Radiomics and artificial intelligence for predicting response to neoadjuvant drug therapy in patients with breast cancer: a review
摘要
Breast cancer remains one of the most pressing challenges in modern oncology and is the most common malignant neoplasm among women worldwide. Breast cancer treatment requires a comprehensive approach, including surgery, chemotherapy, radiation therapy, targeted therapy, and hormone therapy. A particularly important role in current clinical practice belongs to neoadjuvant therapy—an approach administered prior to surgery, aimed at reducing tumor size, increasing the likelihood of breast-conserving surgery, and evaluating the tumor’s individual sensitivity to drug therapy. Neoadjuvant therapy is the standard of care for locally advanced, initially inoperable invasive breast cancer. It is also recommended as a first-line treatment for patients with initially operable but biologically aggressive tumor subtypes, such as triple-negative and HER2-positive breast cancer. However, individual responses to therapy vary significantly: some patients demonstrate a good response to neoadjuvant treatment, which markedly improves their prognosis, whereas in others the treatment may prove ineffective. Early prediction of therapeutic response to neoadjuvant treatment helps to avoid unnecessary drug dose exposure, reduce the financial burden on the healthcare system, and minimize the risk of adverse effects. In recent years, radiomics and artificial intelligence methods have been actively developed to analyze medical imaging and detect hidden biomarkers associated with treatment response. This review analyzes articles from recent decades in which diverse prognostic models were developed to evaluate neoadjuvant treatment response through the application of radiomics and artificial intelligence methods. Special attention is given to papers demonstrating the potential of machine learning and deep data analysis aimed at personalizing breast cancer therapy. These innovative approaches offer new opportunities for improving treatment effectiveness and patient survival.