The Use of MR-Guided Radiation Therapy for Head and Neck Cancer and Recommended Reporting Guidance

Adaptive radiotherapy for oropharyngeal cancer with daily adapt-to-shape workflow on 1.5 T MRI-linac: Preliminary outcomes and comparison with helical tomotherapy

Introduction

MR-linac could offer several advantages for radical radiochemotherapy (RCHT) in oropharyngeal squamous cell carcinoma (OPSCC) patients. Currently, only a few case series have been published and no comparison with other techniques have been performed.

Methods

Data of 34 consecutive patients treated from September 2022 to May 2024 at a single Institution with RCHT on Unity® MR-linac for OPSCC with daily adaptive radiotherapy (RT) according to the adapt-to-shape (ATS) workflow were prospectively analyzed. A comparative cohort of 34 patients with similar characteristics treated with helical treatment on Radixact® was retrieved.

Results

Characteristics were well balanced across the two groups. Maximal toxicity grade ≥2 rate was borderline higher at RT end in MRI-linac group (p 0.049), but lower one month after RT (76.5 % vs 91.2 %; p = 0.257).Non-significantly lower rates of grade ≥2 xerostomia and dysgeusia were reported in Unity® group one and three months after RT. Higher rates of hospitalizations were reported in Radixact group at 20 fractions and at RT end (64.1 % vs 35.3 %; p = 0.015). Mean Karnofsky performance status (KPS) was higher in Unity group three months after RT (87.67 vs 83.87; p = 0.038).After a median follow up of 361.5 days, local complete response was reported for 93.6 % of patients treated with Unity® and 96.8 % of patients treated with Radixact®.

Conclusions

Results of this analysis support the feasibility of an ATS MR-linac workflow for RCHT in OPSCC. Compared with tomotherapy, treatment with Unity® resulted in significantly lower rates of hospitalization and higher KPS three months after RT. Grade 2 xerostomia and dysgeusia rates were non-significantly lower in Unity group. Optimal results in terms of local control were reported for both techniques.

Non-surgical organ preservation and new technologies in laryngeal radiation

The term "larynx organ preservation" (LOP) has become a synonym for non-surgical laryngeal cancer treatment based on chemotherapy and radiation multimodality therapy [simultaneous chemoradiation (CRT) or neoadjuvant chemotherapy followed by radiotherapy (NCT+RT)]. Currently, the distinction between good and bad candidates for LOP is not clear, and the decision for surgical or non-surgical treatment depends on the patient's needs and desires, the experience and recommendation of the surgeon, the philosophy of the institution, and others. Nevertheless, the major disadvantage of LOP by CRT and NCT+RT is the potential need for salvage surgery due to tumor persistence after the application of full per-protocol treatment. Head and neck surgeons worldwide complain that in principle, salvage surgery is frequently possible after CRT but causes major complications and is not feasible in a relevant number of patients. While NCT+RT is globally used to select responders for LOP, NCT alone has not been shown to improve overall survival. Therefore, this procedure has lost its influence in standard head and neck cancer treatment beyond LOP. Recently, NCT as part of the perioperative transoral surgical treatment concept in head and neck cancer is gaining interest again. In addition to conventional chemotherapy, the combination with immune checkpoint inhibitors as a neoadjuvant concept has shown to be effective in non-controlled trials by opening a new door of encouraging treatment options for LOP.

Benchmark of Deep Encoder-Decoder Architectures for Head and Neck Tumor Segmentation in Magnetic Resonance Images: Contribution to the HNTSMRG Challenge

Radiation therapy is one of the most frequently applied cancer treatments worldwide, especially in the context of head and neck cancer. Today, MRI-guided radiation therapy planning is becoming increasingly popular due to good soft tissue contrast, lack of radiation dose delivered to the patient, and the capability of performing functional imaging. However, MRI-guided radiation therapy requires segmenting of the cancer both before and during radiation therapy. So far, the segmentation was often performed manually by experienced radiologists, however, recent advances in deep learning-based segmentation suggest that it may be possible to perform the segmentation automatically. Nevertheless, the task is arguably more difficult when using MRI compared to e.g. PET-CT because even manual segmentation of head and neck cancer in MRI volumes is challenging and time-consuming. The importance of the problem motivated the researchers to organize the HNTSMRG challenge with the aim of developing the most accurate segmentation methods, both before and during MRI-guided radiation therapy. In this work, we benchmark several different state-of-the-art segmentation architectures to verify whether the recent advances in deep encoder-decoder architectures are impactful for low data regimes and low-contrast tasks like segmenting head and neck cancer in magnetic resonance images. We show that for such cases the traditional residual UNetbased method outperforms (DSC = 0.775/0.701) recent advances such as UNETR (DSC = .617/0.657), SwinUNETR (DSC = 0.757/0.700), or SegMamba (DSC = 0.708/0.683). The proposed method (lWM team) achieved a mean aggregated Dice score on the closed test set at the level of 0.771 and 0.707 for the pre- and mid-therapy segmentation tasks, scoring 14th and 6th place, respectively. The results suggest that proper data preparation, objective function, and preprocessing are more influential for the segmentation of head and neck cancer than deep network architecture.

