Risk stratification of papillary thyroid cancers using multidimensional machine learning.

BACKGROUND: Papillary thyroid cancer (PTC) is one of the most common endocrine malignancies with different risk levels. However, preoperative risk assessment of PTC is still a challenge in the worldwide. Here, the authors first report a Preoperative Risk Assessment Classifier for PTC (PRAC-PTC) by multidimensional features including clinical indicators, immune indices, genetic feature, and proteomics. MATERIALS AND METHODS: The 558 patients collected from June 2013 to November 2020 were allocated to three groups: the discovery set [274 patients, 274 formalin-fixed paraffin-embedded (FFPE)], the retrospective test set (166 patients, 166 FFPE), and the prospective test set (118 patients, 118 fine-needle aspiration). Proteomic profiling was conducted by FFPE and fine-needle aspiration tissues from the patients. Preoperative clinical information and blood immunological indices were collected. The BRAFV600E mutation were detected by the amplification refractory mutation system. RESULTS: The authors developed a machine learning model of 17 variables based on the multidimensional features of 274 PTC patients from a retrospective cohort. The PRAC-PTC achieved areas under the curve (AUC) of 0.925 in the discovery set and was validated externally by blinded analyses in a retrospective cohort of 166 PTC patients (0.787 AUC) and a prospective cohort of 118 PTC patients (0.799 AUC) from two independent clinical centres. Meanwhile, the preoperative predictive risk effectiveness of clinicians was improved with the assistance of PRAC-PTC, and the accuracies reached at 84.4% (95% CI: 82.9-84.4) and 83.5% (95% CI: 82.2-84.2) in the retrospective and prospective test sets, respectively. CONCLUSION: This study demonstrated that the PRAC-PTC that integrating clinical data, gene mutation information, immune indices, high-throughput proteomics and machine learning technology in multicentre retrospective and prospective clinical cohorts can effectively stratify the preoperative risk of PTC and may decrease unnecessary surgery or overtreatment.

Preoperative risk stratification for patients with ≤ 1 cm papillary thyroid carcinomas based on preoperative blood inflammatory markers: construction of a dynamic predictive model.

Objective: The aim of this study was to investigate the relationships and predictive value of preoperative peripheral blood inflammatory markers as a means by which to assess risk for patients with ≤ 1 cm papillary thyroid carcinomas (PTCs). In addition, a preoperative risk stratification predictive model was constructed and validated. Methods: Clinical and pathologic data, as well as preoperative blood specimens, were collected from patients who underwent initial thyroid cancer surgery at the Hangzhou First People's Hospital, from January 2014 to January 2023. Risk assessment was performed based on postoperative pathology according to the 2015 ATA guidelines for recurrence risk stratification. Using univariate analysis and multivariate logistic regression, we identified independent risk factors associated with risk stratification. A predictive model was established and its discriminative and calibration abilities were validated. An independent validation dataset was used to verify the model, and the model was deployed as an online calculator. Results: A total of 1326 patients were included in the study, with 1047 cases (79.0%) classified as low risk and 279 cases (21.0%) classified as intermediate to high risk. The modeling group consisted of 981 cases, through univariate analysis and multivariate logistic regression analysis, preoperative blood Neutrophil/Lymphocyte Ratio (NLR), gender, tumor diameter, and multifocality were identified as independent risk factors that distinguished between low and intermediate to high risk patients with ≤ 1 cm PTCs. The clinical predictive model exhibited an AUC of 0.785, specificity of 70.6%, and sensitivity of 75.8%. For the independent validation group of 345 patients, the AUC was 0.813, specificity was 83.8%, and sensitivity was 70.4%. The calibration curve and clinical decision curve indicate that the model demonstrates excellent calibration performance. Conclusion: A dynamic clinical predictive model based on preoperative blood NLR and clinical information for patients with ≤ 1 cm PTCs was established. The model is useful for preoperative risk assessment of patients with ≤ 1 cm PTCs.

Blood immune indexes can predict lateral lymph node metastasis of thyroid papillary carcinoma.

