UNASSIGNED: Magnetic resonance imaging (MRI)-only workflow is used in photon radiotherapy (RT) today, but not yet for protons. To bring MRI-only proton RT into clinical use, proton dose calculation on MRI-derived synthetic CT (sCT) must be validated. We evaluated proton dose calculation accuracy of prostate cancer proton plans using a commercially available sCT generator already validated for photon planning.
UNASSIGNED: The retrospective planning study included 10 prostate cancer patients who underwent MRI and planning CT (pCT) before RT. sCT were generated from the MRI with MRI Planner v2.3, and compared to pCT using structural mean absolute error (MAE). The pCT was used to create one-arc volumetric modulated arc therapy (VMAT) photon plan and two-field intensity modulated proton therapy (IMPT) proton plan. Each plan was recalculated on the sCT and compared to pCT doses. Dose volume histogram parameters, gamma analyses and range differences were evaluated.
UNASSIGNED: Median MAE for the body contour was 71 HU. Dose differences between pCT and sCT were small and similar for VMAT and IMPT plans. Median (range) gamma pass rates were lower for IMPT plans with 95.8 (89.3-98.7) % compared to VMAT plans with 99.4 (91.2-99.6) %. The proton range difference was 1.0 (interquartile range -0.1 - 0.2) mm deeper for sCT compared to the reference.
UNASSIGNED: MRI-only IMPT planning for prostate cancer seems feasible in a clinical setting for the evaluated beam arrangement and sCT generator. More patients and evaluation of other beam arrangements are needed for a more general conclusion.

[This corrects the article DOI: 10.3389/fonc.2022.879167.].

The rapid development of artificial intelligence (AI) has gained importance, with many tools already entering our daily lives. The medical field of radiation oncology is also subject to this development, with AI entering all steps of the patient journey. In this review article, we summarize contemporary AI techniques and explore the clinical applications of AI-based automated segmentation models in radiotherapy planning, focusing on delineation of organs at risk (OARs), the gross tumor volume (GTV), and the clinical target volume (CTV). Emphasizing the need for precise and individualized plans, we review various commercial and freeware segmentation tools and also state-of-the-art approaches. Through our own findings and based on the literature, we demonstrate improved efficiency and consistency as well as time savings in different clinical scenarios. Despite challenges in clinical implementation such as domain shifts, the potential benefits for personalized treatment planning are substantial. The integration of mathematical tumor growth models and AI-based tumor detection further enhances the possibilities for refining target volumes. As advancements continue, the prospect of one-stop-shop segmentation and radiotherapy planning represents an exciting frontier in radiotherapy, potentially enabling fast treatment with enhanced precision and individualization.

OBJECTIVE: In radiotherapy of the head and neck (H&N) it is common for the clinical target volume (CTV) to extend to the patient\'s skin. Adding a margin for set-up uncertainty and delivery creates a planning target volume (PTV) that extends beyond the patient surface. This can result in excessive fluence being delivered to the build-up region and therefore the skin. This study evaluates four different planning methods used when inverse-planning H&N radiotherapy treatments with CTV extending to the skin. The aim of the study was to determine which planning method gives superior plan quality.
METHODS: Ten H&N cancer patients with a CTV contoured to the skin were inverse-planned using four planning methods. The planning methods compared were: cropping the optimization PTV back from the skin surface by 5.0, 3.0, and 0.0 mm and a virtual bolus method. For each planning method, the increased fluence at the skin surface was analyzed. The CTV coverage and skin doses were compared. Plan robustness was evaluated by applying an isocenter shift of ±3.0 mm in the major axes.
RESULTS: The planning method cropping the PTV 0.0 mm from the skin surface results in an increased fluence in the build-up region. The average volume of CTV receiving 98% of the prescription dose was 89.6% ± 3.4%, 91.6% ± 2.4%, and 93.5% ± 1.7% when cropped 5.0, 3.0, and 0.0 mm, respectively, and 93.4% ± 2.1% for the virtual bolus method. Introducing plan uncertainty affects CTV coverage the most when cropping 5.0 mm. When plan uncertainties are considered the methods of cropping 5.0, 3.0 mm, and the virtual bolus method have the same average skin dose within ±0.3%.
CONCLUSIONS: This study shows that a virtual bolus planning method results in no increased fluence at the patient\'s surface, improves CTV coverage, and is the most robust to changes in setup and patient anatomy.

