Stereotactic radiosurgery for spinal metastases from renal cell carcinoma

Transcription

1 J Neurosurg Spine 3: , 2005 Stereotactic radiosurgery for spinal metastases from renal cell carcinoma PETER C. GERSZTEN, M.D., M.P.H., STEVEN A. BURTON, M.D., CIHAT OZHASOGLU, PH.D., WILLIAM J. VOGEL, R.T.T., WILLIAM C. WELCH, M.D., JOSEPH BAAR, M.D., AND DAVID M. FRIEDLAND, M.D. Departments of Neurological Surgery, Radiation Oncology, and Oncology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Object. The role of stereotactic radiosurgery in treating renal cell carcinoma (RCC) metastases to the spine has previously been limited. In this study the authors evaluated the clinical outcome in patients with spinal RCC who underwent single-fraction radiosurgery. Methods. Forty-eight patients with 60 RCC metastases to the spine (six cervical, 26 thoracic, 18 lumbar, and 10 sacral) were treated with a single-fraction radiosurgery technique and were followed for a period of 14 to 48 months (median 37 months). All patients were successfully treated in an outpatient setting. The tumor volume ranged from 5.5 to 203 cm 3 (mean 61.9 cm 3 ). Forty-two of the total 60 lesions had been previously treated with external-beam radiation therapy (EBRT). The maximum tumor dose was maintained at 17.5 to 25 Gy (mean 20 Gy). The volume of the spinal cord exposed to greater than 8 Gy ranged from 0.01 to 3 cm 3 (mean 0.64 cm 3 ); the volume of the spinal canal at the cauda equina level exposed to greater than 8 Gy ranged from 0.01 to 2.2 cm 3 (mean 0.65 cm 3 ). No radiation-induced toxicity occurred during the follow-up period. Axial and radicular pain improved in 34 (89%) of 38 patients who were treated primarily for pain. Tumor control was demonstrated in seven of eight patients treated primarily for radiographically documented tumor progression. In time six patients required open surgical intervention for tumor progression that had caused neurological dysfunction after radiosurgery. Conclusions. Spinal radiosurgery can be a successful therapeutic modality for the delivery of large-dose single-fraction radiation to RCC spinal metastases that are often poorly controlled with conventional EBRT modalities. KEY WORDS renal cell carcinoma spinal metastasis CyberKnife image-guided surgery robotic surgery stereotactic radiosurgery R Abbreviations used in this paper: CT = computerized tomography; EBRT = external-beam radiation therapy; IMRT = intensitymodulated RT; LINAC = linear accelerator; MR = magnetic resonance; RCC = renal cell carcinoma; SRS = stereotactic radiosurgery; VB = vertebral body. 288 ENAL cell carcinoma accounts for approximately 1% of all cancers; there are approximately 30,000 new cases in the US annually. 6,28,49 In 30% of patients with RCC distant metastases will be observed at the time of diagnosis, and metastatic disease will eventually develop in 50%. 45 Additionally, metastases to the spine will ultimately develop in 30% of patients with RCC. The spine represents the most common site of osseous involvement of RCC. 25,37,61,62 Spinal cord compression and instability are major causes of morbidity in these patients, manifesting as pain, paraplegia, and incontinence. 52 Furthermore, the overall incidence of RCC is increasing for unknown reasons. 27 The role of radiotherapy in the treatment of spinal metastatic tumors is well established and is often the initial treatment modality; 20,25,35,46,59 however, RCC has traditionally been considered a relatively radioresistant tumor, especially when standard fractionated EBRT modalities are used. 12,62 Renal cell metastases to the spine often progress despite the provision of irradiation, and surgical intervention may be necessary. Spinal metastases from RCC, however, are considered to be difficult to treat using open surgery because of their vascularity, and their surgical treatment is often associated with a higher complication rate than other spinal metastases. 9,10,37,61,62 In the treatment of brain metastases specifically from RCC, the authors of several clinical studies have described a significant benefit of SRS compared with conventional whole-brain fractionated EBRT. 5,11,28,49,53,60 In our previous work we have demonstrated both the feasibility and clinical efficacy of spinal radiosurgery for various benign and malignant lesions when using an image-guided frameless SRS delivery system (CyberKnife; Accuray, Inc., Sunnyvale, CA) ,26 The purpose of this study was to evaluate the clinical outcomes in patients who underwent radiosurgery for the treatment of RCC metastases to the spine to determine if this modality yields efficacious results associated with its use in intracranial RCC metastases.

