Development and Validation of the Predicting Risk of Ischemic Stroke in Malignancy Estimation Tool

Area of Science:

• Oncology
• Neurology
• Epidemiology

Background:

• The first year after cancer diagnosis presents the highest risk for ischemic stroke.
• Currently, no validated tools exist to identify high-risk cancer patients for ischemic stroke.

Purpose of the Study:

• To develop and validate a predictive tool for ischemic stroke risk in newly diagnosed cancer patients.
• To create an accessible online calculator for clinical use.

Main Methods:

• Population-based retrospective cohort study of adults with newly diagnosed cancer in Ontario, Canada (2010-2021).
• Used linked clinical and administrative databases for data.
• Developed a prediction model using multivariable Fine-Gray regression, treating mortality as a competing event.
• Assessed model performance using C statistic and calibration plots.

Main Results:

• The study included 698,566 patients, with 418,911 in the derivation and 279,576 in the validation cohorts.
• The final model incorporated 22 predictors (demographic, cancer, and treatment characteristics).
• The PRIME tool demonstrated good discrimination (C statistic=0.73) and calibration.

Conclusions:

• The PRIME (Predicting Risk of Ischemic Stroke in Malignancy Estimation) tool was successfully derived and validated.
• PRIME effectively predicts ischemic stroke risk in cancer patients within one year of diagnosis.
• An online calculator is available, enabling risk stratification for improved patient care.