Online adaptive proton SBRT was feasible in the phase I PARTY trial
All 50 proton SBRT fractions were completed, with online adaptation used in 84% to restore target coverage or resolve OAR violations.
All 50 proton SBRT fractions were completed, with online adaptation used in 84% to restore target coverage or resolve OAR violations.
Bragg peak FLASH achieved high ultra-high-dose-rate coverage with similar target coverage to conventional IMPT, but produced higher target and CNS hotspots.
Ten-year biochemical control was 64.7% after salvage proton therapy, with 3.2% late grade 3 genitourinary toxicity.
Selective field-based imaging reduced mean setup time by about five minutes with similar geometric accuracy and dosimetric robustness.
FLASH protons largely preserved ocular structure and retinal function after 24 Gy while maintaining short-term tumor control in mice.
Across 187 fractions, surface-guided breath-hold IMPT remained highly stable, with only one patient requiring adaptive replanning.
A six-center end-to-end framework detected anatomy-induced proton dose deterioration and verified near-complete recovery after adaptive replanning.
Boosting selected proton spots reduced simulated prompt-gamma noise by up to 93.3% while maintaining acceptable target and organ-at-risk dosimetry.
Postoperative proton therapy achieved 84% five-year local control in paediatric ependymoma
All 78 deep learning-generated IMPT plans were clinically accepted, with comparable or lower organ doses than manual benchmark plans.
Causal modeling estimated mandibular proton RBE at 1.61, 1.30 and 1.13 at 40, 50 and 60 GyRBE, respectively
Proton IMPRINT kept contralateral lung mean dose at 0.3 Gy, with grade 3 pneumonitis occurring in 6% of patients.
Proton hypofractionation produced 10% one-year local failure despite extensive prior irradiation, with grade 3 toxicity confined to multiply reirradiated patients.
A patient with 22q11.2 deletion developed refractory brainstem necrosis nine months after guideline-concordant proton craniospinal irradiation.
Voxel-wise cumulative EQD2 review prompted reoptimization in 25.5% of proton reirradiation courses and exposed frequent registration problems.