The results of our study, performed in the largest reported homogenous cohort of subjects to date, confirmed that SNPs may be involved in the clinical course of DTC. Among these polymorphisms, the NOD2 rs2066842 polymorphism may distinguish PTC from FTC. It may also be a marker of better response to RAI in DTC patients, as it was linked with a greater decrease in sTg before and after the first RAI administration. Moreover, NFKB1A rs696 may predict a lower stage of FTC and a lower probability of PTC multifocality.

Critical factors influencing thyroid tumorigenesis are genetic. A major role is attributed to the activation of proto-oncogenes, wherein the basal expression level of oncogenes regulates cell growth and differentiation by participating in signal transduction to the cell nucleus. These include growth factors, receptors for growth factors, tyrosine kinases, and transcription factors.

Like the activation of oncogenes, the inactivation of tumor suppressor genes, such as TP53, whose function is to prevent tumorigenesis, is of great importance. Loss or inactivation of both alleles of a tumor suppressor gene leads to tumor development. Mutation of the TP53 gene is the most known genetic alteration involved in human cancer formation. On the other hand, mutations in the TP53 suppressor gene are associated with mutations in the CHEK2 gene and are associated with the Li-Fraumeni syndrome (LFS) phenotype. Mutation in mutator genes that can repair mutations leading to oncogene activation or suppressor gene inactivation may also be a cause of the tumorigenesis process.

Other genetic factors may include abnormalities in growth factors and their receptors, such as tumor necrosis factor (TNF) and other cytokines. Because single nucleotide changes may result in increased protein expression, a variant of the TNF gene has also been of interest to us.

The nucleotide-binding oligomerization domain containing 2 (NOD2) gene, also known as CARD15, is located at chromosomal region 16q21 [44]. NOD2 is a member of the evolutionarily conserved Nod-like receptor (NLR) family and plays a crucial role in detecting elements within microbial cell walls [45, 46]. The gene is a NOD1/APAF-1 family member that encodes proteins with two caspase recruitment domains and six leucine-rich repeats (LRRs) [47]. Studies suggest its involvement in controlling both programmed cell death (apoptosis) and long-term inflammatory disorders. The protein NOD2 is primarily expressed in humans in peripheral blood leukocytes [47]. It plays a vital role in the immune response, especially involving intracellular bacterial lipopolysaccharides (LPSs), by recognizing muramyl dipeptide (MDP), which is derived from bacteria and, in this way, activating the NFκB protein [47]. Interestingly, it has been suggested that the NOD2 protein acts as a regulator of appetite by sensing MDP molecules in a subset of brain neurons: microbial MDP may reach the brain and bind and activate the NOD2 protein in inhibitory hypothalamic neurons, leading to a decrease in neuronal activity, thus regulating satiety perception and probably influencing body temperature [48].

The role of NOD2 has yet to be determined, but it seems that it might be involved in many immunological pathways. It plays an essential role as a regulator of autophagy, especially in dendritic cells, via its interaction with the ATG16L1 protein, possibly through recruiting ATG16L1 at the site of bacterial entry [49]. Studies have also shown that NOD2 activation in the small intestine crypt contributes to intestinal stem cell survival and proper function. Hence, NOD2 may act by promoting mitophagy via its association with ATG16L1, as mentioned above [50, 51]. In addition to its important role in innate immunity, NOD2 controls the adaptive immune system by regulating helper T cells and regulatory T cells (Tregs) [50, 51]. In addition to being involved in pathogen recognition, NOD2 is involved in the endoplasmic reticulum stress response. It may act by sensing and binding to sphingosine-1-phosphate (S1P), a cytosolic metabolite generated in human cells in response to endoplasmic reticulum stress. This event usually initiates an inflammatory process leading to the activation of NF-kappa-B and MAP kinase signaling, which is very often dysregulated in cancer [50, 51].