Gradient Map-Assisted Head and Neck Tumor Segmentation: A Pre-RT to Mid-RT Approach in MRI-Guided Radiotherapy

Radiation therapy (RT) is a vital part of treatment for head and neck cancer, where accurate segmentation of gross tumor volume (GTV) is essential for effective treatment planning. This study investigates the use of pre-RT tumor regions and local gradient maps to enhance mid-RT tumor segmentation for head and neck cancer in MRI-guided adaptive radiotherapy. By leveraging pre-RT images and their segmentations as prior knowledge, we address the challenge of tumor localization in mid-RT segmentation. A gradient map of the tumor region from the pre-RT image is computed and applied to mid-RT images to improve tumor boundary delineation. Our approach demonstrated improved segmentation accuracy for both primary GTV (GTVp) and nodal GTV (GTVn), though performance was limited by data constraints. The final DSC agg scores from the challenge's test set evaluation were 0.534 for GTVp, 0.867 for GTVn, and a mean score of 0.70. This method shows potential for enhancing segmentation and treatment planning in adaptive radiotherapy. Team: DCPT-Stine's group.

Overview of the Head and Neck Tumor Segmentation for Magnetic Resonance Guided Applications (HNTS-MRG) 2024 Challenge

Magnetic resonance (MR)-guided radiation therapy (RT) is enhancing head and neck cancer (HNC) treatment through superior soft tissue contrast and longitudinal imaging capabilities. However, manual tumor segmentation remains a significant challenge, spurring interest in artificial intelligence (AI)-driven automation. To accelerate innovation in this field, we present the Head and Neck Tumor Segmentation for MR-Guided Applications (HNTS-MRG) 2024 Challenge, a satellite event of the 27th International Conference on Medical Image Computing and Computer Assisted Intervention. This challenge addresses the scarcity of large, publicly available AI-ready adaptive RT datasets in HNC and explores the potential of incorporating multi-timepoint data to enhance RT auto-segmentation performance. Participants tackled two HNC segmentation tasks: automatic delineation of primary gross tumor volume (GTVp) and gross metastatic regional lymph nodes (GTVn) on pre-RT (Task 1) and mid-RT (Task 2) T2-weighted scans. The challenge provided 150 HNC cases for training and 50 for final testing hosted on grand-challenge.org using a Docker submission framework. In total, 19 independent teams from across the world qualified by submitting both their algorithms and corresponding papers, resulting in 18 submissions for Task 1 and 15 submissions for Task 2. Evaluation using the mean aggregated Dice Similarity Coefficient showed top-performing AI methods achieved scores of 0.825 in Task 1 and 0.733 in Task 2. These results surpassed clinician interobserver variability benchmarks, marking significant strides in automated tumor segmentation for MR-guided RT applications in HNC.

Comparison between flattening filter and flattening filter-free photon beams in head and neck cancer patients using volumetric modulated arc therapy technique

This study aimed to evaluate the dosimetric and clinical outcomes of flattening filter (FF) versus flattening filter-free (FFF) beams in head and neck cancer (HNC) patients treated with volumetric modulated arc therapy (VMAT). Twenty-four patients with 70/59.4/54 Gy dose prescribed in 33 fractions with simultaneous integrated boost treatment were retrospectively analyzed to compare treatment delivery efficiency, target coverage, sparing of organs at risk (OARs), and remaining volume at risk (RVR) in two HNC groups (nasopharyngeal and oropharyngeal). Study findings indicate that FFF beams significantly reduce conformity index (CI) and homogeneity index (HI) by p-values (0.008, < 0.001, 0.002, 0.015) for PTV70 CI, PTV70 HI, PTV60 HI, and PTV54 HI, respectively. Gradient dose was significantly improved in FFF mode, and monitor units (MU) were increased (p < 0.001). In terms of OARs, the study revealed superior performance of FFF in most of structures and RVR especially in the oropharyngeal group. OARs sparing is notably enhanced for structures distant from the target (eyes, lenses, and optic pathway). Additionally, brainstem sparing shows significant improvement in oropharyngeal cases when using FFF plans (p = 0.046); however, FF plans demonstrate superior results in nasopharyngeal cases (p = 0.026). It is concluded that both FF and FFF photon beams are effective for treating HNC patients. VMAT plans using FFF mode offer clinically acceptable outcomes, demonstrating a significant reduction in gradient and integral dose. However, FF plans exhibit superior target homogeneity and reduced MU requirements. Therefore, the choice between these techniques should be based on a comprehensive evaluation of all relevant parameters.

