Search Results (56)

Background: By integrating an MRI scanner and a linear accelerator, MR-linac systems provide superior soft tissue imaging and allow to perform adaptive radiotherapy adjusted on daily anatomical changes. The advent of this technology represents a revolution in radiation oncology and could improve treatment accuracy and clinical outcomes. We performed a comprehensive bibliometric analysis with the aim of displaying the available scientific literature and trends regarding MR-linac. Methods: Scopus database was investigated, considering documents published up to 6 April 2025. Keywords encompassed terms related to “MR-linac” or “MRI-linac” and possible combinations and acronyms. BibTeX data file was imported into Biblioshiny (Bibliometrix package—v. 4.1.4) and analysis was conducted using R code (R version 4.3.2) and the Bibliometrix package (version 4.1.4). Results: A total of 1624 articles on MR-linac were identified. The number of annual publications gradually increased from 21 in 2008, peaking at 211 in 2022 and then remaining substantially stable in subsequent years. Most of the papers were original articles (79.2%) and the majority was published by the 10 journals with the largest output. Remarkably, of 6385 identified authors, over 85% were from one of the 10 most represented countries (including European, North American and Asian nations). Consistently, the 10 institutions with the larger output were North American, Australian or European and provided over 60% of the articles. International co-authorship was found in only 23.6% of the articles. Keyword and co-occurrence analyses identified MR-guided radiotherapy, SBRT, dosimetry, and adaptive strategies as core themes, with emerging trends in radiomics, diffusion metrics, and deep learning. Conclusions: Bibliometric analysis identified trends and patterns of scientific publications regarding MR-linac, highlighting a growing interest in the topic. Nonetheless, it should be considered that the majority of the papers were published by a few journals and over 85% of authors were from 10 countries, demonstrating an evident disparity across nations. Multicentric international research protocols and common frameworks could foster the transition towards collaborative practice-changing studies. Full article

Objective: Inadequate radiation delivery to recurrent pelvic and abdominal tumors is frequently attributable to the dose limitations of surrounding normal structures, particularly the intestines. Radiotherapy guided by magnetic resonance imaging (MRI) significantly enhances the accuracy of soft-tissue delineation. The purposes of this study were to demonstrate the feasibility and effectiveness of MR-Linac Adaptive stereotactic body radiotherapy in patients with pelvic–abdominal recurrent or metastatic gynecological malignancies with or without systemic therapies. Methods: Patients with pelvic–abdominal recurrent or metastatic gynecological malignancies are eligible for MR-Linac Adaptive stereotactic body radiotherapy. Systemic therapies, including chemotherapy, immunotherapy, and targeted therapy, are considered acceptable treatment options. The safety, tolerability, and efficacy of MR-Linac Adaptive stereotactic body radiotherapy were assessed. Results: Between October 2019 and May 2025, 15 patients were subjected to MR-Linac Adaptive stereotactic body radiotherapy. With a median follow-up period of 4.67 months (range, 0.73–20.10 months), the 6-month overall survival (OS), progression-free survival (PFS), and local control (LC) rates were 93.3%, 66.0%, and 92.3%, respectively. The 12-month OS, PFS, and LC rates were 83.8%, 37.7%, and 70.5%, respectively. The best objective response rate (ORR = CR + PR) for the irradiated lesions was 73.3% (11/15 patients). MR-Linac Adaptive stereotactic body radiotherapy led to objective responses in 73.3% (11/15) of the patients. As of the data cutoff (28 May 2025), one patient experienced dose-limiting toxicity (an enteric fistula). Another patient developed grade 4 thrombocytopenia during treatment; it was considered chemotherapy-induced. Conclusions: These findings suggest that MR-Linac Adaptive stereotactic body radiotherapy is relatively effective and safe and can be an important treatment option for patients with pelvic–abdominal recurrent or metastatic gynecological malignancies. MR-Linac Adaptive stereotactic body radiotherapy exhibited acceptable tolerability, promising efficacy, and a favorable local control rate with regard to heavily pretreated advanced solid tumors. Full article

