Development and validation of a nomogram for risk of pulmonary metastasis in non-papillary thyroid carcinoma: A SEER-based study

Establishment and verification of the first prognostic nomograms in locally advanced thyroid cancer based on the analysis of clinical and follow-up information on 2396 patients

Background and objectives

The purpose of this research was to develop and validate the first prognostic nomograms for 3-, 5-, and 10-year cancer-specific survival (CSS) and overall survival (OS) in patients diagnosed with locally advanced thyroid cancer (LATC) by evaluating independent predictors of prognosis in a population of LATC patients.

Methods

Demographics, clinicopathologic characteristics, treatment, and follow-up of 2396 LATC patients in the surveillance, epidemiology, and end results database from 2004 to 2015 were retrospectively analyzed and compared with patients with LATC according to staging. We randomized all LATC patients into training and validation groups in a 7:3 ratio. Cox regression analyses helped us to derive independent prognostic factors for LATC patients. According to these results, we established and validated the first prognostic nomograms and risk stratification.

Results

In our research, the clinical information of LATC patients was compared and significant differences were found in the relevant variables including CSS and OS (P < 0.05), with CSS of 82.0 % and 49.0 %, and OS of 70.6 % and 40.0 %, respectively. Cox regression analyses showed that age at diagnosis, tumor diameter, presence of DM, extrathyroidal extension sites, histological type, thyroidectomy scope, radiotherapy status, and chronological sequence of radiotherapy and surgery were observably correlated with CSS in LATC patients, and in addition to the above factors, gender, marital status, and chemotherapy status were also observably correlated with OS in LATC patients. The prognostic predictive power of the above factors is visualized by the Kaplan-Meier survival curve. The concordance index of nomograms for CSS and OS were 0.933, 0.925, and 0.926 (CSS), 0.918, 0.909, and 0.906 (OS), respectively, and the time-dependent receiver operating characteristic curve, area under curve, calibration curve and decision curve analysis curve indicate that the nomograms have good discriminatory ability, accuracy and clinical applicability in both the training and validation groups.

Conclusions

In these findings, we drawed a conclusion that there were significant differences in clinical information between patients with T4a and T4b LATC, and we established and validated the first prognostic nomograms and risk stratification of CSS and OS for LATC patients at 3, 5, and 10 years, which will help clinicians to individualize their postoperative treatment and individualized follow-up.

Machine learning based on SEER database to predict distant metastasis of thyroid cancer

OBJECTIVE

Distant metastasis of thyroid cancer often indicates poor prognosis, and it is important to identify patients who have developed distant metastasis or are at high risk as early as possible. This paper aimed to predict distant metastasis of thyroid cancer through the construction of machine learning models to provide a reference for clinical diagnosis and treatment.

MATERIALS & METHODS

Data on demographic and clinicopathological characteristics of thyroid cancer patients between 2010 and 2015 were extracted from the National Institutes of Health (NIH) Surveillance, Epidemiology, and End Results (SEER) database. Our research used univariate and multivariate logistic models to screen independent risk factors, respectively. Decision Trees (DT), ElasticNet (ENET), Logistic Regression (LR), Extreme Gradient Boosting (XGBoost), Random Forest (RF), Multilayer Perceptron (MLP), Radial Basis Function Support Vector Machine (RBFSVM) and seven machine learning models were compared and evaluated by the following metrics: the area under receiver operating characteristic curve (AUC), calibration curve, decision curve analysis (DCA), sensitivity(also called recall), specificity, precision, accuracy and F1 score. Interpretable machine learning was used to identify possible correlation between variables and distant metastasis.

RESULTS

Independent risk factors for distant metastasis, including age, gender, race, marital status, histological type, capsular invasion, and number of lymph nodes metastases were screened by multifactorial regression analysis. Among the seven machine learning algorithms, RF was the best algorithm, with an AUC of 0.948, sensitivity of 0.919, accuracy of 0.845, and F1 score of 0.886 in the training set, and an AUC of 0.960, sensitivity of 0.929, accuracy of 0.906, and F1 score of 0.908 in the test set.

CONCLUSIONS

The machine learning model constructed in this study helps in the early diagnosis of distant thyroid metastases and helps physicians to make better decisions and medical interventions.