Image-guided and intensity-modulated radiosurgery for patients with spinal metastasis




Radiosurgery can deliver a single, large radiation dose to a localized tumor using a stereotactic approach and hence, requires accurate and precise delivery of radiation to the target. Of the extracranial organ targets, the spine is considered a suitable site for radiosurgery, because there is minimal or no breathing-related organ movement. The authors studied spinal radiosurgery in patients with spinal metastases to determine its accuracy and precision.


The spinal radiosurgery program was based on an image-guided and intensity-modulated, shaped-beam radiosurgical unit. It is equipped with micromultileaf collimators for beam shaping and radiation intensity modulation and with a noninvasive, frameless positioning device that uses infrared, passive marker technology together with corroborative image fusion of the digitally reconstructed image from computed tomography (CT) simulation and orthogonal X-ray imagery at the treatment position. These images were compared with the port films that were taken at the time of treatment to determine the accuracy of the isocenter position. Clinical feasibility was tested in 10 patients who had spinal metastasis with or without spinal cord compression. The patients were treated with fractionated external beam radiotherapy followed by single-dose radiosurgery as a boost (6–8 grays) to the most involved portion of the spine or to the site of spinal cord compression.


The accuracy for the isocenter was within 1.36 mm ± 0.11 mm, as measured by image fusion of the digitally reconstructed image from CT simulation and the port film. Clinically, the majority of patients had prompt pain relief within 2–4 weeks of treatment. Complete and partial recovery of motor function also was achieved in patients with spinal cord compression. The radiation dose to the spinal cord was minimal. The maximum dose of radiation to the anterior edge of the spinal cord within a transverse section, on average, was 50% of the prescribed dose. There was no acute radiation toxicity detected clinically during the mean follow-up of 6 months.


Image-guided, shaped-beam spinal radiosurgery is accurate and precise. Rapid clinical improvement of pain and neurologic function also may be achieved. The results indicate the potential of spinal radiosurgery in the treatment of patients with spinal metastasis, especially those with solitary sites of spine involvement, to increase the prospects of long-term palliation. Cancer 2003;97:2013–8. © 2003 American Cancer Society.

DOI 10.1002/cncr.11296

With the development of more effective treatments for malignant disease, the overall survival has increased for patients with many malignant diseases. Because patients with malignancies are now experiencing longer survival, quality of life has become an important factor not only in the decision to treat definitively but also in the decision to offer palliative treatment for metastatic disease. This is particularly true in the treatment of patients with metastatic disease to the vertebral column, because pain and neurologic symptoms further reduce the quality of life. The spine is the most frequent metastatic site for many tumors. It is estimated that 20,000–25,000 patients per year in the U.S. develop spinal cord or root compression as a manifestation of their metastatic disease.1–4 The estimated number of patients diagnosed with spinal metastases has increased with the increased use of computed tomography (CT) and magnetic resonance imaging (MRI) scans.4 The common presenting symptoms of spine involvement are back pain, and the common presenting symptoms of spinal cord compression are weakness and numbness below the level of spinal involvement. Many patients, if they are not treated, become paraplegic or lose control of bowel and bladder function, which results in significant morbidity and poor quality of life. Early treatment prior to the development of significant neurologic deficits improves the chance for patients to remain ambulatory.5

Treatment for patients with spinal metastasis and/or cord compression has been decompressive surgery and/or external beam radiotherapy.6, 7 Surgery usually is offered to patients with a reasonable life expectancy, if a neurologic deficit is evolving rapidly, if spinal instability is present and is causing symptoms, or if the diagnosis of malignant disease has not been established previously.8 Surgery also has been used for more aggressive and relatively radioresistant tumors, either alone or combined with radiation.9 Under other circumstances, external beam radiation therapy is offered, initially combined with steroids, in the vast majority of patients. The effectiveness of external beam radiotherapy has been well established. Varying degrees of pain relief were seen in two-thirds of patients by 3 months after radiation.7 Conventional radiotherapy utilizes generous margins around the involved areas, typically one or two vertebral segments, to compensate for internal organ motion as well as patient motion during the treatment. In view of this, a significant amount of normal tissue, including healthy spinal cord, is included within the treatment volume. To limit the toxicity, the radiotherapy is fractionated and, hence, requires multiple, daily treatments, which often are difficult in this group of critically ill patients.

