Room: AAPM ePoster Library
Purpose: Treatment monitoring using cone-beam CT (CBCT) and an action-guideline for re-simulation for re-plan has been recently developed. This is to report on the initial experience of 20 H&N patients using this tool.
Methods: A new in-house monitoring system to find the optimal timing of Off-line Adaptive Planning (OAP) has been developed using acquired images of daily or weekly CBCTs that were used for patient setup. This system automatically reads newly acquired CBCT images for import into the RayStation treatment planning system (TPS: RaySearch Laboratories AB, Sweden, V. 8A.1). The system then automatically reports if the acquired parameters are above the preset action level.
The action level was composed of two parameters, which are (1) Dosimetric parameters over ±5% from initial plan; D95, D90, D10 and D5 of PTV, and (2) The Pearson Correlation Coefficient (PCC) within the CTV that is using HU differences decreasing below 0.75 between the first CBCT and subsequent CBCTs through the course of treatment.
Results: CBCT images of twenty H&N patients were collected with four of them triggering alerts, which indicated that both dosimetric and PCC tolerances were exceeded. For two of those four patients, the alerts were triggered towards the end of the treatment, and so OAP was not considered to be a practical option. OAP was found to be appropriate for the other two cases.
Conclusion: The offline monitoring system has been successfully implemented as a valuable tool to alert OAP timing. PCC variation was sensitive not only to setup error but also to the variation of CBCT image quality. The PCC not only provides the optimal timing for OAP but also alerts clinical staff to any systematic or random set-up errors.