “cheese” OR “yogurt” OR “yoghurt” OR “butter” OR “butterfat” OR “cream” OR “curd” OR “koumiss” OR “pudding” OR “custard” OR “fromage” OR “whey” OR “casein” OR “lactalbumin” OR “margarine” OR “ghee” OR “kefir” OR “ricotta” OR “quark” OR “muenster” OR “mozzarella” OR “cheddar”) with terms for TC (“Thyroid Neoplasm” OR “Thyroid Neoplasms” OR “Thyroid Carcinoma” OR “Thyroid Carcinomas” OR “Cancer of Thyroid” OR “Thyroid Cancers” OR “Thyroid Cancer” OR “Cancer of the Thyroid” OR “Thyroid Adenoma” OR “Thyroid Adenomas”). The initial search yielded a total of 298 relevant studies. The complete search strategies for all databases are provided in Appendix 1.

Study selection

Two researchers (X.S. and J.L.) independently screened titles and abstracts for eligibility, extracted relevant data, and assessed the quality of the included studies. Any discrepancies were resolved through discussion with a third reviewer (M.S.). Studies were included if they met the following criteria: (I) were observational studies with cohort or case-control design; (II) provided data on dairy intake (any type); (III) compared different levels of dairy intake; (IV) reported TC risk as an outcome; and (V) provided effect estimates for the association between dairy intake and TC, e.g., odds ratio (OR), relative risk (RR), or hazard ratio (HR) with corresponding 95% confidence intervals (CIs). Studies fulfilling all five criteria were included. Ultimately, eight articles were selected for the final analysis after rigorous screening. Although our search strategy was broad and did not explicitly exclude cross-sectional studies, no such studies that fulfilled all other eligibility criteria were identified.

Inclusion criteria

This review employed the PICOS strategy. All studies investigating dairy product consumption in relation to TC incidence were included (Table 1).

Exclusion criteria

In accordance with our predefined PICOS criteria, we excluded non-observational studies, including reviews, meta-analyses, editorials, comments, case reports, basic science studies, and animal studies. Additionally, duplicate publications, studies with overlapping data, and those from which complete data could not be extracted were excluded.

Data extraction

Two researchers (X.S. and J.L.) independently extracted data using a standardized form. Extracted information included: first author, year of publication, country, study design, study duration, sample size, Newcastle-Ottawa Scale (NOS) scores, etc. For studies that reported multiple effect estimates for different dairy products (e.g., milk, cheese, butter), we applied the following pre-specified criteria to select one estimate for inclusion in the main meta-analysis: first, we prioritized the estimate for total dairy product consumption; if this was not available, we selected the estimate for the most commonly reported individual dairy product (e.g., milk) or the estimate representing the broadest category of dairy intake as defined in the original study. This process was conducted independently by two reviewers, with any discrepancies resolved by consensus.

Quality assessment

The quality of eligible studies was independently assessed by two researchers (X.S. and J.L.) using the NOS (16). The NOS evaluates studies based on three domains: selection of study groups, comparability of groups, and ascertainment of exposure or outcome. A maximum of nine stars can be awarded. The studies were classified as high quality (≥7 stars), moderate quality (4–6 stars), or low quality (1–3 stars). This approach helps ensure the reliability of the pooled results by systematically assessing the risk of bias in individual studies. Disagreements were resolved through consultation with the third reviewer (M.S.).

Statistical analysis

In this meta-analysis, the included studies reported associations between dairy product consumption and TC risk using various effect measures (OR, RR, and HR). Given the low incidence of TC, we applied the rare disease assumption, whereby RRs and HRs approximate ORs. All effect estimates were pooled on the log-OR scale for quantitative synthesis using STATA 19. For the binary data, pooled ORs with 95 % CI were calculated. A P<0.05 was considered statistically significant. Heterogeneity was assessed using the I² statistic, where I² ≤50% indicated no observed heterogeneity and I²>50 % indicated significant heterogeneity. Subgroup and meta-regression analyses were performed to explore the sources of heterogeneity. Sensitivity analysis was conducted to assess the stability of the pooled results. If a sufficient number of studies were available, funnel plots were used to examine the risk of publication bias. Depending on the degree of heterogeneity, a fixed-effects or random-effects model was applied for the pooled analysis.

