Sequential monoscopic image-guided motion compensation in tomotherapy stereotactic body radiotherapy (SBRT) for prostate cancer

Lan Lu, Edward Chao, Tong Zhu, Andrew Zhuang Wang, Jun Lian

Research output: Contribution to journalArticlepeer-review


Purpose: To manage intra-fractional motions, recent developments in tomotherapy enable a unique capability of adjusting MLC/jaw to track the moving target based on the intra-fractional motions detected by sequential monoscopic imaging. In this study, we evaluated the effectiveness of motion compensation with a realistic imaging rate for prostate stereotactic body radiotherapy (SBRT). The obtained results will guide optimizing treatment parameters and image-guided radiation therapy (IGRT) in tomotherapy using this approach. Methods: Ten retrospective prostate cases with actual prostate motion curves previously recorded through the Calypso system were used in this study. Based on the recorded peak-to-peak motion, these cases represented either large (> 5 mm) or median (≤ 5 mm) intra-fractional prostate motions. All the cases were re-planned on tomotherapy using 35 Gy/5 fractions SBRT regimen and three different jaw settings of 1 cm static, 2.5 cm static, and 2.5 cm dynamic jaw. Two motion compensation methods were evaluated: a complete compensation that adjusted the jaw and MLC every 0.1 s (the same rate as the Calypso motion trace), and a realistic compensation that adjusted the jaw and MLC at an average imaging interval of 6 s from sequential monoscopic images. An in-house 4D dose calculation software was then applied to calculate the dosimetric outcomes from the original motion-free plan, the motion-contaminated plan, and the two abovementioned motion-compensated plans. During the process, various imaging rates were also simulated in one case with unusually large motions to quantify the impact of the KV-imaging rate on the effectiveness of motion compensation. Results: The effectiveness of motion compensation was evaluated based on the PTV coverage and OAR sparing. Without any motion-compensation, the PTV coverage (PTV V100%) of patients with large prostate motions decreased remarkably to 55%–82% when planning with the 1 cm jaw but to a less level of 67–94% with the 2.5 cm jaw. In contrast, motion compensation improved the PTV coverage (>92%) when combined with the 2.5 cm jaw, but less effective, around 75%–94%, with the 1 cm jaw. For OAR sparing, the bladder D1cc, bladder D10cc, and rectum D1cc all increased in the motion-contaminated plans. Motion compensation improved OAR sparing to the equivalent level of the original motion-free plans. For patients with median prostate motion, motion-induced degradation in PTV coverage was only observed when planning with the 1 cm jaw. After motion compensation, the PTV coverage improved to better than 94% for all three jaw settings. Additionally, the effectiveness of motion compensation depends on the imaging rate. Motion compensation with a typical rate of two KV images per gantry rotation effectively reduces motion-induced dosimetric uncertainties. However, a higher imaging rate is recommended when planning with a 1 cm jaw for patients with large motions. Conclusion: Our results demonstrated that the performance of sequential monoscopic imaging-guided motion compensation on tomotherapy depends on the amplitude of intra-fractional prostate motion, the plan parameter settings, especially jaw setting, gantry rotation, and the imaging rate for motion compensation. Creating a patient-specific imaging guidance protocol is essential to balance the effectiveness of motion compensation and achievable imaging rate for intra-fractional motion tracking.

Original languageEnglish
Pages (from-to)518-528
Number of pages11
JournalMedical physics
Issue number1
StatePublished - Jan 2023


  • monoscopic imaging
  • realistic motion compensation
  • tomotherapy SBRT


Dive into the research topics of 'Sequential monoscopic image-guided motion compensation in tomotherapy stereotactic body radiotherapy (SBRT) for prostate cancer'. Together they form a unique fingerprint.

Cite this