Stereotactic Radiosurgery for Malignant Intradural and Intramedullary Tumors of the Spine

Abstract

Malignant intradural (ID) and intramedullary (IM) tumors are challenging to treat with conventional methods. Furthermore, since these tumors are exceedingly rare, limited information is available on how to best manage the debilitating symptoms they often cause, such pain and neurological deficits. Due to the limited information available regarding the treatment of these tumors, we analyzed the role of stereotactic radiosurgery (SRS) in the treatment of malignant ID and IM tumors of the spine. 41 lesions (n=21) were treated with SRS for the management of their malignant ID or IM tumors at a single institution between 06/2001 and 12/2014. Electronic medical records were retrospectively reviewed in this Institutional Review Board approved analysis. Computed Tomography & Magnetic Resonance Images were used to evaluate the primary endpoints: pain, neurologic, and radiographic response 17 patients (81%) were deceased. Median survival post SRS was 13.6 m (4 d – 166 m). 11 (52%) patients were female. Ethnicities included: Caucasian - 14, African American - 3, & other - 4. Primary tumor locations included the following: breast - 4 patients, lung - 4, CNS - 11, other – 2. SRS was salvage treatment in 17 tumors (41%) [14 tumors had prior conventional radiation therapy (EBRT), 2 had previous surgery and 1 had both EBRT & surgery prior to SRS]. Median SRS dose was 16 Gy/1 fx (10 – 20 Gy) and median tumor volume was 3.64 cc (0.372 cc - 144.26 cc). Median follow-up was 7.6 m (21 d – 42 m). Of the 22 tumors (54%) causing pain at presentation, follow up was available for 14 (64%). Pain progressed after SRS in 7%. Pain was stable or improved in 93% [50% stable, 14% complete response (CR), and 29% partial response (PR)]. 18 tumors (44%) caused neurological deficits at the time of presentation; follow up was available for 15 (83%) of those. Following SRS, the neurological deficit progressed in 7%. 93% were stable or improved following treatment (73% stable, 7% PR, and 13% CR). Radiographic follow up was available for 26 tumors (63%). 23% of them progressed. Radiographic tumor control was achieved in 77% (42% stable, 31% PR, and 4% CR). 6 recurrences occurred; median time to recurrence was 5.5 m. Radiographic evidence of necrosis developed in the area of a previously treated (16 Gy/1 fx) C4-C5 leptomeningeal hemangioblastoma lesion 12.2 m after SRS. This patient also developed right extremity weakness 2.6 m after a subsequent treatment (16 Gy/1 fx) to C5-C6. Etiology was unclear, however SRS could not be excluded as the cause. There were no other SRS related toxicities. Our response rates, and the minimal toxicities observed, suggests that SRS should be considered as a treatment option for ID and IM tumors. Larger, studies and prospective trials exploring multimodal treatments options will be necessary to make more definitive treatment recommendations. To the best of our knowledge, this is the largest reported series on SRS for malignant ID and IM spine lesions.