Advantages

• Visualizing the Risk of Stroke Specific to Cancer Patients: High predictive accuracy that surpasses existing thrombosis prediction models


• Calculated Using Only Clinical Data: Composed solely of indicators obtained during routine clinical practice, such as blood tests and clinical cancer staging


• High Versatility and Reliability: Developed and validated using large-scale data from Osaka University and the Osaka International Cancer Center

Technology & Background

Cancer patients are prone to blood clotting disorders, and thromboembolic events such as cerebral infarction and venous thrombosis can adversely affect their prognosis. While the “Khorana score” is widely used to predict venous thrombosis, there has been no reliable risk assessment tool specifically tailored to cerebral infarction associated with cancer (cancer-related cerebral infarction).
To address this challenge, the inventor identified eight factors associated with cancer-related cerebral infarction (high-risk cancer, advanced age, hypertension, atrial fibrillation, advanced cancer, adenocarcinoma histology, neutrophil-to-lymphocyte ratio (NLR), and D-dimer) and developed a new predictive model, the “AHANDS score,” by optimally combining these factors. The use of this score is expected to facilitate the early identification of patients at high risk of developing the condition, thereby contributing to the implementation of appropriate preventive interventions and personalized medicine.

Data & Future Plans

• Comparison of Prediction Accuracy: In terms of predictive ability for cancer-related cerebral infarction (assessed by ROC-AUC), the AHANDS score (0.71) demonstrated significantly higher accuracy than the Khorana score (0.54) (p < 0.0001).


• Validation of External Validity: Validation using independent data from a different institution (Osaka International Cancer Center) than the development cohort (Osaka University) confirmed that the model maintains a high discriminatory ability with an AUC of 0.75, demonstrating consistent generalizability and universality.


• Future Directions: Conducting a multicenter prospective study; Exploring physician-initiated clinical trials targeting high-risk patients; Further refining the model using AI (deep learning) (joint research with the Institute of Scientific and Industrial Research (ISIR), Osaka University), etc.

Principal Investigator

Tomohiro Kawano, MD, PhD (Assistant Professor, Department of Neurology, Osaka University Graduate School of Medicine)

Patents and Publications

Patents
- PCT application filed (unpublished).
Publications
- Kawano T and Mackman N, Thromb. Res. (2024) 237, 155–162.
- Kawano T et al. Sci Rep. (2025) 15(1):39560.

Expectations

TECH MANAGE is currently seeking companies interested in this invention on behalf of Osaka University. We can arrange detailed discussions regarding this technology through direct meetings with the Principal Investigator (PI), as well as disclose unpublished data and other information upon signing a confidentiality agreement with Osaka University. Please feel free to contact us with any inquiries.