Switching From Type B to Type C Dose Calculation Algorithm: Maintaining Treatment Planning Continuity in Head and Neck Treatments
V Feygelman1*, K Latifi1, M Bowers1, K Greco1, E Moros1, M Isacson2, A Angerud2, J Caudell1, (1) H. Lee Moffitt Cancer Center, Tampa, FL, (2) RaySearch Laboratories AB, Stockholm, Sweden
Presentations
(Saturday, 3/26/2022) [Central Time (GMT-5)]
Purpose: The switch to Monte Carlo (MC) calculations in head and neck planning is particularly challenging due to the heterogeneous nature of the site. The differences in dose to tissue were widely discussed in the literature, but seldom in the context of reoptimization with a MC dose engine. The target dose homogeneity received little attention. In clinical practice we adhere to stricter target coverage and dose homogeneity constraints than typically reported. Considering favorable complications rates achieved when this precise planning was performed with a collapsed cone (CC) algorithm, it was imperative to maintain the planning approach as much as possible.
Methods: Twenty-one clinical VMAT plans previously developed in Pinnacle with the CC algorithm were recalculated “as is” with RayStation (RS) MC algorithm and then reoptimized in RS with both CC and MC algorithms. Introduction of the dose to tissue conversion in the optimization pencil beam (PB) algorithm was evaluated. The optimal MC simulation statistical uncertainty (SU) was determined.
Results: The reported PTV D0.03cc hot spot is difficult to limit to the previous goal of 105% of the prescription. The average hot spot with reoptimized RayStation MC plans was statistically significantly higher compared to Pinnacle and RS CC algorithms by 1.2 and 1.0%, respectively. With the SU of 0.3%, only a small portion of this difference can be attributed to the statistical blurring of the target DVH. The 95% confidence interval (CI) indicates that a hot spot of =107% is achievable with MC. Compared to the CI for the historical CC plans recalculated with MC (upper limit 108%), in real terms this result is better. Changes to the PB optimization dose algorithm had no effect.
Conclusion: Precision planning previously employed at our institution can be successfully duplicated with a MC dose engine reporting dose-to-tissue. The effective hot spot is reduced.
Funding Support, Disclosures, and Conflict of Interest: A. Angerud and M. Isacson are employees of RaySearch Laboratories whose product is discussed.
ePosters
- 174-60669-16011654-182599-1541755936.pdf (V Feygelman)
Keywords
Monte Carlo, Convolution/superposition, Dose Uniformity
Taxonomy
TH- External Beam- Photons: treatment planning/virtual clinical studies
Contact Email