2939 - Root Cause Analysis of Unscheduled Replans across 19,249 Radiation Therapy Plans: A Five-Year Single-Institution Review
Presenter(s)
M. Abdou1, J. M. Stark2, S. Kim3, B. Zegers4, and B. M. Anderson5; 1UCSD Radiation Medicine and Applied Sciences, La Jolla, CA, United States, 2University of California San Diego, San Diego, CA, 3UC San Diego, Moores Cancer Center, Department of Radiation Medicine and Applied Sciences, La Jolla, CA, 4UC San Diego, La Jolla, CA, 5Department of Radiation Oncology, University of North Carolina, Chapel Hill, NC
Purpose/Objective(s): Unscheduled replans in radiation therapy impose significant burden on clinical operations, requiring additional effort from dosimetrists, physicists, and physicians. Despite this impact, root causes of replans are not well characterized. This study performed a retrospective review of unscheduled replans at a single academic institution to identify primary drivers, categorize them by etiology and treatment site, and inform targeted interventions to reduce replan frequency and clinical burden.
Materials/Methods: A retrospective review was conducted of all radiation therapy plans at UC San Diego from 2021 through 2025, including VMAT, 3D conformal photon, and electron treatments. Unscheduled replans were identified and classified into three primary etiologies: Machine Down (unit unavailability requiring adaptation to an alternate machine), Physician Driven Change (target or fractionation modifications following chart rounds or imaging), and Planning Error (deficiencies in the original plan). Replans were further subcategorized by treatment site including head and neck, brain, breast, and pelvis. Annual trends were analyzed to identify temporal patterns and intervention opportunities.
Results: Of 19,249 plans across 15,160 unique patients, 241 (1.3%) required an unscheduled replan. Among 189 fully categorized cases, 35 (14.5%) were attributed to machine downtime, 115 (47.7%) to physician driven changes, and 39 (18.2%) to planning errors, with 52 cases under active review. Physician driven changes represented the largest category and increased from 2021 to 2025, with ongoing analysis by body site and treatment technique aimed at identifying clinical scenarios driving these modifications. Planning errors remained low and stable throughout the study period (range 4-8 per year), consistent with early adoption of templated planning approaches. Machine downtime events varied annually with increases in 2021 and 2025.
Conclusion: While representing only 1.3% of plans, unscheduled replans impose disproportionate burden on clinical resources. Physician driven changes account for nearly half of categorized replans, underscoring the need for characterization by treatment site and technique to identify actionable patterns. The low rate of planning errors supports continued use of templated workflows. Ongoing site-specific analysis will enable targeted interventions such as prioritizing early chart rounds for high-risk treatment sites, helping institutions develop evidence-based strategies that improve operational efficiency and patient care.