Development and Validation of a Prediction Model for Thyroid Dysfunction in Patients During Immunotherapy.

Area of Science:

• Oncology
• Endocrinology
• Immunology

Background:

• Immune checkpoint inhibitors (ICIs) like PD-1/PD-L1 inhibitors can cause immune-related adverse events (irAEs).
• Thyroid dysfunction is a significant irAE associated with ICI therapy.
• Predictive models are needed to identify patients at risk of ICI-induced thyroid toxicity.

Purpose of the Study:

• To develop and validate a predictive model for assessing the risk of thyroid toxicity in patients treated with ICIs.
• To identify key predictors of thyroid irAEs.
• To create an intuitive tool for clinical risk assessment.

Main Methods:

• Retrospective analysis of 586 patients treated with PD-1/PD-L1 inhibitors.
• Development of a predictive model using logistic regression on a training cohort (70%).
• Internal validation using K-fold cross-validation on a validation cohort (30%), assessing discrimination and calibration.

Main Results:

• A nomogram was developed incorporating baseline thyrotropin (TSH), thyroglobulin antibody (TgAb), thyroid peroxidase antibody (TPOAb), and platelet count.
• The model demonstrated excellent discrimination with AUCs of 0.863 (training) and 0.885 (validation).
• Calibration curves showed good fit, and decision curve analysis confirmed clinical utility.

Conclusions:

• The developed nomogram is an effective tool for predicting thyroid irAEs in patients receiving ICIs.
• This tool assists clinicians in making individualized treatment decisions.
• Early identification of at-risk patients can potentially mitigate thyroid toxicity.