Session Item

In oesophageal cancer, less toxicity is expected when treating patients with intensity-modulated proton therapy (IMPT) compared to volumetric modulated arc therapy (VMAT). In The Netherlands, this complication risk reduction is estimated using multivariable models and only patients who are expected to benefit clinically are selected for IMPT. Due to the increased sensitivity to density changes of IMPT, there is a general concern of plan robustness. In this study, we evaluated the first oesophageal cancer patients treated with IMPT in our clinic in terms of target and organs-at-risk (OARs) dose robustness, as compared to VMAT.

For 20 consecutive oesophageal cancer patients, clinical IMPT and VMAT plans were created on the averaged planning 4DCT. Weekly verification 4DCTs were acquired to evaluate the robustness of target coverage over the course of IMPT and VMAT treatments. If target coverage was degraded, replanning was performed and evaluations were done for both the initial and the replan. Retrospectively, target coverage was robustly evaluated for complete trajectories with and without replanning, based on dose warping and accumulation. Three radiation oncologists evaluated for each trajectory if accumulated target coverage was sufficient. Accumulation was additionally performed using the nominal doses on all repeated CTs for the adaptive IMPT and VMAT trajectories to evaluate the treatment course mean lung dose (MLD) and mean heart dose (MHD). The accumulated MLD and MHD were then compared to the planned dose to evaluate OARs dose robustness.

Replanning was performed more often for IMPT (15x) than for VMAT (8x). For VMAT, replanning was necessary only due to diaphragm displacements. Replanning for IMPT was not always indicated for these cases, but IMPT suffered from other variations like target deformations (Fig.1). Both adaptive treatment trajectories resulted in adequate accumulated target dose coverage (Fig.2). Adaptation after the first treatment week did not change the target coverage outcome. Overall, replanning could have been omitted 9 times for IMPT and 4 times for VMAT to still achieve an adequate accumulated target dose. Replanning in the first week of treatment had the most clinical impact, as anatomical changes resulting in insufficient accumulated target coverage were already observed at this stage. No differences were found in MLD between the planned dose and the accumulated dose. Accumulated MHD differed from the planned dose due to variations in diaphragm position, but these differences were consistent between IMPT and VMAT (mean MHD planning/accumulated 8.9 Gy/9.9 Gy [p>0.005] and 17.1 Gy/18.1 Gy [p<0.005], respectively).

Anatomical changes in oesophageal cancer patients were already observed in the first treatment week, affecting target and heart doses for both IMPT and VMAT. Adaptive clinical workflows assured adequate target dose coverage over the course of treatment.