Introduction

Thyroid cancer is the most common malignancy of the endocrine system worldwide, ranking seventh in incidence among all cancers1. In China, its epidemiological significance is particularly notable: recent data indicate that thyroid cancer is the fourth most frequently diagnosed cancer in women and the leading cancer type among women under 30 years of age2. The predominant pathological subtype, differentiated thyroid cancer (DTC)--which includes papillary (PTC) and follicular (FTC) carcinomas–generally has favorable outcomes, with 5-year survival rates exceeding 98.5%3.

Despite this optimistic prognosis, postoperative recurrence remains a significant clinical challenge. Studies report recurrence rates as high as 10–30% over long-term follow-up4,5,6, necessitating secondary interventions that substantially affect patients’ quality of life and increase healthcare burdens. Current risk stratification models, such as those endorsed by the American Thyroid Association (ATA)7, incorporate factors like tumor size, lymph node metastasis, and extrathyroidal extension. However, these models have limitations in precision. For instance, the prognostic significance of molecular markers (e.g., BRAF V600E), demographic variables (e.g., gender disparities), and the extent of surgery remain incompletely defined, particularly in ethnically diverse populations like those in Western China.

This study addressed gaps in the existing literature or confirmed previous findings through the following approaches:

  • Evaluating understudied risk factors—including multifocality, BRAF V600E mutation status, and the extent of gland resection—alongside conventional clinicopathological variables.

  • Providing region-specific data from a large cohort in Xinjiang, China, where demographic and environmental factors may affect recurrence patterns.

  • Developing a clinically actionable nomogram to quantify individualized recurrence risk, thereby enabling tailored surveillance and treatment.

By integrating these elements, our work aims to enhance postoperative management strategies and reduce recurrence by enabling the early identification of high-risk patients.

Results

Analysis of recurrence

The follow-up period extends from the day of the patient’s surgery until June 30, 2024, with a follow-up duration of 36 to 60 months. Among 1,118 cases of DTC patients, 46 cases experienced recurrence during the follow-up period, with the time interval from surgery to recurrence ranging from 3 to 59 months. Based on recurrence status, patients were categorized into two groups: the recurrence group (46 cases) and the non-recurrence group (1,072 cases). During the early postoperative period, the frequency of outpatient follow-up visits was relatively high. Visits were conducted at 1 month, 3 months and 6 months after the operation. Laboratory blood tests were performed at 1 month and 3 months, and blood tests combined with ultrasound examinations were conducted at 6 months. Subsequently, telephone follow-ups were conducted every 3 months, and outpatient follow-ups were combined with blood tests and ultrasound examinations every 6 months. During the outpatient follow-ups, detailed physical examinations were also conducted for the patients. For patients with suspicious recurrence suggested by routine ultrasound, further puncture biopsy was performed. For patients with malignant pathology results from the puncture biopsy, secondary surgical treatment was carried out.The postoperative routine pathological diagnosis of malignancy confirmed the recurrence.

For patients with confirmed recurrence, RAI treatment was evaluated. Endocrine treatment was more standardized. The follow-up frequency was higher than that of patients without recurrence within 1 year after the operation. After outpatient follow-up 1 month after surgery, outpatient follow-up every 3 months, blood sampling and ultrasound examination were performed, and the amount of TSH inhibition therapy was adjusted in time. 1 year later, telephone follow-up was performed every 3 months, and outpatient follow-up was performed every 6 months.

Chi-square test for patients with DTC

Patients in the recurrence group and the non-recurrence group exhibited statistically significant differences in gender, tumor diameter, number of tumor lesions, cervical lymph node metastasis status, local tumor invasion, extent of gland resection, BRAF V600E gene mutation, and postoperative serum Tg levels (all P < 0.05). However, there were no statistically significant differences in age, BMI, tumor location, pathological stage, presence of HT, preoperative serum Tg, or preoperative and postoperative serum TSH, Tgab, and TPOab levels between the recurrence and non-recurrence groups (all P > 0.05). Please refer to Table 1 for further details.

