Failure Modes and Effects Analysis for Online Adaptive Radiotherapy

Abstract

Radiotherapy treatments are increasingly challenged by anatomical variations that occur over time scales ranging from sub-seconds to days including organ motion, deformation, and tumor shrinkage. These dynamic changes can compromise the precision of conventional radiotherapy plans, particularly in mobile anatomical regions such as the abdomen and pelvis. [1] To address this, Online Adaptive Radiotherapy (OART) has emerged as a transformative approach that enables daily plan adaptation based on images acquired whilst the patient is on the treatment table. OART represents a paradigm shift from static, pre-planned treatments to real-time, anatomy-informed re-optimization. Enabled by both Magnetic Resonance Imaging (MRI)-guided systems and Cone Beam Computed Tomography (CBCT)-guided platforms, OART offers improved dose conformity by better tumor targeting and/or better sparing of organs at risk. However, this dynamic and time-constrained workflow also introduces novel failure modes that challenge traditional quality assurance protocols. To proactively address these risks, Failure Modes and Effects Analysis (FMEA) has been increasingly applied to OART workflows. This systematic review synthesizes the existing literature on FMEA for both MR- and CBCTguided OART systems. We catalog and compare failure modes across institutions, identify recurring high-risk process steps, and highlight mitigation strategies. Our aim is to support the development of tailored, robust and efficient QA frameworks that can guide safe and effective implementation of OART in diverse clinical settings.