A dynamic nomogram and risk stratification system for predicting cancer-specific survival in patients with locally advanced differentiated thyroid cancer: a population-based study

Introduction

Thyroid cancer (TC) is a common type of endocrine cancer worldwide, and its incidence has significantly increased over the past decade. In 2022, there were approximately 821,000 new cases of TC globally, ranking seventh in cancer incidence and posing a significant global healthcare burden (1,2). The vast majority of histological subtypes of TC is differentiated thyroid cancer (DTC), including papillary TC (~84%), follicular TC (~4%), and oncocytic TC (~2%) (3).

Locally advanced differentiated thyroid cancer (LADTC) refers to a stage of DTC in which the tumor extends beyond the thyroid gland and invades nearby structures, such as the larynx, trachea, esophagus, recurrent laryngeal nerve, blood vessels or lymph nodes. Currently, there is no clear definition for LADTC in the American Thyroid Association guidelines or the National Comprehensive Cancer Network guidelines. Referring to the American Joint Committee on Cancer (AJCC) staging system and previous studies, those DTC patients with T4 stage were classified as LADTC in this study, regardless of whether they had lymph node metastases or distant metastases (4-7). Previous research has indicated that approximately 20% of patients with DTC have tumors that directly extend and invade neighboring tissues (5). Some of these tumors are inoperable or difficult to remove completely, leading to a major cause of mortality among patients with LADTC. The 10-year cancer-specific mortality rate of LADTC can reach 41–42% (8). This aggressive form of DTC presents significant challenges in terms of diagnosis, treatment, and prognosis, and remains a focal point of difficulty in its treatment.

Traditionally, the AJCC staging system has been a commonly used tool for evaluating the prognosis of cancer patients (9). However, there are several limitations, including low accuracy, overlooking important demographic and clinicopathological characteristics (such as age, therapy, or marital status), and poor performance in predicting individual survival outcomes. In recent years, there has been growing interest in the development of predictive models and tools that can provide individualized prognostic information. Nomograms, in particular, have emerged as valuable tools for predicting cancer outcomes by integrating multiple prognostic factors into a single predictive model. The use of nomograms allows for the estimation of patient-specific risks, which can aid in treatment decision-making and patient counseling. Nomograms have become increasingly popular as predictive tools in the field of oncology, gradually replacing the traditional AJCC staging system (10-13), and numerous nomograms have been developed and utilized for identifying TC (14-16). However, for LADTC, which is characterized by a high incidence and poor prognosis, there is still a lack of reliable and effective nomograms for predicting patient prognosis. Therefore, the development of an individualized prognostic nomogram is crucial for patients with LADTC.

The aim of this study was to develop a prognostic nomogram by utilizing data obtained from the Surveillance, Epidemiology, and End Results (SEER) database. Specifically, we aimed to develop a novel predictive tool that incorporates both clinical and pathological variables to improve the accuracy of prognostic predictions. This tool has the potential to be a valuable clinical tool for risk stratification and to assist clinicians in making informed decisions for patients with LADTC. Its application can improve patient outcomes and provide a scientific basis for guiding clinical practice. We present this article in accordance with the TRIPOD reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-2025-111/rc).

Methods

Data source

The data for this study were extracted from the SEER database, a registry funded by the National Cancer Institute. This multicenter and multipopulation registry is not subject to medical ethics review and does not require informed consent as patient data are anonymized and publicly available. The specific data used for this study were extracted from “Incidence-SEER Research Data, 17 Registries, Nov 2020 Sub (2000-2020)” using the SEER*Stat software version 8.4.2 (https://seer.cancer.gov/data-software/). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Patients with TC were classified into different histological subtypes using the histology codes from the International Classification of Diseases for Oncology, 3rd edition (ICD-O-3), including DTC and undifferentiated TC (UTC). The histological subtypes were identified using specific ICD-O-3 codes. Classic papillary thyroid cancer (C-PTC) was identified by the codes 8050/3, 8260/3, and 8343/3, while variant papillary thyroid cancer (V-PTC) was identified by the codes 8340/3, 8350/3, 8344/3, 8052/3, 8130/3, and 8342/3. Follicular thyroid cancer (FTC) was identified by the codes 8330/3, 8331/3, 8332/3, and 8335/3. Medullary thyroid cancer (MTC) was identified by the codes 8345/3, 8510/3, 8346/3, and 8347/3. Anaplastic thyroid cancer (ATC) was identified by the code 8021/3. C-PTC, V-PTC, and FTC were classified as DTC, whereas MTC and ATC were classified as UTC (17). This study focused specifically on collecting data related to the C-PTC, V-PTC, and FTC subtypes.