Results

Characteristics of the study protocol

Study selection process

The study selection process is detailed in Figure 1. The initial search identified 298 records from the PubMed, EMBASE, and Web of Science databases. After removing 73 duplicates, 225 records remained for screening. Based on title and abstract review, 207 records were excluded. The remaining 18 full-text articles were assessed for eligibility. Of these, 5 articles were excluded for being review articles and 5 for irrelevant exposure data. Consequently, 8 publications met all the criteria and were included in the final meta-analysis (13,14,17-22).

Figure 1 Flowchart of literature search and study selection.

Study characteristics

The meta-analysis included eight studies examining the relationship between dairy product intake and TC risk (Table 2). These studies varied in their design, geographic location, and duration, being conducted in Korea, Japan, Italy, the USA, France, Sweden, and Norway.

Five studies were cohort designs, with follow-up periods ranging from 7.6 years to an average of 16.7 years, while the remaining three were case-control studies. Sample sizes in these studies varied widely, from 323 to 492,810 participants, with the number of TC cases ranging from 106 to 930.

Dairy intake was primarily assessed using the Food Frequency Questionnaire (FFQ), including semi-quantitative and cohort-specific versions. Exposure definitions varied, including frequency of consumption (e.g., “≥5 vs. 3–4 times/week”), quantity of intake (e.g., “>194.9 vs. ≤40.6 g/day”), or comparisons between highest and lowest consumption groups.

Quality assessment

Quality assessment using the NOS revealed generally high quality among the included studies, with the majority (n=6) scoring between 7 and 9. Two studies received a slightly lower score of 6. The detailed scores for each study are presented in Table 3.

In the “Selection” domain, most studies demonstrated methodological rigor, with all eight scoring at least 3 stars, indicating adequately defined cases and representative populations. Two studies (25%) achieved the maximum score of 4 stars, primarily for explicitly confirming the absence of the outcome at baseline or for establishing a well-defined control group.

For “Comparability”, all studies attained the maximum score of 2 stars. This consistent high rating resulted from adjustments for key confounding factors, such as age and sex, ensuring a high degree of comparability between the study groups.

In the “Outcome” domain, most studies were awarded 2 stars, and one study received a perfect score of 3. These ratings stemmed from the use of objective and reliable methods for outcome assessment, notably pathological confirmation, coupled with adequate follow-up periods with minimal loss. Conversely, some studies, such as Park 2009 and Galanti 1997, scored lower as their methodologies did not explicitly state whether the exposure assessment was conducted in a blinded or unbiased manner.

The overall high NOS scores across the included studies suggest a generally low risk of bias, thereby strengthening the reliability of the aggregated evidence in this meta-analysis.

Quantitative data synthesis

The meta-analysis revealed no statistically significant association between dairy product consumption and TC risk. The pooled OR was 0.92 (95% CI: 0.72–1.17), with the CI including the null value of 1.0. A high degree of heterogeneity was observed among the studies (I²=67.9%, P=0.003), suggesting significant variability in the effect sizes across the included studies (Figure 2).

Figure 2 Forest plot of the overall association between dairy consumption and thyroid cancer risk. CI, confidence interval; OR, odds ratio.

Subgroup analysis

Subgroup analyses were conducted to explore heterogeneity based on study design, geographical location, study quality, and adjustment for key confounders (total energy intake and age) (Figure 3).

Figure 3 Summary of subgroup analyses. CI, confidence interval; OR, odds ratio.

Stratified by study design, cohort studies showed a pooled OR of 0.85 (95% CI: 0.61–1.17), while case-control studies yielded a pooled OR of 1.04 (95% CI: 0.70–1.53), showing no statistically significant difference (P=0.47).