2 Spinal radiosurgery for renal cell metastases Clinical Material and Methods This study involved the prospective evaluation of 60 lesions in 48 consecutive patients with RCC metastatic to the spine who underwent CyberKnife radiosurgery in which the Dynamic Tracking System (version 3.0; Accuracy, Inc.) was used. All patients were treated at the University of Pittsburgh Medical Center and the protocol was approved by the university s institutional review board. The study was designed to test the hypothesis that spinal radiosurgery is safe, feasible, and clinically effective for the treatment of spinal metastatic RCC. There were 28 men and 20 women who ranged in age from 45 to 84 years (mean 62 years). Table 1 provides a summary of the characteristics of the treatment group, and Table 2 provides a summary of the primary indications for spinal SRS that were used for patient selection in this study. Exclusion criteria for CyberKnife radiosurgery were: 1) evidence of overt spinal instability; and 2) neurological deficit resulting from osseous compression of neural structures. The CyberKnife System The CyberKnife consists of a 6-MV compact LINAC that is smaller and lighter in weight than LINACs used in conventional radiotherapy. 2,29,50,51 Its smaller size allows the device to be mounted on a computer-controlled sixaxis robotic manipulator that permits a much wider range of beam orientations than can be achieved with conventional radiotherapy devices. 4,14,16,58 Two diagnostic radiographic cameras are positioned orthogonally (90 offset) to acquire real-time images of the patient s internal anatomy during treatment. The images are processed to identify radiographic features (skull bone landmarks or implanted fiducials) and then automatically compared with the CT treatment planning study. The precise tumor position is communicated through a real-time control loop to a robotic manipulator, which aligns the radiation beam with the intended target. 58 In an analysis of the accuracy of the CyberKnife system, one group found that the machine has a clinically relevant accuracy of mm when using a 1.25-mm CT scan slice thickness. 16 Overview of Treatment TABLE 1 Characteristics obtained in 48 patients with 60 RCC metastases undergoing single-dose SRS Characteristic No. of Patients/Value age (yrs) mean 62 range sex male 28 female 20 no. of cases of multiple lesions 6 prior EBRT 42 levels treated cervical 6 thoracic 26 lumbar 18 sacral 10 skull tracking 6 fiducial tracking 54 tumor vol (cm 3 ) mean 61.9 range max tumor dose (Gy) mean 20 range mean vol receiving 8 Gy (cm 3 ) spinal canal 0.64 cauda equina 0.65 CyberKnife spinal radiosurgery consists of three distinct components: 1) CT scan acquisition based on skull bone landmarks or implanted bone fiducials; 2) treatment planning; and 3) the treatment itself. 22,24,26 All cervical lesions to the C-7 level were tracked relative to skull bone landmarks. All patients with cervical lesions were fitted with a noninvasive molded Aquaplast facemask (WRF/Aquaplast Corp., Wyckoff, NJ) that stabilized the head and neck on a radiographically transparent headrest. In all other cases the lesions were tracked relative to fiducial markers placed within the bone adjacent to the lesion. Because these implanted fiducials have a fixed relationship with the bone in which they are implanted, any movement in the vertebrae would be detected as movement in the fiducials, and this movement is detected and compensated for by the Cyber- Knife. Computerized tomography scans were acquired using 1.25-mm-thick slices from the top of the skull to the bottom of the cervical spine. In all other lesions performed fluoroscopically guided percutaneous placement of four to six gold fiducial markers (Alpha-Omega Services, Inc., Bellflower, CA) into the pedicles immediately adjacent to the lesion to be treated; this was undertaken using a standard Jamshidi bone marrow biopsy needle (Allegiance Healthcare Corp., McGraw Park, IL) as previously described. 26 The fiducial placement procedure was performed in the operating room in an outpatient setting prior to undertaking the CT planning, which was performed using the aforementioned slice thickness. The patient was placed in a conformal alpha cradle during CT scanning and during treatment. We obtained 1.25-mm-thick CT slices that included the lesion of interest and all fiducials. The second component of the CyberKnife treatment is the development of the radiosurgical treatment plan. Each treatment plan was devised jointly by a team of neurosurgeons, radiation oncologists, and radiation physicists. In each case, the radiosurgical treatment plan was designed based on tumor geometry, its proximity to spinal cord, and its location. Treatment planning was performed using the Accuray treatment planning system (Dynamic Tracking System running version 3.0 software). The tumor dose was determined based on spinal cord tolerance and quantity of previously administered radiation. Dose prescription was based on currently accepted single-fraction doses used for intracranial sites. We also considered cases involving prior radiation exposure and those with lesions close to the spinal cord. The prescription dose was independent of the tumor volume. In each case, the spinal cord and/or cauda equina was outlined as a critical structure. At the level of the cauda equina, the spinal canal was outlined. Therefore, at the level of the cauda equina, the critical volume is the entire spin- 289