The following specific criteria were utilized for inclusion in the study: (I) the primary site was the thyroid gland, labelled C73.9; (II) the patient was diagnosed from 2004 to 2020; (III) the ICD-0-3 codes were 8050/3, 8260/3, 8343/3, 8340/3, 8350/3, 8344/3, 8052/3, 8130/3, 8342/3, 8330/3, 8331/3, 8332/3 and 8335/3; (IV) the T stage was T4. The following criteria were used for exclusion: (I) surgery or radiation unknown; (II) node (N) stage or metastasis (M) stage unknown; (III) tumor size unknown; (IV) survival month <1; (V) race or marital status unknown.

Variable evaluation and definition

In our retrospective cohort study, a total of 4,856 patients diagnosed with LADTC were included. To develop and validate the nomogram, these patients were randomly divided into training and validation cohorts at a 7:3 ratio. Various clinicopathological characteristics were gathered and subsequently transformed into categorical variables for analysis, and these included age at diagnosis (classified as <55 and ≥55 years), sex, race (classified as white, black, and other), marital status (classified as married or single: including single, divorced, separated, domestic partner, widowed and unmarried), histology (classified as C-PTC, V-PTC, and FTC), surgery, radiation, chemotherapy, tumor size (classified as ≤2, 2–4, and >4 cm), N stage (classified as N0 and N1), M stage (classified as M0 and M1), AJCC stage (classified as I, II, III, IV, with all patients reclassified according to the 8th edition of the AJCC staging system), survival month, and cancer-specific death. The main outcome of this study was cancer-specific survival (CSS), defined as the period from LADTC diagnosis to death specifically caused by LADTC.

Statistical analysis

Summary statistics were utilized to detail the demographic characteristics and clinical variables at baseline. The enumeration data were presented as patient numbers and percentages [n (%)]. The Chi-squared test or Fisher’s exact test was used to analyze the categorical variables. In the training cohort, the least absolute shrinkage and selection operator (LASSO) regression analysis was used to screen for potential prognostic factors, and these factors were subsequently included in the multivariate Cox regression analysis to identify the independent prognostic factors associated with the survival of patients with LADTC. The results were reported as hazard ratios (HRs) and 95% confidence intervals (95% CIs). Using the independent prognostic factors mentioned above, a prognostic nomogram was constructed to predict the 3-, 5-, and 10-year CSS of patients with LADTC. Subsequently, the discriminating ability of the nomogram was assessed using the concordance index (C-index), time-dependent receiver operating characteristic (ROC) curve, and area under the curve (AUC). Additionally, calibration plots were generated to compare the associations between the observed CSS and the predicted CSS. The clinical benefits of the nomogram were evaluated using the decision curve analysis (DCA). The improvement of the nomogram compared with the AJCC staging system was evaluated using the C-index, net reclassification index (NRI), and integrated discrimination improvement (IDI). In addition, to promote the integration of the nomogram into clinical practice, an interactive web-based dynamic nomogram application was constructed using the open source R Shiny Server. Finally, to construct a risk stratification system, patients in the training and validation cohorts were categorized into three risk subgroups according to the tri-sectional quantiles of the total scores obtained from the nomogram. The differences in CSS among the three risk subgroups were then analyzed using Kaplan‒Meier curve with the log-rank test. The data processing and analysis were carried out using SPSS 27.0 and R software (version 4.3.0). P<0.05 was considered to indicate statistical significance.

Results

Demographic and clinicopathological characteristics

A total of 4,856 patients diagnosed with LADTC were included in our study from 2004 to 2020. Of these, 3,399 patients (70%) were assigned to the training cohort, while 1,457 patients (30%) were assigned to the validation cohort. Among the entire study population, 56.98% of the patients were classified as elderly (age ≥55 years), and 64.72% of the patients were female. Regarding race, 78.56% of the patients were identified as white, making up the vast majority of the sample. The marital status distribution revealed that a greater percentage of patients were married than single patients. In terms of histology, C-PTC was the most prevalent type (72.36%), followed by V-PTC (23.87%) and FTC (3.77%). Additionally, the majority of patients had undergone surgery (95.59%) and received radiation treatment (76.19%), while only a few patients had received chemotherapy (5.54%). The majority of patients who received radiotherapy underwent surgery and/or chemotherapy at the same time, and only a few patients received radiotherapy alone. The distributions of tumor size, N stage, M stage, and AJCC stage were also delineated, along with their respective P values. There were no significant differences in demographic and clinicopathological characteristics between the training and validation cohorts, as shown in Table 1.

