Session Item

A limited number of beam directions, often non-coplanar, is commonly used when planning sinonasal cancer patients. While one of the advantages of proton therapy is that there is no exit dose, there is still entrance dose, and using just a few beam directions could result in a high dose to the organs at risk (OARs) they are traversing. In this study, we compared the commonly used planning strategy with a strategy using 11 coplanar beams in a pseudo-arc setup, and evaluated and compared the effect in the dose to the OARs.

Retrospective proton (IMPT) plans were made for 24 sinonasal cancer patients in Eclipse v15.6. Dose was 66-68Gy/60-66Gy for primary/postoperative radiotherapy. The strategies evaluated were: S1, which consisted of 3-4 beams angles; and S2, 11 beams angles spaced 20 degrees through the front (avoiding the nose). A range shifter of 5 cm was used for S1 for beams where the water equivalent distance between skin surface and target in beams-eye-view was <4cm. For S2 the range shifter was avoided if acceptable target coverage could be achieved without, but if needed was used for all fields.  Beam configurations are shown in Figure 1. All plans were optimized with robust optimization (RO) and multifield optimization. For RO setup uncertainty of ± 2mm in all cardinal directions and ± 3.5% range uncertainty were used (14 scenarios in total). For robustness evaluation (RE), the same parameters were used. Dose to OARs was evaluated and compared between the strategies.   

For 9/24 patients, it was necessary to use range shifter for S2, in order to reach acceptable coverage for all RE scenarios.

Mean dose to OARs in the ipsilateral side was lower for S2 than for S1 (Figure 2a). For example, mean doses to the ipsilateral posterior eye were 24.7Gy/14.8G for S1/S2 (population median). Organs farther away from the target or in the contralateral side, received a lower dose with S1 compared to S2. For example, mean doses to the contralateral posterior eye were 3.3Gy/4.5Gy for S1/S2 (population median).

The same tendency was observed for maximum doses (Figure 2b). Maximum doses to the ipsilateral anterior eye were 37.6Gy/21.0Gy for S1/S2 (population median), while maximum doses to the contralateral optic nerve were 12.7Gy/16.5Gy for S1/S2 (population median).

Even when range shifter (RS) had to be used in S2, there was still a benefit for some ipsilateral organs – see maximum dose (population median) in table below.

OARs in the proximity of the target benefited from using multiple beams in a pseudo-arc, but this resulted in an increased low dose for OARs farther away from the target. The use of range shifter gave a lower benefit of using multiple beams. An evaluation of the effect of the day-to-day anatomical variations during the treatment of the different planning strategies is presently being performed.