3 P. C. Gerszten, et al. TABLE 2 Primary clinical indications for SRS in 48 patients with RCC metastases* al canal and not actual neural tissue. The maximum dose was defined as that delivered to a single pixel. Several patients had previously undergone resection and placement of stabilizing hardware; thus, imaging artifacts were present. On axial images exhibiting a significant amount of spray artifact, the spinal cord and/or cauda equina volume was contoured by interpolation from the spinal cord or cauda equina rosral and caudal to the instrumentation-produced artifact. In some cases, myelography allowed for the critical structures to be more easily identified. The third component of the CyberKnife treatment is the actual treatment delivery. All treatments were performed using a single-fraction technique. The patients were placed supine on the CyberKnife treatment table with the appropriate immobilization device. Preoperative analgesic, sedation, or steroid agents were not routinely given. During the treatment, real-time digital radiographs were obtained. The location of the VB being treated was established from these images and was used to determine tumor location as previously described. The patient was observed throughout the treatment by closed-circuit television. Results No. of Cases W/ Favorable Indication Overall Outcome pain imaging-based tumor progression 8 7 primary treatment modality 6 6 postop radiation boost 6 6 progressive neurological deficit 2 0 * Tabular values reflect the overall number of lesions (60), not patients. Table 1 provides a summary of clinical characteristics and treatment of the 60 lesions in 48 patients. The followup duration ranged from 14 to 48 months (median 37 months). The tumor dose was maintained at 14 to 20 Gy to the 80% isodose line which was contoured to the edge of the target volume (mean 16 Gy). The planning treatment volume was defined as the gross tumor volume within the margin. The dose was prescribed to the 80% isodose line, which covered the planning treatment volume in all cases. The maximum intratumoral dose ranged from 17.5 to 25 Gy (mean 20 Gy). In six patients (all those in whom lesions were limited to the cervical spine) osseous landmarks were used for image guidance. In the remaining 54 cases (those with thoracic, lumbar, and sacral lesions) fiducial tracking was used. Tumor volume ranged from 5.5 to cm 3 (mean 61.9 cm 3 ). Forty-two of the 60 lesions had been treated previously with EBRT with spinal cord doses thought to preclude further conventional EBRT. The tumor dose was not decreased in a uniform manner in these previously irradiated areas. Instead, the maximum dose to the spinal cord or cauda equina was more strictly limited, as allowed for by the CyberKnife s inverse treatment planning system. The combination of a steep dose gradient and high conformality of the CyberKnife treatment allows for such high doses to be delivered so close to the adjacent critical structures (that is, the spinal cord). The doses delivered to tumors at the level of the spinal cord and the cauda equina were analyzed separately. The mean maximum dose delivered to cervical and thoracic tumors (32 cases) at the level of the spinal cord was 20 Gy; the mean volume in these 32 cases was 40.1 cm 3. The dose received by the spinal cord during the treatment ranged from 2.42 to Gy (mean 9.68 Gy). The volume of the spinal cord or cauda equina receiving greater than 8 Gy was recorded for each patient. The spinal cord volume receiving greater than 8 Gy in these same patients ranged from 0.01 to 3.06 cm 3 (mean 0.64 cm 3 ). The mean maximum dose delivered to lumbar and sacral tumors (28 cases) at the level of the cauda equina was 19 Gy and the mean volume in these 28 cases was 85.8 cm 3, twice that of the tumors at the level of the spinal cord. The intraoperative dose received by the cauda equina ranged from 0.51 Gy to Gy (mean 9.15 Gy). The volume of the cauda equina receiving greater than 8 Gy for these same patients ranged from 0.01 to 2.2 cm 3 (mean 0.65 cm 3 ). Given the kidneys relative radiosensitivity, a limit of 2 Gy was set as the maximum dose received by each. Furthermore, in many patients only a single kidney remained. Thirty-one patients had previously undergone a nephrectomy. Within the cohort, no patient experienced an exacerbation of symptoms, hemorrhage, or new neurological deficit in the immediate posttreatment period. There were no cases of radiation-induced myelopathy or radiculopathy during the follow-up period. The primary indications for CyberKnife treatment and overall clinical outcomes are listed in Table 2. The primary indication for radiosurgery in 38 patients was significant pain secondary to the lesion. Thirty-seven (95%) of 38 patients reported an improvement in pain measured using a 10-point pain scale at the 1-month follow-up examination. Long-term pain improvement was demonstrated in 34 cases (89%). In 28 of the 38 cases treated primarily for pain, the symptomatic lesion had received prior irradiation. There was no association between prior irradiation and symptomatic improvement after radiosurgery. Analgesic medication usage was also documented. Three patients underwent a repeated Cyber- Knife treatment for recurrence of same-level pain at 6, 7, and 18 months after their initial SRS. The maximum treatment doses in these three cases were 17.5 Gy for two and 20 Gy for the third. Eight patients were treated primarily for an increase in tumor size demonstrated on follow-up imaging; there was no associated neurological deficit or significant pain. All eight patients (eight lesions) had previously undergone EBRT. During a follow-up period ranging from 20 to 29 months, the lesions in seven of these patients were stable at the location of treatment, with none progressing in size within the treatment volume on MR imaging. In the remaining patient, an elective T-10 corpectomy was performed 11 months after SRS to treat progressive kyphosis. In six cases, spinal SRS was performed as the primary treatment modality for the lesion. These lesions were usually asymptomatic and were observed on imaging studies acquired in patients who were undergoing open decom- 290