LASSO and Cox regression analyses

Age, sex, race, marital status, histology, surgery, radiation, chemotherapy, tumor size, N stage, and M stage were considered potential prognostic factors and were included in LASSO regression with tenfold cross-validation. The results showed that age, sex, surgery, radiation, chemotherapy, tumor size, N stage, and M stage with nonzero coefficients were associated with the CSS of patients with LADTC (Figure 1). These factors were subsequently included in the multivariate Cox regression analyses, and the results revealed that age, sex, surgery, radiation, chemotherapy, tumor size, N stage, and M stage were all identified as independent prognostic factors for CSS in patients with LADTC. These results are tabulated in Table 2.

Figure 1 LASSO regression analysis of clinical variables in patients with LADTC. (A) LASSO regression coefficient path plot; (B) tenfold cross validation. LADTC, locally advanced differentiated thyroid cancer; LASSO, least absolute shrinkage and selection operator.

Nomogram construction and validation

Eight independent prognostic factors (age, sex, surgery, radiation, chemotherapy, tumor size, N stage, and M stage) were incorporated to construct a prognostic nomogram for predicting the 3-, 5-, and 10-year CSS in patients with LADTC, which is illustrated in Figure 2 and available online (https://handyuan.shinyapps.io/dynnomapp/). Each variable in the nomogram was assigned a score from 0 to 100 based on its contribution to this nomogram, with the total score being the sum of all variable scores. This total score was subsequently used to estimate the CSS for each individual. The performance of the nomogram was assessed using the C-index, time-dependent ROC curve, AUC, and calibration plots. The C-index of the nomogram was 0.833 (95% CI: 0.819–0.847) in the training cohort and 0.824 (95% CI: 0.800–0.848) in the validation cohort. The AUC values of the nomogram for predicting the 3-, 5-, and 10-year CSS were 0.873 (95% CI: 0.855–0.891), 0.873 (95% CI: 0.856–0.889), and 0.893 (95% CI: 0.877–0.910), respectively, in the training cohort, and 0.858 (95% CI: 0.828–0.888), 0.859 (95% CI: 0.831–0.887), and 0.869 (95% CI: 0.840–0.897), respectively, in the validation cohort. The results of the time-dependent ROC curve analysis are shown in Figure 3. Additionally, the results of the calibration plots showed strong agreement in the probability of 3-, 5-, and 10-year CSS between the observed CSS and the predicted CSS in the training and validation cohorts (Figure 4). DCA is a method for evaluating the clinical benefit of model and was applied to nomograms by quantifying net benefits at different threshold probabilities. In our study, the results of DCA demonstrated that the nomogram had good clinical benefits for predicting the 3-, 5-, and 10-year CSS in patients with LADTC, as it added more net benefits for almost all threshold probabilities in both the training and validation cohorts (Figure 5).

Figure 2 Nomogram for predicting the 3-, 5-, and 10-year CSS in patients with LADTC in the training cohort. CSS, cancer-specific survival; LADTC, locally advanced differentiated thyroid cancer; M, metastasis; N, node.

Figure 3 ROC curves of the nomogram for predicting 3-, 5-, and 10-year CSS in patients with LADTC in the training cohort (A) and validation cohort (B). AUC, area under the curve; CI, confidence interval; CSS, cancer-specific survival; LADTC, locally advanced differentiated thyroid cancer; ROC, receiver operating characteristic.

Figure 4 Calibration plots of the nomogram for predicting 3-, 5-, and 10-year CSS in patients with LADTC in the training cohort (A) and validation cohort (B). CSS, cancer-specific survival; LADTC, locally advanced differentiated thyroid cancer.

Figure 5 DCA of the nomogram for predicting 3- (A), 5- (B), and 10-year (C) CSS in the training cohort and 3- (D), 5- (E), and 10-year (F) CSS in the validation cohort. CSS, cancer-specific survival; DCA, decision curve analysis; M, metastasis; N, node.