Table 1 Comparison of basic information and clinical characteristics of patients [M(P25, P75), n(%)]
Full size table

Univariate Cox regression analysis of postoperative recurrence in patients with DTC

Univariate analysis revealed that patients’ gender, tumor diameter, number of tumor lesions, neck lymph node metastasis, local invasion of the tumor, extent of glandular lobectomy, and mutation of BRAF V600E significantly influenced the postoperative recurrence of thyroid cancer (P < 0.05). In contrast, age, BMI, tumor location, pathological stage, the presence of HT, and preoperative and postoperative serum levels of Tg, TSH, Tgab, and TPOab did not have a significant effect on postoperative recurrence (P > 0.05). Please refer to Table 2.

Table 2 Univariate Cox regression analysis of relapse correlation.
Full size table

Multifactorial Cox regression analysis of postoperative recurrence in patients with DTC

Please refer to Table 3. The indicators suggest that significant factors influencing prognosis in the univariate analysis will be utilized as independent variables, while recurrence will serve as the dependent variable in the multivariate regression analysis. The results indicate that male gender, tumor diameter, the number of tumor lesions, preoperative lymph node metastasis, and BRAF V600E gene mutation are independent risk factors for the recurrence of DTC (P < 0.05). Conversely, bilateral total thyroidectomy is identified as an independent protective factor against the recurrence of DTC (P < 0.05). Recurrence curves are presented separately for each of the following variables: excision scope (Fig. 1), central lymph node metastasis (Fig. 2), lateral lymph node metastasis (Fig. 3), and BRAF V600E gene mutation (Fig. 4).

Table 3 Cox regression analysis of multiple factors related to recurrence.
Full size table
Fig. 1
figure 1

Recurrence risk function with respect to excision scope.

Full size image
Fig. 2
figure 2

Recurrence risk function with or without central lymph node metastasis.

Full size image
Fig. 3
figure 3

Recurrence risk function with or without lateral lymph node metastasis.

Full size image
Fig. 4
figure 4

Recurrence risk function with or without BRAF V600E Gene Expression.

Full size image

Discussion

The impact of gender on prognosis

Studies have shown that the recurrence rate of thyroid cancer after surgery is higher in males than in females8,9. However, one study indicated no difference in prognosis between the two genders10. Our data demonstrate that male patients have a worse prognosis than female patients, and gender is an independent prognostic factor. A multicenter retrospective study also confirmed that males are a significant independent risk factor for specific mortality in patients with DTC associated with the BRAF V600E mutation11.

The impact of tumor size and pathological stage on prognosis

This study demonstrates that tumor diameter is an independent risk factor for the recurrence of DTC (P < 0.05). Recent research has also confirmed that tumor size serves as an independent predictor of survival without recurrence in DTC patients12. However, the specific survival rate for T3b stage DTC patients with a tumor size of ≤ 2 cm is not statistically different from that of T1 stage patients13. Additionally, this study indicates that there is no statistically significant difference in pathological staging between the recurrence group and the non-recurrence group.This may be due to the fact that in patients under the age of 55, regardless of tumor size or the presence of lymph node metastasis in the neck, they are classified as stage I as long as there is no distant metastasis. However, this classification does not accurately represent the quantitative relationship between pathological staging and prognosis. Furthermore, it is possible that the generally favorable prognosis of DTC leads to improved outcomes for patients across all stages, from stage I to stage IV, following surgical treatment.

The impact of tumor multifocality, local invasion, and neck lymph node metastasis on prognosis

Patients with multifocal and locally invasive DTC have a higher risk of central and lateral neck lymph node metastasis14,15. Consistent with relevant studies16, this research suggests that preoperative lymph node metastasis and the presence of multiple lesions are independent risk factors for the recurrence of DTC (P < 0.05), which may result in the existence of additional hidden lesions in the body. Although the lesions and lymph nodes identified prior to surgery may be completely excised, it remains challenging to entirely eliminate hidden lesions, potentially leading to an increased risk of postoperative recurrence. In this study, local tumor invasion is identified as a risk factor influencing the recurrence of thyroid cancer after surgery; however, it is not considered an independent risk factor.

