Predicting and assessing the response of borderline resectable pancreatic cancer to neoadjuvant chemotherapy: Methodological approaches

Abstract Borderline resectable pancreatic cancer (BRPC) represents a complex clinical entity characterized by anatomical, biological, and conditional heterogeneity. Accurate assessments of resectability and the response to neoadjuvant chemotherapy (NAC) are crucial for determining surgical eligibility and improving overall survival. This review comprehensively summarizes current methods for predicting and evaluating the NAC response in patients with BRPC. A structured literature search was conducted in the PubMed, Embase, the Cochrane Library, and ClinicalTrials.gov, focusing on imaging-based evaluations, serum and molecular biomarkers, and emerging radiomic and radiogenomic techniques. This review is a narrative synthesis of the current evidence. Imaging modalities such as contrast-enhanced computed tomography, magnetic resonance imaging, and positron emission tomography/computed tomography remain fundamental for assessing vascular involvement and the therapeutic response, whereas dynamic biomarkers, including carbohydrate antigen 19-9, circulating tumor DNA, and exosomal markers, provide complementary insights into biological activity. Radiomics-based models further increase the precision of the predictions by integrating quantitative imaging features with molecular signatures. However, despite promising advances, the current evidence is limited by heterogeneity in patient selection, small sample sizes, lack of standardized thresholds, and insufficient validation in prospective studies. A multidisciplinary strategy integrating imaging, biomarker, and clinical parameters within standardized assessment frameworks is essential for refining the resectability evaluation and guiding the personalized management of patients with BRPC.