Magnetic resonance imaging (MRI) offers superior soft-tissue contrast and MRI-only radiotherapy (RT) eliminates uncertainties associated with computed tomography (CT)-to-MRI image registration. An MRI-only RT workflow in patients with hip prostheses remains challenging since delineation, synthetic CT (sCT) generation, and marker identification ideally rely on MR images, which are sensitive to distortions. Metal-artifact reduction sequences (MARS) mitigate metal distortions but often obscure gold markers. The aim of this study is to present a proof-of-concept 3 T MRI-only RT workflow for prostate cancer patients with hip prostheses, using only MARS MAVRIC-SL (GE HealthCare) for delineation, sCT generation and fiducial marker identification. Five prostate cancer patients with one or two hip prostheses and intraprostatic fiducial markers underwent CT, standard 3 T MRI and additional MAVRIC-SL MRI. All images were used retrospectively and did not affect patient treatment. Delineation and sCT generation were performed using MAVRIC-SL. Intraprostatic markers were not visible in MAVRIC-SL due to artifact suppression; instead, raw data were reconstructed into spectral-bin images to manually identify the markers. To prevent incoming radiation from passing through the implants and reduce the associated uncertainties in dose calculation and treatment delivery, beam avoidance sectors (AS) were applied. Conventional CT-based avoidance sectors were registered to the sCT and defined as AS1. Prosthesis-induced signal voids in MAVRIC-SL were used to define AS2. Two treatment plans (42.7 Gy, 7 fractions) were created per patient using AS1 and AS2. Plans were compared using dose-volume histogram (DVH) parameters. AS2-plans were recalculated on registered CTs for DVH-comparison and gamma analyses (2

BACKGROUND:Precise radiotherapy relies on accurately targeting tumours while minimising exposure to healthy tissue, yet patient and organ motion complicate treatment delivery. To address intra-fractional motion, multi-leaf collimator (MLC) tracking systems have recently been adopted, adapting beam shapes in real-time. The American Association of Physicists in Medicine (AAPM) Task Group 264 (TG-264) provides guidelines for safely commissioning such tracking systems, yet these guidelines were initially developed for conventional linear accelerators and require evaluation, especially for newer platforms such as Radixact Synchrony. PURPOSE:This study aimed to: (i) evaluate the clinical performance and dosimetric accuracy of the Radixact Synchrony MLC tracking system according to AAPM TG-264 guidelines, from commissioning to clinical implementation; and (ii) critically assess and suggest practical refinements to these guidelines based on experiences with this novel tracking technology. METHODS:Commissioning followed TG-264 recommendations, adapted for Radixact Synchrony, utilizing three tracking modes: fiducial-based, markerless adaptive, and marker-based adaptive tracking. Performance was assessed with multiple test systems, including the Delta4 Phantom+, HexaMotion, Quasar platform, and film dosimetry. Measurements included geometric accuracy of phantom trace tracking, dosimetric accuracy of delivered dose to movable phantom, and system latency. Clinical protocols established treatment planning, quality assurance (QA), safety procedures, and clinical decision pathways, focusing on prostate and lung cancer treatments. RESULTS:The Synchrony system demonstrated substantial improvements in geometric accuracy compared to non-MLC-tracking approaches. Fiducial-based tracking achieved a root mean square error (RMSE) of 0.76 ± 0.27 mm compared to 3.99 ± 2.84 mm without tracking (p = 0.008), with a mean absolute error (MAE) reduction to 0.36 ± 0.12 mm. Markerless adaptive tracking resulted in similar accuracy (RMSE 0.80 ± 0.15 mm, MAE 0.68 ± 0.15 mm). Dosimetric evaluations revealed consistent improvements, with gamma pass rate ≥ 95% (criteria 2%/2 mm) for tracked plans, significantly outperforming static plans under dynamic conditions (V = 7.0, p = .037). System latency was measured one time at approximately 630 ms for fiducial tracking without external breathing monitoring, slightly exceeding TG-264's ideal threshold (500 ms), yet well within the manufacturer's tolerance (1.5 s). Clinical cases confirmed feasibility, showing median deviations of 2.0-3.9 mm for prostate tracking and around 3.3 mm for markerless lung tracking. Safety protocols and clinical pathways developed during implementation ensured treatment robustness. CONCLUSIONS:The Radixact Synchrony MLC tracking system successfully met TG-264 guidelines, significantly improving geometric and dosimetric accuracy during real-time tumour tracking. However, practical implementation highlighted necessary adaptations to TG-264 recommendations for non-standard platforms such as Radixact, specifically regarding QA protocols, latency tolerance, and handling of the system's unique characteristics (pneumatic MLC, jaw tracking, and flattening-filter-free beams). Our findings underscore the importance of maintaining conservative margins initially, rigorous QA, specialized staff training, and careful patient selection strategies. Further clinical trials focusing on safe margin reduction strategies are essential for optimizing the clinical benefits of advanced tracking technologies.