4 Spinal radiosurgery for renal cell metastases pression surgery for other symptomatic lesions. Fiducial markers were implanted adjacent to these asymptomatic lesions and would be used as landmarks when the patients recovered from their surgical procedure, to avoid a protracted course of EBRT. There were no cases of wound healing problems associated with radiosurgery in these open surgery treated cases. In one case, fiducials were implanted at the time of percutaneous biopsy sampling in a patient with previously unknown metastatic disease. The lesion had been discovered on routine imaging for another indication. Spinal SRS was undertaken in five patients who had previously undergone an open decompressive surgery. All five had previously undergone EBRT of the lesion prior to progression of the neoplasm that caused neurological deficit and subsequent need for surgery. In three cases, follow-up imaging had revealed tumor progression at the resection site of the lesion. In all five cases, surgical reexploration was thought to be contraindicated, and spinal radiosurgery was elected. Reexploration was believed to be contraindicated because of the poor performance status documented in two patients, previously placed posterior spinal instrumentation that would have complicated a second surgery because of the anterior compression of the thecal sac in two patients, and medical comorbidities that precluded a transthoracic corpectomy in one patient. In four patients primarily treated with radiosurgery for pain neurological deficits subsequently developed after the treated tumor progressed during the follow-up period (range months, median 37 months). All four patients underwent open decompressive surgery. Progression to complete paraplegia occurred during the follow-up period in two patients with lesions treated with radiosurgery. Two patients underwent SRS for progressive neurological deficit as their primary indication. In both patients spinal cord compression was caused by myelopathy. Myelopathy initially resolved in each but later progressed, and the patients required open decompressive surgery and instrumentation-augmented fixation within 6 months of the CyberKnife treatment. In both cases the maximum tumor dose was 17.5 Gy. In both of these patients progression to paraparesis ultimately occurred. The mean tumor volume for these six patients with post-srs tumor progression was cm 3, 40% greater than the overall mean tumor volume of 61.9 cm 3. Case 1 Illustrative Cases This 59-year-old woman with RCC had previously undergone 14 fractions of 35 Gy EBRT and had experienced temporary improvement in symptoms. Five months after EBRT she presented with recurrent severe left sacral pain as well as pain and paresthesias radiating into her left leg. A percutaneous fiducial marker was placed 1 week prior to radiosurgery. The radiosurgery plan was designed to treat the tumor with a prescribed dose of 14 Gy that was calculated to the 80% isodose line (Fig. 1). The maximum tumor dose was 17.5 Gy, and the tumor volume was 157 cm 3. The spinal canal received a maximum dose of 9.9 Gy. The patient experienced complete pain relief within 1 month that has lasted 30 months. FIG. 1. Case 1. Isodose lines of the treatment plan obtained in a 59-year-old woman with a sacral RCC metastasis. She had previously undergone 35-Gy EBRT in 14 fractions and had experienced improvement of symptoms for 5 months. Thereafter, she presented with recurrent severe left sacral pain and pain and paresthesias radiating into her left leg. The treatment plan was designed to treat the tumor with a prescribed dose of 14 Gy to the 80% isodose line. The maximum tumor dose was 17.5 Gy, and the tumor volume was 157 cm 3. The spinal canal received a maximum dose of 9.9 Gy. The patient experienced complete pain relief within 1 month, and this was maintained for 30 months. Follow-up imaging has revealed no evidence of tumor progression. Case 2 This 57-year-old woman harbored T-11 and T-12 RCC metastases. She had undergone surgical treatment of a symptomatic T-6 metastasis resulting in a pathological fracture. At the time of that procedure, fiducial markers were percutaneously implanted. The T-11 and T-12 metastases had not received prior radiation, and these were the only other areas of spinal involvement demonstrated on MR imaging. The intention was to treat these lesions definitively by using primary single-fraction radiosurgery. The patient was neurologically intact. The radiosurgery plan was designed to treat the tumors with a prescribed dose of 18 Gy that was calculated to the 80% isodose line (Fig. 2). The maximum tumor dose was 2.5 Gy, and the tumor volume was 54 cm 3. The spinal cord received a maximum dose of 9.8 Gy, and 0.04 cm 3 of the spinal cord received greater than 8 Gy. Follow-up MR imaging revealed no evidence of tumor progression at 11 months after CyberKnife treatment, and the patient remains asymptomatic. Discussion Despite advances during the past decade in the surgical treatment of patients with localized RCC, 50% of these 291

5 P. C. Gerszten, et al. patients initially present with metastases or metastases develop after nephrectomy for Stages I and II tumors. 38,61 In 30% of patients with RCC spinal metastases will ultimately develop. The spine represents the most common site of osseous involvement for RCC. 26,37,62 Spinal cord compression and spinal instability due to metastases are major causes of morbidity among such patients. 52 Standard treatments for RCC metastases to the spine include radiotherapy alone, radiotherapy combined with systemic chemotherapy, or decompressive surgery and/or spinal stabilization followed by radiotherapy. 9,10,25,30,37,38,40, 41,52,56,61,62 Most patients with RCC metastatic to the spine may be treated by nonoperative measures. The goals of local radiotherapy in the treatment of these spinal tumors have been palliation of pain, prevention of pathological fractures, and cessation of progression or reversal of neurological compromise; however, RCC is a relatively radioresistant tumor and failure to respond to radiotherapy or progressive growth after irradiation is not uncommon. 