KEY POINTS
- This cross-sectional survey by the International MR-Linac Consortium collected responses from 36 experts across four continents, including 24 physicians, seven medical physicists, two radiation therapists, and three nonclinical researchers. Most respondents (24/36, 67%) worked in academic institutions, with a median of 10 years of head and neck radiation oncology experience. Although 31/36 (86%) reported access to an MR-Linac, only 10/36 (28%) currently treated head and neck patients using the platform, with a median of 2.5 patients treated per respondent.
- Adaptive treatment strategies varied considerably among the ten active head and neck MR-Linac users. Four used adapt-to-position (ATP), four performed online adapt-to-shape (ATS), and two used offline ATS. Among nine respondents reporting adaptation frequency, four adapted daily and five weekly. In conventional linac practice, 50% of 30 respondents used 5-mm PTV margins, whereas 67% of nine MR-Linac users reported 3-mm margins. These differences describe current practice and do not establish the safety of smaller margins or superiority of a particular adaptation frequency.
- Treatment delivery efficiency was the dominant implementation concern. Faster delivery was requested by 27/30 respondents (90%), while 28/30 (93%) considered VMAT capability important for head and neck MR-Linac treatment. Specific requested improvements included faster beam delivery, higher dose rates, shorter imaging and optimization time, and improved treatment-plan quality. AI-assisted contouring was also prioritized by 7/24 respondents (29%) answering the planning-system question. Long treatment sessions and limited clinical staffing were repeatedly identified as obstacles to broader adoption.
- Dose accumulation and automated quality assurance remain important unmet technical needs. Among respondents addressing cumulative dose reporting, 13/18 (72%) did not use it routinely outside research settings. In contrast, 25/25 (100%) wanted standardized automated QA for adaptive plans, incorporating structure checks, independent dose calculation, OAR constraint verification, and rapid plan-quality assessment. Functional MRI was another priority, with 12/20 respondents (60%) identifying diffusion-weighted imaging as their preferred functional MR sequence.
- MR-guided reirradiation attracted strong interest, particularly stereotactic approaches. All 24 respondents answering the relevant question considered MR-Linac technology useful for head and neck reirradiation, and 20/23 (87%) supported SBRT as a strategy for future reirradiation trials. The median previously delivered dose overlap that respondents considered acceptable was 35 Gy EQD2 (IQR 23–55 Gy), but this represents expert opinion rather than a validated cumulative dose limit. Oropharyngeal cancer was the most frequently prioritized site for future MR-Linac trials (19/24, 79%), followed by hypopharyngeal (63%) and nasopharyngeal cancer (58%).
- Experts prioritized early response assessment and selective adaptation over routine daily full replanning. Early identification of treatment responders received the highest trial-priority score (mean importance score 4.38/5), followed by weekly tumor adaptation (4.15/5) and margin reduction (4.14/5). Daily tumor adaptation scored substantially lower (3.14/5). Additionally, 15/15 respondents addressing central trial QA supported standardized central review, and 11/12 (92%) considered standardized dose accumulation reporting important for prospective studies.
- The survey is not a representative global audit of MR-Linac practice. Respondents were selected from a specialized consortium, participation was voluntary, and the overall response rate could not be determined. Only ten clinicians or centers reported active head and neck MR-Linac use, with individual questions answered by smaller and varying subsets. The study did not measure treatment times directly, compare delivered doses, assess toxicity, or demonstrate that VMAT, AI automation, or weekly adaptation improves clinical outcomes.
CLINICAL TAKEAWAY
MR-Linac technology remains relatively underused for head and neck radiotherapy even among specialized centers with access to the platform, largely because of workflow complexity, treatment duration, and resource demands. Faster delivery, validated automation, standardized QA, and reliable dose accumulation were consistently identified as priorities. The findings can inform technical development and prospective trial design but do not establish the optimal adaptive workflow or clinical superiority over contemporary conventional image-guided radiotherapy.