Background: Stereotactic body radiotherapy (SBRT) for prostate cancer delivers high doses in few fractions but poses challenges in sparing adjacent organs at risk (OARs), particularly the intra-prostate urethra, bladder, rectum and penile bulb. Magnetic resonance-guided radiotherapy (MRgRT) using MR-Linac offers superior soft-tissue visualization and daily adaptive planning, potentially reducing OAR dose while maintaining target coverage. This study aimed to compare dose–volume parameters among MR-Linac (ML), volumetric modulated arc therapy (VMAT), and Tomotherapy (HT) plans for prostate SBRT. Methods: Thirty patients with localized prostate cancer were retrospectively analyzed. For each patient, three plans were generated: ML, VMAT and HT, using identical prescription and planning objectives. Dose–volume histogram (DVH) metrics were evaluated for clinical target volume (CTV), planning target volume (PTV), and OARs. Statistical comparisons were performed using non-parametric Friedman’s Test with post hoc Bonferroni test, with significance set at a p < 0.05. Results: CTV coverage was comparable across all modalities. ML achieved significantly higher PTV Dmin and near-maximum doses compared to VMAT and HT. Notably, ML provided substantial urethral sparing, reducing Dmax and Dmean by approximately 3.3 Gy compared to both VMAT and HT (p < 0.001). Rectal dose metrics were also lower with ML, while bladder and penile bulb doses showed minor increases (<3.5 Gy), considered clinically negligible. Femoral head doses were reduced in ML plans. Conclusions: MR-Linac planning for prostate SBRT offers meaningful dosimetric advantages, particularly in intra-prostate urethra urethral dose reduction, without compromising target coverage. These findings support incorporating MR-guided adaptive workflows into SBRT protocols to enhance OAR protection and potentially reduce treatment-related toxicity. Full article

Oesophageal cancer remains a significant global health burden with poor survival outcomes despite multimodal treatment. Recent advances in magnetic resonance imaging (MRI) have opened opportunities to improve radiotherapy delivery. This review examines the role of MRI and MR-guided radiotherapy (MRgRT) in oesophageal cancer, focusing on applications in staging, treatment planning, and response assessment, with particular emphasis on magnetic resonance linear accelerator (MR-Linac)-based delivery. Compared to computed tomography (CT), MRI offers superior soft-tissue contrast, enabling more accurate tumour delineation and the potential for reduced treatment margins. Real-time MR imaging during treatment can facilitate motion management, while daily adaptive planning can accommodate anatomical changes throughout the treatment course. Functional MRI sequences, including diffusion-weighted and dynamic contrast-enhanced imaging, offer quantitative data for treatment response monitoring. Early clinical and dosimetric studies demonstrate that MRgRT can significantly reduce radiation dose to critical organs while maintaining target coverage. However, clinical evidence for MRgRT in oesophageal cancer is limited to small early-phase studies, with no phase II/III trials demonstrating improvements in survival, toxicity, or patient-reported outcomes. Long-term clinical benefits and cost-effectiveness remain unproven, highlighting the need for prospective outcome-focused studies to define the role for MRgRT within multimodality treatment pathways. Full article

