Quality and Safety of Real-Time Brachytherapy Treatments: Clinical Implementation of a Multipoint Scintillation Detector for Treatment Monitoring

Overview

This project addresses a key safety gap in High-Dose-Rate (HDR) brachytherapy, where very high, localized radiation doses mean that small geometric deviations in catheter placement or radioactive source positioning can cause clinically significant errors. Incident reports from the Radiation Oncology Incident Learning System (RO-ILS) of the American Society for Radiation Oncology (ASTRO) and guidelines from the American Association of Physicists in Medicine Task Group 59 (AAPM TG-59) identify these events-often occurring between imaging and treatment delivery-as among the most critical. Unlike external beam radiotherapy, brachytherapy lacks a prior phantom-based verification step, making in vivo dosimetry the only real-time method to confirm delivered dose and detect anomalies in dwell time or source position. International bodies such as the International Atomic Energy Agency (IAEA) recommend this approach for quality management and accident prevention. Multipoint plastic scintillation detectors (mPSD) offer water-equivalent, real-time measurements and can enable three-dimensional (3D) radioactive source tracking, but their clinical use remains limited due to integration challenges and the absence of well-defined alert thresholds. The project aims to implement a real-time in vivo monitoring solution using an mPSD at the Centre régional intégré de cancérologie (CRIC) of Lévis, Québec (QC) for HDR gynecological and prostate brachytherapy. Objectives include: (1) demonstrating clinical feasibility and integration within existing clinical workflows; (2) validating the accuracy and precision of dose, dwell-time, and source-position measurements under clinical conditions; and (3) developing practical alert thresholds to identify common deviations. The central hypothesis is that mPSD technology can provide reliable, reproducible real-time measurements without lengthening procedures or disrupting clinical workflow. The methodology relies on adapting the HYPERSCINT® platform for HDR in vivo use. A custom mPSD, consisting of several scintillators along a single optical fiber, will be insertable into a 6-French (6F) catheter. Phase 1 focuses on feasibility and integration, including HDR-specific calibration, definition of detector insertion scenarios, integration of acquisition hardware, and staff training. Phase 2 validates performance in at least 20 patients. Planned treatment parameters and time-stamped detector signals acquired at 1-10 hertz (Hz) will be used to reconstruct delivered dose, dwell times, and source-to-detector distances. Measurement accuracy and reproducibility will be quantified using mixed-effects statistical models. Phase 3 establishes preliminary alert thresholds based on deviations in dose, time, and position, along with visualization tools such as Bland-Altman plots, although these thresholds will not be applied clinically during the study. Expected outcomes include demonstrated feasibility of in vivo mPSD monitoring, quantitative performance metrics with confidence intervals, identification of influential factors and operational recommendations, preliminary alert thresholds with reporting templates, and a roadmap for sustainable deployment in Lévis and other centers. Scientific outputs will also be produced for dissemination across the Santé Québec network.