Background and purpose:One of the main challenges in stereotactic planning is achieving a steep dose gradient to spare nearby structures. This study aimed to develop and validate a knowledge-based (KB) model for automated planning of single brain lesions using a robotic stereotactic system, achieving plan quality comparable or superior to manual plans. Materials and methods:Sixty retrospective plans were used to train the model. A relationship was established between the Planning Target Volume (PTV) radius and the effective radii of multiple isodose volumes. These regressions were used to generate patient-specific dose shell structures and automated optimization templates. Model performance was assessed through internal (15 cases) and external (13 cases) validation. Conformity Index (CI), Dose Gradient Index (DGI), healthy brain dose, and delivery parameters were compared between KB-generated and clinical plans. Results:The predictive accuracy of the model was ≥0.98 for all isodose levels. For both validation cohorts, KB plans achieved a significantly steeper dose fall-off. A median DGI of 87.3 vs 81.5 (p = 0.001) and of 88.5 vs 79.2 (p < 0.0001) was found, respectively. A significantly (p = 0.01) lower CI was found for external validation and a similar CI (p = 0.08) for the internal cohort. Automatic plans reduced irradiated healthy brain volume, particularly at 50% and 30% isodose levels. Significantly fewer beams were obtained: -58 (p = 0.004) and - 34 (p = 0.03), respectively for both cohorts. Conclusions:The proposed KB model enables automated stereotactic planning with high quality, efficiency, and standardization.

Radiotherapy (RT) is a standard curative treatment for prostate cancer (PCa) and there is growing evidence of the high efficacy of moderate and ultra-hypofractionated RT. Reducing treatment duration to one week or less is a major advance, but very few studies have explored single-fraction therapy. This study evaluates the feasibility, safety, and efficacy of single-fraction stereotactic body RT (SBRT) while delivering the entire procedure in one day, with a potentially high benefit in terms of patient comfort and therapy cost and logistics. This prospective, non-randomized monocentric trial uses Robotic Radiosurgery (CyberKnife v.7 system) to deliver a single 24 Gy fraction to the prostate (± seminal vesicles) with a “urethral sparing HDR-like” technique, and target tracking. The first phase will enroll 13 PCa patients following Simon’s optimal design. Treatment is to be stopped if ≥ 2 patients develop ≥ G3 toxicity (CTCAE v5.0) within a month from RT end; otherwise, 52 more patients will be added, totaling 65. To account for minimal drop-out, 5 extra patients will be enrolled, reaching 70. All procedures are performed in a single day, including fiducial implantation, imaging acquisition, contouring, planning, dosimetry quality control, and treatment. Apart from treatment feasibility in terms of one-month acute toxicity, secondary endpoints include late toxicity, biochemical and clinical control. Few others have investigated the 24 Gy single-fraction schedule using different delivery modalities (not including tracking), which has proved to be non-inferior to 5 fraction SBRT. Our approach aims to maintain (and possibly improve) the previously reported acute, subacute and late toxicity as well as disease control, adding evidence in favor of single-fraction delivery. Another significant goal of the study is the demonstration that all the complex treatment procedures can be safely delivered in a single day. This would be especially appealing for patients far from radiotherapy centers and those with work commitments not allowing daily hospital visits. The study of response to RT can also provide useful information about PCa radiobiology. Planned additional analyses may help in better assessing the clinical value of PSMA PET/CT in the selection of high-risk patients with true limited disease, and in identifying radiomic features associated to outcome. Trial registration: The study was prospectively registered at clinicaltrials.gov (NCT 05936736).

Background and purpose:In order to optimize the radiotherapy treatment and minimize toxicities, organs-at-risk (OARs) and clinical target volume (CTV) must be segmented. Deep Learning (DL) techniques show significant potential for performing this task effectively. The availability of a large single-institute data sample, combined with additional numerous multi-centric data, makes it possible to develop and validate a reliable CTV segmentation model. Materials and methods:Planning CT data of 1822 patients were available (861 from a single center for training and 961 from 8 centers for validation). A preprocessing step, aimed at standardizing all the images, followed by a 3D-Unet capable of segmenting both right and left CTVs was implemented. The metrics used to evaluate the performance were the Dice similarity coefficient (DSC), the Hausdorff distance (HD), and its 95th percentile variant (HD_95) and the Average Surface Distance (ASD). Results:The segmentation model achieved high performance on the validation set (DSC: 0.90; HD: 20.5 mm; HD_95: 10.0 mm; ASD: 2.1 mm; epoch 298). Furthermore, the model predicted smoother contours than the clinical ones along the cranial-caudal axis in both directions. When applied to internal and external data the same metrics demonstrated an overall agreement and model transferability for all but one (Inst 9) center. Conclusion:. A 3D-Unet for CTV segmentation trained on a large single institute cohort consisting of planning CTs and manual segmentations was built and externally validated, reaching high performance.

