Immobilisation accuracy of double shell positioning system for stereotactic radiotherapy in patients with brain tumors.

Introduction: Non-invasive frameless systems have paved its way for stereotactic radiotherapy treatments compared to gold standard invasive rigid frame-based systems as they are comfortable to patients, do not have risk of pain, bleeding, infection, frame slippage and have similar treatment efficacy.

Aim and objective: To estimate immobilisation accuracy (interfraction and intrafraction) and PTV margins with double shell positioning system (DSPS) using daily image guidance for stereotactic radiotherapy in patients with brain tumors.

Materials and method: A prospective study was done in 19 cranial tumor patients with KPS ≥70, immobilized by the DSPS with mouth bite and treated with LINAC based image guided stereotactic radiotherapy. A PTV of 2 mm was given from the tumor. Patients were positioned by aligning the treatment room lasers to the marked isocentre on the DSPS. For all patients 3D-image registration (automatic bony anatomy) was performed by matching 1st CBCT images with the simulation reference CT (simCT) images to measure the 3D target displacement prior to the treatment delivery every day. The initial setup deviation/ interfraction motion- translational (medio-lateral-X, cranio-caudal-Y, anterior-posterior-Z) displacements in mm and rotational axis (pitch, roll, yaw) in degrees were documented. All transitional errors were corrected online. For residual Interfraction motion a 2nd CBCT was done after correction of initial setup errors and matched with simCT and treatment executed. To evaluate the intrafraction motion CBCT was done at end of every fraction and compared with 2nd CBCT images. Systematic and random errors were calculated and planning target volume (PTV) margins were estimated using van Herk formula.

Results: A total of 95 CBCT image data sets were evaluated. The initial setup relocation accuracy -mean (±SD) displacements for translational X, Y and Z directions were 1.2 (0.6), 1.0 (0.9), 0.5 (0.6) mm respectively and rotations were 0.6 (± 0.5), 0.1 (± 0.4), 0.60 (± 0.6) degrees for pitch, roll and yaw respectively. Post correction, the residual interfraction mean displacements in X, Y and Z directions were 0.1 (± 0.3), 0.2 (± 0.6), 0.3 (± 0.4) mm respectively. The population systematic and random translational errors were 0.2, 0.3, 0.3 and 0.6, 0.4, 0.4 respectively. For intrafraction motion, the mean (±SD) displacements were 0.3 (± 0.2), 0.3 (± 0.5), 0.4 (± 0.2) mm in X, Y and Z directions respectively with minimal rotations in all axis. The intrafraction population systematic and random errors were <0.5 mm for all displacements. The online corrections decreased the interfraction PTV margins to 1.1, 1.1 and 1.2 mm in X, Y and Z directions respectively.

Conclusion: Frameless DSPS system with mouth bite using image guidance achieved a setup accuracy of a millimeter for stereotactic treatment in cranial tumors with submillimeter intrafraction motion. A decrease in PTV margins of 1.1 mm was achieved for future patients undergoing brain SRT.