30,61 When a spinal tumor causes spinal cord or other neural element compression, decompressive surgery is often necessary with or without the placement of spinal fixation based on the extent of spinal column destruction and spinal instability. 25 Indications for surgery in patients with metastatic spinal disease include intractable pain, spinal instability, or progressive neurological dysfunction secondary to dural compression in cases in which radiotherapy has failed. 37 Patients with extensive osseous destruction of the spinal column should undergo decompressive surgery and stabilization to correct or prevent progressive spinal deformity. Although resection of RCC spinal metastases and stabilization have been shown to improve patients neurological function and pain, these lesions are difficult to treat because of their vascularity. 9,10,37,61 Conventional EBRT for RCC spinal metastases is usually administered in 10 to 20 daily fractions to a total dose of 30 to 40 Gy. 20,30,46 A primary factor that limits radiation dose in this conventional radiotherapy is the relatively low tolerance of the spinal cord to radiation. 42 Conventional EBRT is not precise enough to deliver large single-fraction doses of radiation to the spine near radiosensitive structures such as the spinal cord. It is the spinal cord s low tolerance to radiation that often confines the treatment dose to one that is far below the optimal therapeutic dose. 20,43,58 Precise limitation of the radiation dose to the treatment volume, as is the case for intracranial radiosurgery, should maximize the control of RCC spine metastases. Since Hamilton, et al., 31 first described the possibility of LINAC-based spinal SRS in 1995, investigators at multiple centers have undertaken large-fraction conformal irradiation spinal lesions by using various technologies. 7,13,19,33, 48,55 Other authors have reported the effectiveness of protons for spinal and paraspinal tumors. 34 In recent studies involving hypofractionated or single-dose treatments for spinal metastases, investigators have reported results that were comparable with those of conventional fractionation. 7,13,18,33 With advances in conformal treatment techniques involving multileaf collimators such as IMRT, Chang, et al., 13 found that intensity-modulated, near-simultaneous, CT scanning guided stereotactic radiotherapy was a feasible and highly precise modality for the noninvasive treatment of spinal metastases. Bilsky, et al., 7 found successful tumor control in 13 of 15 patients with spinal metastases who underwent IMRT. Milker-Zabel and colleagues 48 documented tumor control in 95% of their patients. 48 The CyberKnife was first developed for the treatment of benign and malignant intracranial lesions. 3,4,17 Treatment outcome has been similar to that associated with conventional frame-based radiosurgery. 58 With the ability to treat lesions outside of the skull via fiducial tracking, CyberKnife treatment of spinal lesions has emerged. 22,50,58 Because of the spatial precision with which the Cyber- Knife can deliver radiation to the spine, it is theoretically feasible to administer a tumoricidal radiation dose in a single fraction similar to that in intracranial radiosurgery. In SRS, high-dose radiation is delivered in a single fraction to a well-defined intra- or extracranial target. 64 Stereotactic radiosurgery has become an important treatment modality in the management of a wide variety of intracranial and, more recently, extracranial lesions. It has been demonstrated to be an effective treatment for brain metastases, either with or without whole-brain radiotherapy, with an 85 to 95% control rate. 15,21,39,44 The emerging technique involved in spinal radiosurgery represents a logical extension of the current state-of-the-art radiotherapy. Unlike conventional radiotherapy that delivers a full dose to both the VB and the spinal cord, the CyberKnife can deliver a single high-dose fraction of radiation to the targeted tissue while sparing most of the adjacent spinal cord. The treatment plan can create a high gradient dose falloff to the targeted tissue that should significantly reduce the possibility of radiation-induced myelopathy. This is one advantage of SRS for spinal tumors. It therefore has the potential for significantly improving local control of spinal metastases, which can in turn translate into more effective palliation, delaying progression of neurological deficits requiring open surgical intervention, and potentially leading to longer survival. There have been no large-scale studies involving spinal radiosurgery or hypofractionated radiotherapy in which their respective optimal doses have been established. At other centers, in which clinicians use intensity-modulated near-simultaneous CT scanning guided SRS, 6- to 30-Gy doses are delivered in one to five fractions. 7,13,57 One difference between our technique and the use of IMRT for conformal spinal radiotherapy is the delivery of the dose in a single fraction rather than in a fractionated treatment scheme. There is simply not enough literature-based evidence regarding the treatment of spinal RCC metastases to allow direct comparison of clinical outcomes obtained using different radiation delivery technologies. Our overall clinical outcomes are similar to those reported by others who have performed hypofractionated radiotherapy. 7,47, 57,58 In our series, the maximal tumor dose was maintained at 17.5 to 25 Gy and delivered in a single fraction. Spinal radiosurgery was found to be safe when the doses were comparable with those used in intracranial radiosurgery, without any incidence of radioinduced neural injury. There were no cases of radiation-induced toxicity during a follow-up period of 14 to 48 months (median 37 months), a duration long enough for such events to have occurred. 1,8,32,36,47,54,63 In this series, pain was the primary indication of radiosurgery treatment. This, of course, is different from the primary indication for intracranial radiosurgery for RCC metastases. Ninety-five percent of patients reported im- 292