Purpose or Objective: To evaluate the feasibility and clinical utility of integrating sequential PSMA-PET imaging into an offline–online adaptive workflow for response-based dominant intraprostatic lesion (DIL)-boosting high-risk prostate cancer treated with stereotactic ablative radiotherapy (SABR). Materials and Methods: As part of a prospective trial, patients were treated on MR- or CBCT-guided adaptive radiotherapy (ART) systems with prostate/pelvic node 5-fraction SABR (36.25 Gy/25 Gy) with DIL boost (50 Gy). Whereas traditional DIL boost volumes delineate full pre-therapy imaging-defined disease (GTVinitial), this study serially refined DIL boost volumes based on treatment response defined by PSMA-PET scans after neoadjuvant androgen deprivation therapy (nADT, GTVmb1) and fraction 3 SABR (GTVmb2). DIL delineation employed PET-PSMA fusion to CT/MR simulation and was guided by a rule-based %SUVmax threshold approach. Comparisons of GTV volumes and OAR dosimetry were performed between plans using GTVinitial versus GTVmb1/GTVmb2 for DIL boost, for each of the initial cohorts of five patients from the initially treated cohorts. Results: Five patients treated on MR-Linac (n = 3) or CBCT-based ART (n = 2) were analyzed. Three patients exhibited complete imaging response after nADT, omitting GTVmb boosts. Offline GTVmb refinements based on PSMA-PET were seamlessly integrated into ART workflows without introducing additional treatment time. DIL GTV volumes significantly decreased (p = 0.03) from an initial mean of 11.4 cc (GTVinitial) to 4.1 cc (GTVmb1) and 3.0 cc (GTVmb2). Dosimetric analysis showed meaningful reductions in OAR doses: rectal wall D0.035 cc decreased by up to 12 Gy, while bladder wall D0.035 cc and V18.3 Gy reduced from 52.3 Gy and 52.3 cc (Plan_initial) to 42.9 Gy and 24.9 cc (Plan_mb2), respectively. Urethra doses remained stable, with minor reductions. Sigmoid and femoral head doses remained within acceptable limits. Online adaptation efficiently addressed daily anatomical variations, enabling simulation-free plan re-optimization. Conclusion: PSMA-PET-guided adaptive microboosting for HRPCa SABR is feasible and effective. Standard MR-Linac and CBCT systems offer practical alternatives to BgRT platforms, enabling biology-driven dose personalization and potentially reducing toxicity. Full article

Background/Objectives: MR-guided radiation therapy (MRgRT) has rapidly evolved into an important treatment modality, with the Elekta Unity and ViewRay MRIdian systems being two major MR-linac platforms. Despite the shared concept of MRgRT, the two platforms elected different system designs that could potentially impact the dosimetric characteristics and quality of a treatment. In this study, we aim to perform a comparative dosimetric investigation between these two MR-linac systems in prostate and pancreas cancers. Methods: Dosimetric characteristics were evaluated by retrospectively re-creating 20 clinical prostate and pancreas cases originally treated on MRIdian using the Unity system, adhering to MIRAGE and SMART clinical trial constraints. Treatment plans were re-created with matching planning images, structures, beam geometry, and dose parameters. To ensure comparison consistency, all Unity treatment plans were normalized to match the target coverage of the MRIdian counterparts, and the organ-at-risk (OAR) dose was investigated. Results: Most OARs’ dose-volume metrics showed no statistically significant differences. For prostate patients, Unity demonstrated lower rectum V36Gy (p = 0.0095), V38Gy (p = 0.0043), V40Gy (p = 0.0469), and lower left (p = 0.0137) and right femur V20Gy (p = 0.0020). For pancreas patients, Unity plans had a lower mean liver dose (p = 0.0371). All Unity plans had a Gamma passing rate > 90%, confirming the clinical deliverability. Mean delivery times were 12.78 ± 1.68 and 13.53 ± 1.88 min for MRIdian and Unity prostate plans, respectively, and 14.58 ± 2.78 and 17.40 ± 3.77 min for MRIdian and Unity pancreas plans, respectively. Conclusions: Overall, comparable treatment quality and delivery times were observed between the two platforms. Full article