Radiosurgery delivers a highly conformal, large radiation dose to a localized tumor by a stereotactic approach. This requires accurate targeting and immobilization of the target organ during irradiation. The difficulty of applying radiosurgery to the extracranial site is mainly due to organ motion associated with breathing and/or lack of immobilization. Among the extracranial organs, the spinal cord and vertebrae are the organs with the least breathing-related organ movement. This makes the spinal cord and vertebrae particularly suitable for stereotactic radiosurgery. Spinal radiosurgery has been used in limited numbers of patients.10 The method used by Hamilton et al. was an invasive procedure that required anchoring of the stereotactic frame to the spinous process under general anesthesia.11, 12 Metastatic lesions involving the vertebra are usually very irregular in shape and often are associated with a paravertebral mass. This target irregularity makes the use of radiosurgery more difficult. To avoid radiation-induced injury of the spinal cord, the irregular lesion should be targeted with great accuracy. Image-guided, shaped-beam radiosurgery is capable not only of shaping the radiation beam by using micromultileaf collimation but also of localizing the lesion by using a new, noninvasive, infrared marker technology for positioning. In addition, intensity modulation of radiation increases the conformality of radiation to the tumor while minimizing the radiation dose to the spinal cord. Therefore, a study was performed in our institution to determine the accuracy and precision of shaped-beam radiosurgery and the potential role of spinal radiosurgery for the treatment of patients with spinal metastasis and cord compression.


Ten patients were enrolled in this study at Henry Ford Hospital between April 2001 and December 2001. All patients had pathologically proven, primary malignant neoplasms and signed the Institutional Review Board-approved informed consent form. All patients had one or two contiguous vertebral metastases with or without spinal cord compression that could be identified on CT or MRI scans. Two patients were treated for two contiguous vertebral bodies. Patient characteristics and the primary sites of malignancy are shown in Table 1. All patients received external beam radiation therapy (25 grays [Gy] in 10 fractions) followed by radiosurgery boost (6–8 Gy single dose) to the site of the spine involvement or spinal cord compression. External beam radiotherapy was given in the conventional method with a single posterior port and included two vertebral bodies above and below the radiographic lesion.

Table 1. Patient Characteristics and Radiosurgery Doses with Clinical Outcome
DiagnosisPathologySiteRadiosurgery (Gy)Outcomea
  • Gy: grays.

  • a

    Numeric values indicate motor strength or pain scores.

Lung cancerSquamous cellL26Pain relief with reduced medication
Multiple myelomaPlasma cellT7–T86Pain reduced at 4 weeks, complete relief at 4 months
Breast cancerInvasive ductal cellT76Pain relief 7/10→4/10, progressed at other sites
Multiple myelomaPlasma cellT6–T76Pain relief at 1 week, progressed at other sites
Hodgkin diseaseType undeterminedT68Pain relief at 1 week, motor recovery 0/5→5/5
PlasmacytomaPlasma cellT26Pain relief with reduced medication
Prostate cancerAdenocarcinomaL16Pain relief at 2 weeks, motor recovery 0/5→3/5
Prostate cancerAdenocarcinomaT116Pain relief 9/10→3/10
Breast cancerInvasive ductal cellT36Pain relief by surgery
Breast cancerInvasive ductal cellT108Pain relief at 1 week

The image-guided, shaped-beam radiosurgery at Henry Ford Hospital uses the Novalis system (BrainLAB, Inc, Munich, Germany). It is equipped with a built-in micromultileaf collimator with 6-megavolt (MV) photon.13 The dosimetric characteristics of this treatment unit have been presented previously.14, 15 The positioning devices consist of ExacTrac, an automated patient positioning device, and Novalis Body, an image-guided, target-localization device. The ExacTrac device includes two infrared cameras and a computerized control system. Its major functions are to detect infrared-sensitive markers that are placed on the patient skin, to automatically compare marker locations with the reference information, and to move the treatment couch to the preplanned position. The Novalis Body device includes two 80–100 kiloelectron volt (keV) X-ray tubes, an aSi digital detector, and a computerized control and image analysis system. For the verification of patient position, the system compares internal structures by fusion of the two images taken by the Novalis Body device and the simulator. The information of isocenter deviation is forwarded automatically to the ExacTrac system for adjustment of position.

