Influence of anatomical changes in robust optimized proton plans for bilateral head and neck cancer targets

Robust optimization in proton therapy considers uncertainties in patient setup and particle range during the optimization process. However, anatomical changes that may occur during the treatment course are neglected. The aim of this study was to quantify the influence of anatomical changes on the dose distributions for head and neck cancer (HNC) patients scheduled for bilateral neck irradiation.

Datasets from eight bilateral HNC patients, consisting of a planning computed tomography (CT) and weekly control CTs, were used. Intensity-modulated proton therapy plans were calculated with minimax robust optimization, account-ing 3 mm and 3.5% for setup and range uncertainty, respectively. The dose to the low- and high-risk clinical target vol-umes (CTV) consisted of 57 and 70 Gy(RBE), respectively, in 33 fractions. Organs at risk, e.g. spinal cord, brainstem, parotid glands, larynx, pharyngeal constrictor and esophageal inlet muscle, were considered for plan optimization and analysis. The cumulative dose during 33 fractions was checked weekly taking the anatomy of the control CTs into ac-count, and compared with the nominal plan.

Nominal plans fulfilled the clinical specifications of D98 ≥ 95% of the prescribed dose to the CTVs (range: 95.8-98.8% for low-risk CTV and 96.2-98.9% for high-risk CTV). During the treatment course, anatomical changes lead to reduced D98 values in five patients, with minimum of 87.3% in the low-risk CTV and 91.3% in the high-risk CTV. Maximum doses to spinal cord and brainstem remained below 45 Gy and 54 Gy, respectively. Mean doses to the contralateral pa-rotid gland remained below 26 Gy, except in one patient (maximum mean dose = 27.2 Gy).

For some patients, robust optimization prior to treatment initiation is insufficient to account for anatomical changes occurring during the treatment course. The results for a total cohort of 17 patients, including robustness analysis and plan adaptation strategies will be presented.