To further evaluate the improvement of the nomogram compared to the AJCC staging system, we used the C-Index, NRI, and IDI to compare the discriminatory ability and accuracy of the nomogram and the AJCC staging system. The C-indexes for nomogram-related predictions in the training and validation cohorts were 0.833 (95% CI: 0.819–0.847), and 0.824 (95% CI: 0.800–0.848), respectively. In contrast, the C-indexes for the AJCC staging system were 0.757 (95% CI: 0.739–0.775) and 0.771 (95% CI: 0.747–0.795) in the training and validation cohorts, respectively. When comparing the nomogram to the AJCC staging system in the training cohort, the NRIs for the 3-, 5-, and 10-year CSS were 0.367 (95% CI: 0.273–0.463), 0.339 (95% CI: 0.246–0.436), and 0.230 (95% CI: 0.140–0.315), respectively, and the IDIs for the 3-, 5-, and 10-year CSS were 0.105 (95% CI: 0.073–0.135, P<0.001), 0.105 (95% CI: 0.079–0.137, P<0.001), and 0.095 (95% CI: 0.059–0.126, P<0.001), respectively. These results were corroborated in the validation cohort (Table 3).

Risk stratification based on the nomogram

To construct a risk stratification system, all variables in the nomogram were assigned a score, and the total score of each patient was calculated based on their contribution to the nomogram. These patients were subsequently categorized into the a low-risk subgroup (total score ≤95), a moderate-risk subgroup (95< total score <171), and a high-risk subgroup (total score ≥171) based on their total scores in the nomogram. The Kaplan-Meier survival analysis demonstrated significant differences in CSS among the three risk subgroups in both the training and validation cohorts (P<0.001) (Figure 6).

Figure 6 Kaplan-Meier survival analysis in patients with LADTC at different risk subgroups stratified by the nomogram in the training cohort (A) and validation cohort (B). LADTC, locally advanced differentiated thyroid cancer.

Discussion

According to SEER data, TC accounted for 2.3% of all new cancer cases in 2022, with a five-year relative survival rate of 98.4% (18). LADTC refers to a severe stage of DTC with a relatively poor prognosis. Approximately 24% of patients with DTC die from airway obstruction caused by local tumor dilation, while 28% die from respiratory failure caused by lung metastasis or malignant pleural effusion (19). Despite this, most patients with LADTC can still be cured after active treatment. However, studies investigating LADTC are rare, and a precise and efficient tool for predicting mortality in individual patients is lacking in clinical practice. Therefore, the need to develop an accurate and suitable predictive tool for patients with LADTC is urgent. Previous studies have indicated that various factors, including age, sex, marital status, histology, surgery, radiation, tumor size, N stage, and M stage, could impact the survival of patients with DTC (20-22). Therefore, we considered these factors in detail and included them in our comprehensive analysis using LASSO and Cox regression analyses. We subsequently constructed and internally validated a relatively discriminating and accurate nomogram for predicting CSS by incorporating variables derived from the LASSO and Cox regression analyses. In addition, to enable clinicians to conveniently evaluate patient prognosis using the nomogram, we constructed a web-based dynamic nomogram using the open-source R Shiny Server (https://handyuan.shinyapps.io/dynnomapp/).

In our study, the prognosis of patients with LADTC was assessed using the commonly used and objective index known as the CSS. The LASSO and Cox regression analyses revealed that eight variables (age, sex, surgery, radiation, chemotherapy, tumor size, N stage, and M stage) were significantly associated with the CSS in patients with LADTC. However, we found that race, marital status, and histology were not identified as independent prognostic factors. These eight independent prognostic factors were selected based on their clinical significance and were used to construct and validate a prognostic nomogram with the potential to provide the basis for future clinical decisions.

This study has demonstrated that age was the most significant prognostic factor, followed by tumor size, chemotherapy, M stage, surgery, N stage, radiation, and sex. Age is widely known to be a prognostic indicator for many diseases (10,23,24). The eighth edition of the AJCC Staging system has set 55 years of age as a critical threshold for staging (9). Yang et al. reported a significant correlation between patient age and cause-specific mortality in advanced PTC (25). Kong et al. investigated the association between age and cancer-specific prognosis in patients with ATC and confirmed that patients older than 70 years had a poorer CSS prognosis (23). Additionally, Peng et al. proposed that age is the most significant predictive factor for CSS prognosis in patients with DTC, with younger patients (age ≤67 years) demonstrating better outcomes compared to older patients (age >67 years) (26). The results of this study also emphasized the significance of age in predicting outcomes in patients with LADTC.