The shaped-beam radiosurgery procedure deserves a brief description here: Infrared-sensitive markers are placed on the patient's skin before simulation. Vacuum bags or other positioning devices, as needed, can be used to assist patient positioning. Simulation CT is obtained with cross-sections measuring 3 mm in thickness without spacing, and the images are sent to the dedicated treatment-planning system. The target tumor volume and the critical organs, such as the spinal cord, are drawn. The radiosurgery uses multiple (five to nine) intensity-modulated radiation beams to minimize the dose to the critical organs. Repositioning of the patient is performed with the assistance of the ExacTrac system and the Novalis Body system, as described earlier. Prior to the delivery of radiation, orthogonal portal films are obtained for the final verification and to determine the accuracy of the isocenter.

The endpoint of the study was to measure the isocenter variation to determine the accuracy and precision of the spinal radiosurgery. This was achieved by comparing the fusion images of the CT simulation with the orthogonal films obtained at the time of treatment. Although it was not the primary endpoint, patients also were evaluated for relief of pain or neurologic symptoms.


The precision of spinal radiosurgery was defined as the degree of variation between the isocenters of the CT simulation and of the portal films at the time of radiation delivery. It was measured by image fusion of simulation and portal films. The precision for a given isocenter between the simulation and the actual treatment position was within 1.36 mm ± 0.11 mm. We also measured the radiation dose at the isocenter using the same positioning parameters of the individual patients in a phantom with a microion chamber. The average deviation of the measured dose from the estimated dose was 2%.

The radiation dose to the adjacent normal spinal cord was calculated using the computerized isodose calculation program. A representative example of an intensity-modulated treatment plan is shown in Figure 1. The radiation dose was prescribed to the volume included by the 90% isodose line. In a transverse section at the isocenter level, 20% of the spinal cord volume immediately adjacent to the diseased vertebra received > 50% of the prescribed radiation dose. A diagram of the average radiation distribution and a dose-volume histogram are shown in Figure 2.

Figure 1.

A representative intensity-modulated radiosurgery plan with radiation isodose curves. The patient had multiple myeloma involving the seventh and eighth thoracic vertebrae.

Figure 2.

(A) Representative diagram of average radiation dose distribution to the spinal cord in a transverse plane at the isocenter level. (B) Dose-volume histogram of spinal cord with average cord dose (circle) of the entire cord volume at the treated vertebral level and the maximum point dose at a transverse section (square).

Although assessment of efficacy was not the primary goal of this study, pain relief and neurologic improvement were analyzed. All patients had varying degrees of pain relief. One patient had pain relief after laminectomy before radiosurgery. The other nine patients presented with significant pain before they started radiotherapy. Complete pain relief was noted in five of nine patients, and the remaining four patients were able to reduce pain medication. Two patients later developed progressive metastases at different vertebrae, but the regions of the spine treated with radiosurgery were clinically stable. Time to pain relief usually was 2–4 weeks after boost radiosurgery. Two patients presented with paraplegia (motor strength, 0/5) before treatment. One patient had complete motor recovery (motor strength, 5/5) from the Hodgkin disease with restoration of ambulation within 1 week after completion of treatment. This patient had complete radiographic disappearance of the epidural tumor, as seen in the follow-up MRI scan (Fig. 3). The other patient with metastatic prostate carcinoma had partial recovery (motor strength, 3/5). The patient characteristics and radiosurgery dose parameters with clinical outcomes are summarized in Table 1. There was no acute radiation toxicity detected clinically during a mean follow-up of 6 months (range, 3–12 months), although the duration of follow-up was not long enough to detect any late spinal cord toxicity.

Figure 3.

Transverse and sagittal sections of magnetic resonance image (MRI) scans from a patient with Hodgkin disease (A) before treatment and (B) after 12 months. Arrows indicate the site of spinal cord compression before treatment. Follow-up MRI shows complete control of the enhancing lesion at 12 months after treatment.


