Jun Deng, Presenter: Nothing to Disclose

Zhe Chen PhD, Abstract Co-Author: Nothing to Disclose

Jonathan P.S. Knisely MD, Abstract Co-Author: Nothing to Disclose

Roy H. Decker, Abstract Co-Author: Nothing to Disclose

Veronica Chiang MD, Abstract Co-Author: Nothing to Disclose

Ravinder Nath PhD, Abstract Co-Author: Nothing to Disclose

The purpose of this work is to investigate the dose discrepancy between the Monte Carlo method and a commercial treatment planning system (TPS) for image-guided intensity-modulated radiosurgery (IMRS) of spinal metastases and its clinical implications.

Two patients with spinal metastases treated by Varian Trilogy 6 MV at 1000 MU/min were included in this study. Treatment plans were generated for each patient using the Eclipse TPS (Pencil Beam Convolution, version 8.0.05, no heterogeneity correction). Retrospective Monte Carlo dose calculations were performed on the real patient anatomy and beam setup using the EGS4/MCSIM code. An in-house software tool was used to convert the DICOM images and DICOM RT structures into an EGS4 phantom for Monte Carlo dose calculations. 3-D dose distributions and dose-volume histograms (DVH) were analyzed and compared between the plans calculated with Eclipse and the Monte Carlo method.

Generally speaking, the Monte Carlo calculated plan was not as good as the Eclipse calculated plan in terms of target dose coverage. The homogeneity index has been found to be worse in the Monte Carlo calculated plans by about 9%, in comparison to the Eclipse PBC plans. The target dose was over-estimated in the PBC plans with the mean dose to the PTV on average 5% higher than that calculated by the Monte Carlo. This is due to the fact that almost all of the beams have to transport through the low-density lung tissues and/or air cavities to reach the target around spinal cord, producing much fewer secondary electrons to deposit energy in the target. On the other hand, the mean dose to the spinal cord was also over-estimated in the PBC by about 6%, compared to the Monte Carlo results. This is because the photon beams have a dose builddown as photons exit the bony structure of the target and reach the spinal cord behind. The over-estimation of the PTV dose may lead to under-treatment of the tumor and increase the risk of recurrence of primary cancers in patients with spinal metastases.

Due to the involvement of complex intensity modulation, large number of segments with small fields, inhomogeneous mediums, and extremely high dose in the intensity-modulated radiosurgery of spinal metastases, caution should be exercised when clinical decision is to be made based on the dose distributions from the treatment planning system which uses simple dose calculation algorithms such as pencil beam convolution. Large dose discrepancies have been observed in the target dose coverage between the Monte Carlo and the PBC calculations. If possible, Monte Carlo method should be used to provide more accurate dose distributions and dose-volume histogram analysis for intensity-modulated radiosurgery of spinal metastases.

Deng, J, Chen, Z, Knisely, J, Decker, R, Chiang, V, Nath, R, Monte Carlo Investigation of High Dose Single-Fraction Image-guided Intensity-modulated Radiosurgery of Spinal Metastases.  Radiological Society of North America 2008 Scientific Assembly and Annual Meeting, February 18 - February 20, 2008 ,Chicago IL. http://archive.rsna.org/2008/7001043.html