Sci-Fri AM: YIS-06: On-line adaptive radiation therapy based on the intra-fractional digital tomosynthesis images

To investigate the feasibility of performing on-line adaptive radiation therapy (ART) based on the intra-fractional digital tomosynthesis (DTS) images.

Intra-fractional DTS images were reconstructed as the gantry rotated between treatment positions. An edge detection algorithm was used to automatically segment the DTS images as the gantry arrived at each treatment position. The original treatment plan was then re-optimized for the most recent DTS image contours and dose was delivered from each treatment position based on the newly re-optimized plan. Plan re-optimization was performed using modified direct aperture optimization (DAO). To test our system, a model representing typical prostate, bladder and rectum anatomy was generated. First, a treatment plan based on this original anatomy was created using our DAO system. To simulate prostate deformations, three clinically relevant deformations (small, medium and large) were modeled by systematically deforming the original anatomy. The ability of our approach to adapt the original treatment plan and account for the anatomy deformations was investigated.

Based on the dose-volume constraints from the RTOG 0415 prostate protocol, the original treatment plan would have been clinically unacceptable for all three deformations. Using our approach to on-line ART, the original treatment plan was successfully adapted to arrive at a clinically acceptable plan for all three anatomy deformations.

We have shown that performing on-line ART based on intra-fractional DTS images is feasible. The advantages are reduced treatment time and the ability to detect and account for patient motion during the treatment fraction.