Objective: To explore the clinical significance of blood immune indexes in predicting lateral lymph node metastasis (LLNM) of thyroid papillary carcinoma (PTC). Methods: The pathological data and preoperative blood samples of 713 patients that underwent thyroid surgery at affiliated Hangzhou First People's Hospital Zhejiang University School of Medicine from January 2013 to June 2021 were collected as the model group. The pathological data and preoperative blood samples of 177 patients that underwent thyroid surgery in the same hospital from July 2021 to October 2021 were collected as the external validation group. Univariate and multivariate logistic regression analyses were used to determine the independent risk factors of LLNM in PTC patients. A predictive model for assessing LLNM in PTC patients was established and externally validated using the external data. Results: According to univariate and multivariate logistic regression analyses, tumor diameter (P < 0.001, odds ratios (OR): 1.205, 95% confidence interval (CI): 1.162-1.249) and the preoperative systemic immune-inflammation index (SII) (P = 0.032, OR: 1.001, 95% CI: 1.000-1.002) were independent risk factors for distinguishing LLNM in PTC patients. When the Youden index was the highest, the area under the curve (AUC) was 0.860 (P < 0.001, 95% CI: 0.821-0.898). The externally validated AUC was 0.827 (P < 0.001, 95% CI: 0.724-0.929), the specificity was 86.4%, and the sensitivity was 69.6%. The calibration curve and the decision curve indicated that the model had good diagnostic value. Conclusion: Blood immune indexes can reflect the occurrence of LLNM and the biological behavior of PTC. The predictive model established in combination with SII and tumor diameter can effectively predict the occurrence of LLNM in PTC patients.

Total tumor diameter is a better indicator of multifocal papillary thyroid microcarcinoma: A propensity score matching analysis.

Background: Accurate evaluation of the risk of papillary thyroid microcarcinoma (PTMC) is the key to treatment. However, the maximum diameter (MD), which is currently used in various staging systems, may not truly reflect the aggressiveness of multifocal tumors. Methods: Clinical and pathological data for 1001 patients with papillary thyroid carcinoma who underwent surgery at the Hangzhou First People's Hospital were retrospectively analyzed. First, the relationship between total tumor diameter (TTD) and clinicopathological features in multifocal PTMC was explored. Then, patients were divided into subgroups according to the TTD. The baseline was consistent after using the propensity score matching method, and the differences between groups were compared. In addition, the effectiveness of TTD and MD in evaluating central lymph node metastasis (CLNM) was analyzed and compared. Results: TTD is associated with a range of clinicopathological features, including lymph node metastasis, extrathyroidal extension, and risk stratification. Assuming the same MD and number of foci, the invasiveness of multifocal PTMC with TTD >1 cm was significantly higher than that with TTD <1 cm, and even higher than unifocal non-PTMC. Moreover, the efficiency of TTD in predicting CLNM was also significantly higher than that of MD. Conclusion: For multifocal PTMC, TTD is a more realistic indicator of tumor biological characteristics than MD. The aggressiveness of PTMC with TTD >1 cm was significantly enhanced, and surgical treatment should be actively sought in such cases.

Risk factors of skip lateral cervical lymph node metastasis in papillary thyroid carcinoma: a systematic review and meta-analysis.

OBJECTIVE: To analyze and explore the risk factors of skip lateral cervical lymph node metastasis (SLLNM) in papillary thyroid carcinoma (PTC). METHODS: PubMed, Web of Science, Embase, Cochrane, Wanfang, China National Knowledge Infrastructure, and China Science and Technology Journal databases, updated to April 4, 2021, were systematically searched for literature on the risk factors of SLLNM in PTC. The meta-analysis was completed using Stata 15.0 software after quality evaluation. The odds ratio (OR) and 95% confidence interval (CI) of each variable were calculated using fixed or random-effects models, and the publication bias was evaluated by the Egger's test. RESULTS: A total of 28 studies with 10,682 cases were included in our meta-analysis; 1592 (14.90%) cases were positive for SLLNM. The meta-analysis showed that female sex (OR = 1.16, 95% CI = 1.02-1.31, P = 0.021), age ≥45 (OR = 1.60, 95% CI = 1.19-2.15, P = 0.002), tumor diameter ≤10 mm (OR = 2.23, 95% CI = 1.62-3.06, P < 0.001), and upper location of tumor (OR = 3.60, 95% CI = 2.65-4.89, P < 0.001) were risk factors for SLLNM in PTC patients. Hashimoto's thyroiditis (OR = 1.02, 95% CI = 0.88-1.19, P = 0.777), multifocality (OR = 0.98, 95% CI = 0.75-1.28, P = 0.873), bilateral tumors (OR = 0.92, 95% CI = 0.70-1.19, P = 0.515), extrathyroidal extensions (OR = 1.07, 95% CI = 0.83-1.39, P = 0.598), and capsular invasion (OR = 0.93, 95% CI = 0.65-1.31, P = 0.660) were not closely related to SLLNM risk. CONCLUSION: This study confirmed significant associations between SLLNM and female sex, age ≥45, tumor diameter ≤10 mm, and upper location of the tumor.

