Pushing the Dose to the Spinal Cord in Spine SBRT - Cord and Thecal Sac Dosimetric Correlation with Toxicity

Abstract

Purpose/Objectives: Spine SBRT is widely performed and the reported risk of myelopathy is very low. Currently, guidelines vary on the maximum tolerable dose levels and designation of spinal cord vs thecal sac as the dose limiting structure. The current single institution study examines dose metrics to both structures and correlations with toxicity for patients treated with IMRT/VMAT. Materials/Methods: The radiation plans and DVH parameters were exported for 46 patients treated with SBRT for spinal metastases between April 2008 and December 2010 at the Mayo Clinic Rochester. Diagnostic MRI and where applicable, CT myelograms, from each patient were fused with the CT planning set and used to contour the spinal cord and the thecal sac. High resolution PRV structures were created in 1 mm increments for the cord (1-7 mm) and the thecal sac (1-2 mm), to examine dose gradients. Using an alpha/beta of 2 Gy, the biologically equivalent 2 Gy dose maximum (Max(2)[EQ2 Gy]) and high dose sub-volumes (Dxcc(2)[EQ2 Gy]) were calculated for xZ 0.1 cc to 1.0 cc in 0.1 cc increments from the dose volume histograms (DVHs). Toxicities for pain, nausea, myelitis, fatigue, fracture and radionecrosis were assessed (CTCAE v4.0). ROC analysis was used to define thresholds for constructing 2 x 2 contingency tables indicating toxicity using fisher exact test for significance (p < 0.01). Differences in mean values for groups with/without toxicity were assessed using t-tests for significance (p < 0.05). Results: Median values of Max(2)[EQ2 Gy] for the spinal cord and thecal sac were 38.5 (range, 7.9-67.9 Gy) and 67.7 Gy (range, 15.5-155.8 Gy), respectively. There were no cases of myelopathy (median follow-up 14 months). Median values for high dose sub volumes, Dxcc(2)[EQ2 Gy], were 2 times higher than the doses for 5% predicted grade 3 cord toxicity recommended by Sahgal et al. (IJORBP: 2013;85). Five patients had pain (4 with grade 1, 1 with grade 2). Four patients had nausea (3 with grade 1, 1 with grade 2). Cord D0.1cc(2)[EQ2 Gy] 23.8 Gy was significant (p Z 0.007) as an indicator for pain 1. Thecal sac D2.0cc(2)[EQ2 Gy] 29.3 Gy was significant (pZ 0.008) as an indicator for nausea 1. Distribution of Max(2)[EQ2 Gy] values for the thecal sac corresponded most closely to spinal cord + 3.5 mm margin. Median survival was 14 months (range 1-64 months). Conclusions: Current guidelines may overestimate the risk of myelopathy from spine SBRT. The current study’s population included patients who were both radiation naive and retreated after conventional palliative radiotherapy. The actual risk of myelopathy may be much lower than predicted. For patients with limited survival, it may be appropriate to allow higher doses to cord if it would be expected to result in better QOL/ functional outcomes/pain relief for the duration of their life. Author Disclosure: D. Owen: None. C.S. Mayo: None. L.S. Song: None. K. Ahmed: None. N.N. Laack: None. K.R. Olivier: None.