BACKGROUND: Fibroblast activation protein (FAP) is expressed in the tumor microenvironment (TME) of various cancers. In our analysis, we describe the impact of dual-tracer imaging with Gallium-68-radiolabeled inhibitors of FAP (FAPI-46-PET/CT) and fluorodeoxy-D-glucose (FDG-PET/CT) on the radiotherapeutic management of primary esophageal cancer (EC).
METHODS: 32 patients with EC, who are scheduled for chemoradiation, received FDG and FAPI-46 PET/CT on the same day (dual-tracer protocol, 71%) or on two separate days (29%) We compared functional tumor volumes (FTVs), gross tumor volumes (GTVs) and tumor stages before and after PET-imaging. Changes in treatment were categorized as \"minor\" (adaption of radiation field) or \"major\" (change of treatment regimen). Immunohistochemistry (IHC) staining for FAP was performed in all patients with available tissue.
RESULTS: Primary tumor was detected in all FAPI-46/dual-tracer scans and in 30/32 (93%) of FDG scans. Compared to the initial staging CT scan, 12/32 patients (38%) were upstaged in nodal status after the combination of FDG and FAPI-46 PET scans. Two lymph node metastases were only visible in FAPI-46/dual-tracer. New distant metastasis was observed in 2/32 (6%) patients following FAPI-4 -PET/CT. Our findings led to larger RT fields (\"minor change\") in 5/32 patients (16%) and changed treatment regimen (\"major change\") in 3/32 patients after FAPI-46/dual-tracer PET/CT. GTVs were larger in FAPI-46/dual-tracer scans compared to FDG-PET/CT (mean 99.0 vs. 80.3 ml, respectively (p < 0.001)) with similar results for nuclear medical FTVs. IHC revealed heterogenous FAP-expression in all specimens (mean H-score: 36.3 (SD 24.6)) without correlation between FAP expression in IHC and FAPI tracer uptake in PET/CT.
CONCLUSIONS: We report first data on the use of PET with FAPI-46 for patients with EC, who are scheduled to receive RT. Tumor uptake was high and not depending on FAP expression in TME. Further, FAPI-46/dual-tracer PET had relevant impact on management in this setting. Our data calls for prospective evaluation of FAPI-46/dual-tracer PET to improve clinical outcomes of EC.

OBJECTIVE: Within a multi-institutional project, we aimed to assess the transferability of knowledge-based (KB) plan prediction models in the case of whole breast irradiation (WBI) for left-side breast irradiation with tangential fields (TF).
METHODS: Eight institutions set KB models, following previously shared common criteria. Plan prediction performance was tested on 16 new patients (2 pts per centre) extracting dose-volume-histogram (DVH) prediction bands of heart, ipsilateral lung, contralateral lung and breast. The inter-institutional variability was quantified by the standard deviations (SDint) of predicted DVHs and mean-dose (Dmean). The transferability of models, for the heart and the ipsilateral lung, was evaluated by the range of geometric Principal Component (PC1) applicability of a model to test patients of the other 7 institutions.
RESULTS: SDint of the DVH was 1.8 % and 1.6 % for the ipsilateral lung and the heart, respectively (20 %-80 % dose range); concerning Dmean, SDint was 0.9 Gy and 0.6 Gy for the ipsilateral lung and the heart, respectively (<0.2 Gy for contralateral organs). Mean predicted doses ranged between 4.3 and 5.9 Gy for the ipsilateral lung and 1.1-2.3 Gy for the heart. PC1 analysis suggested no relevant differences among models, except for one centre showing a systematic larger sparing of the heart, concomitant to a worse PTV coverage, due to high priority in sparing the left anterior descending coronary artery.
CONCLUSIONS: Results showed high transferability among models and low inter-institutional variability of 2% for plan prediction. These findings encourage the building of benchmark models in the case of TF-WBI.