The Prediction of Metastases of Lateral Cervical Lymph Node in Medullary Thyroid Carcinoma.

Purpose: Development and validation of a nomogram for the prediction of lateral lymph node metastasis (LLNM) in medullary thyroid carcinoma (MTC). Methods: We retrospectively reviewed the clinical features of patients with MTC in the Surveillance, Epidemiology, and End Results (SEER) database between 2010 and 2017 and in our Department of Surgical Oncology, Hangzhou First People's Hospital between 2009 and 2019. The log-rank test was used to compare the difference in the Kaplan-Meier (K-M) curves in recurrence and survival. The nomogram was developed to predict the risk of LLNM in MTC patients. The prediction efficiency of the predictive model was assessed by area under the curve (AUC) and concordance index (C-index) and calibration curves. Decision curve analysis (DCA) was performed to determine the clinic value of the predictive model. Result: A total of 714 patients in the SEER database and 35 patients in our department were enrolled in our study. Patients with LLNM had worse recurrence rate and cancer-specific survival (CSS) compared with patients without LLNM. Five clinical characteristics including sex, tumor size, multifocality, extrathyroidal extension, and distant metastasis were identified to be associated with LLNM in MTC patients, which were used to develop a nomogram. Our prediction model had satisfied discrimination with a C-index of 0.825, supported by both training set and internal testing set with a C-index of 0.825, and 0.816, respectively. DCA was further made to evaluate the clinical utility of this nomogram for predicting LLNM. Conclusions: Male sex, tumor size >38mm, multifocality, extrathyroidal extension, and distant metastasis in MTC patients were significant risk factors for predicting LLNM.

Extranodal Extension Is an Independent Prognostic Factor in Papillary Thyroid Cancer: A Propensity Score Matching Analysis.

Purpose: To investigate the prognostic significance of extranodal extension (ENE) in papillary thyroid cancer (PTC). Methods: Seven hundred forty-three PTC patients were enrolled in the study from January 2014 to December 2017. The patients were dichotomized according to the presence of ENE. Logistic analysis was used to compare differences between the two groups. Kaplan-Meier (K-M) curve and propensity score matching (PSM) analyses were used for recurrence-free survival (RFS) comparisons. Cox regression was performed to analyze the effects of ENE on RFS in PTC. Results: Thirty-four patients (4.58%) had ENE. Univariate analysis showed that age, tumor size, extrathyroidal extension, and nodal stage were associated with ENE. Further logistic regression analysis showed that age, extrathyroidal extension, and nodal stage remained statistically significant. Evaluation of K-M curves showed a statistically significant difference between the two groups before and after PSM. Cox regression showed that tumor size and ENE were independent risk factors for RFS. Conclusions: Age ≥55 years, extrathyroidal extension, and lateral cervical lymph node metastasis were identified as independent risk factors for ENE. ENE is an independent prognostic factor in PTC.

A Novel Scoring System for Predicting the Metastases of Posterior Right Recurrent Laryngeal Nerve Lymph Node Involvement in Patients With Papillary Thyroid Carcinoma by Preoperative Ultrasound.

Objective: Our goal was to investigate the correlation between papillary thyroid carcinoma (PTC) characteristics on ultrasonography and metastases of lymph nodes posterior to the right recurrent laryngeal nerve (LN-prRLN). There is still no good method for clinicians to judge whether a patient needs LN-prRLN resection before surgery, and we also wanted to establish a new scoring system to determine whether patients with papillary thyroid carcinoma require LN-prRLN resection before surgery. Patients and Methods: There were 482 patients with right or bilateral PTC who underwent thyroid gland resection from December 2015 to December 2017 recruited as study subjects. The relationship between the PTC characteristics on ultrasonography and the metastases of LN-prRLN was analyzed by univariate and logistic regression analyses. Based on the risk factors identified in univariate and logistic regression analysis, a nomogram-based LN-prRLN prediction model was established. Result: LN-prRLN were removed from all patients, of which 79 had LN-prRLN metastasis, with a metastasis rate of 16.39%. Multivariate logistic regression analysis revealed that LN-prRLN metastasis was closely related to sex, age, blood supply, larger tumors (> 1 cm) and capsular invasion. A risk prediction model has been established and fully verified. The calibration curve used to evaluate the nomogram shows that the consistency index was 0.75 ± 0.065. Conclusion: Preoperative clinical data, such as sex, age, abundant blood supply, larger tumor (> 1 cm) and capsular invasion, are positively correlated with LN-prRLN metastasis. Our scoring system can help surgeons non-invasively determine which patients should undergo LN-prRLN resection before surgery. We recommend that LN-prRLN resection should be performed when the score is above 103.1.

