Applying a commercial atlas-based synthetic Computed Tomography algorithm to patients with hip prostheses for prostate Magnetic Resonance-only radiotherapy. (April 2019)

• Record Type: Journal Article
• Authors: Wyatt, Jonathan; McCallum, Hazel
• Is Part Of: Radiotherapy and oncology. Volume 133(2019)
• Page Start: 100
• Page End: 105
• Publication Date: 2019-04
• Subjects: MR-only radiotherapy planning -- Magnetic Resonance imaging -- Radiotherapy -- Prostate -- Hip prostheses; Oncology -- Periodicals; Radiotherapy -- Periodicals; Tumors -- Periodicals; Medical Oncology -- Periodicals; Neoplasms -- radiotherapy -- Periodicals; Radiotherapy -- Periodicals; Radiothérapie -- Périodiques; Cancérologie -- Périodiques; Tumeurs -- Périodiques; Electronic journals; 616.9940642
• DOI: 10.1016/j.radonc.2018.12.029
• Languages: English
• ISSNs: 0167-8140

Abstract

Highlights: Magnetic Resonance-only radiotherapy requires generating synthetic Computed Tomography (sCT) images. All current sCT software designed for patients without hip prostheses. This study investigated applying sCT software to patients with hip prostheses. Mean dosimetric differences were small (<0.5%) and similar to results for patients without prostheses. Abstract: Background and purpose: Magnetic Resonance (MR)-only prostate radiotherapy has recently been clinically implemented using commercial synthetic Computed Tomography (sCT) algorithms. However patients with hip prostheses have been excluded from all MR-only research to date and assumed to require dedicated sCT algorithms. This study aimed to investigate the dosimetric accuracy of applying a commercial sCT algorithm, based on an atlas of patients without hip prostheses, to patients with prostheses. Materials and methods: 18 patients with unilateral hip prostheses received MR and CT scans in the radiotherapy position. sCTs were generated from the MR using a commercial algorithm. The clinical Volumetric Modulated Arc Therapy (VMAT) plan, consisting of partial arcs which avoided the prosthesis, was recalculated using the sCT and the dose distribution compared. Results: The mean isocentre dose difference was Δ D = (−0.4 ± 0.2)% (mean ± standard error of the mean (sem), range – 1.9%, 1.1%) and the mean differences in Planning Target Volume, bladder and rectum mean doses were ≤0.3%. The 3D global gamma pass rate with dose difference 1% and distance to agreement 1 mm within the body was Γ Body 1 / 1 = (95.0 ± 0.5)% (sem) and within the 50% isodose volume, which excluded the prosthesis, was Γ 50 % 1 / 1 = (98.5 ± 0.4)% (sem). The pass rate within the PTV was Γ PT V 2 / 2 ≥ 99.7% for all patients, although for PTVs close (≤3.5 cm) to the prosthesis Γ PT V 1 / 1 < 85% for three patients. The sCT did not accurately represent the prosthesis with a mean difference in radiological isocentre depth near the prosthesis of Δ d Outside Rad = (15.8 ± 2.6) mm (sem). However inside the treatment plan arc the difference was Δ d Inside Rad = (−1.8 ± 0.5) mm (sem). Conclusions: Using a commercial prostate sCT algorithm for patients with unilateral hip prostheses is dosimetrically accurate (<0.5%) as long as the routine prosthesis-avoidance treatment planning approach is used and the PTV is >3.5 cm from the prosthesis. This suggests MR-only prostate radiotherapy can be extended to patients with hip prostheses without requiring a specific sCT algorithm.