Search

Purpose: We propose a model for estimating dosimetric uncertainties as a consequence of discrepancies in the Deformable Image Registration (DIR) process when accumulating dose for adaptive Pencil Beam Scanned (PBS) proton therapy. Methods: Three components are considered: (i) deformable vector fields (DVFs) derived from different DIR algorithms; (ii) a patient specific linear regression model of DVF uncertainty, and (iii) dose gradients from each fraction dose distribution. The model is constructed for the first fraction only, by correlating the magnitude of DVFs calculated with a reference algorithm (B-spline) to DVF uncertainty, quantified as the voxel-wise difference between the largest and smallest vector magnitudes of the different DVFs resulting from the different DIR algorithms. In subsequent fractions, only the reference (B-spline) DVF is applied, with dose uncertainties from its use being calculated by the multiplication of the model predicted DVF uncertainty with the dose gradients. For validation, adaptive PBS proton therapy plans were accumulated for seven lung cancer patients, for which nine repeated (fraction) CTs were available. As a ‘ground-truth’ (GT), each fraction specific dose distribution was warped back to the planning CT using the five different DIR algorithms, and the voxel-wise max-min dose uncertainty calculated. Voxel-wise max-min dose uncertainties were then calculated using the DVF uncertainty model and were compared to those from this GT scenario. Results: Mean differences between predicted and GT uncertainties were 2.3±6.0% over all seven patients (each with nine CTs and six structures). For all non-zero dose inside the body, uncertainty could be predicted to 2.5±8.0% of the GT uncertainty. Conclusion: We propose a method for estimating DIR induced uncertainties in dose accumulation for adapted PBS proton therapy, with predicted uncertainties being within 3% of the ground truth. The model will be further improved by including additional parameters such as DVF directional uncertainty and image feature information. Funding Support, Disclosures, and Conflict of Interest: This project is funded by Krebsliga Schweiz.

With proton therapy, highly conformal dose can be applied to the tumor while sparing the surrounding organs at risk. Protons are however sensitive to density changes in the beam path: If the patient anatomy is changing, the resulting dose is distorted. To restore the initial plan quality, the plan has to be adapted. In this thesis, the concept of daily adaptive proton therapy (DAPT) is investigated to react to daily anatomical changes and improve the overall treatment quality. With DAPT, a new plan is optimized based on a 3D image acquired directly (some minutes) before each fraction. The whole online workflow must be efficient, not exceeding 5 to 10 minutes, to not increase the occupancy time in the treatment room too much or decrease the patient comfort. DAPT has the potential to improve the treatment quality, but technical and workflow challenges have prevented its clinical integration so far. In this thesis, different aspects about the clinical benefit of DAPT and the sensitivity of DAPT to different uncertainties caused by simplifications in the DAPT workflow, such as analytical dose calculation or structure propagation, are investigated. Additionally, the uncertainties in the evaluation of the overall applied treatment dose and outcome are discussed and the dosimetric benefit of DAPT is shown in a phantom experiment. After a short introduction (chapter 1), more details about the clinical implementation of DAPT at the Paul Scherrer Institute (PSI) are discussed in chapter 2. The potential benefit of DAPT for patients treated in the paranasal sinuses is discussed in chapter 3. There it is shown how DAPT enables novel planning concepts and can reduce the integral dose to healthy tissue. Chapter 4 shows that DAPT brings a dosimetric benefit to the patients, even if the daily plan is optimized with analytical dose calculation algorithms in challenging anatomical sites such as the nose or the lung. Chapter 5 investigates how the use of propagated structures is affecting the dose distribution of the daily plan, because in the strict time span allowed for a DAPT treatment, there is no time for correcting the contours. For each fraction, there is only limited time to check the daily plan, which makes a careful offline dose review and monitoring of the treatment progress necessary. In chapter 6, the uncertainties of deformable image registration for the dose accumulation are investigated and in chapter 7 the effect of DAPT to the clinical outcome (e.g. tumor control, mortality or other side effects) is evaluated. Finally, the dosimetric benefit of the DAPT workflow implemented at PSI is shown in chapter 8 in an end-to-end test over multiple fractions with a newly developed anthropomorphic phantom with anatomical changes. The work described in this thesis puts the benefits of DAPT in relation to different uncertainties occurring during the DAPT workflow. In different planning studies and in an end-to-end experiment it was shown that the dosimetric benefits of DAPT are larger than the risks of the investigated uncertainties. It was also shown that the dosimetric benefit of DAPT can translate into an improved clinical outcome. The first patient treatment with the DAPT workflow tested in this project is planned for the beginning of next year, 2021.

Background: For proton therapy of paranasal tumors, field directions avoiding volumes that might change during therapy are typically used. If the plan is optimized on the daily anatomy using daily adapted proton therapy (DAPT) however, field directions crossing the nasal cavities might be feasible. In this study, we investigated the effectiveness of DAPT for enabling narrow-field treatment approaches. Material and methods: For five paranasal tumor patients, representing a wide patient spectrum, anatomically robust 4-field-star and narrow-field plans were calculated and their robustness to anatomical and setup uncertainties was compared with and without DAPT. Based on the nominal planning CTs, per patient up to 125 simulated CTs (simCTs) with different nasal cavity fillings were created and random translations and rotations due to patient setup uncertainties were further simulated. Plans were recalculated or re-optimized on all error scenarios, representing non-adapted and DAPT fractions, respectively. From these, 100 possible treatments (60 GyRBE, 30 fx) were simulated and changes in integral dose, target and organs at risk (OARs) doses evaluated. Results: In comparison to the 4-field-star approach, the use of narrow-fields reduced integral dose between 29% and 56%. If OARs did not overlap with the target, OAR doses were also reduced. Finally, the significantly reduced target coverage in non-adapted treatments (mean V95 reductions of up to 34%) could be almost fully restored with DAPT in all cases (differences, Acta Oncologica, 58 (10), ISSN:0284-186X, ISSN:1651-226X