This study revealed that sex served as an independent risk factor for CSS in patients with LADTC. Males had significantly worse outcomes than females, which is consistent with previous studies (16,27). This sex disparity may be attributed to the relatively higher estrogen levels in women, which stimulate the secretion of thyroid-stimulating hormone. It is known that thyroid-stimulating hormone has the potential to suppress the occurrence and development of TC (28).

The significance of tumor size as a determinant of prognosis in patients with TC is widely acknowledged. Schindele et al. identified tumor size as a factor associated with recurrence risk in DTC, and established 25mm as the threshold for increased recurrence risk (29). Moreover, Ho et al. reported that an increase in tumor size directly corresponded to increased mortality in patients with TC (30). These findings were corroborated in the present study, underscoring the inverse relationship between tumor size and prognosis. In addition, this study identified N1 stage and M1 stage as notable risk factors for CSS in patients with LADTC, consistent with the findings of Jin et al. and Ibrahimpasic et al. (21,31).

At present, the approach for treating LADTC relies primarily on previous experience in treating DTC. Surgical intervention remains the preferred method for treating DTC (5,32), while other treatment options include radiation, chemotherapy, targeted therapy, and endocrine suppression therapy. This study demonstrated the benefits of surgery and radiation on the CSS of patients with LADTC. However, the use of chemotherapy posed a risk to the CSS of patients with LADTC (AJCC stages I–IV), consistent with findings from Tang et al. (28). Clinical research has progressed from the use of targeted therapy for TC to its use in clinical practice, resulting in improvements in patients with TC (32). Currently, there is a focus on exploring neoadjuvant therapies. With the emergence of TC medications, neoadjuvant therapies are gradually being integrated into LADTC treatment, thereby enabling some patients with inoperable LADTC to undergo radical surgery and confirming their significant role in treating LADTC (7,33-35).

In this study, we constructed and validated a dynamic nomogram to predict CSS in patients with LADTC. The C-indexes and time-dependent ROC curves indicated that the nomogram exhibited strong discriminative ability in both the training and validation cohorts, as demonstrated by AUC values exceeding 0.85. Furthermore, the calibration plots indicated a strong accuracy between the nomogram-predicted CSS and the actual observed CSS. The DCA curves further demonstrated the potential clinical benefits of our nomogram in accurately predicting survival. The C-index, NRI, and IDI revealed that the prognostic predictive ability and accuracy of our nomogram were notably superior to those of the AJCC staging system for patients with LADTC. In addition, the risk stratification system demonstrated the favorable ability to categorize patients with LADTC into low-risk, moderate-risk, and high-risk subgroups, revealing significant differences among the groups. Notably, these findings were successfully replicated within the validation cohort.

This study has several significant advantages worth noting. First, to the best of our knowledge, this is the first study to construct a prognostic nomogram for predicting the CSS in patients with LADTC. The population-based nomogram displayed the most encouraging outcomes. Furthermore, the present study included a relatively large number of patients, which was sufficient to construct a prognostic nomogram with good performance. Finally, the nomogram allows for risk stratification and identification of high-risk patients who may derive benefit from more aggressive treatment strategies.

Nevertheless, it is important to acknowledge the potential limitations of this study. First, since this is a retrospective study based on the SEER database, the results may have been influenced by selection biases. Moreover, the exclusion of patients with incomplete variable information could have also introduced significant selection biases. Second, this study lacked external validation in an independent dataset. While the validation cohort provided some evidence of generalizability, it is clear that large-sample prospective clinical trials are essential for validating the discriminant ability, accuracy and clinical benefits of the nomogram. Moreover, novel treatments such as targeted therapy are a constantly evolving field, and further research is needed to verify their effectiveness.