Background: Knowledge-based (KB) planning is a promising approach to model prior planning experience and optimize radiotherapy. To enable the sharing of models across institutions, their transferability must be evaluated. This study aimed to validate KB prediction models developed by a national consortium using data from another multi-institutional consortium in a different country. Methods: Ten right whole breast tangential field (RWB-TF) models were built within the national consortium. A cohort of 20 patients from the external consortium was used for testing. Transferability was defined when the ipsilateral (IPSI) lung first principal component (PC1) was within the 10th–90th percentile of the training set. Predicted dose–volume parameters were compared with clinical dose–volume histograms (cDVHs). Results: Planning target volume (PTV) coverage strategies were comparable between the two consortia, even though significant volume differences were observed for the PTV and contralateral breast (p = 0.002 and p = 0.02, respectively). For the IPSI lung, the standard deviation of predicted mean dose/V20 Gy was 1.13 Gy/2.9% in the external consortium versus 0.55 Gy/1.6% in the training consortium. Differences between the cDVH and the predicted IPSI lung mean dose and the volume receiving more than 20 Gy (V20 Gy) were <2 Gy and <5% in 88.7% and 92.3% of cases, respectively. PC1 values fell within the 10th–90th percentile for ≥90% of patients in 6/10 models and 65–85% for the remaining 4. Conclusions: This study demonstrates the feasibility of applying RWB-TF KB models beyond the consortium in which they were developed, supporting broader clinical implementation. This retrospective study was supported by AIRC (Associazione Italiana per la Ricerca sul Cancro) and registered on ClinicalTrials.gov (NCT06317948, 12 March 2024).

PURPOSE:To train and validate KB prediction models by merging a large multi-institutional cohort of whole breast irradiation (WBI) plans using tangential fields. METHODS:Ten institutions (INST1-INST10, 1481 patients) developed their KB-institutional models for left/right WBI (ten models for right and eight models for left). The transferability of models among centers was assessed based on the overlap of the geometric Principal Component (PC1) of each model when applied to other institutions and/or on the presence of significantly different optimization policies. Centers corresponding to transferable models were asked to join the building of two KB-benchmark models for right/left breast. Dose-volume histogram (DVH) prediction bands (lung/heart) were compared against those of the KB-institutional models. RESULTS:All models were transferable except INST6 (right breast) and INST1 (left breast). Planning data from 6 institutions for right breast and 5 institutions for left breast (out of 9 and 7 institutions with transferable models, respectively) were combined, totaling data from 850 patients. Prediction bands on the test cohorts (n = 30/25 right/left) showed a large overlap with bands of each institution model: for the right-breast, the KB-benchmark model predicts slightly lower lung Dmean when compared to KB-institution models, except for INST7. Regarding the left-breast, even greater similarity between KB-benchmark and KB-institution model predictions was found. CONCLUSIONS:Multi-institutional KB-benchmark models for WBI were successfully generated. They may be employed by other users, representing the performances reached in a multi-institutional context of experienced centers. KB-benchmark models can also have significant applications for large-scale automatic plan optimization, QA/audit and tutoring/education purposes.

Purpose: This study analyzed inter-institute conformity and dose gradient variability of CyberKnife (CK) brain SRS/SRT plans. The feasibility of multi-center predictive models was investigated, aiming at guided/automated planning optimization. Methods: Data from 335 clinical plans, delivered for single lesions in 1-5 fractions, were collected by 8 CK centers. Conformity index (CI), Dose Gradient Index (DGI) and the effective radii defined by different isodose volumes (Reff) were computed. Predictability of dose fall-off from PTV dimensions was analyzed. DGI average, 80th and 10thpercentile values were evaluated stratifying plans by PTV size into six groups. Linear regression models were created for Reff as a function of PTV equivalent radius. Results: CI values (range 0.96---2.23) exceeded 1.20 in 88/335 plans, mostly (65 %) collected in 2 of the participating centers. DGI showed an acceptable inter- institute variability and a strong significant correlation (p < 0.0001) with PTV. Ideal and Minimal DGI for each of the six groups were respectively 95 (86), 82 (73), 77 (68), 71 (60), 59 (43) and 50 (29). The rate of DGI values passing the multicenter minimal criteria, considering each center separately, varied from 43 % to 100 %. R2 values for the regression between Reff and PTV radius were >= 0.958, showing an increasing inter-center variability for decreasing isodose values. Conclusion: Observed inter-center differences enhanced the advantages of a multi-institute approach. Multicenter predictive models for dose fall-off in CK brain SR/ SRT planning are feasible and easy to use. Reff models and DGI analysis may permit to partially automate planning optimization avoiding creation of suboptimal plans.

Purpose: 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.