S1541
Interdisciplinary - Quality assurance and risk management
ESTRO 2026
standards, the credentialing systems performed well in detecting 5% or 7% errors (sensitivity >0.92 and >0.97), but only achieved 0.75-0.93 sensitivity to detect 3% dose errors (Figure 1). The harmonized standard: 95% sensitivity to detect a 5% error, was achievable (Figure 2). This resulted in variable specificity across systems, illustrating the trade-off between detecting unacceptable plans and minimizing the number of plans falsely classified as failures.
Fig2: Sensitivity and specificity based on current tolerance and reoptimized to provide 95% sensitivity. Conclusion: This work has established an international standard, and reciprocity agreement, for radiotherapy clinical trial credentialing. Any participating institution may present a credentialing result from the groups listed in Figure 2 that meets the harmonized tolerance for acceptance into clinical trials supported by IROC, TROG and RTTQA. This standard is being reviewed by other global clinical trial groups (JCOG, IAEA, EORTC, CCTG, and IRROG) under the auspices of the Global Harmonization Group of Quality Assurance of Radiation Therapy Clinical Trials. These reciprocity standards reduce duplication of work for institutions, reduce burden of monitoring activities outside local regions for clinical trial support agencies, and thereby help to streamline global participation in clinical trials. Keywords: Credentialing, harmonisation, clinical trials
Digital Poster Highlight 5093
Offline γ-Analysis of Inter-Session Dose Verification Using kV-CBCT DICOM Plan/Dose Sets: Enhancing Patient Safety in Radiotherapy. ABEL RODRIGUEZ ARANDA 1 , ADRIAN ANDRADES 2 1 MEDICAL-PHYSICS, Lanzarote University Hospital, Arrecife, Spain. 2 Radiotherapy, Las Palmas University Hospital, Las Palmas, Spain Purpose/Objective: Ensuring treatment accuracy and patient safety in radiotherapy requires detecting anatomical changes, setup variations, or device misplacement between sessions. This study introduces a robust workflow for inter-session dose verification using kV-CBCT images
Fig1. Performance using current credentialing to detect plans with errors of a) 3%, b) 5%, and c) 7%. AUC isshown in parentheses; dots represent current tolerances.
Made with FlippingBook - Share PDF online