The impact of thyroidectomy extent on prognosis

The extent of initial surgery for thyroid cancer has long been a subject of debate among experts. The latest guidelines in China recommend that patients with tumors larger than 4 cm in diameter, tumors larger than 1 cm located in the isthmus, positive surgical margins, visible extrathyroidal invasion, vascular invasion, bilateral multifocal lesions, clinical lymph node metastasis (number of nodes ≥ 5 or diameter ≥ 3 cm) or distant metastasis should undergo total or near-total thyroidectomy. The American Thyroid Association (ATA) recommends that patients with tumors larger than 4 cm, those exhibiting obvious extrathyroidal invasion, or those with confirmed lymph node metastasis should undergo total or near-total thyroidectomy7. The extent of resection of the primary tumor significantly impacts the recurrence of DTC. In this study, bilateral thyroid lobe resection was identified as an independent protective factor against the recurrence of DTC (P < 0.05). A retrospective analysis of Korean patients with DTC indicated that total thyroidectomy did not provide a survival advantage over lobectomy, except in specific high-risk groups17. Surgeons must balance the risk of postoperative complications, such as hypoparathyroidism and recurrent laryngeal nerve injury18, while considering factors such as tumor size and regional lymph node metastasis, etc. to tailor an individualized surgical strategy that optimizes postoperative quality of life and minimizes the risk of recurrence.

The impact of BRAF V600E gene mutation on prognosis

V600E mutation is the most frequently observed alteration in the BRAF oncogene, exhibiting a high mutation rate19. Although the ATA guidelines indicate that the specificity and sensitivity of the BRAF V600E gene mutation in predicting the risk of recurrence in DTC remain contentious16, study have demonstrated that the BRAF V600E mutation is positively correlated with aggressive factors such as lymph node metastasis and lymphatic vessel invasion in DTC20. Furthermore, several studies have confirmed that the BRAF V600E gene mutation is closely associated with an increased risk of recurrence21,22. This study establishes that the BRAF V600E gene mutation is an independent risk factor for recurrence in DTC (P < 0.05). For patients with DTC harboring the BRAF V600E gene mutation, a more aggressive surgical approach is recommended, followed by postoperative evaluation for 131I clearance therapy, and follow-up should be conducted with greater rigor.

The impact of HT on prognosis

HT is negatively correlated with the aggressive pathological features of DTC; however, it does not significantly impact overall survival or disease-free survival in patients following surgery23,24. Additionally, it is rarely identified as an independent protective factor in multivariate analyses25. In this study, there was no statistically significant difference in the recurrence rates between patients with HT and those without it in both the recurrence and non-recurrence groups (both P > 0.05). However, a limitation of this study is that it did not analyze patients with HT who developed DTC, focusing only on cases where both conditions coexisted. Furthermore, the prior diagnosis and treatment history of HT in the patients were not documented.

The impact of laboratory test results on prognosis

Luca et al. established a decision tree model to predict the recurrence of DTC based on postoperative serum Tg levels26. Meanwhile, Martin et al. indicated through retrospective analysis that preoperative Tg can guide the initial treatment and follow-up management of DTC27. In this study, single-factor Cox regression analysis demonstrated that preoperative and postoperative serum Tg, TSH, Tgab, and TPOab had no significant impact on postoperative recurrence (P > 0.05).

While postoperative serum Tg showed statistical significance in our univariate chi-square test (P < 0.05), its utility as a standalone marker for recurrence is limited in cohorts with mixed surgical approaches. In patients with residual thyroid tissue (e.g., after lobectomy), elevated Tg may reflect physiological remnant activity rather than recurrence. Therefore, Tg trends should be interpreted contextually: in total thyroidectomy patients, a Tg level greater than 1 ng/mL suggests an increased risk of recurrence; in lobectomy patients, a Tg level exceeding 10 ng/mL or rising trends warrant further investigation.

