Thyroid cancer is one of the most prevalent endocrine malignancies, with its annual incidence increasing in numerous countries and regions. However, the mortality rate remains consistently low (1, 2). Differentiated thyroid cancer (DTC) is the most frequently diagnosed type, accounting for over 95% of cases (3). DTC comprises two main subtypes, with papillary thyroid cancer (PTC) account for 90% of cases and follicular thyroid cancer accounting for 7–8% (4–6). In recent years, increasing studies have reported the association of thyroid cancer with various risk factors, including radiation exposure, environmental factors, iodine intake, serum TSH levels and Hashimoto’s thyroiditis (HT) (7–9). Among these factors, thyroid autoimmunity is closely linked to the development of DTC.

Thyroid peroxidase (TPO), an enzyme located in the apical membrane of thyroid follicular cells, is involved in the biosynthesis of thyroid hormone (10). Thyroid peroxidase antibody (TPOAb) is predominantly produced by lymphocytes infiltrating the thyroid gland and is one of the most common autoantibodies against the thyroid (11). It can cause thyroid cell damage through the activation of the complement system and cell cytotoxicity (12, 13). TPOAb is a hallmark thyroid autoimmune antibody in autoimmune thyroiditis (AIT) (14, 15). It is also intricately linked to several other diseases. In 1998, Smyth et al. observed significantly higher TPOAb positivity in breast cancer patients than in those with benign breast disease. Some studies have shown that TPO is expressed in breast cancer and peri-cancerous tissues, and the antigenic and biochemical properties of TPO in breast tissues are similar to those in thyroid tissues, with only minor differences in post-translational modifications (16–18). Furthermore, disease-free and overall survival were longer in TPOAb+ breast cancer patients than in their TPOAb– breast cancer counterparts (19). Numerous subsequent studies have yield similar findings (20–22). Sharma and collaborators proposed that TPOAb might cross-react with lactoperoxidase expressed in breast tissue, potentially explaining the coexistence of TPOAb and breast cancer (23). Additionally, several studies have highlighted a strong association between TPOAb and various adverse pregnancy outcomes (24–26). Furthermore, two separate fine needle aspiration cytology studies from the same center identified TPOAb as an independent risk factor for thyroid malignancy (27, 28). Nonetheless, while many studies have explored the relationship between thyroglobulin antibody (TgAb), another serum marker of AIT, and DTC, the association between TPOAb and DTC remains unclear (29–31).

Previously, a variety of diseases or molecules have been found to be associated with the prognosis of DTC, such as graves’ disease, HT, thyrotropin, thyroglobulin (Tg), BRAF mutation, estrogen receptor and VEGF pathway, etc (32–38). And we aimed to conduct a comprehensive meta-analysis and systematic review to elucidate the relationship between TPOAb and DTC. By analyzing extensive large-scale research data, we expect to provide highly in-depth theoretical support for the clinical application of TPOAb in the early diagnosis, formulation of treatment strategies, and prognosis evaluation of DTC.

Our meta-analysis included 12 original studies encompassing 20,330 patients from six countries for investigating associations between TPOAb and DTC prevalence, along with various prognostic factors. Our findings reveal that TPOAb+ individuals were at a higher risk of developing DTC than TPOAb– individuals (p=0.049). Furthermore, TPOAb+ DTC patients were more likely than TPOAb– DTC patients to present with multifocal and bilateral tumors, and this difference was statistically significant. However, no significant difference in tumor size, recurrence rate, risk of extrathyroidal extension, and lymph node metastasis was observed between TPOAb+ and TPOAb– DTC patients.

We observed an association between TPOAb and DTC prevalence, but the results may be subject to controversy. Upon analysis using Review Manager, we obtained a calculated p-value of 0.05, while recalculation using Stata 14 yielded a p-value of 0.049, which was statistically significant. This discrepancy might be because Review Manager rounded the p-value, and ultimately, the p-value was determined to be 0.049. Nevertheless, both the heterogeneity and sensitivity of this result were relatively high. Several factors may account for this observed heterogeneity. First, variations in the cutoff values for defining TPOAb positivity were evident, ranging from a minimum of >5.61 IU/mL to a maximum of >100 IU/mL. Second, differences existed in the methods and instruments used for TPOAb determination. Third, differences in the study populations across diverse regions may have contributed to heterogeneity. Lastly, the presence or absence of AIT in TPOAb+ patients could have influenced the results. The available data were not sufficient for a comprehensive subgroup analysis. Furthermore, the observed sensitivity issues were likely related to ethnic and regional differences. In the nine included articles, three studies contributing to the elevated sensitivity were from joint research conducted between the United States and Greece, Serbia, and Turkey, while the remaining six studies were conducted in China. Located in Asia, China differs geographically from the other four countries primarily situated in Europe and North America. As DTC prevalence varies across regions, geographic factors might play a pivotal role (51, 52). Although the p-value was <0.05, this does not inherently guarantee result stability or high reliability. Further validation through additional independent studies is essential to establish the reliability of the results.