Background: Stereotactic body radiation therapy (SBRT) has proven effective in controlling spinal lesions with minimal toxicity, primarily due to its ability to limit spinal cord dose. Recent advances in MR-linac (MRL) technology offer superior spinal cord visualization and real-time gating, which can facilitate dose escalation in spinal tumor treatment while maintaining safety. Purpose: This study aimed to optimize motion management for spine SBRT on an MRL by analyzing patient-specific motion dynamics and evaluating the most effective registration structures. We hypothesized that baseline shifts (BLS) would improve delivery efficiency while maintaining spinal cord dose constraints. The goal was to establish displacement thresholds and assess the role of baseline shift correction adaptative planning in improving treatment delivery efficiency. Methods: Twelve patients underwent two MRI sessions on the MRL. The optimal registration structure was identified, and intrafraction motion was assessed to calculate delivery efficiency. Baseline shift (BLS) simulations were applied for five cases that showed significant motion and suboptimal delivery efficiency, and the dosimetric impact of the BLS was evaluated. The simulated BLS-based plan adaptation was implemented via a segment aperture morphing adapt-to-position workflow. Results: The most stable registration structure was the spinal canal plus three adjacent vertebrae. Cine imaging revealed average intrafraction motion (95th to 5th percentiles) of 0.8 ± 0.5 mm in the right-left (RL) direction, 0.9 ± 0.6 mm in the anterior–posterior (AP) direction, and 0.7 ± 0.5 mm in the SI direction. Simulated BLS improved delivery efficiency to >80% in all but one case, with a ±1 mm displacement threshold tolerance. While target coverage remained consistent after BLS simulation, the spinal cord dose increased by 7–60%, exceeding the 14 Gy constraint in three of the five simulated cases. Conclusions: Cine imaging and BLS can enhance delivery efficiency in spine SBRT but may increase spinal cord dose. These findings underscore the need for careful patient selection, advanced motion management, and patient-specific BLS protocols. Full article

Background/Objectives: Non-ablative stereotactic body radiation therapy (SBRT) is commonly employed for locally advanced pancreatic cancer (LAPC) using computed tomography-guided radiotherapy (CTgRT) without online adaptive radiation therapy (oART). The safe delivery of ablative SBRT has been demonstrated using stereotactic magnetic resonance-guided online adaptive radiation therapy (SMART). We performed an in silico comparison of non-adapted CTgRT versus SMART to better understand the potential benefit of oART for ablative pancreatic SBRT. Methods: We retrospectively evaluated original and daily adapted SMART plans that were previously delivered for 20 consecutive LAPC cases (120 total plans across all patients) treated on a 0.35 T MR-linac prescribed to 50 Gy (gross disease) and 33 Gy (elective sites) simultaneously in five fractions. Six comparative CTgRT plans for each patient (one original, five daily treatment) were retrospectively generated with the same prescribed dose and planning parameters as the SMART plans assuming no oART availability. The impact of daily anatomic changes on CTgRT and SMART plans without oART was evaluated across each treatment day MRI scan acquired for SMART. Results: Ninety percent of cases involved the pancreatic head. No statistically significant differences were seen between CTgRT and SMART with respect to target coverage. Nearly all (96%) fractions planned on either CT or MRI platforms exceeded at least one GI organ at risk (OAR) constraint without oART. Significant differences favoring SMART over non-adaptive CTgRT were observed for the duodenum V35 Gy ≤ 0.5 cc (34.2 vs. 41.9 Gy, p = 0.0035) and duodenum V40 Gy ≤ 0.03 cc (37 vs. 52.5 Gy, p = 0.0006) constraints. Stomach V40 Gy trended towards significance favoring SMART (37 vs. 40.3 Gy, p = 0.057) while no significant differences were seen. Conclusions: This is the first study that quantifies the frequency and extent of GI OAR constraint violations that would occur during ablative five-fraction SBRT using SMART vs. CTgRT. GI OAR constraint violations are expected for most fractions without oART whereas all constraints can be achieved with oART. As such, these data suggest that oART should be required for ablative five-fraction pancreatic SBRT. Full article

