Marieke van Son

73 FOCAL SALVAGE TREATMENT: A NARRATIVE REVIEW software is used to register the pre-operatively delineated tumor location to real-time prostate images. Image registration may be either rigid (overlay of images without adjustment for possible prostate deformation during treatment) or non-rigid (using algorithms that compensate for deformation). Some factors that contribute to prostate deformation are unavoidable, such as swelling of the prostate due to catheter insertion during a brachytherapy implant procedure. Prostate motion can also be caused by surrounding organ movement, such as rectal distension due to flatulence or introduc- tion of an ultrasound probe. Evidently, non-rigid registration is challenging: a variety of registration methods using different algorithms have been presented in the search for the most optimal solution [94]. The next step in the development of MRI-guided intervention is the incorporation of live MR-images into the treatment workflow, thereby achieving direct treatment guidance and avoiding any registration errors. Although early experiences with re- al-time MRI-guided brachytherapy date back to 1997, this approach has not been widely adopted yet due to logistical issues such as resource demand and procedural time prolongation [95]. One of the obvious challenges of in-bore intervention is the limited workspace. Open MRI units that provide access to the patient while imaging are avail- able, but these deliver low image quality and need increased scanning time due to the inherently lower signal-to-noise ratio. To overcome these shortcomings, a robotic MRI-compatible implantation device for prostate brachytherapy was developed at our institution (see Figure 3). The robot system fits in a 1.5T MRI scanner and can be placed between the patient’s legs. In 2010, the first clinical proof of principal study was performed with the UMCU robot, success- fully implanting gold fiducial markers into the prostate for external beam radiation [96]. It was shown that the in vivo use of the robot was feasible. After this first clinical test, the UMCU robot was further developed and optimized for the application of brachyther- apy implant procedures. We are currently working on a study investigating the in vivo technical feasibility of robotic insertion of a brachytherapy needle into the prostate. It is expected that this study will be a step forward in the development of MRI-guided focal salvage brachytherapy with a robotic device. In the future, a full MRI-guided robotic implantation procedure may allow for a reduction of needles needed for the implant [97], with expected lower toxicity rates and a reduction of time necessary for the procedure. 4