A meta-analysis of risk factors for lymph node posterior to the right recurrent laryngeal nerve metastasis in papillary thyroid carcinoma.

BACKGROUND: Whether the lymph node posterior to the right recurrent laryngeal nerve (LN-prRLN) should be dissected is still controversial. This meta-analysis aimed to assess the risk factors for LN-prRLN metastasis in papillary thyroid carcinoma (PTC). METHODS: We retrieved relevant studies published before May 2020 from the Wanfang Data, CNKI, PubMed, Embase, Web of Science and Cochrane Library databases. Heterogeneity was assessed with the Q-test and inconsistency index and sensitivity analysis and subgroup analysis were then used to find the source of heterogeneity. Begg test and generate a funnel chart to assess publication bias. RESULTS: We retrieved 236 articles, 14 articles were selected as the subjects of our research. Fourteen studies involving 10,580 patients were analysed in this study. The LN-prRLN metastasis rate was 9.22% (975/10,580). The results of the meta-analysis showed that sex (P<0.001), age (P<0.001), tumour size (P<0.001), multifocality (P<0.001), capsular invasion (P=0.04), extrathyroidal extension (P<0.001), superficial central lymph node (VIa-LN) metastasis (P<0.001), and lateral lymph node (LLN) metastasis (P<0.001) were correlated with LN-prRLN metastasis. DISCUSSION: Male sex, age ≤45 years, tumour size >1 cm, multifocality, capsular invasion or extrathyroidal extension, and VIa-LN metastasis or LLN metastasis in PTC patients were significant risk factors for predicting LN-prRLN metastasis. B ultrasound and CT scans are expected to predict LN-prRLN metastasis in the future. REGISTRATION: This research is registered on the PROSPERO website (registration number: CRD42020200898).

Risk factors for lateral cervical lymph node metastasis in medullary thyroid carcinoma.

: To investigate risk factors of lateral cervical lymph node metastasis (LLNM) in patients with medullary thyroid carcinoma (MTC). : Published studies regarding clinicopathological factors of LLNM in MTC were searched in PubMed, Web of Science, Embase, Cochrane library, Wanfang date and CNKI. Statistical analysis was performed using Stata 14.0 software. The mean and standard deviation from the sample size, range, median, and interquartile range was estimated. Odds ratio () or standard mean difference () with 95% confidence interval () of related factors were analyzed by fixed/random-effects models. Egger's test and Begg's test were applied to assess the publication bias of the literature. This study was registered with PROSPERO (CRD42021254955). : Fifteen studies involving 1424 patients were included in the analysis, among whom 543 cases had LLNM (38.13%). Meta-analysis revealed that an increased risk of LLNM was associated with male gender (1.64, 95%: 1.29-2.09, 4.06, 0.01), tumor diameter≥1cm (5.09, 95%: 2.43-10.67, 4.31, 0.01), multifocality (2.55, 95%: 1.79-3.61, 5.22, 0.01), capsule invasion (7.80, 95%: 4.84-12.55, 8.46, 0.01), extracapsular extension (9.46, : 5.66-15.81, 8.58, 0.01), cervical central lymph node metastasis (23.58, : 9.44-58.87, 6.77, 0.01), elevated preoperative calcitonin (1.17,95%: 0.67-1.67, 4.56, 0.01), spiculated margin on ultrasonography (4.32, 95%: 2.43-7.68, 4.99, 0.01), irregular shape on ultrasonography (6.81, : 3.64-12.73, 6.01, 0.01); while age ≥ 45 years (=1.22, 95%: 0.65-2.29, 0.62, >0.05), elevated preoperative carcinoembryonic antigen (0.95, : -0.48-2.38, 1.30, >0.05) and calcification on ultrasonography (1.28, 95%: 0.75-2.18, 0.92, >0.05) were not associated with LLNM. : Male gender, tumor diameter≥multifocality, capsule invasion, extracapsular extension, central lymph node metastasis, elevated preoperative calcitonin, spiculated margin and irregular shape on ultrasonography are risk factors for LLNM in MTC, when these clinical and ultrasonic features are present, lateral neck lymph node dissection is recommended.