Development of a virtual spacer to support the decision for the placement of an implantable rectum spacer for prostate cancer radiotherapy: Comparison of dose, toxicity and cost-effectiveness

作者：van Wijk Yvonka^*; Vanneste Ben G L; Walsh Sean; van der Meer Skadi; Ramaekers Bram; van Elmpt Wouter; Pinkawa Michael; Lambin Philippe

来源：Radiotherapy and Oncology, 2017, 125(1): 107-112.

DOI：10.1016/j.radonc.2017.07.026

• Prostate cancer
• Radiotherapy
• Implantable rectum spacer
• NTCP models
• Cost-effectiveness
• Decision support

Introduction: Previous studies have shown that the implantable rectum spacer (IRS) is not beneficial for all patients. A virtual IRS (V-IRS) was constructed to help identify the patients for whom it is cost-effective to implant an IRS, and its viability as a tool to tailor the decision of an IRS implantation to be beneficial for the specified patient was assessed. Please watch animation: (https://www.youtube.comiwatch?v=tDlagSXMKqw)
Materials and methods: The V-IRS was tested on 16 patients: 8 with a rectal balloon implant (RBI) and 8 with a hydrogel spacer. A V-IRS was developed using 7 computed tomography (CT) scans of patients with a RBI. To examine the V-IRS, CT scans before and after the implantation of an IRS were used. IMRT plans were made based on CT scans before the IRS, after IRS and with the V-IRS, prescribing 70 Gray (Gy) to the planning target volume. Toxicity was accessed using externally validated normal tissue complication probability (NTCP) models, and the Cost-effectiveness was analyzed using a published Markov model.
Results: The rectum volume receiving 75 Gy (V75) were improved by both the IRS and the V-IRS with on average 4.2% and 4.3% respectively. The largest NTCP reduction resulting from the IRS and the V-IRS was 4.0% and 3.9% respectively. The RBI was cost-effective for 1 out of 8 patients, and the hydrogel was effective for 2 out of 8 patients, and close to effective for a third patient. The classification accuracy of the model, regarding cost-effectiveness, was 100%.
Conclusion: The V-IRS approach in combination with a toxicity prediction model and a cost-effectiveness analyses is a promising basis for a decision support tool for the implantation of either a hydrogel spacer or a rectum balloon implant.

• 出版日期2017-10
• 单位RWTH Aachen