Updates in the management of intradural spinal cord tumors: a radiation oncology focus

Abstract

Primary spinal cord tumors represent a hetereogeneous group of central nervous system malignancies whose management is complex given the relatively uncommon nature of the disease and variety of tumor subtypes, functional neurologic deficits from the tumor, and potential morbidities associated with definitive treatment. Advances in neuroimaging; integration of diagnostic, prognostic, and predictive molecular testing into tumor classification; and developments in neurosurgical techniques have refined the current role of radiotherapy in the multimodal management of patients with primary spinal cord tumors, and corroborated the need for prospective, multidisciplinary discussion and treatment decision making. Radiotherapeutic technological advances have dramatically improved the entire continuum from treatment planning to treatment delivery, and the development of stereotactic radiosurgery and proton radiotherapy provides new radiotherapy options for patients treated in the definitive, adjuvant, or salvage setting. The objective of this comprehensive review is to provide a contemporary overview of the management of primary intradural spinal cord tumors, with a focus on radiotherapy.

Key Points.

1. Multidisciplinary management is crucial to managing spinal cord tumors.

2. Radiation therapy has an important role for patients with spinal cord tumors.

Primary spinal cord tumors represent a heterogeneous group of central nervous system (CNS) malignancies,¹ whose management is based on patient characteristics (eg, age, performance status) as well as tumor subtype and disease extent. These tumors are divided into 3 anatomic categories: intradural intramedullary (including astrocytoma, ependymoma, hemangioblastoma), intradural extramedullary (including myxopapillary ependymoma, meningioma, neurofibroma, schwannoma), and extradural (including chordoma, chondrosarcoma, osteosarcoma, hematopoetic malignancies). The management of patients with spinal cord tumors is complex given the relatively uncommon nature of the disease, functional neurologic deficits from the tumor, and potential morbidities associated with definitive treatment. Although spinal cord tumors have genetic and biologic similarities with their intracranial counterparts,^2,3 molecular analyses show key differences between the intracranial and spinal variants of tumors that are histopathologically similar.^4–6 Given these factors, patients are typically managed at large academic centers, where multidisciplinary experience is critical to appropriate therapeutic selection and where treatment delivery is a function of expertise.

Many recent advances have refined the current role of radiotherapy in the multimodal management of patients with primary spinal cord tumors and corroborated the need for prospective, multidisciplinary discussion and treatment decision making. For example, advances in neuroimaging allow for more accurate determination of disease extent, leading to better target volume delineation for radiotherapy planning. The updated 2016 World Health Organization (WHO) classification of CNS tumors⁷ has integrated diagnostic, prognostic, and predictive molecular testing for establishing a definitive diagnosis, and this underscores the expanding role of neuropathologists in providing accurate and robust histopathologic and molecular reports that guide therapeutic decision making. The reality of small biopsy specimens in the setting of intramedullary spinal cord tumors makes this task especially challenging. Developments in neurosurgical techniques, including use of novel imaging procedures for presurgical planning and intraoperative visualization to improve tumor resectability, have also impacted the role of radiotherapy in the modern era. Finally, radiotherapeutic technological advances have dramatically improved the entire continuum from treatment planning to treatment delivery. Stereotactic radiosurgery (SRS), stereotactic body radiotherapy (SBRT), and proton beam therapy (PBT) have now become robust and mature, providing new radiotherapy options for patients with primary spinal cord malignancies treated in the definitive, adjuvant, or salvage setting. One of the most recent of such advances is the development of an MR linear accelerator which incorporates real-time continuous MR imaging with treatment delivery, thereby improving delivery precision. The objective of this comprehensive review is to provide a contemporary overview of the role of radiotherapy in the management of primary intradural spinal cord tumors.

Intramedullary Spinal Cord Tumors

Astrocytoma

Spinal astrocytomas represent one of the most common intramedullary tumors, accounting for approximately one-third of primary spinal cord tumors in adults, two-thirds of those in adolescents, and 90% of those in children under age 10.¹ As with astrocytic tumors in the brain, astrocytomas in the spine include pilocytic astrocytoma (WHO grade I), diffuse astrocytoma (WHO grade II), anaplastic astrocytoma (WHO grade III), and glioblastoma (WHO grade IV).⁸ Pilocytic astrocytomas account for 11% of pediatric spinal tumors but are rare in adults (0.8%).

