DR-441 A transcriptomic-driven machine learning model to predict the efficacy of CDK4/6-inhibitors in metastatic hormone-receptors positive breast cancer

In metastatic hormone receptor-positive (HR+/HER2-) breast cancer (mBC), cyclin-dependent kinase 4 and 6 inhibitors (CDK4/6-i) in combination with endocrine therapy (ET) emerged as the preferred first- or second-line treatment option. Regardless, clinical responses are not uniform, with about ~20% of patients presenting with primary endocrine resistance. Importantly, while numerous molecular mechanisms promoting secondary endocrine-resistance have been reported, they came short in demonstrating clinical validity in predicting upfront which patients will derive the least benefits from using CDK4/6 inhibitors plus ET.

Accordingly, the present project aims to develop a transcriptomic-driven machine learning model (MLM) to predict the efficacy of CDK4/6-I + ET in HR+/HER2- mBC.

Zhan Yinxiu, Istituto Europeo di Oncologia, Italy