CHAPTER 263 Radiosurgery for Benign Spine Tumors and Vascular Malformations
Stereotactic radiosurgery has become a widely accepted primary and adjuvant treatment of both malignant and benign primary brain tumors, intracranial metastases, and intracranial arteriovenous malformations (AVMs), with large studies demonstrating effective tumor control and low morbidity.1–6 Technologic advances have more recently allowed radiosurgery to transcend the confines of the stereotactic head frame such that extracranial targets are now feasible.7 The current body of evidence supporting the use of extracranial radiosurgery for malignant spine tumors is considerable,8–27 and radiosurgery has only recently become part of the multimodality management of benign spinal tumors and spinal vascular malformations.
Benign spinal tumors represent a group of intradural extramedullary neoplasms that include meningiomas, schwannomas, and neurofibromas. The primary treatment option for most benign spinal neoplasms is microsurgical resection. The safety and effectiveness of such surgery have clearly been documented.7,28–33 The majority of spinal meningiomas, schwannomas, and neurofibromas are not infiltrative and can be completely and safely resected with microsurgical techniques.31,34–36 When complete tumor removal is achieved, recurrence is unlikely.28,31,37,38
In certain circumstances, however, some patients are less than ideal candidates for open standard surgical resection because of age, medical comorbid conditions, and the recurrent nature of their tumor.7 Multiple benign spinal tumors, as commonly seen in familial neurocutaneous disorders, may be a pattern of spinal pathology better suited for the less invasive radiosurgical option. In addition, tumors that have recurred after open surgical resection may make safe surgical resection challenging or impossible. It is in such clinical circumstances that radiosurgery might be an important clinical option for these patients.
Stereotactic radiosurgery for the treatment of a wide variety of benign intracranial lesions has become widely accepted and has excellent long-term outcomes and minimal toxicity.1,3,4,39 Radiation therapy has been used for the treatment of numerous benign diseases since discovery of the therapeutic potential of ionizing radiation.40–44 Although similar benign tumor histologies can be found in and around the spine, the initial radiosurgical instruments were frame based and unable to treat such extracranial targets.7 The recent emergence of frameless image-guided radiosurgery allows the treatment of such tumors throughout the body.23,45,46 There is now a substantial body of literature that supports the use of extracranial radiosurgery for the treatment of a variety of malignant spinal tumors,8–2747 but clinical experience with radiosurgery for the management of benign spine tumors is more limited.7,48 Because patients with benign lesions have prolonged life expectancies in comparison to their malignant counterparts, the potential for delayed radiation-induced myelopathy is a special concern. In addition, benign spine tumors have their own unique clinical manifestations, relationship to the spinal cord, and radiobiologic response to radiosurgery, any of which could represent unique challenges to safe and effective application of radiosurgical ablation.7
Besides treating poor surgical candidates and recurrent or residual tumor after surgery, spinal radiosurgery is most appropriate for well-circumscribed lesions associated with minimal spinal cord compromise and no biomechanical instability.49 Physicians at the Memorial Sloan-Kettering Cancer Center have proposed a four-tiered grading system for spinal tumors based on the degree of epidural involvement and spinal cord compromise; patients with severe grade 2 and 3 spinal cord compromise are deemed more appropriate candidates for surgery than for radiotherapy. Relative contraindications to spinal radiosurgery include evidence of overt spinal instability, neurological deficit resulting from bony compression, or previous radiation treatment to the spinal cord tolerance dose. A theoretical advantage of using spinal radiosurgery as frontline management of spinal tumors is the possibility that such treatment may act as prophylaxis against future spinal instability or compression of neural elements and thus obviate the need for extensive spinal surgery and instrumentation. Moreover, early radiosurgery may obviate the need for large-field external beam radiation, which is known to suppress bone marrow function. Tumor shrinkage and complete obliteration aside, the minimally invasive technique of spinal radiosurgery may develop into an effective palliative strategy solely through local tumor control. Finally, the ability to perform spinal radiosurgery in the outpatient setting is an advantage that may spare patients with spinal tumors both time and the morbidity of hospitalization.
