Introduction
In head-and-neck IMPT, trigger-based offline plan adaptation (Offline trigger-based) is often used. Our goal was to compare this to four alternative adaptive strategies for dosimetry, workload and treatment time, considering also foreseen further technological advancements, including anticipated automation.

Materials and methods
Alternative strategies included weekly offline re-planning (Offline weekly), daily plan selection from a library (Library static and Library progressive) and a fast, approximate daily online re-optimization approach (Online re-opt). Impact on CTV coverage and NTCPs was assessed by simulations based on repeat-CTs from 15 patients. Full daily re-planning was used as dosimetric benchmark. Increases in workload and treatment time were estimated.

Results
Both for coverage and NTCPs, fast Online re-opt performed as well as full re-planning. Compared to current practice, Online re-opt showed enhanced probabilities for high coverage, and resulted in reductions in grade ≥ II NTCPs of 4.6 ± 1.7 %-point for xerostomia and 4.2 ± 2.3 %-point for dysphagia. Offline weekly and library strategies did not show coverage enhancements and resulted in smaller NTCP improvements. Further automation can largely limit workload and treatment time increases. With anticipated further automation, adaptation-related workload of Offline weekly, Library static, Library progressive, and Online re-opt was expected to increase by 3, 8, 21, and 66 h for 35 fraction treatment courses compared to Offline trigger-based. The corresponding adaptation-related prolonged treatment times were estimated to be 0, 4, 6, and 29 min/fraction.

Conclusion
Online adaptive strategies could approach dosimetric quality of full re-planning at the cost of additional workload and prolonged treatment time compared to the current offline adaptive strategy. Automation needs to play a key role in making more complex adaptive approaches feasible.@en

Purpose: To develop and validate a robust and accurate registration pipeline for automatic contour propagation for online adaptive Intensity-Modulated Proton Therapy (IMPT) of prostate cancer using elastix software and deep learning. Methods: A three-dimensional (3D) Convolutional Neural Network was trained for automatic bladder segmentation of the computed tomography (CT) scans. The automatic bladder segmentation alongside the computed tomography (CT) scan is jointly optimized to add explicit knowledge about the underlying anatomy to the registration algorithm. We included three datasets from different institutes and CT manufacturers. The first was used for training and testing the ConvNet, where the second and the third were used for evaluation of the proposed pipeline. The system performance was quantified geometrically using the dice similarity coefficient (DSC), the mean surface distance (MSD), and the 95% Hausdorff distance (HD). The propagated contours were validated clinically through generating the associated IMPT plans and compare it with the IMPT plans based on the manual delineations. Propagated contours were considered clinically acceptable if their treatment plans met the dosimetric coverage constraints on the manual contours. Results: The bladder segmentation network achieved a DSC of 88% and 82% on the test datasets. The proposed registration pipeline achieved a MSD of 1.29 ± 0.39, 1.48 ± 1.16, and 1.49 ± 0.44 mm for the prostate, seminal vesicles, and lymph nodes, respectively, on the second dataset and a MSD of 2.31 ± 1.92 and 1.76 ± 1.39 mm for the prostate and seminal vesicles on the third dataset. The automatically propagated contours met the dose coverage constraints in 86%, 91%, and 99% of the cases for the prostate, seminal vesicles, and lymph nodes, respectively. A Conservative Success Rate (CSR) of 80% was obtained, compared to 65% when only using intensity-based registration. Conclusion: The proposed registration pipeline obtained highly promising results for generating treatment plans adapted to the daily anatomy. With 80% of the automatically generated treatment plans directly usable without manual correction, a substantial improvement in system robustness was reached compared to a previous approach. The proposed method therefore facilitates more precise proton therapy of prostate cancer, potentially leading to fewer treatment-related adverse side effects.

@en

The aim of this study was to investigate the feasibility of using prompt gamma (PG) ray emission profiles to monitor changes in dose to the planning target volume (PTV) during pencil beam scanning (PBS) proton therapy as a result of day-to-day variation in patient anatomy.
 For 11 prostate patients, we simulated treatment plan delivery using the patients' daily anatomy as observed in the planning CT and 7-9 control CT scans, including the detected PG profiles resulting from the 5%, 10%, and 20% most intense proton pencil beams. For each patient, we determined the changes in dosimetric parameters for the high- and low-dose PTVs between the simulations performed using the planning CT scan and the different control CT scans and correlated these to changes in the PG emission profiles.
 Changes in coverage of the high- and low-dose PTV correlated most strongly with the median and mean absolute PG emission profile shifts of the 5% most intense pencil beams, respectively. With a mean Pearson correlation coefficient of -0.76 (SD: 0.17) for the high-dose PTV and of -0.60 (SD: 0.51) for the low-dose PTV.
 We showed, as a proof of principle, that PG emission profiles obtained during PBS proton therapy could be used to detect changes in PTV coverage due to day-to-day anatomical variation.@en