Conclusions

We identified eight variables (age, sex, surgery, radiation, chemotherapy, tumor size, N stage, and M stage) as independent prognostic factors for predicting the CSS in patients with LADTC based on the LASSO and Cox regression analyses. Subsequently, we constructed and validated a dynamic nomogram by incorporating these factors to predict the 3-, 5-, and 10-year CSS in these patients, which has strong discriminant ability, accuracy and clinical benefits. In addition, we constructed a risk stratification system based on the nomogram to evaluate prognostic risk of these patients. The dynamic nomogram and risk stratification system have the potential to assist clinicians in evaluating prognostic risk and devising personalized treatment strategies for patients with LADTC.

Acknowledgments

We would like to thank the SEER program for providing open access to the database.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-2025-111/rc

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-2025-111/prf

Funding: This work was supported by the Science-Health Joint Medical Scientific Research Project of Chongqing (2020MSXM016).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-2025-111/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. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

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. Zhou T, Wang X, Zhang J, et al. Global burden of thyroid cancer from 1990 to 2021: a systematic analysis from the Global Burden of Disease Study 2021. J Hematol Oncol 2024;17:74. [Crossref] [PubMed]
2. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
3. Boucai L, Zafereo M, Cabanillas ME. Thyroid Cancer: A Review. JAMA 2024;331:425-35. [Crossref] [PubMed]
4. Metere A, Aceti V, Giacomelli L. The surgical management of locally advanced well-differentiated thyroid carcinoma: changes over the years according to the AJCC 8th edition Cancer Staging Manual. Thyroid Res 2019;12:10.
5. Enomoto K, Inohara H. Surgical strategy of locally advanced differentiated thyroid cancer. Auris Nasus Larynx 2023;50:23-31. [Crossref] [PubMed]
6. Roka R. Surgical treatment of locally advanced thyroid cancer. Innov Surg Sci 2020;5:27-34. [Crossref] [PubMed]
7. Huang NS, Wei WJ, Xiang J, et al. The Efficacy and Safety of Anlotinib in Neoadjuvant Treatment of Locally Advanced Thyroid Cancer: A Single-Arm Phase II Clinical Trial. Thyroid 2021;31:1808-13. [Crossref] [PubMed]
8. Lamartina L, Godbert Y, Nascimento C, et al. Locally unresectable differentiated thyroid cancer: outcomes and perspectives. Endocrine 2020;69:133-41. [Crossref] [PubMed]
9. Amin MB, Greene FL, Edge SB, et al. The Eighth Edition AJCC Cancer Staging Manual: Continuing to build a bridge from a population-based to a more "personalized" approach to cancer staging. CA Cancer J Clin 2017;67:93-99.
10. Zhuo X, Xia L, Tang W, et al. A practical nomogram and risk stratification system for predicting survival outcomes in neuroblastoma patients: a SEER population-based study. J Cancer Res Clin Oncol 2023;149:12285-96. [Crossref] [PubMed]
11. Xie L, Zhang Y, Niu X, et al. A nomogram for predicting cancer-specific survival in patients with locally advanced unresectable esophageal cancer: development and validation study. Front Immunol 2025;16:1524439. [Crossref] [PubMed]
12. Cao M, Hu C, Pan S, et al. Development and validation of nomogram for predicting early recurrence after radical gastrectomy of gastric cancer. World J Surg Oncol 2024;22:21. [Crossref] [PubMed]
13. Peiyuan G, Xuhua H, Ganlin G, et al. Construction and validation of a nomogram model for predicting the overall survival of colorectal cancer patients. BMC Surg 2023;23:182. [Crossref] [PubMed]
14. Aksoy YA, Xu B, Viswanathan K, et al. Novel prognostic nomogram for predicting recurrence-free survival in medullary thyroid carcinoma. Histopathology 2024;84:947-59. [Crossref] [PubMed]
15. Zhou J, Zhao B, Liu L, et al. A nomogram for individualized prediction for cervical lymph node metastasis of papillary thyroid carcinoma. Gland Surg 2024;13:1965-76. [Crossref] [PubMed]
