One addition to this topic from the practice. We just had one stereotactic lung patient treated with AlignRT in BH. The target located longitudinally near diaphragm, laterally and vertically in the midline of the lung. We created ROI for AlignRT on to bony chest wall, caudal parts were weighted in the ROI delineation, still excluding the diaphragm, just like advised in the VisionRT ROI reference guide. We acquired half arc CBCT and matched the images. Small shifts were done in online match patient in BH, based on tumor match. Both tumor and bony structures were in their correct location. Treatment went really well based on AlignRT. After treatment, we acquired half arc CBCT, just to verify tumor location after treatment. Tumor was shifted more than 1 cm in both VRT and LNG from its correct location, bony structures (ribs, sternum, vertebra) were still in their correct location without need for corrections just like our AlignRT ROI suggested. This means that ROI on the bony chest wall did not correlate with the diaphragm, and thereby the tumor near diaphragm was able to shift remarkably without chest wall ROI detecting it. For the next 4 fractions we created two ROIs for AlignRT, bony chest wall ROI for setup and ROI which included both diaphragm and caudal parts of bony chest wall for treatment. Additionally, half arc CBCT was acquired three times, at the beginning, at the middle of treatment and at the end of the treatment. With that treatmentROI, tumor located inside 3 mm in all the CBCTs all the time. Small couch shifts were done also based on CBCTs taken at the middle of the treatment when needed. Possibilty to simultaneously monitor both lower chest wall ROI and diaphragm ROI could be optimal in the future. User could have a possibility to select which ROI interrupts the beam, or both.
This was just one warning example that external bony chest wall 6D location and internal tumor location near diaphragm are not necessarily well correlated in BH. Great presentation: Latest research and evidence moving from block-based to SGRT SABR treatments – Stewart Gaede. Next time we have to use triggered kV-imaging to ensure the correlation between the surface and internal position and manually turn off the beam if the tumor comes outside the volume.
Best regards Marko