6 Spinal radiosurgery for renal cell metastases FIG. 2. Case 1. Axial (left) and sagittal (right) projections of the treatment plan obtained in a 57-year-old woman with T-11 and T-12 RCC metastases. The patient had undergone surgical treatment of a symptomatic T-6 metastasis resulting in a pathological fracture. The T-11 and T-12 metastases had not been previously irradiated and were the only other areas of spinal involvement. The treatment plan was designed with a prescribed dose of 18 Gy to the 80% isodose line. The maximum tumor dose was 22.5 Gy, and the tumor volume was 54 cm 3. The spinal cord received a maximum dose of 9.8 Gy, and 0.04 cm 3 of the spinal cord received greater than 8 Gy. Note the conformality of the isodose line around the spinal cord. provement in their pain after radiosurgery; at the latest follow-up examination, this decreased to 89%. This is similar to the success reported by others who performed hypofractionated radiotherapy. 7,48,57,58 Conventional EBRT may yield less than optimal pain relief because the total dose is limited by the tolerance of adjacent tissues (for example, the spinal cord). Spinal radiosurgery may improve pain control and increase the duration of pain control because of its larger overall radiobiological doses. In the three cases in which the patients pain recurred after radiosurgery, the maximum tumor doses were in the lower range of 17.5 Gy. Perhaps higher radiation doses will translate into better long-term pain control in this patient population. A direct comparison of efficacy between this patient cohort and clinical outcomes achieved using conventional radiotherapy cannot be performed because radiosurgery was chosen in the majority of cases because prior irradiation precluded further conventional radiation therapy or because the patients were not candidates for open surgery techniques. In the absence of radiation-induced toxicity, our trend has been to increase the tumor prescription dose. The mean maximum dose delivered to the tumors at the level of the cauda equina was 19 Gy, whereas that delivered to the tumors at the spinal cord was only slightly greater, at 20 Gy. We are somewhat more comfortable with treating cauda equina lesions with higher doses even if the cauda equina is subjected to more overall radiation than the spinal cord when the same dose is administered. The maximum dose to the cauda equina has exceeded 14 Gy without adverse consequences; however, the mean volume of cauda equina lesions was 86 cm 3 compared with that of 40 cm 3 for spinal cord lesions. This twofold increase in tumor volume is not surprising given the larger size of the lumbar VBs and sacrum compared with the cervical and thoracic VBs. In the treatment plans for these larger-volume lesions a slight decrease in maximum tumor dose was required. Furthermore, the treatment plans involving lumbar lesions were further complicated by the restriction of the kidneys or often a single remaining kidney as well as the limitations of possible bowel toxicity. Six patients eventually required open decompressive surgery and stabilization after spinal radiosurgery during the 48-month follow-up period (median 37 months). Four of these patients had been treated initially for pain, whereas the other two were treated for progressive neurological dysfunction. In the latter two patients, myelopathy initially resolved but it later progressed, necessitating open decompressive surgery within 6 months. In both of these cases, the maximum tumor dose of 17.5 Gy was administered. In the absence of delayed radiation-induced toxicity, we have continued to increase the maximum dose delivered to the tumor, and further experience will show if treatment failures can be delayed or possibly prevented as the dose is increased. Careful follow-up examination will be necessary to ensure that delayed-onset toxicity does not occur in these patients. Because patients with RCC survive longer because of the disease s more successful systemic treatment, delayed radiation-induced toxicity will be more of a concern. In the six cases with post-srs tumor progression the mean tumor volume was cm 3, which was 40.7 cm 3 greater than the overall mean volume of 61.9 cm 3. The overall mean tumor volume of these lesions was therefore 40% greater than the others. Furthermore, these treatment failures often occurred at the margin of the contoured treatment volume. Additional experience has led us to increase the planned treatment volume to include more of the adjacent normal-appearing VB. In the future, the capacity may exist to dose paint the dose, which is to say to deliver 293

7 P. C. Gerszten, et al. higher doses to hypoxic portions of the tumor or area of active growth based on functional imaging studies. One clinical advantage of single-fraction radiosurgery is that the treatment can be completed in a single day rather than over a course of several weeks, which is not inconsequential for patients with a limited life expectancy. The modality may be useful to capitalize on possible advantages of radiosensitizers. In addition, patients with cancer may have difficulty visiting an oncology facility for a prolonged period of daily fractionated therapy. A large single fraction of radiation may be more radiobiologically advantageous in treating certain tumors such as RCC compared with prolonged fractionated radiotherapy. A clinical response such as reduced pain or improved neurological function may also be more rapidly elicited by this form of radiosurgery. Such clinical outcomes are difficult to measure in an outpatient setting. Finally, compared with open surgery the procedure is minimally invasive and can be performed in an outpatient