Machine learning based anoikis signature predicts personalized treatment strategy of breast cancer.

BACKGROUND

Breast cancer remains a leading cause of mortality among women worldwide, emphasizing the urgent need for innovative prognostic tools to improve treatment strategies. Anoikis, a form of programmed cell death critical in preventing metastasis, plays a pivotal role in breast cancer progression.

METHODS

This study introduces the Artificial Intelligence-Derived Anoikis Signature (AIDAS), a novel machine learning-based prognostic tool that identifies key anoikis-related gene patterns in breast cancer. AIDAS was developed using multi-cohort transcriptomic data and validated through immunohistochemistry assays on clinical samples to ensure robustness and broad applicability.

RESULTS

AIDAS outperformed existing prognostic models in accurately predicting breast cancer outcomes, providing a reliable tool for personalized treatment. Patients with low AIDAS levels were found to be more responsive to immunotherapies, including PD-1/PD-L1 inhibitors, while high-AIDAS patients demonstrated greater susceptibility to specific chemotherapeutic agents, such as methotrexate.

CONCLUSIONS

These findings highlight the critical role of anoikis in breast cancer prognosis and underscore AIDAS's potential to guide individualized treatment strategies. By integrating machine learning with biological insights, AIDAS offers a promising approach for advancing personalized oncology. Its detailed understanding of the anoikis landscape paves the way for the development of targeted therapies, promising significant improvements in patient outcomes.