Feasibility of Ultrasound-Based Intra-Fraction Motion Monitoring for Prostate Stereotactic Body Radiation Therapy

Abstract

Real-time transperineal ultrasound offers a non-invasive option to monitor the inter- and intra-fraction motion of the prostate. This prospective study evaluated the feasibility of using transperineal ultrasound for treatment setup and intra-fraction motion tracking for prostate stereotactic body radiation therapy (SBRT). 25 prostate SBRT patients treated at our institution from March 2016 to December 2019 were included. All patients received a total dose of 36.25 Gy in 5 fractions using volumetric modulated arc therapy. At simulation, volumetric ultrasound images were acquired with the patient at the treatment position and fused with the planning CT. The prostate volume was contoured on the ultrasound images and then used as the reference structure for ultrasound-based motion monitoring during treatment. Prior to each treatment, the patient was first set up using skin marks and then shifted using KV-CBCT. Ultrasound images were acquired continuously during treatment to track the prostate motion in real-time. The treatment beam would be paused if the prostate motion exceeded 2mm in any direction for 5 seconds. For patient setup, the shifts calculated by the ultrasound images were different from the shifts from the KV-CBCT. The Differences were 1.1 +/-5.4 mm in lateral, -1.7+/-4.9 mm in vertical and 0.2+/-4.0 mm in longitudinal directions. There were moderate correlations between the shifts of the two systems (R = 0.54 in lateral, 0.51 in vertical and 0.39 in longitudinal directions). Real-time motion monitoring was successfully performed in 120 fractions. The average intra-fraction prostate motion was 0.1+/-0.6 mm in lateral, -0.2+/-0.8 mm in vertical and 0.1+/-0.9 mm in longitudinal directions. The prostate motion exceeded 2 mm in any direction in 6.4% (ranging from 0% to 60.5% for individual patients) of the time, and >3 mm in 3.2% of the time. Beam pauses happened in 13 of the 25 patients and in 26 of the 120 fractions. The prostate motion was transient in the majority of the time except in 12 fractions from 7 patients, additional CBCT was required to reposition the patient. Ultrasound system does not agree with CBCT for pre-treatment patient setup. However, it is feasible to use ultrasound for intra-fraction monitoring without the need for inserting fiducial markers in the prostate or giving additional imaging dose to patients. Approximately half of the prostate SBRT patients require intra-fraction motion monitoring to ensure that the target stays within the PTV margin, and one-third of the patients require repositioning during treatment.