On the pre-clinical validation of a commercial model-based optimisation engine: application to volumetric modulated arc therapy for patients with lung or prostate cancer

Abstract

Purpose: To evaluate the performance of a model-based optimisation process for volumetric modulated arc therapy applied to advanced lung cancer and to low risk prostate carcinoma patients.

Methods and materials: Two sets each of 27 previously treated patients, were selected to train models for the prediction of dose-volume constraints. The models were validated on the same sets of plans (closed-loop) and on further two sets each of 25 patients not used for the training (open-loop).

Results: Quantitative improvements (statistically significant for the majority of the analysed dose-volume parameters) were observed between the benchmark and the test plans. In the pass-fail analysis, the rate of criteria not fulfilled was reduced in the lung patient group from 11% to 7% in the closed-loop and from 13% to 10% in the open-loop studies; in the prostate patient group it was reduced from 4% to 3% in the open-loop study.

Conclusions: Plans were optimised using a knowledge-based model to determine the dose-volume constraints. The results showed dosimetric improvements when compared to the benchmark data, particularly in the sparing of organs at risk. The data suggest that the new engine is reliable and could encourage its application to clinical practice.

Keywords: Knowledge based planning; Locally advanced lung cancer; Low risk prostate cancer; RapidArc; RapidPlan.