Abstract Background Metastasis-directed stereotactic body radiotherapy (MD-SBRT) has shown promise in retrospective and phase II studies for oligometastatic hormone-sensitive prostate cancer. However, prospective randomized phase III data—particularly in newly diagnosed cases and in combination with androgen deprivation therapy and next-generation androgen receptor pathway inhibitors—are limited. The METRO trial investigates the addition of MD-SBRT to standard of care in patients with prostate-specific membrane antigen (PSMA) PET/CT-detected oligometastatic disease. Methods METRO is a multicentre, double arm, open-label, phase III randomized trial comparing MD-SBRT plus standard of care versus standard of care alone in patients with one to three PSMA PET/CT-detected distant metastases. The PSMA-RADS scale is used to support inclusion, and only patients with PSMA-RADS 4 or 5 lesions in bone or non-regional lymph nodes are eligible. Standard of care includes time-limited androgen deprivation therapy and/or androgen receptor pathway inhibitor, as well as local radiotherapy to the prostate or prostate bed. Patients are stratified by disease type (synchronous or metachronous) and metastasis location (lymph node/bone). The primary endpoint is biochemical progression-free survival; secondary endpoints include time to castration-resistant prostate cancer, adverse events, and health-related quality of life. The intervention is prescribed either 30 Gy in 3 fractions or 40 Gy in 5 fractions and delivered by stereotactic treatment principles. Discussion The METRO trial investigates the added value of combining MD-SBRT with time-limited intensified hormonal therapy in both synchronous and metachronous oligometastatic hormone-sensitive prostate cancer staged by PSMA‑PET/CT. The use of the PSMA-RADS scale for inclusion ensures a standardized and reproducible approach for patient selection. Trial registration ClinicalTrials.gov Identifier: NCT04983095.

BACKGROUND:HYPO-RT-PC is a phase 3 trial comparing ultra-hypofractionated and conventionally fractionated radiotherapy in intermediate-to-high-risk localised prostate cancer. This 10-year update reports long-term efficacy and toxicity outcomes. METHODS:In this open-label, randomised, phase 3, non-inferiority trial done in ten centres in Sweden and two in Denmark, we recruited men aged 75 years or younger with intermediate-risk or high-risk prostate cancer and a WHO performance status between 0 and 2. Previous or current androgen deprivation therapy was not permitted. Patients were randomly assigned (1:1) to ultra-hypofractionated radiotherapy (42·7 Gy in seven fractions, 3 days per week for 2·5 weeks) or conventionally fractionated radiotherapy (78·0 Gy in 39 fractions, 5 days per week for 8 weeks). Randomisation was performed with a minimisation algorithm balancing T stage, Gleason score, prostate-specific antigen, and trial centre. The primary endpoint was failure-free survival, defined as time from randomisation to the first occurrence of biochemical failure, evidence of clinical progression, initiation of androgen deprivation therapy, or death from prostate cancer, analysed in the per-protocol population. The non-inferiority margin was 4% at 5 years and had previously been met, corresponding to a critical hazard ratio (HR) limit of 1·338. Toxicity was assessed using the Radiation Therapy Oncology Group morbidity scale. Here, we report long-term efficacy and safety results at 10 years. The trial is registered with the ISRCTN registry, ISRCTN45905321, and is closed. FINDINGS:Between July 1, 2005, and Nov 4, 2015, 1200 patients were randomly assigned to conventional fractionated radiotherapy (n=602) or ultra-hypofractionated radiotherapy (n=598). Ten patients withdrew consent, eight were found to be ineligible, and two died of reasons unrelated to prostate cancer. 