TU-F-CAMPUS-J-04: Setup Uncertainties in the Mediastinum Area for IMRT Treatment of Lymphoma Patients

Authors


Abstract

Purpose:

Despite daily image guidance setup uncertainties can be high when treating large areas of the body. The aim of this study was to measure local uncertainties inside the PTV for patients receiving IMRT to the mediastinum region.

Methods:

Eleven lymphoma patients that received radiotherapy (breath-hold) to the mediastinum were included in this study. The treated region could range all the way from the neck to the diaphragm. Each patient had a CT scan with a CT-on-rails system prior to every treatment. The entire PTV region was matched to the planning CT using automatic rigid registration. The PTV was then split into 5 regions: neck, supraclavicular, superior mediastinum, upper heart, lower heart. Additional auto-registrations for each of the 5 local PTV regions were performed. The residual local setup errors were calculated as the difference between the final global PTV position and the individual final local PTV positions for the AP, SI and RL directions. For each patient 4 CT scans were analyzed (1 per week of treatment).

Results:

The residual mean group error (M) and standard deviation of the inter-patient (or systematic) error (Σ) were lowest in the RL direction of the superior mediastinum (0.0mm and 0.5mm) and highest in the RL direction of the lower heart (3.5mm and 2.9mm). The standard deviation of the inter-fraction (or random) error (σ) was lowest in the RL direction of the superior mediastinum (0.5mm) and highest in the SI direction of the lower heart (3.9mm) The directionality of local uncertainties is important; a superior residual error in the lower heart for example keeps it in the global PTV.

Conclusion:

There is a complex relationship between breath-holding and positioning uncertainties that needs further investigation. Residual setup uncertainties can be significant even under daily CT image guidance when treating large regions of the body.

Ancillary