OBJECTIVE: The capacity for machine learning (ML) to facilitate radiation therapy (RT) planning for primary brain tumors has not been described. We evaluated ML-assisted RT planning with regard to clinical acceptability, dosimetric outcomes, and planning efficiency for adults and children with primary brain tumors.
METHODS: In this prospective study, children and adults receiving 54 Gy fractionated RT for a primary brain tumor were enrolled. For each patient, one ML-assisted RT plan was created and compared with 1 or 2 plans created using standard (\"manual\") planning procedures. Plans were evaluated by the treating oncologist, who was blinded to the method of plan creation. The primary endpoint was the proportion of ML plans that were clinically acceptable for treatment. Secondary endpoints included the frequency with which ML plans were selected as preferable for treatment, and dosimetric differences between ML and manual plans.
RESULTS: A total of 116 manual plans and 61 ML plans were evaluated across 61 patients. Ninety-four percent of ML plans and 93% of manual plans were judged to be clinically acceptable (P = 1.0). Overall, the quality of ML plans was similar to manual plans. ML plans comprised 34.5% of all plans evaluated and were selected for treatment in 36.1% of cases (P = .82). Similar tumor target coverage was achieved between both planning methods. Normal brain (brain minus planning target volume) received an average of 1 Gy less mean dose with ML plans (compared with manual plans, P < .001). ML plans required an average of 45.8 minutes less time to create, compared with manual plans (P < .001).
CONCLUSIONS: ML-assisted automated planning creates high-quality plans for patients with brain tumors, including children. Plans created with ML assistance delivered slightly less dose to normal brain tissues and can be designed in less time.

OBJECTIVE: To investigate differences in uptake regions between methyl-11C-L-methionine positron emission tomography (11C-MET PET) and gadolinium (Gd)-enhanced magnetic resonance imaging (MRI), and their impact on dose distribution, including changing of the threshold for tumor boundaries.
METHODS: Twenty consecutive patients with grade 3 or 4 glioma who had recurrence after postoperative radiotherapy (RT) between April 2016 and October 2017 were examined. The study was performed using simulation with the assumption that all patients received RT. The clinical target volume (CTV) was contoured using the Gd-enhanced region (CTV(Gd)), the tumor/normal tissue (T/N) ratios of 11C-MET PET of 1.3 and 2.0 (CTV (T/N 1.3), CTV (T/N 2.0)), and the PET-edge method (CTV(P-E)) for stereotactic RT planning. Differences among CTVs were evaluated. The brain dose at each CTV and the dose at each CTV defined by 11C-MET PET using MRI as the reference were evaluated.
RESULTS: The Jaccard index (JI) for concordance of CTV (Gd) with CTVs using 11C-MET PET was highest for CTV (T/N 2.0), with a value of 0.7. In a comparison of pixel values of MRI and PET, the correlation coefficient for cases with higher JI was significantly greater than that for lower JI cases (0.37 vs. 0.20, P = 0.007). D50% of the brain in RT planning using each CTV differed significantly (P = 0.03) and that using CTV (T/N 1.3) were higher than with use of CTV (Gd). V90% and V95% for each CTV differed in a simulation study for actual treatment using CTV (Gd) (P = 1.0 × 10-7 and 3.0 × 10-9, respectively) and those using CTV (T/N 1.3) and CTV (P-E) were lower than with CTV (Gd).
CONCLUSIONS: The region of 11C-MET accumulation is not necessarily consistent with and larger than the Gd-enhanced region. A change of the tumor boundary using 11C-MET PET can cause significant changes in doses to the brain and the CTV.

UNASSIGNED: Glioblastoma is one of the most common and aggressive primary brain tumours in adults. Though radiation therapy (RT) techniques have progressed significantly in recent decades, patient survival has seen little improvement. However, an area of promise is the use of fluorine-18-fluoroethyltyrosine positron-emission-tomography (18F-FET PET) imaging to assist in RT target delineation. This retrospective study aims to assess the impact of 18F-FET PET scan timing on the resultant RT target volumes and subsequent RT plans in post-operative glioblastoma patients.
UNASSIGNED: The imaging and RT treatment data of eight patients diagnosed with glioblastoma and treated at a single institution were analysed. Before starting RT, each patient had two 18F-FET-PET scans acquired within seven days of each other. The information from these 18F-FET-PET scans aided in the creation of two novel target volume sets. The new volumes and plans were compared with each other and the originals.
UNASSIGNED: The median clinical target volume (CTV) 1 was statistically smaller than CTV 2. The median Dice score for the CTV1/CTV2 was 0.98 and, of the voxels that differ (median 6.5 cc), 99.7% were covered with a 5 mm expansion. Overall organs at risk (OAR) and target dosimetry were similar in the PTV1 and PTV2 plans.
UNASSIGNED: Provided the 18F-FET PET scan is acquired within two weeks of the RT planning and a comprehensive approach is taken to CTV delineation, the timing of scan acquisition has minimal impact on the resulting RT plan.

[This corrects the article DOI: 10.3389/fonc.2022.879167.].