Localization and monitoring of the target during stereotactic radiotherapy: comparation between varian portal vision™ technology with IR camera exactrac® and Brainlab 6d exactrac®/Novalis Tx System integrated with Brainlab’s Robotics Couch

Abstract

Generally, treatment for patients of secondary brain lesions involves surgery. When this is not possible, stereotactic radiotherapy confirmed its standing as the elective treatment choice. Intra/inter-fraction movements and set up errors, have always been a problem for radiation therapy. To take into account organ motions and to be sure the dose coverage of planning target volume, currently Image-guided radiation therapy (IGRT) has been employed. The use of imaging during radiation therapy improves the precision and accuracy of treatment delivery during the course of each session, especially when dose escalation studies are planned during the treatment.

Purpose

The study aims to describe two different systems concerning patient positioning and intra-fraction motion of intracranial lesions in stereotactic treatments in use at our institution. We show the evolution of the two instruments used, Varian Portal Vision™ with IR camera ExacTrac® and BrainLAB 6D ExacTrac® / Novalis Tx System integrated with BrainLAB's robotics 6D couch, as they constitute a particularly significant factor effectiveness in radiation therapy treatment.

The study demonstrated some limitations of the Varian Portal Vision™ in comparison with IR camera ExacTrac®. We drafted a protocol that has been and still is in use in our institution for the marker position during simulation phase using anatomical landmarks easily identifiable in all patients. Furthermore, we have verified that the position of the patient can be corrected by adjusting the actual marker position to the planned marker position thanks to infrared camera systems in the linear accelerator room. Lastly we evaluated the validity of the online verification system.

Materials and methods

We used a CT simulator Siemens SOMATOM® Emotion® in radiotherapy treatment planning, Varian Linear Accelerator equipped with a 2D-EPID system and ExacTrac® positioning system by BrainLAB. Standardization process of the reflecting spheres positioning, as per protocol, was based on thermoplastic mask, markers supplied by BrainLAB, and infrared camera system in the linear accelerator room. Process validation was realized as the collection and evaluation of data related to intrafraction setup variability recorded for all patient on a daily basis. At the end of each treatment four portal images were acquired with gantry angle 0° and couch rotation 45°, 90°, 270°, 315° to verify the accuracy of the BrainLAB system. The four portal images were matching with the corresponding DRR obtained from a reconstructed planning CT projection images. The displacements obtained by the BrainLAB system were estimated as significant (> 1mm) and compared with the displacements obtained by 2D-EPID images (assumed as Gold Standard).

Results

In the first phase of research, the correlation between treatment beams and standardized markers arrangement was showed in a scatter plot for each patient. The results showed that 94 beams out of 107 examined (approximately 88%) were visible. In particular, for 2 out of 10 patients were estimated a correctly displayed position of the treatment beams. In the remaining 2 patients we have seen a greater difficulty in the signal detection. In one case the localization of the brain tumor was too deep, for which the Gantry overlapped with the infrared camera. In the other case a loss of signal from IR camera occurred due to usury of the reflecting surface of X-ray markers. To validate the BrainLAB device as online verification system, a total of 168 images for 13 patients were acquired and subsequently matched with with the DRRs. The logging and statistics of the data collected are summarized below: sensibility 50%, specificity 89%, positive predictive value of 52%, negative predictive value 88% and accuracy (65%).

Conclusions

After verifying quantitative systems reliability parameters, we reached the conclusion that both systems have a high potential for reproducibility and this has allowed us to create a standard for the arrangement of reflective markers. Data from our study show that BrainLAB system was unreliable in respect to the online verification, obtaining a low sensitivity (50%) and a positive predictive value of 52, indicating that in approximately half the displacements requested by BrainLAB system, it presented a potential source of errors. Both systems are valid for patient localization in the treatment of intracranial lesions stereotactic radiation therapy. We also highlighted that technology developments from the first system to the second, newer, system allowed us to monitor and solve some limits of the old system. Nonetheless, the Varian Portal Vision™ and IR camera ExacTrac® represents a very useful instrument, being a cheaper and less complex system for stereotactic radiotherapy.