• Machine Learning Prediction of...

*Result*: Machine Learning Prediction of Recurrence in Pediatric Thyroid Cancer: Malignant Endocrine Tumors Cohort Analysis Using XGBoost and SHAP.

Title:
Machine Learning Prediction of Recurrence in Pediatric Thyroid Cancer: Malignant Endocrine Tumors Cohort Analysis Using XGBoost and SHAP.
Authors:
Redlich A; Department of Pediatrics, Pediatric Hematology/Oncology, Otto von Guericke-University, Magdeburg 39120, Germany., Pfaehler E; Institute for Neuroscience and Medicine 4, INM-4, Forschungszentrum Jülich GmbH, Juelich 52428, Germany., Kunstreich M; Department of Pediatrics, Pediatric Hematology/Oncology, Otto von Guericke-University, Magdeburg 39120, Germany.; Pediatrics and Adolescent Medicine, Faculty of Medicine, University of Augsburg, Augsburg 86156, Germany., Schmutz M; Haematology and Oncology, Faculty of Medicine, University of Augsburg, Augsburg 86156, Germany.; Bavarian Cancer Research Centre, Augsburg 86156, Germany., Lapa C; Bavarian Cancer Research Centre, Augsburg 86156, Germany.; Nuclear Medicine, Faculty of Medicine, University of Augsburg, Augsburg 86156, Germany., Kuhlen M; Pediatrics and Adolescent Medicine, Faculty of Medicine, University of Augsburg, Augsburg 86156, Germany.; Bavarian Cancer Research Centre, Augsburg 86156, Germany.
Source:
The Journal of clinical endocrinology and metabolism [J Clin Endocrinol Metab] 2026 Feb 20; Vol. 111 (3), pp. e844-e852.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Oxford University Press Country of Publication: United States NLM ID: 0375362 Publication Model: Print Cited Medium: Internet ISSN: 1945-7197 (Electronic) Linking ISSN: 0021972X NLM ISO Abbreviation: J Clin Endocrinol Metab Subsets: MEDLINE
Imprint Name(s):
Publication: 2017- : New York : Oxford University Press
Original Publication: Springfield, Ill. : Charles C. Thomas
MeSH Terms:
Thyroid Neoplasms*/pathology , Thyroid Neoplasms*/diagnosis , Thyroid Neoplasms*/epidemiology , Thyroid Neoplasms*/therapy , Neoplasm Recurrence, Local*/diagnosis , Neoplasm Recurrence, Local*/epidemiology , Neoplasm Recurrence, Local*/pathology , Machine Learning*, Risk Assessment/methods ; Humans ; Child ; Male ; Female ; Retrospective Studies ; Adolescent ; Child, Preschool ; Prognosis ; Follow-Up Studies ; Registries ; ROC Curve ; Boosting Machine Learning Algorithms
Grant Information:
DKS 2021.11, DKS 2024.16, and DKS 2025.07) Deutsche Kinderkrebsstiftung; Mitteldeutsche Kinderkrebsforschung
Contributed Indexing:
Keywords: SHAP; XGBoost; children and adolescents; differentiated thyroid carcinoma; prediction
Entry Date(s):
Date Created: 20250901 Date Completed: 20260219 Latest Revision: 20260219
Update Code:
20260220
DOI:
10.1210/clinem/dgaf487
PMID:
40890050
Database:
MEDLINE

*Further Information*

*Context: Pediatric differentiated thyroid carcinoma (DTC) often presents with advanced disease but generally has excellent long-term survival. However, recurrence or failure to achieve remission remains relatively frequent, underscoring the need for improved early risk stratification.
Objective: To develop and evaluate an interpretable machine learning model for predicting recurrence or nonremission in pediatric DTC using routine clinical and biochemical variables.
Design and Setting: Retrospective analysis of 250 pediatric patients (aged <18 years) enrolled in the German Pediatric Oncology Hematology-Malignant Endocrine Tumors Registry (1997-2023). Inclusion required known age at diagnosis and ≥24 months of follow-up. The composite study endpoint was structural recurrence or failure to achieve remission within 24 months of initial therapy.
Methods: An extreme gradient boosting classifier was trained on 80% of the data, with the remaining 20% used as an independent test set. Model generalizability was assessed via 50 randomized stratified train-validation splits of the training dataset. SHapley Additive exPlanations (SHAP) were used to interpret feature contributions.
Results: The final model achieved an area under the receiver operating characteristic curve (AUROC) of 0.86 on the independent test set. Across 50 validation splits, the mean AUROC was 0.82 (SD ± 0.05), sensitivity 0.81 (SD ± 0.09), and specificity 0.64 (SD ± 0.06). SHAP analysis identified younger age at diagnosis (<10 years), elevated postoperative thyroglobulin levels, and distant metastases as the most influential predictors.
Conclusion: This interpretable machine learning model reliably predicts early recurrence or nonremission in pediatric DTC and may complement current risk stratification systems to support personalized, risk-adapted treatment decisions.
(© The Author(s) 2025. Published by Oxford University Press on behalf of the Endocrine Society.)*