Head and neck radiation therapy relies on accurate imaging for target delineation and normal-tissue sparing given the complex anatomy of the skull base, oral cavity, pharynx, and larynx. The authors report a clinically implemented MR-simulation protocol developed through multidisciplinary collaboration to provide reproducible, high-quality magnetic resonance imaging (MRI) sequences that integrate into radiation oncology workflows.
The protocol was implemented prospectively in a cohort of 41 patients with head and neck cancer plus 3 volunteers. Imaging was performed on either a 1.5T or 3T MR-simulation scanner. Participants were enrolled under protocols approved by the University of Texas MD Anderson Cancer Center Institutional Review Board (volunteer protocol PA15-0418; patient protocol 2024-0406) with written informed consent obtained from all subjects.
The implemented sequence set spans several technical approaches to address differing tissue contrast and clinical needs. Major families described are Dixon-based techniques for water/fat separation, SPAIR-based fat suppression strategies, short T2–sensitive sequences for tissues with rapid signal decay, and various quantitative imaging methods. The manuscript provides sequence categories and practical notes to aid local optimization.
Dixon-based methods were included to provide robust water–fat separation, which can improve soft-tissue contrast and consistency for contouring. The authors describe using Dixon approaches within the MR-simulation environment and outline steps to optimize acquisition parameters on both 1.5T and 3T platforms. Specific parameter values and stepwise optimization details are reported in the manuscript to support reproducible implementation in other clinics.
SPAIR-based sequences were implemented as an alternative fat-suppression strategy, particularly where uniform fat suppression is required across the head and neck. The team describes sequence selection and tuning to mitigate typical artifacts and achieve consistent fat nulling across the complex anatomy of the skull base, oral cavity, and pharynx/larynx regions.
Short T2–sensitive sequences were incorporated to visualize tissues or structures with rapid transverse relaxation that may be poorly depicted on conventional sequences. The authors outline how to optimize these sequences within a radiation planning context to capture anatomical detail that supports target definition and organ-at-risk delineation.
The protocol set also includes quantitative imaging approaches. These methods were described as supplementary techniques to provide measurable tissue parameters that may aid characterization and reproducibility. The manuscript reports inclusion of quantitative sequences and provides practical guidance on their integration into a clinical MR-simulation exam.
Developing a clinical MR-simulation protocol required collaboration across disciplines: radiation oncologists, neuroradiologists, medical physicists, and radiation therapists. The authors detail steps to optimize sequence choices, adapt vendor-specific implementations for 1.5T and 3T scanners, and integrate the protocol into routine simulation workflows. Emphasis is placed on reproducibility and practical measures that facilitate clinical throughput while achieving diagnostic-quality images for planning.
The work followed institutional ethical guidelines and received IRB approval for both volunteers and patients. All participants provided written informed consent. The manuscript states that all data produced in the present work are contained within the manuscript, and supplementary material and data/code links are available from the preprint record.
Multiple funding sources are declared, including National Cancer Institute grants, National Center for Advancing Translational Sciences training support, and institutional and philanthropic fellowships. Industry collaborators and vendors are listed among funders. One author (CDF) disclosed unrelated travel, honoraria, and royalties involving multiple imaging and oncology vendors. The authors caution that the report is a preprint and has not been peer reviewed; therefore, it should not be used to directly guide clinical practice without further validation.
This report supplies a practical, multi-sequence MRI protocol tailored for head and neck MR-simulation across skull base, oral cavity, and larynx/pharynx targets. By sharing sequence families, optimization steps, and workflow notes for both 1.5T and 3T scanners, the authors aim to facilitate adoption and reproducibility at other radiation oncology centers. Readers are reminded that specific parameter values and detailed optimization steps are provided within the manuscript and supplementary material; because this is a preprint, those implementing changes should consider local validation and peer-reviewed evidence when altering clinical practice.