The rs2066842 missense mutation is one of the most studied polymorphisms. It was initially found to be associated with an increased risk of Blau syndrome, as well as Crohn’s disease and ulcerative colitis [52]. It is located at coding regions and might affect the expression and function of NOD2 by altering amino acids. For the first time, NOD2 polymorphisms were linked to the risk of colorectal cancer [53]. Subsequently, further studies revealed an association between NOD2 polymorphisms and the risk of various cancers, including breast cancer and ovarian, endometrial, gastric, and laryngeal cancers [54]. However, the findings across individual studies varied and lacked consistency. A meta-analysis investigating overall cancer risk in relation to NOD2 polymorphisms showed that the NOD2 rs2066844 C/T, rs2066845 C/G, and rs2066847 (3020insC) polymorphisms might be associated with increased cancer risk. However, no significant association was observed between the NOD2 rs2066842 C/T polymorphism and cancer risk. However, thyroid cancer was not included in the analysis [54]. Only one group has studied the role of rs2066842 in thyroid cancer, and no increase in the occurrence of this gene was detected [55].

Our study showed a difference between FTC and PTC and was even more clinically relevant. Our study showed that the NOD2 rs2066842 polymorphism was linked to a better response to the RAI. NOD2 rs2066842 was linked to a greater decrease in the sTg marker after the first RAI.

The sTg measurement is the cornerstone of modern management of differentiated thyroid cancer, and clinical decisions on treatment and follow-up are based on the results of such measurements [35]. Tg production occurs primarily within normal and well-differentiated malignant thyrocytes, rendering Tg an ideal “tumor marker” after elimination of both healthy and pathological tissue. The introduction of highly sensitive thyroglobulin assays has significantly enhanced analytical sensitivity and stability in measuring Tg within the lower detection range. This has substantial implications for interpreting results in contemporary clinical practice [35]. A significant correlation existed between decreasing concentrations of sTg and the disappearance of thyroid tissue and between decreasing concentrations of sTg and RAI treatment efficacy.

RAI ablation success is the main predictor of DTC overall prognosis. Therefore, markers of RAI efficacy are highly needed. According to our study, NOD2 rs2066842 may be an easy and stable marker for the prediction of RAI efficacy.

It is not clear which mechanism, NOD2 rs2066842, may facilitate RAI action. One possible hypothesis is that the inflammatory process in which the NOD2 pathway participates may be involved in RAI action or may be protective against the progression of DTC. RAI activity is determined by host genetic background rather than the environment [56]. Therefore, SNPs may demonstrate prognostic value in DTC. Based on our data, the SNP NOD2 rs2066842 seems to be the most feasible candidate for determining the efficacy of RAI in DTC patients, thus affecting prognosis. Increased expression of NOD2 has been documented in various human metabolic disorders and chronic diseases linked to mitochondrial dysfunction [57]. It may manifest through direct signaling pathways, indirect modulation of cellular stress responses, or exacerbation of inflammatory processes [57]. RAI therapy in patients with DTC may increase proliferative lymphocyte responses and interferon-γ levels, demonstrating its proinflammatory function [58, 59].

The ongoing discourse revolves around the pivotal role of the cancer microenvironment. Cunha et al. proposed that the protective effect of CLT against the spread of DTC could be attributed to the involvement of various immune cells, such as CD4 + , CD8 + , CD201 + , Th17, and regulatory T cells (Tregs) [60]. Additionally, the interleukin-1 (IL-1) secreted by infiltrating lymphocytes in the CLT might exert an antitumorigenic effect by influencing the differentiation and replication of DTC cells [56, 61, 62]. Immune-mediated destruction of thyroid tissue may involve the NOD2 pathway, and our findings revealed the possible role of NOD2 rs2066842.

