In vivo dosimetry is recommended in radiotherapy to avoid major treatment errors and to improve accuracy. Large scale clinical implementation however is not always easy. Recently, commercially automated systems have made it feasible to perform transit dosimetric quality assurance on a very large scale. We started using such a system for all our machines and all our patients in 2018 using transit EPID images. We analyzed results in both 2019 and 2020 for more than 7000 patients in total, including causes and actions for failed fractions. Ten different sets of parameters for gamma analysis, depending on treatment site, were empirically determined, balancing the rate between detection of clinically relevant problems and the number of false positive results. The choice of tolerance levels was based on an AMARA-principle: the goal of detecting errors and deviations, but only As Many As Reasonably Achievable, taking into account economic and societal factors.
In-vivo analysis in 2019 showed 16% of fractions failed, of which 6% were false positives and 10% were caused by patient related issues. The number of false positives depended on machine type. Causes for failed in-vivo analysis included deviations in patient positioning (4,9%) and anatomy change (4,3%). In addition, errors in planning, imaging, treatment delivery, simulation, breath hold and positioning devices were detected. Actions for failed fractions were mostly to repeat the measurement while taking extra care in positioning and to intensify imaging procedures. Four percent of failed fractions initiated plan adjustments, showing the potential of the system as a trigger for adaptive planning. Based on these in vivo results, some permanent actions were taken for improving quality, e.g. extra imaging for boost breast, extra education to the RTT’s and extra help from dietitians. When analyzing results in 2020, it was seen these actions had resulted in a reduction of failed fractions from 16% to 13%, including a drop from 4,9% to 3,0% in patient positioning deviations and a drop from 4,3% to 3,3% in anatomy changes.
In conclusion, large scale clinical implementation of EPID-based in-vivo transit dosimetry using a commercially available automated system is feasible and it efficiently reveals a wide variety of deviations. Evaluating results on a regular basis can offer important insights in the quality of treatments and indicate possible items for improvement.