Preoperative serum Tg, TSH, Tgab, and TPOab can be combined with relevant imaging examinations to assist in the diagnosis of thyroid tumors. Additionally, postoperative serum levels can aid in assessing prognosis, with serum Tg being a crucial indicator for monitoring residual thyroid tissue after total thyroidectomy or radioactive iodine (131I) therapy for DTC. Although HT does not worsen the prognosis of DTC24, it is still recommended to monitor TPOab levels both preoperatively and postoperatively to mitigate the uncertain effects of HT.

Clinical application of the nomogram

The nomogram developed from our multivariate Cox regression model serves as a practical tool for assessing individualized recurrence risk in DTC patients. Clinicians can determine a patient’s probability of recurrence by summing the points assigned to key predictors. Please refer to Fig. 5.

Fig. 5
figure 5

Nomogram for recurrence risk in DTC patients. (To use it, locate each predictor variable, draw a vertical line to the “Points” axis to assign points, sum all points, and project the total to the “Linear Predictor” axes. Higher total points indicate increased recurrence risk.)

Full size image

Clinical Utility:

  • Risk Stratification: This process identifies high-risk patients (e.g., those with total points exceeding 100) who require intensive surveillance. As mentioned earlier, frequent follow-ups were conducted.

  • Treatment Guidance: Provides support for decisions regarding bilateral resection or adjuvant RAI therapy in high-risk cases. (Current guidelines7 strongly recommend adjuvant RAI for high-risk DTC, such as cases with gross extrathyroidal extension or lateral N1b metastasis.)

  • Patient Counseling: Effectively conveys personalized recurrence risk through visual communication.

Conclusions

Combined with our data, the developed nomogram provides a clinically actionable tool for recurrence risk quantification. We recommend that high-risk patients with differentiated thyroid cancer—such as those with bilateral lymph node metastasis identified through imaging examinations, BRAF V600E gene mutations, and those with large tumor volumes and multifocal lesions revealed by rapid intraoperative pathological examination—be considered for bilateral thyroid lobectomy. The nomogram objectively identifies these high-risk profiles and quantifies how surgical choices influence recurrence probability. For the management of cervical lymph nodes, therapeutic cervical lymph node dissection is advised. A comprehensive perioperative evaluation should be conducted, and preoperative communication with patients must be prioritized.

There are still some limitations in this experiment. First, the follow-up period in this study was limited to 36 to 60 months. Although this duration is sufficient to capture early recurrences, it may not be adequate for identifying late recurrences, which can occur many years after initial treatment. Longer follow-up periods are necessary to fully understand the long-term recurrence patterns and outcomes in DTC patients. Second, the study did not document the specific locations of tumor occurrences within the thyroid gland. The precise location of the tumor (e.g., upper pole, lower pole, isthmus) may influence the risk of recurrence and should be considered in future studies. Despite these limitations, the findings of this study contribute to the growing body of evidence regarding the risk factors for postoperative recurrence of DTC. Addressing these limitations in future research will provide a more comprehensive understanding of DTC recurrence and help refine clinical guidelines for the management and follow-up of DTC patients.

Methods

Selection and characteristics of the patients

Based on the established inclusion and exclusion criteria, a total of 1,118 patients diagnosed with DTC after undergoing thyroid cancer surgery between June 2019 and June 2021 were included in this study with a follow-up period of 36 to 60 months (i.e., 3 to 5 years). **Inclusion criteria: ** 1)Patients with a postoperative pathological diagnosis of differentiated thyroid carcinoma (DTC), including papillary thyroid carcinoma (PTC) and follicular thyroid carcinoma (FTC); 2) Patients with complete clinical data who consented to participate in follow-up visits, either by telephone or in-person outpatient visits. **Exclusion criteria: ** (1) Pathological diagnosis of undifferentiated cancer, medullary cancer, or benign tumors; (2) Patients with concomitant malignant tumors at other sites or distant metastases to the liver, lungs, etc.; (3) Patients with incomplete clinical data, those who refused follow-up visits, or those lost to follow-up (i.e., failed to attend outpatient reviews or provided incorrect contact information).