MR-linac: role of artificial intelligence and automation

The integration of artificial intelligence (AI) into radiotherapy has advanced significantly during the past 5 years, especially in terms of automating key processes like organ at risk delineation and treatment planning. These innovations have enhanced consistency, accuracy, and efficiency in clinical practice. Magnetic resonance (MR)-guided linear accelerators (MR-linacs) have greatly improved treatment accuracy and real-time plan adaptation, particularly for tumors near radiosensitive organs. Despite these improvements, MR-guided radiotherapy (MRgRT) remains labor intensive and time consuming, highlighting the need for AI to streamline workflows and support rapid decision-making. Synthetic CTs from MR images and automated contouring and treatment planning will reduce manual processes, thus optimizing treatment times and expanding access to MR-linac technology. AI-driven quality assurance will ensure patient safety by predicting machine errors and validating treatment delivery. Advances in intrafractional motion management will increase the accuracy of treatment, and the integration of imaging biomarkers for outcome prediction and early toxicity assessment will enable more precise and effective treatment strategies.

A systematic review of the role of artificial intelligence in automating computed tomography-based adaptive radiotherapy for head and neck cancer

Purpose

Adaptive radiotherapy (ART) may improve treatment quality by monitoring variations in patient anatomy and incorporating them into the treatment plan. This systematic review investigated the role of artificial intelligence (AI) in computed tomography (CT)-based ART for head and neck (H&N) cancer.

Methods

A comprehensive search of main electronic databases was conducted until April 2024. Titles and abstracts were reviewed to evaluate the compliance with inclusion criteria: CT-based imaging for photon ART of H&N patients and AI applications. 17 original retrospective studies with samples sizes ranging from 37 to 239 patients were included. The quality of the studies was evaluated with the Quality Assessment of Diagnostic Accuracy Studies-2 and the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) tools. Key metrics were examined to evaluate the performances of the proposed AI-methods.

Results

Overall, the risk of bias was low. The average CLAIM score was 70%. A major finding was that generated synthetic CTs improved similarity metrics with planning CT compared to original cone-beam CTs, with average mean absolute error up to 39 HU and maximum improvement of 80%. Auto-segmentation provided an efficient and accurate option for organ-at-risk delineation, with average Dice similarity coefficient ranging from 80 to 87%. Finally, AI models could be trained using clinical and radiomic features to predict the effectiveness of ART with accuracy above 80%.

Conclusions

Automation of processes in ART for H&N cancer is very promising throughout the entire chain, from the generation of synthetic CTs and auto-segmentation to predict the effectiveness of ART.

Overview of the Head and Neck Tumor Segmentation for Magnetic Resonance Guided Applications (HNTS-MRG) 2024 Challenge

Magnetic resonance (MR)-guided radiation therapy (RT) is enhancing head and neck cancer (HNC) treatment through superior soft tissue contrast and longitudinal imaging capabilities. However, manual tumor segmentation remains a significant challenge, spurring interest in artificial intelligence (AI)-driven automation. To accelerate innovation in this field, we present the Head and Neck Tumor Segmentation for MR-Guided Applications (HNTS-MRG) 2024 Challenge, a satellite event of the 27th International Conference on Medical Image Computing and Computer Assisted Intervention. This challenge addresses the scarcity of large, publicly available AI-ready adaptive RT datasets in HNC and explores the potential of incorporating multi-timepoint data to enhance RT auto-segmentation performance. Participants tackled two HNC segmentation tasks: automatic delineation of primary gross tumor volume (GTVp) and gross metastatic regional lymph nodes (GTVn) on pre-RT (Task 1) and mid-RT (Task 2) T2-weighted scans. The challenge provided 150 HNC cases for training and 50 for testing, hosted on grand-challenge.org using a Docker submission framework. In total, 19 independent teams from across the world qualified by submitting both their algorithms and corresponding papers, resulting in 18 submissions for Task 1 and 15 submissions for Task 2. Evaluation using the mean aggregated Dice Similarity Coefficient showed top-performing AI methods achieved scores of 0.825 in Task 1 and 0.733 in Task 2. These results surpassed clinician interobserver variability benchmarks, marking significant strides in automated tumor segmentation for MR-guided RT applications in HNC.

