Abstract
METHODS AND MATERIALS
Thirty-three SCRT with daily repeat CT scans and 30 LCRT patients with daily scans during the first week followed by weekly scans were included. The CTV was delineated on all scans and local shape variation was calculated with respect to the planning CT. Margin estimation was done using the local shape variation to assure 95% minimum dose for at least 90% of patients.
CONCLUSIONS
Heterogeneous shape variation demands anisotropic PTV margins. Required margins were substantially larger in the anterior direction compared to current clinical margins. These larger margins were, however, based on strict delineated CTVs, resulting in smaller PTVs compared to current practice.
RESULTS
Using 482 CT scans, systematic and random CTV shape variation was heterogeneous, ranging from 0.2 cm close to bony structures up to 1.0 cm SD at the upper-anterior CTV region. A significant reduction in rectal volume during LCRT resulted in an average 0.5 cm posterior shift of the upper-anterior CTV. Required margins ranged from 0.7 cm close to bony structures up to 3.1 and 2.3 cm in the upper-anterior region for SCRT and LCRT, respectively.
PURPOSE
To quantify the inter-fraction shape variation of the CTV in rectal-cancer patients treated with 5 × 5 (SCRT) and 25 × 2 Gy (LCRT) and derive PTV margins.