In this initial feasibility study, we determined the accuracy and precision of spinal radiosurgery. This was achieved by noninvasive patient positioning and by the ability to conform the target tumor by image-guided, shaped-beam radiosurgery. To deliver accurate radiosurgery treatment, reproducible and reliable patient positioning is required. The patient also should feel comfortable with the given body position. Accurate patient positioning was achieved using an automated positioning device with infrared marker technology and image-guided target localization. This process was entirely noninvasive. The image fusion of the reconstructed CT simulation and the orthogonal Novalis Body X-ray image provided further, integrated verification of the precise positioning using the internal nonmoving structures. The precision of the isocenter obtained in this study was acceptable and was achieved easily for spinal radiosurgery. Previous reports describing spinal radiosurgery for patients with spinal metastases used surgical fixation as part of the treatment to perform high-dose irradiation.11, 12 It was an invasive procedure that required general anesthesia. The procedure included surgical attachment of the stereotactic frame to the spinous process (bone screw fixation) for immobilization and localization. The requisite prone position also could be problematic because it would allow greater motion of the vertebral body due to abdominal wall movement associated with breathing. Despite its invasiveness, the clinical outcome was encouraging. More recently, other radiosurgery technology used a robotic arm to deliver radiation in a wide range of beam orientations, except for the posterior region where the X-ray detectors are located.16 This also appears to have an invasive component, with the implantation of a few metal markers in the patient to help determine the target.

Tolerance of the spinal cord to a single dose of radiation has not been defined well. Figure 2 shows that the dose to the spinal cord generally was < 50% of the prescribed radiation dose in this feasibility study. The volume that received a dose that was higher than this was < 20% of the anterior portion of the spinal cord. Note that the radiation doses to the spinal cord from conventional external beam radiotherapy (with 3 Gy per fraction) usually are higher than the dose prescribed. Although there are instances of patients who have spinal cord tolerance from reirradiation to the same cord level,17 information regarding the tolerance of the human spinal cord to a single, large dose of radiosurgery is not available. We plan to explore the single-dose tolerance of the spinal cord in an ongoing radiosurgery treatment study with a dose-escalation paradigm. The challenge for a successful spinal radiosurgery is how to immobilize the patient effectively, localize the tumor, and create highly conformal dose distribution to spare radiation damage to critical organs (i.e., the spinal cord). Three-dimensional shaping of the radiation beam to conform to the irregular tumor was achieved by a computerized dynamic movement of the micromultileaf collimator during the radiosurgery procedure. In addition, intensity-modulated radiation added homogenous radiation within the target tumor while limiting the radiation to the spinal cord. These capabilities added more precision to minimize the radiation to the critical normal tissue.

This study demonstrated the potential of spinal radiosurgery for the treatment of patients who had spinal metastasis with or without spinal cord compression. Only modest doses of radiosurgery were used, and the assessment of efficacy was not the primary goal of the study. However, all evaluable patients had a significant degree of pain relief. These results are encouraging. Although the role of boost radiosurgery in reducing pain remains speculative, the relatively rapid pain relief seemed quicker than expected compared with external beam radiation, for by which the time to pain relief was as long as 3 months after treatment.7 It may be possible that the rapid pain relief and neurologic improvement were due to the relatively radioresponsive type of tumor treated in this study. Rapid pain relief from radiosurgery may improve quality of life further. Therefore, this study offers the potential to extend the radiosurgery as the primary treatment of patients with spinal metastasis and spinal cord compression. A single hospital visit or a minimal number of hospital visits would be more convenient for the patients. Radiosurgery has become a viable treatment option for patients with brain metastasis, either alone or in combination with external beam radiotherapy. Definitive treatment with radiosurgery alone for patients with single or multiple brain metastases has resulted in excellent local control of the tumor with improved neurologic function.18 It is possible that similar tumors that spread to single sites in the spine, potentially with spinal cord compression, may be treated in the same fashion. An efficient and effective system for tumor control by spinal radiosurgery will require a noninvasive, accurate, reproducible technique with the ability to target irregularly shaped masses while avoiding damage to the adjacent spinal cord. The use of intensity-modulated radiation with a precise beam-shaping technique and noninvasive positioning device makes radiosurgery possible for the treatment of patients with spinal metastasis, and spinal radiosurgery has the potential to improve the clinical outcome of these patients.