Special Considerations for Benign Tumors
Radiosurgery has been documented to be an effective treatment modality for a variety of malignant tumors of the spine and spinal cord.8–27 Because the life expectancy of most of these patients is measured in months and radiation injury can take years to become evident,7 there is more controversy regarding radiosurgery for the management of benign tumors of the spine. The low radiation tolerance of the spinal cord is the primary limiting factor in the dose of therapeutic ionizing radiation that can be used for spinal tumors. Conventional external beam radiotherapy lacks the precision to deliver large single-fraction doses to benign spinal tumors in close proximity to the spinal cord. In fact, the radiosensitivity of the spinal cord has often required that the treatment dose be far below the optimal therapeutic dose.23,50,51 Spinal radiosurgery aims to deliver a highly conformal radiation dose to the treatment volume, which should increase the likelihood of successful tumor control and minimize the risk for spinal cord injury.8,9,23,25 Especially in patients with greater survival and a longer life span because of benign spinal tumor histology, the potential for delayed radiation-induced myelopathy must be entertained. The intradural nature of benign spinal tumors engenders a close proximity between the tumor and spinal cord or cauda equina, an anatomic relationship that may influence the risk for neural toxicity. Moreover, because benign spinal tumors are prone to late recurrence and the late toxicity of radiation delivered to the spinal cord may take years to develop, evaluation of radiosurgical treatment in terms of efficacy, safety, and durability will necessitate longer follow-up than what has been granted to patients with metastatic spine tumors.52 Finally, because benign spinal tumors are rarely life-threatening or destabilizing to the spinal column, treatment with high-dose ionizing radiation or by any other means creates more controversy than such treatment in the context of malignant spinal tumors; morbidity from any intervention, whether immediate or delayed, becomes less acceptable in this younger, healthier patient population.
Technical and Dose Considerations
Accurate geometric visualization of the tumor target on imaging is essential for radiosurgical contouring and treatment planning. The diagnostic imaging modality of choice for benign spinal tumors is magnetic resonance imaging (MRI). However, optimal radiosurgical treatment planning often rests on the modality of computed tomography (CT). Modern commercial radiosurgery systems use CT-based imaging for planning and delivery, and benign extramedullary spinal tumors are typically well visualized with contrast-enhanced CT. Most systems today also permit MRI-CT image fusion. Such fusion may improve the target definition for spinal tumors, especially when the neoplasms exhibit heterogeneous contrast enhancement. MRI-CT image fusion in the spine is usually more challenging than that required for intracranial radiosurgery because it is more dependent on the technical aspects of image acquisition. Quality spinal image fusion often requires that the patient’s imaging position closely match the intended treatment position. Issues of spatial distortion need to be considered if MRI is used directly for planning. Because signal intensities in MRI do not reflect a direct relationship with electron densities, unless attenuation coefficients are manually assigned to the region of interest, spatial distortion limits the accuracy of using MRI directly for radiosurgical planning.53 Virtually all extramedullary spinal tumors show some degree of contrast enhancement, so contrast-enhanced CT is sometimes used to directly define the target. CT myelography is an alternative imaging strategy that can provide superior visualization of the tumor and spinal cord. For spinal AVMs, the use of three-dimensional angiography has improved the ability to visualize these vascular lesions.54
The goal of spinal radiosurgery for benign spinal tumors is to deliver a clinically significant radiation dose to the tumor via a plan that respects the radiation dose tolerance limits of the nearby spinal cord, cauda equina, and surrounding organs such as the intestines, esophagus, kidneys, larynx, and liver. By virtue of their origin along the dura and spinal nerve roots, extramedullary spinal tumors can significantly impinge on the spinal cord or cauda equina and thus make contouring difficult. The single-fraction doses prescribed for spinal radiosurgery have varied from 12 to 20 Gy (Table 263-1).7,8,14,55–57 Although spine radiosurgery is often performed with a single–radiation fraction technique, the formal definition includes hypofractionated dosing plans with a maximum of five treatment sessions. Doses as high as 30 Gy have been reported in the setting of hypofractionation, and even higher total cumulative doses have been administered when one considers patients who have undergone conventional radiotherapy before spinal radiosurgery. The linear quadratic equation remains the most accepted mathematical model of cell kill secondary to ionizing radiation.47 The equation has been used to compare various dose fractionation schedules.
