PV QA 3 - Poster Viewing Q&A 3
Purpose/Objective(s): Adaptive re-planning is frequently required in IMRT involving large GTV. However; the time at which it has to be initiated has not been standardized. The fluence map of the IMRT exit beam reflect the changes in the target. This study reports the standardization of serial gamma analyses and clinical implementation of serial exit fluence mapping to predict the need for adaptive re-planning.
Materials/Methods: A standard head-and-neck IMRT plan, 70 Gy in 33 fractions to a custom-made tissue-equivalent tumor attached to a RANDO phantom with setup (0,0,0) was used to generate a baseline exit fluence map. The aS1000 EPID was used to acquire the portal fluence maps on the Clinac 2100 C/D linear accelerator. Portal fluence maps were acquired with setup errors introduced incrementally from 1-10 mm in all six directions and the gamma-index method (2 mm distance-to-agreement and 2% dose difference) with the Linac's Portal Dosimetry software was used to evaluate the changes from baseline. The same method was used to study the variation in fluence map due to changes in tumor volume (120 cc, 80 cc and 0 cc). The standardization process involved quantifying the change in fluence with respect to change in tumor volume after accounting for setup error. Serial weekly fluence maps were acquired prospectively for 8 patients with T4 and/or N3 primary undergoing IMRT and gamma analysis was performed. A planning CT was acquired at the end of RT to measure the change in dose volume parameters of GTV and OARs. Descriptive statistics were used for analysis.
Results: In the phantom experiments, mean area-Gamma values showed >95% congruence for a setup variation amounting upto 5 mm. A quadratic relationship was found with setup errors and gamma changes. With a shift of 10 mm in either direction the mean area gamma value dropped to 88.5%. A linear change in gamma values was observed with decrease in tumor volume. The exit fluence maps of 80 cc and 0 cc volumes when compared to the 120 cc volume resulted in area gamma values of 94.5% and 92.7% respectively, after negating the setup errors. All 8 patients who had serial fluence mapping had considerable fall in gamma values with respect to time. The area gamma value of the last day of treatment was median 74.3%, (23.7 - 90.5) when compared to the baseline. This can be compared with a median reduction in GTV volume of 90% (26.3 - 91.4). No under dose of the target volume or overdose of the critical structures was found when dosimetric analysis of original plan was done on CT acquired at the end of the radiation treatment.
Conclusion: Serial exit dose fluence analyses were standardized for set up errors and volume changes. Setup errors within 5 mm did not result in a gamma reduction > 5%. Reduction in target volume by 1/3rd reduced the area gamma > 5%. A larger clinical series evaluating the quantum of gamma change to trigger an adaptive re-planning is currently in progress to assess the clinical utility.
The asset you are trying to access is locked. Please enter your access key to unlock.