The new WHO classification system for CNS tumors includes an integrated histologic-molecular diagnosis for astrocytomas.⁷ However, while the genetics of intracranial astrocytomas have been well described, fewer studies have investigated the genetics of spinal astrocytomas. Some common mutations observed in cranial astrocytomas are also noted in spinal astrocytomas, including mutations in the p16 gene, PTEN, BRAF, p53, and the replication-independent histone 3 variant H3.3 gene. Specifically, p53 mutations are frequent in spinal cord astrocytomas^9,10; however, isocitrate dehydrogenase mutations are seen in low frequencies.¹¹ The spinal cord is one of the more common locations for astrocytomas mutated by histone H3 on lysine 27 (H3K27) in adults (Supplementary Fig. 1).¹² Traditionally, we have associated H3K27 mutation with a dismal outcome¹³; however, a recent study reported this as a favorable factor in spinal tumors.¹⁴ Spinal astrocytomas have also been found to harbor the BRAF-KIAA1549 fusion gene and BRAF^V600E mutation.^3,15 This predictive molecular marker may bring about treatment of selected patients with BRAF inhibitors; related data are limited yet promising, and at this time upfront therapy with a BRAF inhibitor despite a BRAF mutation would be considered investigational. The integration of molecular biology and whole genome sequencing into clinical care has allowed for identification of potential targets for more personalized treatment options directed at these aberrant pathways or tumor-specific mutations and are currently under investigation.

Accurate determination of the disease extent is critical to the appropriate management of patients with spinal astrocytomas, including evaluation of the subarachnoid space and imaging of the craniospinal axis for potential multifocal disease, which occurs more commonly than with intracranial astrocytomas.¹⁶ The general treatment paradigm includes maximum safe resection as the primary treatment¹⁷; however, the true value of aggressive resection in patients with higher-grade disease is unknown, as these tumors are more infiltrative and have poorly defined resection planes.^18,19 Therefore, in clinical practice, patients with WHO grade I astrocytomas undergo primary resection, whereas those with higher-grade tumors often undergo biopsy alone. Adjuvant therapy recommendations are based on the extent of surgery, disease course extent (initial or recurrent disease), performance status, age of the patient, and WHO grade.

Patients with pilocytic astrocytomas who undergo gross total resection and follow-up imaging without radiographic evidence of residual disease are observed, given the favorable outcomes in this subgroup and the lack of benefit with adjuvant treatment (either radiotherapy or chemotherapy).²⁰ Similarly, patients with diffuse astrocytomas (WHO grade II) of the spinal cord can also be followed with observation alone following gross total resection.²¹ However, complete resection of WHO grade II spinal tumors is uncommonly achieved, ranging 10–20%.^17,22–26 For patients who undergo subtotal resection or experience relapse, even in the setting of low-grade histology (WHO grades I–II), adjuvant radiotherapy is recommended (Fig. 1).²¹ A review of 242 patients across 6 institutions demonstrated an improvement in progression-free survival (PFS) with postoperative radiotherapy in patients with grades I–II tumors.²⁷ Primary anaplastic astrocytoma (WHO grade III) and glioblastoma (WHO grade IV) of the spinal cord are rare and associated with poor survival. After maximal safe resection (including biopsy alone for most patients), the primary treatment consists of definitive radiotherapy, and systemic therapy is often used as per intracranial management guidelines.^28,29 In one retrospective series, 58% of patients with infiltrative astrocytomas received chemotherapy and on multivariate analysis, chemotherapy was associated with a significant improvement in PFS (hazard ratio [HR] = 0.22, P = 0.0075) but not with overall survival (OS) (HR = 0.89, P = 0.83).³⁰ Other case series and reports have shown small benefits of a variety of other treatment regimens, predominantly in the pediatric population.^31–33 Of note, the success of immune checkpoint inhibitors such as pembrolizumab and nivolumab in treating a variety of malignancies has sparked interest in these agents for treatment of high-grade gliomas. However, early experience with these agents in recurrent glioblastomas has been less promising.³⁴

Fig. 1.

(A) Sagittal T2-weighted MRI of a patient with a low-grade (WHO grade II) astrocytoma extending from the cervicomedullary junction to the T4 level. (B) Corresponding sagittal CT scan shows the isodose distribution of the definitive radiotherapy treatment plan to a dose of 54 Gy in 30 fractions delivered with IMRT.