The Evolving Role of Novel Imaging Techniques for Radiotherapy Planning

The ability to visualise cancer with imaging has been crucial to the evolution of modern radiotherapy (RT) planning and delivery. And as evolving RT technologies deliver increasingly precise treatment, the importance of accurate identification and delineation of disease assumes ever greater significance. However, innovation in imaging technology has matched that seen with RT delivery platforms, and novel imaging techniques are a focus of much research activity. How these imaging modalities may alter and improve the diagnosis and staging of cancer is an important question, but already well served by the literature. What is less clear is how novel imaging techniques may influence and improve practical and technical aspects of RT planning and delivery. In this review, current gold standard approaches to integration of imaging, and potential future applications of bleeding-edge imaging technology into RT planning pathways are explored.

Target volumes comparison between postoperative simulation magnetic resonance imaging and preoperative diagnostic magnetic resonance imaging for prone breast radiotherapy after breast-conserving surgery

BACKGROUND

This study investigated the differences in target volumes between preoperative magnetic resonance imaging (MRIpre) and postoperative MRI (MRIpost) for breast radiotherapy after breast-conserving surgery (BCS) using deformable image registration (DIR).

METHODS AND MATERIALS

Seventeen eligible patients who underwent whole-breast irradiation in the prone position after BCS were enrolled. On MRIpre, the gross tumor volume (GTV) was delineated as GTVpre, which was then expanded by 10 mm to represent the preoperative lumpectomy cavity (LC), denoted as LCpre. The LC was expanded to the clinical target volume (CTV) and planning target volume (PTV) on the MRIpre and MRIpost, denoted as CTVpre, CTVpost, PTVpre, and PTVpost, respectively. The MIM software system was used to register the MRIpre and MRIpost using DIR. Differences were evaluated regarding target volume, distance between the centers of mass (dCOM), conformity index (CI), and degree of inclusion (DI). The relationship between CILC /CIPTV and the clinical factors was also assessed.

RESULTS

Significant differences were observed in LC and PTV volumes between MRIpre and MRIpost (p < 0.0001). LCpre was 0.85 cm3 larger than LCpost, while PTVpre was 29.38 cm3 smaller than PTVpost. The dCOM between LCpre and LCpost was 1.371 cm, while that between PTVpre and PTVpost reduced to 1.348 cm. There were statistically significant increases in CI and DI for LCpost-LCpre and PTVpost-PTVpre (CI = 0.221, 0.470; DI = 0.472, 0.635). No obvious linear correlations (p > 0.05) were found between CI and GTV, primary tumor volume-to-breast volume ratio, distance from the primary tumor to the nipple and chest wall, and body mass index.

CONCLUSIONS

Despite using DIR technology, the spatial correspondence of target volumes between MRIpre and MRIpost was suboptimal. Therefore, relying solely on preoperative diagnostic MRI with DIR for postoperative LC delineation is not recommended.

Assessing the Impact of Charged Particle Radiation Therapy for Head and Neck Adenoid Cystic Carcinoma: A Systematic Review and Meta-Analysis

Purpose: Head and neck adenoid cystic carcinoma (HNACC) is a radioresistant tumor. Particle therapy, primarily proton beam therapy and carbon-ion radiation, is a potential radiotherapy treatment for radioresistant malignancies. This study aims to conduct a meta-analysis to evaluate the impact of charged particle radiation therapy on HNACC. Methods: A comprehensive search was conducted in Pubmed, Cochrane Library, Web of Science, Embase, and Medline until December 31, 2022. The primary endpoints were overall survival (OS), local control (LC), and progression-free survival (PFS), while secondary outcomes included treatment-related toxicity. Version 17.0 of STATA was used for all analyses. Results: A total of 14 studies, involving 1297 patients, were included in the analysis. The pooled 5-year OS and PFS rates for primary HNACC were 78% (95% confidence interval [CI] = 66-91%) and 62% (95% CI = 47-77%), respectively. For all patients included, the pooled 2-year and 5-year OS, LC, and PFS rates were as follows: 86.1% (95% CI = 95-100%) and 77% (95% CI = 73-82%), 92% (95% CI = 84-100%) and 73% (95% CI = 61-85%), and 76% (95% CI = 68-84%) and 55% (95% CI = 48-62%), respectively. The rates of grade 3 and above acute toxicity were 22% (95% CI = 13-32%), while late toxicity rates were 8% (95% CI = 3-13%). Conclusions: Particle therapy has the potential to improve treatment outcomes and raise the quality of life for HNACC patients. However, further research and optimization are needed due to the limited availability and cost considerations associated with this treatment modality.