setting. In cases in which the lesions were implanted with fiducials at the time of initial surgery, there is no need to endure the prolonged period normally required for wound healing. Given the negligible dose of radiation to the fascia and skin, radiosurgery can be undertaken without delay soon after open surgery. Conclusions In this study with long-term follow-up data, we found that single-fraction spinal SRS for RCC metastases is safe and provides good symptomatic response. The major potential benefits of SRS of spinal lesions are the relatively short treatment time in an outpatient setting, potentially better local tumor control, and minimal risk of side effects. Such factors may translate into better palliation of symptoms and a longer survival period, and the morbidity associated with open surgery of these highly vascular lesions is potentially avoided. In addition, this modality allows for the treatment of lesions previously irradiated using conventional EBRT. This modality offers an important new therapy for spinal RCC metastases. Further experience with higher-dose irradiation and improved tumor imaging will likely lead to even better clinical outcomes. References 1. Abbatucci JS, Delozier T, Quint R, Roussel A, Brune D: Radiation myelopathy of the cervical spinal cord: time, dose and volume factors. Int J Radiat Oncol Biol Phys 4: , Adler JR Jr, Chang SD, Murphy MJ, Doty J, Geis P, Hancock SL: The CyberKnife: A frameless robotic system for radiosurgery. Sterotact Funct Neurosurg 69: , Adler JR, Cox RS, Kaplan I, Martin DP: Stereotactic radiosurgical treatment of brain metastases. J Neurosurg 76: , Adler JR Jr, Murphy MJ, Chang SD, Hancock SL: Image-guided robotic radiosurgery. Neurosurgery 44: , Amendola B, Wolf AL, Coy SR, Amendola M, Bloch L: Brain metastases in renal cell carcinoma: management with gamma knife radiosurgery. Cancer J 6: , Bennington J: Proceedings: Cancer of the Kidney: etiology, epidemiology and pathology. Cancer 31: , Bilsky MH, Yamada Y, Yenice KM, Lovelock M, Hunt M, Gutin PH, et al: Intensity-modulated stereotactic radiotherapy of paraspinal tumors: a preliminary report. Neurosurgery 54: , Boden G: Radiation myelitis of the cervical spinal cord. Br J Radiol 21: , Bowers TA, Murray JA, Charnsangavej C, Soo CS, Chuang VP, Wallace S: Bone metastases from renal carcinoma. The preoperative use of transcatheter arterial occlusion. J Bone Joint Surg Am 64: , Broaddus WC, Grady MS, Delashaw JB Jr, Ferguson RD, Jane JA: Preoperative superselective arteriolar embolization: a new approach to enhance resectability of spinal tumors. Neurosurgery 27: , Brown PD, Brown CA, Pollock BE, Gorman DA, Foote RL: Stereotactic radiosurgery for patients with radioresistant brain metastases. Neurosurgery 51: , Cannady SB, Cavanaugh KA, Lee SY, Bukowski RM, Olencki TE, Stevens GH, et al: Results of whole brain radiotherapy and recursive partitioning analysis in patients with brain metastases from renal cell carcinoma: a retrospective study. Int J Radiat Oncol Biol Phys 58: , Chang EL, Shiu AS, Lii M-F, Rhimes LD, Mendel E, Mahajan A, et al: Phase I clinical evaluation of near-simultaneous computed tomographic image-guided stereotactic body radiotherapy for spinal metastases. Int J Radiat Oncol Biol Phys 59: , Chang S, Adler J: Current status and optimal use of radiosurgery. Oncology 15: , Chang SD, Adler JR Jr, Hancock SL: The clinical use of radiosurgery. Oncology 12: , Chang SD, Main W, Martin DP, Gibbs IC, Heilbrun MP: An analysis of the accuracy of the CyberKnife: a robotic frameless stereotactic radiosurgical system. Neurosurgery 52: , Chang SD, Murphy MJ, Martin DP, Adler JR Jr: Image-guided robotic radiosurgery: clinical and radiographic results with the CyberKnife, in Kondziolka D (ed): Radiosurgery: 4th International Sterotactic Radiosurgery Society Meeting, Sydney, February Basel: Karger, Vol. 3, 2000, pp Colombo F, Benedetti A, Pozza F, Zanardo A, Avanzo RC, Chierego G, et al: Stereotactic radiosurgery utilizing a linear accelerator. Appl Neurophysiol 48: , Colombo F, Pozza F, Chierego G, Casentini L, De Luca G, Francescon P: Linear accelerator radiosurgery of cerebral arteriovenous malformations: an update. Neurosurgery 34:14 21, Faul CM, Flickinger JC: The use of radiation in the management of spinal metastases. J Neurooncol 23: , Flickinger JC, Kondziolka D, Lunsford LD, Coffey RJ, Goodman ML, Shaw EG, et al: A multi-institutional experience with stereotactic radiosurgery for soliatary brain metastasis. Int J Radiat Oncol Biol Phys 28: , Gerszten PC, Ozhasoglu C, Burton S, Kalnicki S, Welch WC: Feasibility of frameless single-fraction stereotactic radiosurgery for spinal lesions. Neurosurg Focus 13(4):e2, Gerszten PC, Ozhasoglu C, Burton SA, Vogel W, Atkins BA, Kalnicki S, et al: CyberKnife frameless single-fraction stereotactic radiosurgery for benign tumors of the spine. Neurosurg Focus 14(5):e16, Gerszten PC, Ozhasoglu C, Burton SA, Vogel WJ, Atkins BA, Kalnicki S, et al: CyberKnife frameless stereotactic radiosurgery for spinal lesions: clinical experience in 125 cases. Neurosurgery 55:89 99, Gerszten PC, Welch WC: Current surgical management of metastatic spinal disease. Oncology 14: , Gerszten PC, Welch WC: CyberKnife radiosurgery for the spine. Techn Neurosurg 9: , Godley P, Ataga K: Renal cell carcinoma. Curr Opin Oncol 12: , Goyal LK, Suh JH, Reddy CA, Barnett GH: The role of whole brain radiotherapy and stereotactic radiosurgery on brain metastases from renal cell carcinoma. Int J Radiat Oncol Biol Phys 47: ,

