gms | German Medical Science

Introduction: In treatment of (para-) spinal tumors the radiation sensitivity of the spinal cord inhibits the application of high doses needed for long term local control. In this work we first describe a novel image-guided radiation therapy (IGRT) protocol with on-line correction of patient positioning errors.

Methods: Six patients have been treated according to this protocol so far. Indications were spinal metastases (n=5) and a paraspinal tumor (n=1). In all cases the planning target volume (PTV) was of concave shape wrapped around the spinal cord. Intensity modulated radiotherapy treatment (IMRT) plans were generated. On-line verification of target and organ-at-risk position was done before every treatment fraction: a cone beam CT (CBCT) was acquired in treatment position, registered with the planning CT study and the 3D positioning error was calculated. This 3D set-up error was corrected using a robotic treatment couch that allows movement in six degrees of freedom (translation and rotation). The influence of the translational and rotational set-up errors on target coverage and doses to the spinal cord was investigated.

Results: The treatment protocol was carried out in a time frame of about 15min total treatment time. Five patients were successfully treated according this IGRT protocol, one patient died of pneumonia during therapy. Mean dose to the PTV was 57Gy in 20 to 30 fractions; the maximum dose to the spinal cord did not exceed 45Gy. Based on the CBCT studies the systematic (mean) set-up errors were small, less than 1mm and less than 1o. Random (standard deviation) translational errors were 4.2mm in longitudinal, 2.9mm in lateral and 2.8mm in vertical direction, random rotational errors of 2.2o, 1.9o and 1.9o were measured, respectively. Translational errors greater than 2mm were calculated in 25% and rotational errors greater than 1o in 16% of all measurements. Without correction, translational and rotational errors would have resulted in a 30% increase of the maximum dose to the spinal cord compared to the initial treatment plan.

Discussion: This IGRT protocol for treatment of (para-) spinal tumors was successfully implemented into day-to day clinical practise. Both translational and rotational set-up errors were shown to result in decreased target coverage and highly increased doses to the spinal cord. The on-line correction of these set-up errors allows a safe dose escalation and will hopefully result in increased rates of local control.

Figure 1 [Fig. 1], Table 1 [Tab. 1].