Is the Motion Causing a Commotion? Two-Fraction Prostate SBRT on the MR-Linac.

Abstract

In HERMES (NCT04595019) men with localized prostate cancer are treated on the Unity MR-Linac platform (MRL, Elekta AB, Stockholm) and randomized between stereotactic body radiotherapy (SBRT) with 36.25 Gy in 5 fractions and 24 Gy in 2 fractions. Patients randomized to two fractions receive 24 Gy to the high risk PTV, 20 Gy to the low risk PTV and a boost of 27 Gy to the dominant intraprostatic lesion. This study explores dose received by the target and organs at risk (OARs) when considering intrafraction motion in two fraction SBRT. Targets and OARs were delineated and a reference plan generated on Monaco v5.40.01 (Elekta). An Adapt-to-Shape (ATS) workflow was used. Contours were propagated to the session MRI (MRIsession) and edited accordingly. Prior to delivery, a verification MRI (MRIverif) was acquired with baseline shifts corrected for using the Adapt-to-Position (ATP-of-ATS) workflow. A post treatment MRI (MRIpost) was acquired after delivery. Men in the 2-fraction arm received each fraction in 2 sub-fractions sequentially on the same day, to mitigate intrafraction motion. The plans of 5 men receiving 2 fraction SBRT were analyzed. The targets, urethra, bladder and rectum were recontoured on the MRIverif and MRIpost. Delivered plans were recalculated on the corresponding MRIverif and MRIpost. The percentage of optimal and mandatory target dose constraints met were calculated. Accumulated OAR doses were calculated by averaging their respective dose statistics across all sub-fractions, conservatively assuming that the same area of the OAR receives the maximum dose each fraction. Analysis was carried out separately for MRIverif and MRIpost as the true 'delivered dose' most likely lies between these two estimates. There was good coverage across all fractions. The mandatory constraints of CTVpsv V24.0 Gy > 95% and CTVsv V20.0 Gy > 95% were met in 100% of fractions and V2700cGy > 95% in 90% on the MRIpost. Table 3 shows OAR dose. This work demonstrates that target coverage is good, even for the GTV where no margin is applied. With our conservative dose calculation approach, we found dose constraints are exceeded for some patients. However, treatment has been well tolerated, suggesting that that our current dose constraints may be cautious. Once Elekta's True Tracking and automated gating software is implemented at our center we will be able to further improve OAR clinical goal compliance.