16. Li WH, Yu WY, Du JR, et al. Nomogram prediction for cervical lymph node metastasis in multifocal papillary thyroid microcarcinoma. Front Endocrinol (Lausanne) 2023;14:1140360. [Crossref] [PubMed]
17. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin 2019;69:7-34. [Crossref] [PubMed]
18. Yu X, Deng Q, Gao X, et al. A prognostic nomogram for distant metastasis in thyroid cancer patients without lymph node metastasis. Front Endocrinol (Lausanne) 2025;16:1523785. [Crossref] [PubMed]
19. Park H, Park J, Park SY, et al. Clinical Course from Diagnosis to Death in Patients with Well-Differentiated Thyroid Cancer. Cancers (Basel) 2020;12:2323. [Crossref] [PubMed]
20. Yang T, Hu T, Zhao M, et al. Nomogram Predicts Overall Survival in Patients With Stage IV Thyroid Cancer (TC): A Population-Based Analysis From the SEER Database. Front Oncol 2022;12:919740. [Crossref] [PubMed]
21. Jin S, Liu H, Yang J, et al. Development and validation of a nomogram model for cancer-specific survival of patients with poorly differentiated thyroid carcinoma: A SEER database analysis. Front Endocrinol (Lausanne) 2022;13:882279. [Crossref] [PubMed]
22. Wang W, Shen C, Yang Z. Nomogram individually predicts the risk for distant metastasis and prognosis value in female differentiated thyroid cancer patients: A SEER-based study. Front Oncol 2022;12:800639. [Crossref] [PubMed]
23. Kong N, Xu Q, Zhang Z, et al. Age Influences the Prognosis of Anaplastic Thyroid Cancer Patients. Front Endocrinol (Lausanne) 2021;12:704596. [Crossref] [PubMed]
24. Chen X, Gong H, Chen J, et al. Development and validation of a prognostic nomogram for predicting mortality risk in adult rheumatoid arthritis: an analysis of NHANES 1999-2018 data. Front Immunol 2025;16:1592958. [Crossref] [PubMed]
25. Yang Y, He X, Qu X, et al. Identification of Risk Factors for Cause-specific Mortality in Advanced Papillary Thyroid Cancer and Construction of a Competing Risk Model: A SEER-Based Study. Cancer Control 2025;32:10732748251336412. [Crossref] [PubMed]
26. Peng H, Zheng M, Li JY, et al. Analysis of the ideal cutoff age as a predictor of differentiated thyroid cancer using the Surveillance, Epidemiology, and End Results database. Transl Cancer Res 2024;13:4278-89. [Crossref] [PubMed]
27. Zahedi A, Bondaz L, Rajaraman M, et al. Risk for Thyroid Cancer Recurrence Is Higher in Men Than in Women Independent of Disease Stage at Presentation. Thyroid 2020;30:871-7. [Crossref] [PubMed]
28. Tang J, Zhanghuang C, Yao Z, et al. Development and validation of a nomogram to predict cancer-specific survival in middle-aged patients with papillary thyroid cancer: A SEER database study. Heliyon 2023;9:e13665. [Crossref] [PubMed]
29. Schindele A, Krebold A, Heiß U, et al. Interpretable machine learning for thyroid cancer recurrence predicton: Leveraging XGBoost and SHAP analysis. Eur J Radiol 2025;186:112049. [Crossref] [PubMed]
30. Ho AS, Luu M, Zalt C, et al. Mortality Risk of Nonoperative Papillary Thyroid Carcinoma: A Corollary for Active Surveillance. Thyroid 2019;29:1409-17. [Crossref] [PubMed]
31. Ibrahimpasic T, Ghossein R, Shah JP, et al. Poorly Differentiated Carcinoma of the Thyroid Gland: Current Status and Future Prospects. Thyroid 2019;29:311-21. [Crossref] [PubMed]
32. Haddad RI, Bischoff L, Ball D, et al. Thyroid Carcinoma, Version 2.2022, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2022;20:925-51. [Crossref] [PubMed]
33. Alshehri K, Alqurashi Y, Merdad M, et al. Neoadjuvant lenvatinib for inoperable thyroid cancer: A case report and literature review. Cancer Rep (Hoboken) 2022;5:e1466. [Crossref] [PubMed]
34. Pitoia F, Abelleira E, Román-González A, et al. Neoadjuvant Treatment of Locally Advanced Thyroid Cancer: A Preliminary Latin American Experience. Thyroid 2024;34:949-52. [Crossref] [PubMed]
35. Chen JY, Huang NS, Wei WJ, et al. The Efficacy and Safety of Surufatinib Combined with Anti PD-1 Antibody Toripalimab in Neoadjuvant Treatment of Locally Advanced Differentiated Thyroid Cancer: A Phase II Study. Ann Surg Oncol 2023;30:7172-80. [Crossref] [PubMed]