The same NOD2 polymorphism was found to be protective against multiple sclerosis and ameliorate the response to interferon-beta therapy, most likely as a part of a complex multidirectional system of genetic, immunological and environmental factors [63]. The NOD2 pathway may facilitate the action of pattern recognition receptors, which are germline-encoded host sensors that detect pathogens and play a crucial role in the proper function of the innate immune system [44].

According to the results of our study, NOD2 rs2066842 may serve as a marker for response to RAI, complementing the established role of Tg.

We suggest the additional use of NOD2 rs2066842 as a marker for response to RAI in patients with CLT and high concentration of aTg, where we suspect possible interference of Tg and aTg [35], potentially confounding and biasing the role of Tg as a marker of DTC. NOD2 rs2066842 may offer superior prognostication for response to RAI compared to Tg alone and facilitate better planning of RAI treatment. In patients who are NOD2 rs2066842-negative, it should be considered whether administering a higher activity of RAI would achieve the same effect as in NOD2 rs2066842-positive patients. The inflammatory process mediated through the NOD2 pathway and the activation of NF-kappa-B and MAP kinase signaling may enhance RAI efficacy. However, this hypothesis requires further investigation. In the text of our publication, we have clarified the suggested clinical role of NOD2 rs2066842 and its relationship with Tg.

Among the analyzed SNPs, NFKB1A rs696 may predict a lower stage of FTC and a lower probability of PTC multifocality. In inflammation, nuclear factor kappa B (NFKB) potentially exerts an oncogenic influence by fostering the proliferation and survival of numerous solid tumors [64]. Its protective role in DTC is consistent with the finding from the meta-analysis of Zhang et al. that NFKB1A rs696 seems to be associated with decreased susceptibility to cancer, especially Hodgkin lymphoma [65].

The nuclear factor-κB (NF-κB) gene, NFKB1A, resides on chromosome 14q13. The polymorphic variants within this gene rs2233406, rs3138053, and rs696 are positioned in regions that bind to CCAAT/enhancer binding protein and GATA binding protein 2. These variations potentially modulate the expression of IκBα, consequently impacting the activation of NF-κB. Moreover, these polymorphisms (rs2233406, rs3138053, and rs696) are directly associated with processes such as apoptosis, irregular immune cell maturation, and slowed cell growth [65, 66]. NF-κB belongs to a family of transcription factors that play crucial roles in inflammatory and immune reactions [67]. The protective effect of NFKB1A rs696 against higher stages of FTC and multifocality may be involved in these reactions.

Our findings are affected by a factor that restricts their interpretation: selection bias. It is crucial to acknowledge this influence before drawing any conclusions. To reduce bias in selection and control for external factors, our study group primarily comprises a homogenous population (with just 4.7% nonethnic Poles within Poland). This places our group in the seventh position among 159 countries worldwide and third in Europe in terms of homogeneity [68].

Another limitation was the difference in methods used for sTg measurement (as different assays have been used in recent years [35]); however, every patient was a control for themselves, and only the difference in sTg concentration before and after RAI was considered significant in our study.

Additionally, there was a notable difference in the prevalence of patients diagnosed with PTC compared to those diagnosed with FTC in the study group. However, this disproportion is reflective of the respective ratios of these diagnoses within the general population [28].

There were missing data in our study, including genetic analysis of SNPs, particular parameters of tumor histopathology, and particular measurements of thyroglobulin. To address these issues, we excluded missing data from the analysis by pairwise deletion (we used available data for each specific analysis).

Acknowledging the limitations of this study, it is pertinent to note that SNP analysis may be considered somewhat outdated in the context of the multiomics era. Consequently, conducting research involving genome sequencing is a more advanced and beneficial approach. However, we decided to analyze SNPs because of the ease of their performance in laboratories and the relatively low price of these methods, which make it possible to perform these tests on all patients in many countries; additionally, these methods are reimbursed by national insurance systems, which is still not true for multiomics. Due to its clinical utility and cost-effectiveness, SNP analysis has also been used to study thyroid cancer [55, 69, 70].