Subgroup analysis by geographical location showed that studies from Asia had a pooled OR of 0.74 (95% CI: 0.51–1.08). In contrast, studies from both Europe and the USA reported a pooled OR of 1.04 (95% CI: 0.70–1.53 and 0.72–1.17, respectively). No significant difference was found among these regional subgroups (P=0.27).

When studies were grouped by quality, high-quality studies demonstrated a pooled OR of 0.86 (95% CI: 0.65–1.13), while moderate-quality studies had a pooled OR of 1.09 (95% CI: 0.64–1.86), presenting no statistically significant difference (P=0.39).

Further subgroup analyses examined the impact of adjusting for key confounders. For studies with adjustment for total energy intake, the pooled OR was 0.89 (95% CI: 0.65–1.22), compared to 0.94 (95% CI: 0.58–1.53) for studies without (P=0.81). Similarly, studies that were adjusted for age resulted in a pooled OR of 0.84 (95% CI: 0.68–1.04), whereas those that were not adjusted for age yielded an OR of 1.01 (95% CI: 0.64–1.60), with no significant difference between groups (P=0.46). Notably, in our analysis, studies that adjust for age showed no significant statistical heterogeneity (I²=0.0%, P=0.41), suggesting that differences in age adjustment may be a contributing factor to the overall heterogeneity observed. However, the interpretation of this finding requires caution due to the limited number of studies.

Sensitivity analysis

To confirm the stability and reliability of our findings, a sensitivity analysis was conducted using the leave-one-out method. The results demonstrated that the pooled OR and its corresponding 95% CI remained stable and within the original overall CI (0.72–1.17) upon sequential removal of each study, confirming the robustness of the findings (Figure 4).

Figure 4 Sensitivity analysis using the leave-one-out method. CI, confidence interval.

Publication bias

Visual inspection of the funnel plot suggested an asymmetrical distribution, which might indicate the possibility of publication bias (Figure 5). However, given the small number of studies included (n=8), formal statistical tests for publication bias have limited power. Consequently, neither Begg’s rank correlation test (P=0.54) nor Egger’s regression test (P=0.056) reached statistical significance (P<0.05) (Figure 6). Thus, the evidence for publication bias in this meta-analysis remains inconclusive.

Figure 5 Funnel plot for the assessment of publication bias. OR, odds ratio.

Figure 6 Results of Begg’s and Egger’s tests. Coef., coefficient; Conf. Interval, confidence interval; Std. Err, standard error; Std_Eff, standardized effect.

Discussion

In this meta-analysis, we summarized nearly 30 years of observational studies to evaluate the potential relationship between dairy consumption and TC risk. After rigorous screening, a total of eight studies, comprising five cohort and three case-control design, were finally included. The primary finding was the absence of a statistically significant association between dairy intake and TC risk.

Subgroup analyses reinforced this null association. No significant differences were observed when stratifying by study design, geographical region (Asia, Europe, USA), or study quality. Adjustments for total energy intake and age also did not materially alter the results. A key finding was that adjustment for age substantially reduced and explained the significant heterogeneity observed in the overall analysis (I² decreased to 0.0%). Age-related differences in thyroid cell biology, such as variations in the expression of iodine metabolism-related proteins and thyrotropin (TSH) levels, may contribute to differential susceptibility to external risk factors, including potential dietary influences (23). Furthermore, the inclusion of populations potentially exposed to radiation from events like the 1986 Chernobyl nuclear accident, where the increased risk of TC manifested after a latency period and persisted for decades (24), could interact with age and confound the results. Additional sources of heterogeneity likely include differences in study design (cohort vs. case-control), regional variations across three continents, methodological disparities in controlling for confounding factors, and differences in the assessment of both exposure (dairy intake) and outcome (TC). It is important to interpret these subgroup and meta-regression findings with caution. The statistical power to detect genuine moderators is limited when the number of studies is small, as in our analysis. Therefore, while age adjustment appears to be associated with lower heterogeneity in our dataset, this observation should not be over-interpreted as the sole or definitive explanation for the between-study differences.

