Abstract 194P
Background
Proton therapy is sensitive to density changes, which occur in patients with large tumor motion. Individual margins, or internal target volumes (ITV) to account for motion, are used for the primary tumor (GTVp), while the nodes (GTVn) are treated with a fixed margin. The aim of this study was to evaluate the effect of nodal motion on the dose coverage in proton therapy.
Methods
The study included 58 lung cancer patients with at least an involved upper mediastinal lymph node (4), infracarinal lymph node (7), and/or hilar lymph node (10). All patients had a 4DCT. The total GTVn was delineated on the planning phase and expanded with 5 mm to the clinical target volume (CTVn). The treatment plan was robustly optimized to ensure a coverage (V95%) of at least 95%. The total GTVn was divided into separate structures for GTVn4, GTVn7 and GTVn10 and propagated to the other phases of the 4DCT. For each GTVn, the amplitude was determined using the center of mass method. To evaluate the effect of the nodal motion on the planned dose, the treatment plan was evaluated on all phases of the 4DCT. For each lymph node station, the V95% of their respective CTV was compared using the 3D and 4D evaluation doses.
Results
The motion amplitude was highest in the superior-inferior direction (table). The largest average amplitudes were 1.9 mm (left-right), 1.7 mm (anterior-posterior) and 5.7 mm (superior-inferior). V95% in the 3D-evaluation was >95% for the total GTVn, but ranged from 88–93% for the individual node stations. Loss of coverage due to breathing motion (4D-3D) was at most 1.8% (CTVn7). V95% for the hilar node stations was 93% (10L and 10R), and remained >92% in the 4D evaluation.
Table: 194PAmplitude in three directions and dose coverage (3D and 4D) of each lymph node station
| Lymph node stations | Number of patients | Average amplitude ± SD (mm) | Dose coverage V95% (%) | |||||
|---|---|---|---|---|---|---|---|---|
| left-right | anterior-posterior | superior-inferior | 3D evaluation | 4D evaluation | Difference 4D - 3D | |||
| Mediastinal | 4L | 15 | 1.6 ± 1.5 | 1.0 ± 0.7 | 2.6 ± 2.2 | 88.3 ± 14.1 | 87.5 ± 16.6 | −0.8 |
| 4R | 41 | 1.9 ± 1.6 | 1.7 ± 1.0 | 3.4 ± 2.7 | 92.0 ± 6.7 | 91.4 ± 7.0 | −0.6 | |
| 7 | 37 | 1.0 ± 0.8 | 1.2 ± 1.0 | 4.5 ± 2.9 | 88.3 ± 16.3 | 86.5 ± 17.4 | −1.8 | |
| Hilar | 10L | 20 | 1.5 ± 1.8 | 1.0 ± 0.8 | 4.5 ± 3.2 | 92.9 ± 4.7 | 92.1 ± 4.9 | −0.8 |
| 10R | 36 | 1.7 ± 1.6 | 1.4 ± 1.2 | 5.7 ± 3.2 | 93.2 ± 5.6 | 92.3 ± 5.7 | −0.9 | |
| All lymph nodes | 58 | - | - | - | 95.2 ± 1.1 | 94.6 ± 1.5 | −0.6 |
Conclusions
The average effect of nodal motion on the proton dose coverage was minimal. Although the superior-inferior amplitude of the hilar node stations was higher compared to the mediastinal nodes, the dose coverage of the hilar lymph nodes (V95%) remained adequate.
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.