Purpose/Objective: The clinical implementation of MR-guided radiotherapy on MR-linacs (MRL) hasrapidly increased in recent years. The advantages represented by the MR-based daily online plan adaptation and real-time monitoring have been exploited for different tumor sites. Nevertheless, some concerns remain, mainly related to the longer treatment time and limited patient eligibility. We report here the experience of our center, where a 1.5T MRL was clinically implemented in 2019 and, since then, more than 1200 patients have been treated. Material and Methods: The first 1000 patients treated at the MRL in our department were selected. Technical information such as treatment time and adaptive technic have been prospectively recorded, while toxicity data were retrospectively collected. Results: Between October 2019 and June 2024, 1000 patients for a total of 1061 treatment courses were included. Prostate and prostate bed were irradiated in 57.1% and 10.2% of the cases, respectively, including regional pelvic lymphnodes in 4.7%. Other frequent treated sites were lymph node metastases, pancreas and liver. The most frequent prescribed doses were 36.25 Gy (31%), 35 Gy (28.3%) and 30 Gy (9.4%) in five fractions. On a total of 9076 administered fractions, 80.8% were performed with adapt-to-shape and 19.2% with adapt-to-position method. The mean in-room time was 38 min (range, 18–103), with 74.4% of patients completing the session within 40 min. Acute grade (G) 3 toxicity was recorded in 1.6% of the cases, while, on a total of 858 patients available for late toxicity, G3 was recorded in 0.3% of the cases, with no >G3. Conclusions: Our real-world experience of an early-adopter center confirms that MRL treatments are feasible for different tumor entities in several anatomical sites. We showed that most of the patients could be treated within 40 min and showed low toxicity rates. Protocols for dose escalation and margin reduction, by adopting new comprehensive motion monitoring strategies, are under development. Full article

Radiotherapy (RT) has undergone transformative advancements since its inception over a century ago. This review highlights the most promising and impactful innovations shaping the current and future landscape of RT. Key technological advances include adaptive radiotherapy (ART), which tailors treatment to daily anatomical changes using integrated imaging and artificial intelligence (AI), and advanced image guidance systems, such as MR-LINACs, PET-LINACs, and surface-guided radiotherapy (SGRT), which enhance targeting precision and minimize collateral damage. AI and data science further support RT through automation, improved segmentation, dose prediction, and treatment planning. Emerging biological and targeted therapies, including boron neutron capture therapy (BNCT), radioimmunotherapy, and theranostics, represent the convergence of molecular targeting and radiotherapy, offering personalized treatment strategies. Particle therapies, notably proton and heavy ion RT, exploit the Bragg peak for precise tumor targeting while reducing normal tissue exposure. FLASH RT, delivering ultra-high dose rates, demonstrates promise in sparing normal tissue while maintaining tumor control, though clinical validation is ongoing. Spatially fractionated RT (SFRT), stereotactic techniques and brachytherapy are evolving to treat challenging tumor types with enhanced conformality and efficacy. Innovations such as 3D printing, Auger therapy, and hyperthermia are also contributing to individualized and site-specific solutions. Across these modalities, the integration of imaging, AI, and novel physics and biology-driven approaches is redefining the possibilities of cancer treatment. This review underscores the multidisciplinary and translational nature of modern RT, where physics, engineering, biology, and informatics intersect to improve patient outcomes. While many approaches are in various stages of clinical adoption and investigation, their collective impact promises to redefine the therapeutic boundaries of radiation oncology in the coming decade. Full article

Background/Objectives: Radiotherapy for tumors of the central nervous system (CNS) could be improved by incorporating advanced imaging techniques into treatment planning and response assessment. The objective of this narrative review is to highlight the recent developments in magnetic resonance imaging (MRI) and positron emission tomography (PET) for applications in CNS radiotherapy. Methods: Recent articles were selected for discussion, covering the following topics: advanced imaging on MRI-linear accelerators for early response assessment in glioma; PET for guiding treatment planning and response assessment in glioma; and contrast-enhanced imaging and metabolic imaging for differentiating tumor progression and radiation necrosis for brain metastasis treatment. Where necessary, searches of scholarly databases (e.g., Google Scholar, PubMed) were used to find papers for each topic. The topics were chosen based on the perception of promise in advancing specific applications of CNS radiotherapy and not covered in detail elsewhere. This review is not intended to be comprehensive. Results: Advanced MRI sequences and PET could have a substantial impact on CNS radiotherapy. For gliomas, the tumor response to therapy could be assessed much earlier than using the conventional technique of measuring changes in tumor size. Using advanced imaging on combined imaging/therapy devices like MR-Linacs would enable response monitoring throughout radiotherapy. For brain metastases, radiation necrosis and tumor progression might be reliably differentiated with imaging techniques sensitive to perfusion or metabolism. However, the lack of level 1 evidence supporting specific uses for each imaging technique is an impediment to widespread use. Conclusions: Advanced MRI and PET have great promise to change the standard of care for CNS radiotherapy, but clinical trials validating specific applications are needed. Full article