1180 patients constituted the per-protocol population (591 in the conventional fractionation group and 589 in the ultra-hypofractionation group). After a median follow-up of 10·6 years (IQR 9·0-13·0) in the conventional fractionation group and 10·7 years (9·1-12·7) in the ultra-hypofractionation group, 205 and 178 primary events were observed, respectively. 10-year failure-free survival was 65% (95% CI 61-69) in the conventionally fractionated group and 72% (68-76) in the ultra-hypofractionated group. The adjusted HR for the primary endpoint was 0·84 (95% CI 0·69-1·03; Cox regression analysis), confirming non-inferiority. The 10-year cumulative incidence of late grade 2 or worse genitourinary toxic effects was 30% (95% CI 26-34) in the conventional fractionation group and 28% (24-32) in the ultra-hypofractionated group (HR 1·01, 95% CI 0·81-1·25; p=0·95). For late grade 2 or worse gastrointestinal toxic effects, the corresponding figures were 14% (95% CI 11-18) and 14% (11-17; HR 0·94, 95% CI 0·70-1·28; p=0·72). INTERPRETATION:This 10-year follow-up confirms the non-inferiority of the ultra-hypofractionated radiotherapy regimen compared with the conventionally fractionated, with similar toxicity profiles. The findings support the seven-fraction schedule as a safe, effective, and practical standard-of-care option for patients with intermediate-risk prostate cancer. FUNDING:The Nordic Cancer Union, the Swedish Cancer Society, the Swedish Research Council, the Swedish Prostate Cancer Association, and Cancerforskningsfonden i Norrland.

Background and Purpose:Synthetic computed tomography (sCT) enables magnetic resonance imaging (MRI)-only radiotherapy by providing electron density information for dose calculation. While sCT-based planning has been proven sufficiently accurate for convolution-based dose calculation algorithms, linear Boltzmann transport equation (LBTE)-based approaches exhibit higher sensitivity to tissue heterogeneities. This study evaluated the accuracy of both dose calculation algorithm types for MRI-only in prostate and glioma radiotherapy. Materials and methods:Clinical treatment plans for thirty-nine prostate cancer patients and seventeen glioma patients were recalculated on atlas- and deep learning (DL)-based sCTs and conventional CT using convolution- and LBTE-based algorithms. Target dose metrics were analyzed as relative differences, while organs of interest (OOI) dose criteria were evaluated as absolute differences between sCT- and CT-based calculations. Statistical significance was assessed using paired t-tests (α = 0.05). Results:For target volumes, statistically significant differences between sCT- and CT-based calculations were observed for both dose calculation algorithms (p < 0.001). In prostate patients, mean differences in target dose metrics were ≤ 1.4% for the LBTE-based algorithm and ≤ 0.5% for the convolution-based algorithm. Slightly smaller dose differences were observed for glioma patients, ≤ 0.8% and ≤ 0.5%, respectively. OOI dose differences were small (≤ 0.6%). The LBTE-based algorithm consistently yielded the largest dose differences between CT and sCT. Conclusions:Although the LBTE-based approach demonstrated higher sensitivity to sCT-related CT number discrepancies than the convolution-based dose calculation algorithm, the resulting dose differences remained within clinically acceptable limits, supporting the robustness of MRI-only radiotherapy for both atlas- and DL-based sCTs.