We initially observed that TPOAb+ DTC patients are more likely to exhibit bilateral and multifocal tumors. However, Wang’s study (47) with a considerably higher weight share in both results had a significant influence. Upon exclusion of the Wang’s study, we found no significant association between TPOAb and DTC multifocality (OR=0.91 [95% CI: 0.56–1.48], p=0.70) and bilaterality (OR=1.41 [95% CI: 0.74–2.66], p=0.29). Consequently, no conclusive significant association was established in this study, emphasizing the need for further exploration of the association of TPOAb positivity with DTC multifocality and bilaterality. Moreover, we found the TPOAb+ and TPOAb– DTC patients exhibited no significant differences in tumor size, recurrence rate, risk of extrathyroidal extension, and lymph node metastasis. Nevertheless, several studies have reported that TgAb+ DTC patients face a higher risk of lymph node metastasis and recurrence (53–57).

Some studies have demonstrated that persistent or recurrent thyroid cancer can induce the production of TgAb, so the time of TPOAb detection is particularly important (58, 59). Other studies have indicated an association between positive TPOAb results 7-10 years before diagnosis and thyroid cancer. However, most of the original studies included in this analysis conducted TPOAb testing prior to surgery without specifying the exact duration of TPOAb positivity before this period, potentially impacting the credibility of the results (60). Further large prospective studies are necessary for validation. Additionally, if patients were identified with thyroid cancer due to testing positive for TPOAb and subsequently included in the selection process, it could have led to an increased rate of DTC diagnosis in TPOAb+ patients, potentially introducing selection bias. Moreover, because patients with autoimmune thyroid diseases tend to be more vigilant about their thyroid function, they may detect small subclinical tumors during ongoing medical surveillance. In contrast, patients without related diseases might not undergo continuous surveillance, potentially leading to the undiagnosed status of some subclinical tumors. This discrepancy could also introduce bias into the results.

Given that TPOAb is a vital diagnostic marker of HT, considering the influence of AIT when examining the association between TPOAb and DTC is necessary. HT is known to cause substantial immune cell infiltration in the thyroid gland (61), and similar immune cell infiltration is observed around DTC (62, 63). The potential interplay between the two and its effect on DTC progression and prognosis pose intriguing questions. HT appears to be linked to an increased risk of thyroid cancer (64). When it coexists with thyroid cancer, it seems to further elevate the likelihood of thyroid cancer development (64–66). However, it may confer a protective effect against lymph node metastasis, extrathyroidal extension, and distant metastasis (64, 67, 68). Additionally, some studies have shown that immune cell infiltration surrounding PTC resulting from HT is positively correlated with a lower recurrence rate, higher overall survival, and reduced risk of extrathyroidal extension and distant metastasis (69, 70). Moreover, some studies have found that HT is associated with smaller tumor size, lower rate of aggressive PTC variants and lower risk, and DTC patients with HT have a higher clinical remission rate and longer recurrence-free survival (71, 72). However, the accuracy of “non-HT status” as a negative prognostic marker is poor, and it cannot improve the specificity of predicting prognosis. In one meta-analysis, TgAb was distinguished from HT, and TgAb+ patients with HT exhibited larger tumor sizes and a higher risk of extrathyroidal extension, tumor multifocality, lymph node metastasis, and cancer persistence than those without HT (29). Therefore, although HT can promote the development of DTC, it seems to be a protective factor for the prognosis of DTC. However, if we consider the single effect of TgAb positivity without considering HT, it seems to become a risk factor. The studies included in our meta-analysis did not differentiate TPOAb from HT, making it challenging to determine whether our conclusions might have been confounded by HT. Further research is warranted to reach a more definitive conclusion.

Our analysis indicates that TPOAb positivity may be associated with an increased prevalence of DTC. However, TPOAb does not serve as a robust prognostic factor for TPOAb+ DTC patients. Our meta-analysis is based on a limited number of included studies, and we anticipate that future research will provide additional large-scale research data to further inform our understanding.

The original contributions presented in the study are included in the article/ Supplementary Material , further inquiries can be directed to the corresponding author.

HZ: Writing – original draft. LT: Writing – review & editing. XW: Writing – review & editing. XS: Writing – review & editing.