Study groups and data collection

We conducted telephone follow-ups, outpatient follow-ups, and other methods to monitor selected patients. If the routine pathological examination results from the second surgery indicate recurrence of thyroid bed or residual thyroid tissue, central zone lymph node metastasis, or lateral neck lymph node metastasis, it is classified as recurrence. We then document the recurrence status of the patients. Patients are categorized into a recurrence group and a non-recurrence group based on the presence of DTC in the pathology results. Data from both groups are collected, including:

  • Basic information includes age, gender, and Body Mass Index (BMI).

  • Pathology-related information includes the TNM stage (refer to the 2015 ATA guidelines for the diagnosis and treatment of thyroid nodules and differentiated thyroid cancers16, the size of the primary tumor (measured by its maximum diameter), the presence of single or multiple tumors, local invasion(the tumor clearly invades surrounding structures and organs, including the recurrent laryngeal nerve, trachea, esophagus, larynx, and the outer banded muscles, such as the hyoid, sternohyoid, sternothyroid, and thyrohyoid muscles), the presence of neck lymph node metastasis, co-occurrence with Hashimoto’s thyroiditis (HT), and the mutation of the BRAF V600E gene (the status of BRAF V600E was clinically assessed in tumor tissues using a PCR-based detection method conducted by the CAP-certified laboratory at this center).

  • Treatment-related factors include the extent of surgical resection (bilateral total thyroidectomy, subtotal thyroidectomy, or unilateral lobectomy), as well as preoperative and postoperative serum levels of thyroid-stimulating hormone (TSH), thyroglobulin (Tg), thyroglobulin antibody (Tgab), and thyroid peroxidase antibody (TPOab).

Thyroid hormone supplementation is initiated on the second day following a total thyroidectomy. The dosage is adjusted based on laboratory results obtained one month post-surgery. In the case of unilateral thyroid lobectomy, the necessity for thyroid hormone supplementation and the appropriate dosage are determined based on laboratory results one month after the surgery. Patients with DTC who are at medium to high risk of recurrence receive radioactive iodine (131I) therapy as an adjuvant treatment. Thyroid hormone supplementation begins 24 to 72 h after RAI treatment, with dosage adjustments made after a follow-up period of 4 to 6 weeks.Medication is administered according to individual patient conditions to maintain normal thyroid function.

Statistical method

Using SPSS 26.0 software for data analysis, count data is represented as n (%), and comparisons between two groups are conducted using the χ² test. Non-normally distributed quantitative data is expressed as M(P 25, P 75), and comparisons between two groups are performed using the Mann-Whitney U test. After conducting univariate Cox regression analysis on correlated data, multivariable Cox regression analysis is employed to assess the risk factors influencing recurrence. A p-value of less than 0.05 is considered statistically significant.Please refer to Table 4 for detailed assignments.

Table 4 Assignment table.
Full size table

The bootstrap method was employed to independently sample 1,000 times for verification. The deviation of the regression equation B was small, and the variables in the model remained significant. Consequently, the model based on this real-world data was more robust.

A statement about ethics

This retrospective study was conducted in accordance with the Declaration of Helsinki and received approval from the Ethics Committee of the First Affiliated Hospital of Xinjiang Medical University (Approval Number: K202412-52).

Due to the retrospective nature of the study, which involved analyzing anonymized existing clinical data without any additional intervention or risk to participants, the Ethics Committee waived the requirement for obtaining individual informed consent. Patient confidentiality was strictly maintained throughout the research process; all personal identifiers were removed or anonymized prior to data analysis.