Currently, the evidence on the association between TC and dairy consumption is still limited, with relatively small sample sizes in some studies and potential for non-publication of non-significant results. However, this meta-analysis did not perform stratified analyses by specific dairy product types. While the funnel plot asymmetry suggested potential publication bias, this was not confirmed by formal statistical tests (Begg’s and Egger’s tests). The stability of the overall result was further confirmed through the sensitivity analysis, which showed that no single study disproportionately influenced the pooled estimate.

In summary, this analysis found no significant association between dairy consumption and TC risk, aligning with the results from a study in South Korea (17). However, a pooled analysis of two prospective studies reported lower dairy and meat consumption in TC patients compared to controls, which may be related to the complex interactions among various nutritional factors within the overall diet (13). It is noteworthy that in a multiethnic population from different geographical regions, dairy intake has been associated with increased risks for other cancer types, such as bladder, prostate, and colorectal cancers (25), highlighting the context-specific nature of dietary risk factors.

The development of TC is closely linked to environmental factors, including diet and nutrition (26). Dairy products are rich sources of calcium, iron, zinc, and other trace elements. Some evidence suggests that these trace nutrients might influence the expression of thyroid hormone receptors in gut epithelial and immune cells via the gut-thyroid axis (27), potentially offering protection against autoimmune thyroid diseases and even TC. In addition, a recent study showed that insulin resistance, which leads to higher circulating levels of insulin-like growth factor (IGF), has been implicated in TC progression (28). Interestingly, a prospective population-based study suggested that a high intake of dairy products might reduce insulin resistance syndrome (29). The pronounced gender disparity in TC incidence, with women being at much higher risk, invites speculation about shared pathways with other hormonally-sensitive cancers. Vitamin D, abundant in dairy products, has known protective effects against breast cancer (30). Given that TC and breast cancer are among the most common malignancies in women and may share estrogen-related pathways (31), a potential protective role for vitamin D in TC is plausible. Supporting this, studies have shown that vitamin D metabolites can bind strongly to vitamin D receptors on the surface of TC cells, thereby inhibiting cell growth and potentially blocking tumor development (32).

A key strength of this study is its novel synthesis of evidence regarding the consumption of multiple types of dairy products and TC risk over a 30-year period. We conducted a comprehensive quality assessment of the included literature and performed multi-dimensional analyses, including stratification by age, intake levels, study design, and geographical location. This contributes to a more nuanced understanding for the clinical multi-target prevention and treatment of TC. It also lays a foundation for leveraging nutrition in chronic disease prevention.

It is important to acknowledge the methodological limitations of this systematic review. First, the study protocol was not prospectively registered in a public database such as PROSPERO. Although the review adhered to established guidelines, the absence of prior registration may reduce transparency and increase the risk of selective reporting bias. Future research in this area would benefit from protocol registration to enhance methodological rigor and reproducibility. Second, the observational designs of the included studies introduce inherent limitations. The three case-control studies are susceptible to recall bias, whereby differential accuracy in reporting past dietary habits between cases and controls could influence the observed associations. In the five cohort studies, non-differential misclassification of dairy intake is possible, as dietary assessment, often conducted only at baseline, may not fully capture long-term consumption patterns. Such misclassification would generally bias results toward the null. Although these biases are common in nutritional epidemiology, they warrant cautious interpretation of our findings. Third, substantial heterogeneity existed across studies in how dairy consumption was defined and assessed, including frequency-based, quantity-based, and extreme-category comparisons. This variability in exposure measurement limits the direct comparability of studies and constitutes a major constraint in interpreting the pooled estimate. Moreover, the present analysis aggregated all dairy products and was unable to conduct dose-response meta-analyses or separate analyses by dairy type (e.g., milk vs. cheese). This precludes more nuanced biological inferences regarding specific components or consumption patterns and highlights an important avenue for future research. Fourth, although stratification by certain factors (e.g., age adjustment) reduced statistical heterogeneity (I²), this does not necessarily imply that all underlying biological or methodological heterogeneity has been resolved. Unmeasured or unexamined sources of variation likely persist. Finally, assessment of publication bias is challenging due to the limited number of included studies. While the funnel plot suggested some asymmetry, statistical tests were underpowered to confirm it. Thus, potential publication bias cannot be ruled out and should be considered when interpreting the results.