Adjuvant radiotherapy is typically initiated 4–6 weeks after resection to allow healing from surgery and, if needed, postoperative rehabilitation. However, for grades III–IV tumors, radiotherapy may be initiated earlier. Pre- and postoperative MR images are co-registered to treatment planning CT scans to delineate the target volume including the residual disease and postoperative cavity, as well as the region at risk for microscopic disease spread (but not including the syrinx associated with the tumor). The general radiotherapy techniques for patients with spinal astrocytomas include intensity-modulated radiotherapy (IMRT), volumetrically modulated arc therapy (VMAT), or PBT to conformally treat the at-risk spinal disease while minimizing radiation dose to surrounding sensitive structures. Daily image guidance is also utilized to reduce planning margins and ensure accurate patient setup during treatment delivery. The prescription doses for patients with spinal astrocytomas typically range from 45 to 54 Gy at 1.8–2 Gy/fraction, based on the histology/grade, location of disease, and extent of residual disease (Fig 2). At these dose ranges, the risk of treatment-related myelopathy remains low (0.2–0.5% at 50 Gy and 1–6% at 60 Gy).^35,36 Blood counts should be monitored during radiotherapy and marrow-sparing PBT may be beneficial in patients if an extended length of the spine is treated.

Fig. 2.

Patient who presented with myelopathy and was found to have an intramedullary mass spanning from C5 to T1 as visualized on (A) the preoperative sagittal T1 post-contrast and (B) fluid attenuated inversion recovery (FLAIR) MR images. A subtotal resection was performed and (C) the sagittal post-contrast MRI is presented. Pathology confirmed glioblastoma positive for H3K27M. The patient was treated with concurrent temozolomide and radiotherapy to a dose of 54 Gy in 30 fractions using IMRT followed by adjuvant temozolomide. (D) A sagittal representation of the radiotherapy plan with representative isodose lines. The residual gross tumor volume (GTV), surgical bed, and entire canal extending at a minimum 1.5 cm cranially and caudally with extension to encompass the FLAIR are the clinical target volume (CTV) and a 0.5 cm planning target volume (PTV) applied to which the treatment is prescribed to cover by at least 95% of the prescribed dose. She survived 18 months, with the cause of death being diffuse intracerebral leptomeningeal disease. The spinal disease had regressed and stayed stable and (E) the sagittal post-contrast MRI at the time of death is presented.

Ependymal Tumors

Ependymal tumors are the most common spinal cord tumors in the pediatric population (22%) and are relatively common in adults (21%). Most pediatric spinal ependymomas occur in the filum terminale or conus.³⁷ In adults, ependymomas most often occur in the cervical spine and filum terminale (therefore extramedullary by location). Ependymal tumors are classified as subependymoma (grade I), myxopapillary ependymoma (grade I), ependymoma (grade II), or anaplastic ependymoma (grade III). In both children and adults, the myxopapillary variant is most prevalent. Although they were previously presumed to originate from ependymal cells in the central canal, both spinal and intracranial ependymomas are now thought to arise from radial glial stem cells.^38,39

Tumor karyotyping using comparative genomic hybridization analysis reveals that intracranial and spinal ependymomas are chromosomally distinct.⁴⁰ Similarly, in a transcriptomic study of 39 ependymoma tumors, including 10 spinal ependymomas, unbiased hierarchical clustering resulted in discrimination of supratentorial, posterior fossa, and spinal tumors. Fusions of v-rel avian reticuloendotheliosis viral oncogene homolog A (RELA) are not found in spinal ependymomas.⁴¹NF2 mutations are observed in 30–40% of spinal ependymomas and are observed only in spinal ependymomas in studies that also included intracranial ependymomas.⁶ These molecular aberrations in ependymoma are interesting from a pathogenesis perspective, but with the exception of RELA, molecular testing is not currently part of the pathologic classification of ependymal tumors and no actionable mutations have been discovered.

It is critical to evaluate the entire craniospinal axis for disease spread (or uncover an intracranial primary in the setting of spinal metastases), and sampling of the subarachnoid space is often performed. The primary treatment for patients with spinal ependymomas is maximum safe resection. Complete resection of spinal ependymal tumors is the most important prognostic factor and is more often achieved.^42–46 Similar to other intramedullary tumors, adjuvant therapy recommendations are based on the extent of surgery and presence of residual tumor (whether the treatment was the initial treatment or for recurrent disease), performance status and age of the patient, and WHO grade. Focal irradiation is considered in patients in whom CSF cytology is negative for malignancy, whereas craniospinal irradiation is recommended in the setting of positive cytology on CSF sampling or radiographic evidence of multifocal disease throughout the spinal cord and/or brain.