The spinal cord is one of the most radiosensitive structures considered in radiosurgical treatment plans. Radiation-induced myelopathy may occur in delayed fashion, and the spinal cord tolerance of single-fraction radiation has not been carefully determined with long-term follow-up studies. Nevertheless, some data point to less than a 5% probability of myelopathy at 5 years when the cord receives a 60-Gy dose via standard fractionation.47 Most centers have tailored plans to ensure that the spinal cord is exposed to no more than 20 Gy during fractionated therapy.58 Borrowing again from the intracranial radiosurgical experience, some physicians have fashioned a cord avoidance strategy based on dosage tolerance data for the optic nerves and chiasm. In contrast to patients with metastatic cancer, patients undergoing radiosurgery for benign paraspinal neoplasms are expected to survive longer with higher functional status. Consequently, when planning treatment of benign tumors, it is prudent to err on underestimating spinal cord tolerance in the event that an unrecognized form of radiation-induced myelopathy may become manifested over a period of decades. Whether previous radiosurgery will sensitize the spinal cord and cauda equina to degenerative insults in aging patients is also unknown.
Clinical Results
There is a growing amount of experience to date with clinical outcomes after spine radiosurgery for benign tumors.7,8,14,53,59,60 Similar to the application of radiosurgery for intracranial pathology, spine radiosurgery has primarily been embraced as a adjunctive treatment technique for tumor recurrence or residual tumor after open surgery, for progression of tumor after failure of conventional radiation treatment, in patients who have significant medical comorbid conditions that preclude open surgery, or for “salvage” therapy when further conventional irradiation or surgery is not appropriate. Spinal radiosurgery will probably with time be explored as the initial treatment of both malignant and benign spinal tumors.
Given their pathologic similarities, it has been speculated that benign spinal lesions would be equally responsive to radiosurgery as their intracranial counterparts.7 It is for this reason that centers experienced in radiosurgery explored this developing technology to treat benign spine tumors. The extramedullary intradural spinal neoplasms treated with radiosurgery have primarily included meningiomas, schwannomas, and neurofibromas. However, the first reported benign spinal tumor treated with radiosurgery was actually a hemangioblastoma.7 Ryu and coauthors published the first description of radiosurgery used to treat such lesions.23 In the initial experience from Stanford using the CyberKnife for 15 benign spinal tumors, no tumor progression was reported with a follow-up period of 12 months.
Dodd and associates, in the largest published series to date for benign intradural extramedullary spinal tumors, reported the results of radiosurgical treatment of 55 such tumors (30 schwannomas, 9 neurofibromas, and 16 meningiomas).7 Total treatment doses ranged from 16 to 30 Gy delivered in one to five fractions to tumor volumes that varied from 0.136 to 24.6 cm3. Less than 1 year after radiosurgery, three patients (one with meningioma, one with schwannoma, and one with neurofibroma) required open surgical resection of their tumor because of persistent or worsening symptoms. Only 1 of these 3 lesions was larger radiographically. Twenty-eight of the 55 lesions had greater than 24 months’ follow-up. The mean follow-up in this group was 36 months. All lesions in this group were either stable (61%) or smaller (39%), with no tumor increasing in size. The following sections, subdivided by histopathology, summarize the available clinical outcome data for benign spinal tumor radiosurgery (Figs. 263-1 and 263-2).