Conclusions

In conclusion, the results of this meta-analysis do not support a significant association between dairy consumption and TC risk. However, future research should prioritize larger prospective cohorts, standardized and detailed exposure assessments, dose-response analyses, and product-specific meta-analyses to explore the effects of different dairy types (e.g., milk, cheese, yogurt), careful control of key confounders (e.g., iodine, energy, age, sex, BMI), and stratified analyses. If feasible, an individual participant data (IPD) meta-analysis would be valuable to investigate potential effects on different histological subtypes of thyroid malignancy.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2643/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2643/prf

Funding: This study was supported by Project of Harbin Science and Technology Bureau (No. 2023ZCZJNS108).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2643/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Tang L, Angell TE. Thyroid cancer. Semin Perinatol 2025;49:152042. [Crossref] [PubMed]
2. Pizzato M, Li M, Vignat J, et al. The epidemiological landscape of thyroid cancer worldwide: GLOBOCAN estimates for incidence and mortality rates in 2020. Lancet Diabetes Endocrinol 2022;10:264-72. [Crossref] [PubMed]
3. Wild CP, Weiderpass E, Stewart BW, editors. World Cancer Report: Cancer research for cancer prevention. Lyon (FR): International Agency for Research on Cancer; 2020.
4. Zhang XZ. Trends of thyroid cancer incidence and mortality in China from 2005 to 2016. Chinese Journal of Cancer Prevention and Treatment 2022;29:1725-33.
5. Karyadi DM, Bogdanova TI, Milder CM, et al. Distinctive molecular features of radiation-induced thyroid cancers. Sci Adv 2025;11:eadw7680. [Crossref] [PubMed]
6. Baş Aksu Ö, Şahin M. Endocrine-disrupting chemicals and thyroid cancer. Turk J Med Sci 2025;55:1648-56. [Crossref] [PubMed]
7. Li S, Yu H. Dietary and Urinary Iodine in Relation to Thyroid Cancer Risk: A Meta-Analysis. Cancer Control 2026;33:10732748261419192. [Crossref] [PubMed]
8. Kang C, Du Y, Li J, et al. Dissecting the causal association of diet with thyroid cancer: a systematic review with meta-analysis and mendelian randomization analysis. Front Nutr 2025;12:1664129. [Crossref] [PubMed]
9. Greca A, Grau L, Arbet J, et al. Anthropometric, dietary, and lifestyle factors and risk of advanced thyroid cancer: The NIH-AARP diet and health cohort study. Clin Endocrinol (Oxf) 2023;99:586-97. [Crossref] [PubMed]
10. Ludgate ME, Masetti G, Soares P. The relationship between the gut microbiota and thyroid disorders. Nat Rev Endocrinol 2024;20:511-25. [Crossref] [PubMed]
11. Barrea L, Pugliese G, Frias-Toral E, et al. Diet as a possible influencing factor in thyroid cancer incidence: the point of view of the nutritionist. Panminerva Med 2021;63:349-60. [Crossref] [PubMed]
12. Sanjulián L, Fernández-Rico S, González-Rodríguez N, et al. The Role of Dairy in Human Nutrition: Myths and Realities. Nutrients 2025;17:646. [Crossref] [PubMed]
13. Tanitame M, Sugawara Y, Lu Y, et al. Dairy consumption and incident risk of thyroid cancer in Japan: a pooled analysis of the Miyagi Cohort Study and the Ohsaki Cohort Study. Eur J Nutr 2023;62:251-9. [Crossref] [PubMed]
14. Fiore M, Cristaldi A, Okatyeva V, et al. Dietary habits and thyroid cancer risk: A hospital-based case-control study in Sicily (South Italy). Food Chem Toxicol 2020;146:111778. [Crossref] [PubMed]
