ASO Author Reflections: Use of Machine Learning to Identify Patients with Intrahepatic Cholangiocarcinoma Who Could Benefit More from Neoadjuvant Therapies

Excerpt

Despite technologic advances and improvements in surgical techniques, prognosis of patients with intrahepatic cholangiocarcinoma (ICC) still remains dismal, with 5-year overall survival (OS) ranging from 20 to 40% after curative-intent resection.1,2 Given the general poor prognosis of patients with ICC, some clinicians have proposed that patients with ICC should receive neoadjuvant therapy prior to surgery, or even be treated with other non-surgical treatment modalities.3 Nevertheless, which patients should not be offered upfront surgery remains largely unknown to date. Machine learning techniques have recently been used increasingly in health care as a means to aid in treatment decision making. By utilizing a machine-based classification and regression tree (CART) model, we sought to identify which patients will derive the most or least benefit from surgery for ICC.4 …