Geometrical variability of esophageal tumors and its implications for accurate radiation therapy
| Authors |
|
|---|---|
| Supervisors |
|
| Cosupervisors |
|
| Award date | 31-01-2019 |
| ISBN |
|
| Number of pages | 217 |
| Organisations |
|
| Abstract |
Esophageal cancer is currently ranked the eighth most common cancer worldwide. Radiation therapy with concurrent chemotherapy is preferred for both preoperative and definitive treatment. One major concern in accurate radiation therapy is the geometrical uncertainty of the clinical target volume. Currently, inter- and intra-observer variation in target delineation are large due to the limited soft-tissue contrast on CT. Fiducial markers implanted at the tumor borders can significantly reduce the delineation variation.
Apart from delineation variation, interfractional tumor position variation has a major contribution to the geometrical uncertainty in the case of bony anatomy-based setup verification. It is most pronounced in the cranial-caudal direction and in the cardia, implying the need of an anisotropic and region-dependent safety margin for uncertainty compensation. Carina-based setup verification cannot reduce the interfractional tumor position variation compared to bony anatomy-based setup verification. Furthermore, gastrointestinal gas volume variation can also degrade the dose distribution. In clinical practice, using density override could mitigate the gas-induced effect. Another source of geometrical uncertainty is respiration-induced tumor motion. It is also most pronounced in the cranial-caudal direction and in the distal esophagus. Moreover, the interfractional variability of respiration-induced tumor motion is limited. Thus, it is recommended to use 4D-CT for treatment planning. In this case, compared to the concept of internal target volume, using the mid-position strategy may reduce dose to organs at risk such as lung and heart by 10%. There are also other approaches to mitigate geometrical uncertainty. Further research needs to be done in the future. |
| Document type | PhD thesis |
| Language | English |
| Downloads | |
| Permalink to this page | |