Studies on Tolerance Level and Sensitivity for Patient-specific Quality Assurance in Radiation Therapy

Abstract

Purpose: The aim of this study was to investigate the tolerance level and sensitivity of patient-specific quality assurance (QA) for intensity modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT). Methods: In order to investigate the tolerance level of patient-specific QA for IMRT in Korea, a multi-institutional study involving 12 radiation therapy institutions was performed. And then, based on the acquired tolerance level from multi-institutional study, sensitivities of various QA methods to detect errors in IMRT and VMAT plans were investigated. The multi-institutional study was performed by (1) point dose measurements using ion chamber at high- and low-dose regions (2) and planar (per-field and composite-field) dose measurements using film or 2-dimensional detector array. The multi-institutional study consisted of two programs, one was mock program and the other was clinic program. For mock program, we employed five mock structures reflecting anatomy of average Korean, while clinical treatment plans were used for clinic program to investigate the tolerance level. With the results of multi-institutional study, patient-specific QAs of point dose measurements with ion chamber, measurement of 2D dose distribution of axial plane with radiochromic film (EBT2), measurements of 2D dose distribution of coronal plane with MatriXX® and measurements of 3D volume dose distribution with COMPASS® has been performed. The results were compared with statistics by dividing into four groups according to delivery technique (IMRT group and VMAT group) and the degree of modulation (prostate group and H&N group). Additionally, a new discretized deliverable VMAT plan (i.e., static arc (SA) plan) was generated and performed 3D QA to evaluate the dose discrepancy in planning target volume (PTV). One of the reasons possible to explain this discrepancy was an arc discretization of VMAT in treatment planning system (TPS). The dose discrepancy of PTV between VMAT and SA plans were evaluated using by gamma test with 3%/3 mm criteria and dose difference at 95% and 5% volume of PTV, respectively. Results: Since the concept of confidence limit (CL) was appropriate for point dose measurement, tolerance level could be acquired. The tolerance level of point dose measurement at high-dose region was ±3% in both two programs. The tolerance level at low-dose region was ±5% in mock program, while it was ±7% and ±5% for linear accelerator (LINAC) and tomotherpy (TOMO) group in clinic program, respectively. On the contrary, for planar dose measurement, the concept of CL was not appropriate because of a large local deviation and a small number of samples. However, the results of planar dose measurement in both programs were well agreed with that of multi-institutional study performed by American Association of Physicists in Medicine (AAPM). In the sensitivity study, the results of point dose measurement followed a normal distribution at all groups. The CL for IMRT, VMAT, prostate, and H&N groups were 3.0%, 2.1%, 1.0%, and 3.4%, respectively. The results of 2D dose measurements in axial and coronal plane showed significant differences at delivery groups (IMRT vs. VMAT) and patient groups (prostate vs. H&N) with the 3%/3 mm criteria. The results of 3D volume dose measurements showed significant differences at delivery groups and patient groups in both criteria. There was no strong correlation between 2D QAs at axial plane and coronal plane. Similarly no correlation was observed between 2D and 3D QAs. Only 3D QA was possible to detect a dose discrepancy in PTV during delivery of VMAT. As the result of evaluation of arc discretization, the gamma passing rate of QA results were 92.1% and 96.8% for VMAT and SA QA, respectively. The dose differences at D95 were 2.61% and 0.97% and at D5 were 2.71% and 0.04% for VMAT and SA QA, respectively. Conclusions: Since the result of a multi-institutional study in Korea was coincident with those of AAPM and European Society for Therapeutic Radiology and Oncology (ESTRO) guidelines, patient-specific QA in Korea could be considered to meet the standards of international guideline. However, a point dose and 2D measurements recommended as patient-specific QA methods by international guidelines seem not to be enough to guarantee of an accurate delivery of IMRT and VMAT plans because no correlation was observed among point dose measurements, 2D QA and 3D QA. Three-dimensional QA was most sensitive to detect errors in treatment plans especially for VMAT. Therefore, it seems to be reasonable to adopt 3D QA as a patient-specific QA method for accurate radiation therapy using IMRT and VMAT techniques.