KEY POINTS
- The prospective FIGS-RT pilot included 24 patients with head and neck cancer undergoing MRI simulation before RT. Diffusion-weighted MRI and apparent diffusion coefficient maps were used to define a biological at-risk volume within each parotid gland.
- The biological subvolume, termed BRV30, was generated from the lowest 30th percentile of parotid ADC values. It occupied approximately 64% of each parotid gland on average and was incorporated as an additional optimization structure during VMAT planning.
- Functional subvolume sparing was dosimetrically achievable without compromising high-dose target coverage. Mean dose was reduced by 3.1 Gy from whole ipsilateral parotid to ipsilateral BRV30 and by a further 10.1 Gy in its dominant BRV sector; corresponding reductions on the contralateral side were 1.4 Gy and 4.7 Gy.
- Acute xerostomia remained common: median peak clinician-reported xerostomia was grade 2, and 50% of patients still had grade 2 xerostomia at 3 months. Saliva, taste, chewing and swallowing patient-reported domains declined during treatment and had not returned to baseline by three months.
- A higher contralateral BRV30 mean dose showed a moderate but non-significant association with cumulative xerostomia burden (r=0.32; p=0.13). The whole contralateral parotid showed a similar trend (r=0.34; p=0.106).
- Higher dose to the dominant contralateral BRV30 sector was significantly associated with later peak dysphagia (r=0.56; p=0.005). Patients in higher-dose groups also had lower overall QoL at 3 months for both the whole contralateral parotid (p=0.047) and contralateral BRV30 (p=0.025).
- Interpretation is limited by the 24-patient sample, 72 exploratory statistical tests, absence of salivary-flow measurements and lack of assessment of treatment-related parotid shrinkage. Long-term follow-up is planned to determine whether the early signals strengthen as salivary function recovers.
CLINICAL TAKEAWAY
This study moves parotid sparing beyond a single whole-gland mean-dose constraint by testing whether DW-MRI can identify more radiosensitive functional subregions. The planning concept is feasible, but the current clinical evidence is too preliminary to justify replacing established parotid constraints with ADC-defined subvolume objectives.