Purpose: Image-guided radiotherapy (IGRT) has been widely implemented in the treatment of prostate cancer, offering a number of advantages regarding the precision of dose delivery. This study provides an overview of factors, clinical and physical alike, that increase treatment accuracy in prostate cancer radiotherapy in the context of IGRT. The following aspects are explored based on recent literature: the radiotherapy technique used in conjunction with IGRT, the type and frequency of IGRT, the impact of radiotherapy technique/IGRT on target dosimetry and organs at risk, the influence of IGRT on planning target volume margins, the impact of treatment time on dosimetric outcome and clinical outcomes using IGRT repositioning or an online adaptive plan. Methods: A systematic search of the literature was conducted within Pubmed/Medline databases to find relevant studies. Of the 152 articles fulfilling the initial search criteria, 79 were selected for final analysis. Results: The frequency of image guidance, the treatment regimen and the radiation technique are important factors that contribute to the optimization and personalization of the treatment plan. The daily anatomy and volume of the bladder and rectum can vary considerably, which can significantly impact the dosimetric effects on these organs. When used in conjunction with volumetric modulated arc therapy, IGRT allows for shaping the dose distribution to avoid nearby critical structures such as the bladder and rectum. Conclusions: Precise tumor targeting via IGRT can result in fewer geometric uncertainties, thereby improving treatment outcome both in terms of superior target coverage and sparing organs at risk. Full article

Real-time tumor-tracked radiotherapy with a linear accelerator-magnetic resonance (linac-MR) hybrid system requires accurate tumor delineation at a fast MR imaging rate. Various autocontouring methods have been previously evaluated against “gold standard” manual contours by experts. However, manually drawn contours have inherent intra- and inter-observer variations. We aim to quantify these variations and evaluate our tumor-autocontouring algorithm against the manual contours. Ten liver, ten prostate, and ten lung cancer patients were scanned using a 3 tesla (T) magnetic resonance imaging (MRI) scanner with a 2D balanced steady-state free precession (bSSFP) sequence at 4 frames/s. Three experts manually contoured the tumor in two sessions. For autocontouring, an in-house built U-Net-based autocontouring algorithm was used, whose hyperparameters were optimized for each patient, expert, and session (PES). For evaluation, (A) Automatic vs. Manual and (B) Manual vs. Manual contour comparisons were performed. For (A) and (B), three types of comparisons were performed: (a) same expert same session, (b) same expert different session, and (c) different experts, using Dice coefficient (DC), centroid displacement (CD), and the Hausdorff distance (HD). For (A), the algorithm was trained using one expert’s contours and its autocontours were compared to contours from (a)–(c). For Automatic vs. Manual evaluations (Aa–Ac), DC = 0.91, 0.86, 0.78, CD = 1.3, 1.8, 2.7 mm, and HD = 3.1, 4.6, 7.0 mm averaged over 30 patients were achieved, respectively. For Manual vs. Manual evaluations (Ba–Bc), DC = 1.00, 0.85, 0.77, CD = 0.0, 2.1, 2.8 mm, and HD = 0.0, 4.9, 7.2 mm were achieved, respectively. We have quantified the intra- and inter-observer variations in manual contouring of liver, prostate, and lung patients. Our PES-specific optimized algorithm generated autocontours with agreement levels comparable to these manual variations, but with high efficiency (54 ms/autocontour vs. 9 s/manual contour). Full article

