Main Session
Sep 28
PQA 03 - Digital Health Innovation and Informatics, Patient Safety & Quality, and Radiation and Cancer Biology

2596 - Analysis of Low-Dose-Rate (LDR) Brachytherapy Events Reported to Radiation Oncology Incident Learning System® Database (RO-ILS) from 2014-2025

10:45am - 12:00pm ET
Poster Hall - Exhibit Hall A
Screen: 30
POSTER

Presenter(s)

Daniel Scanderbeg, PhD Headshot
Daniel Scanderbeg, PhD - University of California, San Diego, La Jolla, CA

D. J. Scanderbeg1, J. G. Bazan Jr2, A. P. Dalton3, A. R. Godley4, L. Hales5, T. K. Kosak6, K. Kujundzic7, B. M. Miller8, J. R. Olsen9, A. Pietrowski10, S. Tripathi11, M. Washington12, and J. Zeng13; 1University of California, San Diego, La Jolla, CA, 2Department of Radiation Oncology, City of Hope National Medical Center, Duarte, CA, 3Department of Radiation Oncology, University of Alabama at Birmingham, Birmingham, AL, 4Medical Artificial Intelligence and Automation (MAIA) Lab, Dallas, TX, 5Roswell Park Cancer Institute, Buffalo, NY, United States, 6Brigham and Women's Hospital, Boston, MA, 7ASTRO, Falls Church, VA, 8Henry Ford Health System, Detroit, MI, 9University of Colorado, Aurora, CO, 10Clarity Group Inc, Downers Grove, IL, 11Marshfield Clinic Health System, Marshfield, WI, United States, 12Department of Radiation Oncology, Winship Cancer Institute of Emory University, Atlanta, GA, 13Department of Radiation Oncology, University of Washington/Fred Hutchinson Cancer Center, Seattle, WA

Purpose/Objective(s):

Low-dose-rate (LDR) brachytherapy (BT) is a well-established and efficacious treatment for many cancers. While generally safe, the complexity of source handling, treatment planning, and delivery creates opportunities for adverse events. From the RO-ILS: Radiation Oncology Incident Learning System® database, 81 LDR incidents were identified that were reported between Q4 2014 and Q3 2025. We summarize the incidence profile, severity distribution, and primary contributing factors, with the goal of informing targeted quality-improvement initiatives.

Materials/Methods:

  • Data Source: RO-ILS event reports (n ˜ 1400 total BT events) filtered for LDR BT. Duplicates ignored, leaving 81 unique LDR incidents
  • Time Period: Q4 2014 – Q3 2025 (˜ 11 years)
  • Classification: Events were grouped by disease site (Prostate, Eye, Other, Unknown) and by Radiation Oncology Healthcare Advisory Council (RO-HAC) severity score (1–5). Self-reported severity and other details were also recorded
  • Root-Cause Analysis: Each event was examined for contributing factors, categorized as hardware/software, communication, policy/procedure lapses, workplace inadequacy, distraction, patient-focused circumstances, or “other”
  • Dose Assessment: Maximum dose deviation to the target was stratified according to self-reported metrics of <5 %, 5-25 %, 25-100 % or >100 %

Results:

  • Incidence Profile: 81 LDR events comprised 5.8 % of all BT reported events. Distribution of disease site was dominated by prostate (40/81, 49 %), followed by the eye (22/81, 27 %)
  • Severity: Five incidents received the highest RO-HAC severity score of 5. Four of these critical events were self-reported as severe, but only two matched the RO-HAC level 5 severity rating, highlighting variability in severity perception.
  • Dose Deviation: Three events exceeded a 25 % deviation with one implant showing >100 % deviation.
  • Contributing Factors: Communication failures were the most frequent root cause (16/38, 42 %). Policy or procedure non-adherence accounted for 8 events, while hardware/software issues contributed to 3. Distractions and workplace inadequacies each featured in 3 incidents.

Conclusion:

The RO-ILS database is a valuable tool in quality, safety and error mitigation since much more data could be gathered from near-miss experiences, rather than severe events that require reporting to state or national agencies. Frequency of occurrence and root cause could help strengthen intervention before further occurrence of common errors or increased severity. LDR BT safety analysis from the RO-ILS database reveals a modest, but clinically significant, burden of high-severity incidents, primarily driven by communication lapses and procedural non-compliance. Dissemination of information from the database as well as possible interventions could help reduce dose deviations and prevent recurrence of critical errors. Continued monitoring through the RO-ILS system will be essential to measure the impact of any quality-improvement measures.