Dose reconstruction for cord compressions retreatments using helical tomotherapy

Abstract

The treatment of patients with previously irradiated cord compressions must combine accurate patient positioning with conformal dose delivery to minimize spinal cord dose. By combining daily image guidance in the treatment room with intensity modulated radiation therapy (IMRT), it is possible to deliver a compete second course of therapy without exceeding the currently accepted spinal cord dose TD5 tolerances of 50 to 55 Gy. The purpose of this work is to 1.) Use daily CT imaging and dose reconstruction to determine the level of CT based image guidance required to accurately retreat spinal cord compressions with helical tomotherapy and 2.) Evaluate the level of conformal avoidance that is achievable for helical tomotherapy delivery. Five patients with cord compressions that had received previous radiation therapy were imaged and treated on a HI-ART helical tomotherapy system. The radiation oncologist contoured the GTVs, which were expanded into PTVs by adding a uniform 3-mm margin around the GTV, with the cord volume at risk subtracted out. The cord volume at risk (VAR) was defined as the entire spinal canal. The maximum dose to the spinal cord was set with the highest priority so that the total cord dose did not exceed 50 Gy. At least 80% of the PTV was to receive the prescribed dose. In addition, the percent volume of the cord receiving over 45 Gy was minimized during optimization. Five patients were treated with daily Megavoltage CT (MVCT) based image-guidance for a total of 55 fractions. MVCT images were taken twice prior to treatment and auto-fused with the treatment planning CT image set. The two image sets provide us with the initial shifts necessary to properly setup the patient and an opportunity to verify the patient setup before treatment. After the completion of treatment, dose distributions were recalculated on the original treatment planning CT incorporating the measured patient shifts. The first set of patient shifts were used to calculated the doses with No Image Guidance, and the second set of patient shifts were used to calculated the doses with Image Guidance. Cumulative isodose distributions and dose volume histograms were created for each of the five patients. Doses were reconstructed for all 5 patients using the measured shifts. Based upon dose calculations for no image guidance, the mean increase in spinal cord dose was 86.4% (range 33 to 250%), making treatment delivery impossible without image-guidance. Dose calculations were also made considering image guidance. These doses represent what should have been delivered using daily MVCT imaging. The mean increase in spinal cord dose was 5.2% (range 0 to 25%), making fractionated treatment delivery to 30 Gy possible even with minimal immobilization. In addition to the image-guidance analysis, the dose gradients were measured with film dosimetry. The measured dose gradient for the spinal cord retreat cases ranged from 5–10% per mm along the PTV / Spinal Cord interface. It is clear from this data that conformal avoidance of the spinal cord is achievable with helical tomotherapy. This study indicates that helical tomotherapy is capable of highly conformal avoidance of the cord with a high degree of accuracy using daily image-guidance for spinal cord compression recurrences. The Dose reconstruction analysis shows that the dose to the spinal cord can be kept below tolerance while delivering an addition 30 Gy to the PTV. To date, all patients responded to the treatment and none have developed acute or chronic complications as a result of the treatment.