BACKGROUND:Fiducial markers in image-guided prostate cancer radiotherapy reduce geometric uncertainty during daily patient setup and enable assessment of target position changes. Diffusion-weighted magnetic resonance imaging (MRI) for target delineation improves prostate cancer localization, beneficial for intraprostatic focal boost. Artifacts from fiducial markers on prostate diffusion-weighted MRI (DWI) need to be investigated, as they could be detrimental for target delineation. This study aims to determine the distances of artifact extensions caused by fiducial markers in DWI and in the apparent diffusion coefficient (ADC) maps and to assess how motion and signal-to-noise ratio (SNR) influence the artifact size in ADC maps. MATERIALS AND METHODS:Three phantoms were used: two homogeneous gel phantoms-one containing three cylindrical gold fiducial markers (GFM) and the other containing three spherical gold anchor (GA) markers-and a third heterogeneous phantom consisting of a piece of sirloin embedded with three GFM and three GA. Diffusion-weighted images were acquired on a 3T MRI system. The artifacts were analyzed along the phase-encoding (PE) and frequency-encoding (FE) directions. Motion was induced and simulated during acquisition, and SNR was varied. The impact of motion and SNR on the artifact extension was evaluated, and the artifact extensions in diffusion images from eight patients were also analyzed. RESULTS:The artifacts were smaller in the ADC maps compared to DWI. The largest artifact extension occurred along the PE-direction. Larger artifact extensions were observed in homogeneous phantom images compared to patient images. In homogeneous phantom images: 13.8  ±  0.4 mm / 9.1  ±  0.4 mm (PE/FE) in DWI with b = 0 s/mm2 and 11.6  ±  0.9 mm / 8.1  ±  0.4 mm (PE/FE) in the ADC map. In patient images: 10.7  ±  1.2 mm / 8.2  ±  1.3 mm (PE/FE) in DWI with b = 0 s/mm2 and 7.3  ±  1.6 mm / 6.8  ±  1.1 mm (PE/FE) in the ADC map. Motion caused larger artifact extensions compared to no motion. A motion of 2 mm increased the artifact from 11.6 ± 0.9 mm / 8.1 ± 0.4 mm (PE/FE) to 14.1 ± 0.8 mm / 9.7 ± 0.4 mm (PE/FE) in homogeneous phantom images and from 10.3 ± 0.8 mm / 8.1 ± 0.4 mm (PE/FE) to 13.1 ± 0.8 mm / 8.4 ± 0.8 mm (PE/FE) in heterogeneous phantom images. Lower SNR resulted in smaller visible artifact extensions. CONCLUSION:This study assessed the distances of artifact extensions in homogeneous phantoms, heterogeneous phantoms, and patient images caused by fiducial markers in DWI and ADC maps. ADC maps had smaller artifact extensions compared to DWI. The artifact extensions were largest in the homogeneous phantom, smaller in the heterogeneous phantom, and the smallest in the patient images. In patient images, the extensions were approximately 7-11 mm (PE) and 7-8 mm (FE). However, extensions reached up to ∼14 mm (PE) and ∼9 mm (FE) in homogeneous phantom images, suggesting that the true artifact extension may be partially obscured in patient images. Further, motion in images caused larger artifact extensions, and lower SNR caused smaller artifact extensions. The study underlines the need for precise marker placement to avoid obscuring critical anatomical structures, especially for delineation of small boost volumes, and distorting ADC values in quantitative analyses of tumors.

Radiotherapy treatment for prostate cancer relies on computed tomography (CT) and/or magnetic resonance imaging (MRI) for segmentation of target volumes and organs at risk (OARs). Manual segmentation of these volumes is regarded as the gold standard for ground truth in machine learning applications, but to acquire such data is tedious and time-consuming. A publicly available clinical dataset is presented, comprising MRI- and synthetic CT (sCT) images, target and OARs segmentations, and radiotherapy dose distributions for 432 prostate cancer patients treated with MRI-guided radiotherapy. An extended dataset with 35 patients is also included, with the addition of deep learning (DL)-generated segmentations, DL segmentation uncertainty maps, and DL segmentations manually adjusted by four radiation oncologists. The publication of these resources aims to aid research in automated radiotherapy treatment planning, segmentation, inter-observer analyses, and DL model uncertainty investigation. The dataset is hosted on the AIDA Data Hub and offers a free-to-use resource for the scientific community, valuable for the advancement of medical imaging and prostate cancer radiotherapy research.

