Osaka Group Reports Its Accelerator BNCT Experience in About 200 Head and Neck Cancer Cases

Aihara and six colleagues set three questions

Aihara and six colleagues, among them Kawata from head and neck surgery and Ono from the BNCT center, frame their report around three questions: the proportion of referred head and neck cases that are eligible for BNCT at all, the efficacy achieved in the cases they treated, and where the modality goes next. The physical argument they restate is that high linear energy transfer particles released when thermal neutrons meet boron atoms taken up by tumor tissue travel a shorter distance than a single cell diameter, which is what allows a large tumor dose beside untouched normal tissue.

Elsevier keeps the article behind a subscription and the indexed abstract stops at the aims, so no response rate, no toxicity grade and no follow-up figure from this particular paper can be quoted here. The numbers that follow come from the same center's open-access publications, cited individually.

Fifty Percent Complete Response in the Oral Cancer Subset

Between June 2020 and June 2024 the Kansai center treated 300 patients under the insurance scheme, of whom 75 had oral cancer. Yoshino and colleagues analyzed 74 of them in Frontiers in Oncology in February, with a data cutoff of May 12, 2025. Recurrent disease accounted for 73 percent of the group, and the complete response rate came to 50 percent. Severe acute toxicity meant oral mucositis, grade 3 in every case, in 26 percent of patients, and the two significant predictors were the maximum oral mucosal dose and the number of dental metals sitting in the irradiation field. Two-year overall survival was 49 percent, locoregional control 52 percent and progression-free survival 29 percent.

One procedural detail sits behind the dental metal finding: no metal was removed or modified before treatment, and any piece lying even partly inside the field was counted, whatever its material.

The Dosimetry Model Did Not Predict Response

Boron drug uptake is normally estimated before treatment from the tumor-to-blood ratio derived from maximum standardized uptake value on fluorine-18 labeled 4-borono-L-phenylalanine PET, and dose planning uses either that individual ratio or a uniform assumption of 2.5. Writing in the Japanese Journal of Radiology on June 2, the same group tested whether any of it predicts outcome in 30 head and neck patients treated in 2020 and 2021. Objective response reached 90 percent and complete response 50 percent, and complete response strongly predicted two-year overall survival, locoregional control and progression-free survival, all at p below 0.05. Tumor SUVmax did not separate responders from non-responders, and neither did doses calculated by the uniform model or the individual model, all comparisons landing above p of 0.05. The authors conclude that SUVmax-based dosimetry needs rework.

Why This Matters to the APO|APE Reader

Eligibility at the Kansai center is decided by a computed tomography pre-plan against a fixed physical target, 20 gray X-ray equivalent delivered to at least 80 percent of the tumor volume, drawn from earlier work and applied before a patient is accepted. Patients judged at high risk of laryngeal edema from the planned field undergo tracheotomy first. Selection on that basis, in a population where surgery and conventional reirradiation have both been ruled out, is what stands behind the survival curves above, and it is the part of the Japanese experience that any center importing an accelerator inherits along with the hardware. The reimbursement rule adds a second filter, since conventional radiotherapy takes precedence whenever it is still expected to work.