8 Spinal radiosurgery for renal cell metastases 29. Guthrie B, Adler J: Computer-assisted preoperative planning, interactive surgery, and frameless stereotaxy. Clin Neurosurg 38: , Halperin EC, Harisiadis L: The role of radiation therapy in the management of metastatic renal cell carcinoma. Cancer 51: , Hamilton AJ, Lulu BA, Fosmire H, Stea B, Cassady JR: Preliminary clinical experience with linear accelerator-based spinal stereotactic radiosurgery. Neurosurgery 36: , Hatlevoll R, Host H, Kaalhus O: Myelopathy following radiotherapy of bronchial carcinoma with large single fractions: a retrospective study. Int J Radiat Oncol Biol Phys 9:41 44, Hitchcock E, Kitchen G, Dalton E, Pope B: Stereotactic LINAC radiosurgery. Br J Neurosurg 3: , Isacsson U, Hagberg H, Johansson KA, Montelius A, Glimelius B: Potential advantages of protons over conventional radiation beams for paraspinal tumours. Radiother Oncol 45:63 70, Kim RY, Smith JW, Spencer SA, Meredith RF, Salter MM: Malignant epidural spinal cord compression associated with a paravertebral mass: its radiotherapeutic outcome on radiosensitivity. Int J Radiat Oncol Biol Phys 27: , Kim Y, Fayos J: Radiation tolerance of the cervical spinal cord. Radiology 139: , King GJ, Kostuik JP, McBroom RJ, Richardson W: Surgical management of metastatic renal carcinoma of the spine. Spine 16: , Kjaer M, Engelholm S: The clinical course and prognosis of patients with renal cell adenocarcinoma with solitary metastasis. Int J Radiat Oncol Biol Phys 8: , Kondziolka D, Patel A, Lunsford LD, Kassam A, Flickinger JC: Stereotactic radiosurgery plus whole brain radiotherapy versus radiotherapy alone for patients with multiple brain metastases. Int J Radiat Oncol Biol Phys 45: , Kuether T, Nesbit G, Barnwell S: Embolization as treatment for spinal cord compression from renal cell carcinoma: case report. Neurosurgery 39: , Lang EK, Sullivan J: Management of primary and metastatic renal cell carcinoma by transcatheter embolization with Iodine 125. Cancer 62: , Leibel SA, Sheline GE: Tolerance of the brain and spinal cord to conventional irradiation, in Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York: Raven Press, 1991, pp Loblaw DA, Laperriere NJ: Emergency treatment of malignant extradural spinal cord compression: an evidence-based guideline. J Clin Oncol 16: , Loeffler J, Alexander E III: Radiosurgery for the treatment of intracranial metastasis, in Loeffler J, Alexander E III, Lunsford LD (eds): Stereotactic Radiosurgery. New York: McGraw- Hill, 1993, pp Maldazys JD, dekernion JB: Prognostic factors in metastatic renal carcinoma. J Urol 136: , Markoe AM, Schwade JG: The role for radiation therapy in the management of spine and spinal cord tumors, in Rea GL (ed): Spine Tumors. Rolling Meadow, IL: American Association of Neurological Surgeons, 1994, pp McCuniff AJ, Liang MJ: Radiation tolerance of the cervical spinal cord. Int J Radiat Oncol Biol Phys 16: , Milker-Zabel S, Zabel A, Thilmann C, Schlegel W, Wannenmaker M, Debus J: Clinical results of retreatment of vertebral bone metastases by stereotactic conformal radiotherapy and intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 55: , Mori Y, Kondziolka D, Flickinger JC, Logan T, Lunsford LD: Stereotactic radiosurgery for brain metastasis from renal cell carcinoma. Cancer 83: , Murphy MJ, Chang S, Gibbs I, Le QT, Martin D, Kim D: Image-guided radiosurgery in the treatment of spinal metastases. Neurosurg Focus 11(6):e6, Murphy MJ, Cox RS: The accuracy of dose localization for an image-guided frameless radiosurgery system. Med Phys 23: , O Reilly G, Kleefield J, Klein LA, Blume HW, Dubuisson D, Cosgrove GR: Embolization of solitary spinal metastases from renal cell carcinoma: alternative therapy for spinal cord or nerve root compression. Surg Neurol 31: , Payne BR, Prasad D, Szeifert G, Steiner M, Steiner L: Gamma surgery for intracranial metastases from renal cell carcinoma. J Neurosurg 92: , Phillips TL, Buschke F: Radiation tolerance of the thoracic spinal cord. Am J Roentgenol Radium Ther Nucl Med 105: , Pirzkall A, Lohr F, Rhein B, Hoss A, Schlegel W, Wannenmacher M, et al: Conformal radiotherapy of challenging paraspinal tumors using a multiple arc segment technique. Int J Radiat Oncol Biol Phys 48: , Rowe DM, Becker GJ, Rabe FE, Holden RW, Richmond BD, Wass JL, et al: Osseous metastases from renal cell carcinoma: embolization and surgery for restoration of function. Work in progress. Radiology 150: , Ryu S, Fang Yin F, Rock J, Zhu J, Chu A, Kagan E, et al: Image-guided and intensity-modulated radiosurgery for patients with spinal metastasis. Cancer 97: , Ryu SI, Chang SD, Kim DH, Murphy MJ, Le QT, Martin DP, et al: Image-guided hypo-fractionated stereotactic radiosurgery to spinal lesions. Neurosurgery 49: , Shapiro W, Posner J: Medical vs surgical treatment of metastatic spinal cord tumors, in Thompson RA, Green J (eds): Controversies in Neurology. New York: Raven Press, 1983, pp Sheehan JP, Sun MH, Kondziolka D, Flickinger JC, Lunsford LD: Radiosurgery in patients with renal cell carcinoma metastasis to the brain: long-term outcomes and prognostic factors influencing survival and local tumor control. J Neurosurg 98: , Sundaresan N, Krol G, Digiacinto C, Hughes J: Metastatic tumors of the spine, in Sundaresan N, Schmidek HH, Schiller AL, et al (eds): Tumors of the Spine: Diagnosis and Clinical Management. Philadelphia: WB Saunders, 1990, pp Sundaresan N, Scher H, DiGiacinto G, Yagoda A, Whitmore W, Choi IS: Surgical treatment of spinal cord compression in kidney cancer. J Clin Oncol 4: , Wara WM, Phillips TL, Sheline GE, Schwade JG: Radiation tolerance of the spinal cord. Cancer 35: , Yu C, Jozsef G, Apuzzo ML, Petrovich Z: Dosimetric comparison of CyberKnife with other radiosurgical modalities for an ellipsoidal target. Neurosurgery 53: , 2003 Manuscript received August 26, Accepted in final form July 5, This study was supported by the Volker Sonntag Award, a research grant awarded by the American Association of Neurological Surgeons and Congress of Neurological Surgeons Section on Disorders of the Spine and Peripheral Nerves. Address reprint requests to: Peter C. Gerszten, M.D., M.P.H., Department of Neurological Surgery, Presbyterian University Hospital, Suite B-400, 200 Lothrop Street, Pittsburgh, Pennsylvania gersztenpc@upmc.edu. 295