BACKGROUND:Deep learning (DL)-based organ segmentation is increasingly used in radiotherapy. While methods exist to generate voxel-wise uncertainty maps from DL-based auto-segmentation models, these maps are rarely presented to clinicians. PURPOSE:This study aimed to evaluate the impact of DL-generated uncertainty maps on experienced radiation oncologists during the manual correction of DL-based auto-segmentation for prostate radiotherapy. METHODS:Two nnUNet DL models were trained with 10-fold cross-validation on a dataset of 434 patient cases undergoing ultra-hypofractionated MRI-only radiotherapy for prostate cancer. The models performed prostate clinical target volume (CTV) and rectum segmentation. Each cross-validation model was evaluated on an independent test set of 35 patient cases. Segmentation uncertainty was calculated voxel-wise as the SoftMax standard deviation (0-0.5, n = 10) and visualized as a fixed scale color-coded map. Four experienced oncologists were asked to: Step 1: Rate the quality of and confidence in the DL segmentations using a four- and five-point Likert scale, respectively, and edit the segmentations without access to the uncertainty map. Step 2: Repeat step 1 after at least 4 weeks, but this time with the color-coded uncertainty map available. Oncologists were asked to blend the uncertainty map with the DL segmentation and MRI volume. Segmentation edit time was recorded for both steps. In step 2, oncologists also provided free-text feedback on the benefits and drawbacks of using the uncertainty map during segmentation. A histogram analysis was performed to compare the number of voxels edited between step 1 and step 2 for different uncertainty levels (bins with 0.1 intervals). RESULTS:The DL models achieved high-quality segmentations with a mean Dice coefficient per oncologist of 0.97-0.99, calculated between edited and unedited segmentation in step 1 for the prostate CTV and rectum. While the overall quality rating for rectum segmentations decreased slightly on a group level in step 2 compared to step 1, individual responses varied. Some oncologists rated the quality higher for the prostate CTV segmentation with the uncertainty map present, while others rated it lower. Similarly, confidence ratings varied across oncologists for prostate CTV and rectum. Decreased segmentation time was recorded for three oncologists using uncertainty maps, saving 1-2 min per patient case, corresponding to 14%-33% time reduction. Three oncologists found the uncertainty maps helpful, and one reported benefit was the ability to identify regions of interest more quickly. The histogram analysis had fewer voxel edits in regions of low uncertainty in step 2 compared to step 1. Specifically, 50% fewer voxel edits were recorded for the uncertainty region 0.0-0.1, suggesting increased trust in the DL model's prediction in these areas. CONCLUSIONS:Presenting DL uncertainty information to experienced radiation oncologists influences their decision-making, quality perception, and confidence in the DL segmentations. Regions with low uncertainty were less likely to be edited, indicating increased reliance on the model's predictions. Additionally, uncertainty maps can improve efficiency by reducing segmentation time. DL-based segmentation uncertainty can be a valuable tool in clinical practice, enhancing the efficiency of radiotherapy planning.

PURPOSE:Neglecting imaging gradients in b-value calculations has been a documented issue for decades and remains unaccounted for in the current postprocessing pipelines. This omission may introduce inaccuracies that propagate into diffusion parameter estimates, such as in ADC and DTI analysis. Because intravoxel incoherent motion (IVIM) makes use of low b-values, these inaccuracies may be of greater importance. This study examines the impact of biased b-values on IVIM analysis in simulations and in vivo. METHODS:In simulations, b-values were calculated for two pulsed gradient spin-echo sequence designs: one with large cross-terms between imaging and diffusion gradients, and one with minimal cross-terms. Biased and unbiased b-values were calculated from sequences with 200 diffusion directions. These b-values were used to generate IVIM signal curves for parameter estimation. Simulations were repeated with varying in-plane resolutions (1-4 mm) and slice thicknesses (2-10 mm). Additionally, 15 prostate exams were analyzed with scanner-provided b-values and actual b-values derived from the gradient waveforms of the full pulse sequence. RESULTS:The magnitude and direction of errors in IVIM parameters depended on pulse sequence design. Errors persisted until the full contribution of imaging gradients was considered. Errors in the in vivo data were coherent with the simulations, showing errors of -0.7% in f, 0.8 μm2/ms in D*, and 0.07 μm2/ms in D. CONCLUSION:Ignoring imaging gradients in b-value calculations introduces unnecessary inaccuracies, making IVIM results spurious and highly dependent on specific pulse sequence design and imaging parameters. These inaccuracies can be corrected by adjusting the b-value calculations, without additional measurements.

