Real-world data of CanAssist Breast- first immunohistochemistry and AI-based prognostic test

Area of Science:

• Oncology
• Biomarker Discovery
• Artificial Intelligence in Medicine

Background:

• Hormone receptor-positive (HR+), HER2/neu-negative (HER2-) breast cancer requires accurate prognostic tools.
• Existing prognostic tests may not be cost-effective or suitable for all populations.
• The CanAssist Breast (CAB) test was developed using immunohistochemistry and AI for Indian patients.

Purpose of the Study:

• To evaluate the real-world utility of the CAB test in prognostication.
• To assess CAB's performance across various clinicopathological parameters and histological subtypes.
• To determine the impact of CAB on treatment planning for early breast cancer patients.

Main Methods:

• Analysis of real-world data from 5926 early breast cancer patients diagnosed between mid-2016 and 2024.
• Utilizing the CanAssist Breast (CAB) test, which combines five protein biomarkers and three clinical parameters.
• Stratification of patients into low-risk (LR) and high-risk (HR) for distant recurrence.

Main Results:

• Overall, 72% of patients were classified as LR and 28% as HR for distant recurrence.
• CAB demonstrated significant differences in HR proportions across histological types (e.g., mucinous, papillary, micropapillary).
• In the intermediate Ki67 group, CAB identified 77% as LR and 23% as HR.

Conclusions:

• CanAssist Breast (CAB) is a novel, AI-based prognostic test for early breast cancer.
• CAB provides valuable prognostic information across diverse patient subgroups and histological types.
• CAB serves as a cost-effective alternative to Western prognostic tests, impacting treatment decisions.