15. Przybylik-Mazurek E, Hubalewska-Dydejczyk A, Kuźniarz-Rymarz S, et al. Dietary patterns as risk factors of differentiated thyroid carcinoma. Postepy Hig Med Dosw (Online) 2012;66:11-5. [Crossref] [PubMed]
16. Wells G. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Non-Randomised Studies in Meta-Analyses. J. Petersen 2014.
17. Kwon YJ, Lee HS, Kang SW, et al. Association between Consumption of Iodine-Rich Foods and Thyroid Cancer Prevalence: Findings from a Large Population-Based Study. Nutrients 2024;16:1041. [Crossref] [PubMed]
18. Nguyen LTD, Gunathilake M, Lee J, et al. Association between dietary habits and incident thyroid cancer: A prospective cohort study. Front Nutr 2023;10:1104925. [Crossref] [PubMed]
19. Braganza MZ, Potischman N, Park Y, et al. Adolescent and mid-life diet and subsequent risk of thyroid cancer in the NIH-AARP diet and health study. Int J Cancer 2015;137:2413-23. [Crossref] [PubMed]
20. Truong T, Baron-Dubourdieu D, Rougier Y, et al. Role of dietary iodine and cruciferous vegetables in thyroid cancer: a countrywide case-control study in New Caledonia. Cancer Causes Control 2010;21:1183-92. [Crossref] [PubMed]
21. Park Y, Leitzmann MF, Subar AF, et al. Dairy food, calcium, and risk of cancer in the NIH-AARP Diet and Health Study. Arch Intern Med 2009;169:391-401. [Crossref] [PubMed]
22. Galanti MR, Hansson L, Bergström R, et al. Diet and the risk of papillary and follicular thyroid carcinoma: a population-based case-control study in Sweden and Norway. Cancer Causes Control 1997;8:205-14. [Crossref] [PubMed]
23. Kotani A, Tatara Y, Sakamoto R, et al. Relationship between amino acid transporter activity and radioactive iodine therapy efficacy in differentiated thyroid cancer. Oncol Rep 2025;54:160. [Crossref] [PubMed]
24. Wålinder R, Isaksson M, Rääf C, et al. Protracted exposure to 134Cs and 137Cs gives substantial contribution to long-term thyroid absorbed dose after nuclear power plant accidents. Radiat Prot Dosimetry 2025;201:899-904. [Crossref] [PubMed]
25. Shulhai AM, Rotondo R, Petraroli M, et al. The Role of Nutrition on Thyroid Function. Nutrients 2024;16:2496. [Crossref] [PubMed]
26. Knuppel A, Fensom GK, Watts EL, et al. Circulating Insulin-like Growth Factor-I Concentrations and Risk of 30 Cancers: Prospective Analyses in UK Biobank. Cancer Res 2020;80:4014-21. [Crossref] [PubMed]
27. Köhrle J. Selenium, Iodine and Iron-Essential Trace Elements for Thyroid Hormone Synthesis and Metabolism. Int J Mol Sci 2023;24:3393. [Crossref] [PubMed]
28. Iuliano S, Mirabelli M, Giuliano S, et al. Insulin resistance and metabolic dysfunction in thyroid nodules and differentiated thyroid cancer. Curr Opin Oncol 2026;38:1-10. [Crossref] [PubMed]
29. Slurink IAL, Voortman T, Ochoa-Rosales C, et al. Dairy Product Consumption in Relation to Incident Prediabetes and Longitudinal Insulin Resistance in the Rotterdam Study. Nutrients 2022;14:415. [Crossref] [PubMed]
30. Shabanian S, Rozbeh A, Mohammadi B, et al. The Association between Vitamin D Deficiency and Fibrocystic Breast Disorder. Curr Mol Med 2024;24:899-905. [Crossref] [PubMed]
31. Halada S, Casado-Medrano V, Baran JA, et al. Hormonal Crosstalk Between Thyroid and Breast Cancer. Endocrinology 2022;163:bqac075. [Crossref] [PubMed]
32. Babić Leko M, Jureško I, Rozić I, et al. Vitamin D and the Thyroid: A Critical Review of the Current Evidence. Int J Mol Sci 2023;24:3586. [Crossref] [PubMed]