Background and purpose: The study aims to evaluate dosimetric properties of hypofractionated treatment plans integrating focal boost, using registered whole-mount histopathology (WMHP) as reference standard. Methods: Fifteen men from the PAMP trial (EudraCT: 2015-005046-55) were included. Participants had ≥ 1 ISUP Grade group ≥ 4 lesion and underwent [68Ga]prostate-specific membrane antigen (PSMA) positron emission tomography/multiparametric magnetic resonance imaging (PET/mpMRI) and [11C]Acetate-PET/computed tomography before radical prostatectomy. Four radiation oncologists delineated gross tumor volumes (GTVs) on PSMA-PET/mpMRI. Sixty treatment plans were optimized, one per GTV and patient. Prostate planning target volumes were prescribed 42.7 Gy in seven fractions, with a simultaneous GTV boost up to 49.0 Gy, prioritizing organs at risk (OARs). Digital WMHP provided Gleason grading and was co-registered with in-vivo imaging. Target coverage for GTVs and voxels sharing Gleason patterns (GPs) was assessed via dose-volume histogram (DVH) analysis. Interobserver agreement in GTV-delineations was quantified with Fleiss’ kappa. Results: The median GTV dose per plan (D50) ranged from 48.3 to 49.1 Gy. For voxels with the highest GP, D50 was 42.9–49.2 Gy, exceeding 47.2 Gy in all except one plan. In lowest pattern voxels, D50 was 42.5–49.3 Gy, and below 43.4 Gy in over half the plans. Significant positive correlations between Fleiss’ kappa and DVH parameters appeared only for GP 5 regions, specifically for Fleiss’ kappa and D50 for two observers and the average D50 across observers. Interpretation: The histologically confirmed tumor was only partially boosted. Regions with more aggressive disease received better coverage. These findings provide a rational for prioritizing OARs in treatment planning.

BACKGROUND AND PURPOSE:HYPO-RT-PC was the first phase 3 trial showing the long-term efficacy and safety of extremely hypofractionated radiotherapy schedules (ultra-hypofractionation or stereotactic body radiotherapy) compared with standard radiotherapy in localised prostate cancer. We aimed to verify their safety in the context of a clinical trial, by comparing events of potentially radiotherapy-related morbidity according to routinely collected health data between the arms and with men unexposed to prostate radiotherapy. MATERIALS AND METHODS:We included Swedish residents in the per-protocol population and unexposed men matched with the ultra-hypofractionation group. Primary outcomes were genitourinary and gastrointestinal events defined by linking the Common Terminology Criteria for Adverse Events version 5.0 to diagnosis and intervention codes in broad (any related condition) and narrow code sets (strongly related conditions) tested separately. Inpatient care, interventions, and causes of death, identified in Swedish national healthcare registers, were events. Outcomes were analysed as time-to-event, adjusted for Charlson Comorbidity Index and socioeconomic status. RESULTS:Five hundred and forty-one participants (92%) from each trial arm and 2705 unexposed men were included. We found no statistically significant differences between the radiotherapy groups (broad code sets: genitourinary events, adjusted hazard ratio 0.95, 95% confidence interval 0.71-1.25; gastrointestinal events, 1.24, 0.91-1.69). Compared with the unexposed men, 10-year excess cumulative incidence was 8% and 5% for genitourinary and 5% and 4% for gastrointestinal events when applying the broad and narrow code sets, respectively, without excess mortality events. CONCLUSION:The results support the HYPO-RT-PC findings by showing similar morbidity between ultra-hypofractionated and standard radiotherapy. Excess events were estimated to 4-8%, without increased associated mortality.