Background/Objectives: Stereotactic body radiotherapy is frequently used in patients with adrenal metastases. Motion of adherent radiosensitive organs at risk (OARs) and tumors influence OAR toxicity and tumor control. Online-adaptive Magnetic Resonance-guided radiotherapy (MRgRT) can address and mitigate interfractional changes. However, the impact of intrafractional variations in adrenal MRgRT is unknown. Methods: A total of 23 patients with 24 adrenal metastases were treated with MRgRT. After daily plan adaptation and before beam application, an additional (preRT) 3d MRI was acquired. PreRT target volumes and OARs were retrospectively recontoured in 200 fractions. The delivered, online-adapted treatment plans, as well as non-adapted baseline plans, were calculated on these re-contoured structures to quantify the dosimetric impact of intrafractional variations on target volume coverage and OAR doses with and without online adaptation. Normal tissue complication probabilities (NTCPs) were calculated. Results: The median time between the two MRIs was 56.4 min. GTV and PTV coverage (dose to 95% of the PTV, D95%, and volume covered by 100% of the prescription dose, V100%) were significantly inferior in the preRT plans. GTV D_mean was significantly impaired in left-sided metastases, but not in right-sided metastases. Compared to non-adapted preRT plans, adapted preRT plans were still significantly superior for all GTV and PTV metrics. Intrafractional violations of OAR constraints were frequent. D0.5cc and the volume exposed to the near-maximum dose constraint were significantly higher in the preRT plans. The volume exposed to the D0.5cc constraints in single fractions escalated up to 1.5 cc for the esophagus, 3.2 cc for the stomach, 5.3 cc for the duodenum and 7.3 cc for the bowel. This led to significantly elevated NTCPs for the stomach, bowel and duodenum. Neither PTV D95%, nor gastrointestinal OAR maximum doses were significantly impaired by longer fraction duration. Conclusions: Intrafractional motion in adrenal MRgRT caused significant impairment of target volume coverage (D95% and V100%), potentially undermining local control. Frequent violation of gastrointestinal OAR constraints led to elevated NTCP. Compared to non-adaptive treatment, online adaptation still highly improved GTV and PTV coverage. Full article

Linear accelerator–magnetic resonance (linac-MR) hybrid systems allow for real-time magnetic resonance imaging (MRI)-guided radiotherapy for more accurate dose delivery to the tumor and improved sparing of the adjacent healthy tissues. However, for real-time tumor detection, it is unfeasible for a human expert to manually contour (gold standard) the tumor at the fast imaging rate of a linac-MR. This study aims to develop a neural network-based tumor autocontouring algorithm with patient-specific hyperparameter optimization (HPO) and to validate its contouring accuracy using in vivo MR images of cancer patients. Two-dimensional (2D) intrafractional MR images were acquired at 4 frames/s using 3 tesla (T) MRI from 11 liver, 24 prostate, and 12 lung cancer patients. A U-Net architecture was applied for tumor autocontouring and was further enhanced by implementing HPO using the Covariance Matrix Adaptation Evolution Strategy. Six hyperparameters were optimized for each patient, for which intrafractional images and experts’ manual contours were input into the algorithm to find the optimal set of hyperparameters. For evaluation, Dice’s coefficient (DC), centroid displacement (CD), and Hausdorff distance (HD) were computed between the manual contours and autocontours. The performance of the algorithm was benchmarked against two standardized autosegmentation methods: non-optimized U-Net and nnU-Net. For the proposed algorithm, the mean (standard deviation) DC, CD, and HD of the 47 patients were 0.92 (0.04), 1.35 (1.03), and 3.63 (2.17) mm, respectively. Compared to the two benchmarking autosegmentation methods, the proposed algorithm achieved the best overall performance in terms of contouring accuracy and speed. This work presents the first tumor autocontouring algorithm applicable to the intrafractional MR images of liver and prostate cancer patients for real-time tumor-tracked radiotherapy. The proposed algorithm performs patient-specific HPO, enabling accurate tumor delineation comparable to that of experts. Full article