Myxopapillary ependymomas (WHO grade I) are low-grade tumors that typically occur at the filum terminale (intradural, extramedullary by anatomic location). Complete resection is difficult to achieve, except in patients with focal disease. One goal of a gross total resection is to limit the risk of disease seeding.^47–49 The local control rates after complete resection are quite high; however, thorough multidisciplinary discussion is warranted to assess the risk of local recurrence even in the setting of a gross total resection, as those who undergo piecemeal resection or have violation of the tumor capsule are at higher risk of disease relapse (Fig. 3).^{24,25,43,50–53} Long-term follow-up is needed for these patients as late relapses can occur.⁴⁹ The role of adjuvant radiotherapy after gross total resection of a myxopapillary ependymoma is controversial. One study reported an improvement in local control with adjuvant radiotherapy, regardless of the extent of surgery,⁴⁹ and updated results of this series continue to demonstrate this benefit in patients who underwent subtotal or gross total resection.⁵⁴ An analysis from the Rare Cancer Network on 85 patients also demonstrated an improvement in PFS with adjuvant radiotherapy,⁵⁵ further confirmed with a combined analysis of these 2 series.⁵⁶ On the other hand, given the excellent local control and long-term prognosis after gross total resection in other series, radiotherapy is commonly reserved for salvage in the setting of disease relapse, where radiotherapy has been demonstrated to improve outcome, even after gross total re-resection.⁵¹ However, pediatric patients may remain at higher risk of disease recurrence, even after gross total resection in the upfront setting, and a thorough discussion of the risks and benefits of adjuvant radiotherapy should be considered in this patient population at initial diagnosis.⁵² Some patients with myxopapillary ependymomas present with extensive disease across multiple spinal levels, and definitive aggressive resections are unable to be performed (Fig. 4). In these patients, maximal safe debulking of disease is followed by definitive radiotherapy.

Fig. 3.

(A) Sagittal T2-weighted MRI of the lumbar spine demonstrates an intradural extramedullary myxopapillary ependymoma, WHO grade I, for which the patient underwent a near-total resection and was found to have disease attached to the L1 nerve root. (B) Corresponding sagittal planning CT scan shows the isodose distribution of the adjuvant radiotherapy treatment plan to a dose of 50.4 Gy in 28 fractions delivered with IMRT.

Fig. 4.

Diffuse spinal leptomeningeal ependymomatosis in a patient who presented with impaired mobility and bladder and bowel function. Resection to debulk the intradural extramedullary mass from T1 to T5 and from L1 to S1 was performed. (A–B) The preoperative post-contrast sagittal images. (C–D) Postoperative sagittal post-contrast MR images. Staging of the brain suggested leptomeningeal nodules involving cranial nerves VII and VIII but could not be confirmed pathologically. Following 6 months of rehabilitation, the patient was treated with (E) whole spine and posterior fossa radiation to a dose of 36 Gy in 20 fractions followed by (F) a boost to a total dose of 54 Gy in 30 fractions to the whole spine inclusive of 1 cm above the most cranial gross disease at the level of the foramen magnum using IMRT. Three years later the patient is stable with no evidence of recurrent or progressive disease within the craniospinal axis.

The propensity of WHO grade II ependymomas to occur in the upper spinal cord allows for easier access and complete resection of these tumors, which is associated with improved outcomes compared with WHO grade I or III spinal ependymomas.⁵⁷ Several studies have failed to demonstrate a benefit to adjuvant radiotherapy after gross total resection of grade II spinal ependymomas,^37,57,58 and therefore adjuvant treatment is generally discouraged. In the setting of a gross total but piecemeal resection, adjuvant radiotherapy should be strongly considered given the increased risk of disease recurrence.⁵⁹ For patients with subtotally resected grade II or any grade III spinal ependymomas, adjuvant radiotherapy is recommended given the higher risk for disease relapse.^27,52,60,61

Patients receiving adjuvant radiotherapy often undergo treatment 6–8 weeks after resection to allow for recovery from surgery, account for radiographic clarity of postoperative imaging, and allow time for postoperative rehabilitation of neurologic function. Pre- and postoperative MR images are co-registered to treatment planning scans to delineate the target volume including the residual disease, postoperative volume, and the ependymal borders at risk for microscopic disease spread. The general radiotherapy techniques for patients with spinal ependymomas without CSF dissemination include IMRT, VMAT, or PBT in order to conformally treat the at-risk spinal disease while minimizing radiation exposure to surrounding sensitive structures. The prescription doses for patients with spinal ependymomas typically range 50–55.8 Gy at 1.8–2 Gy/fraction, based on the histology/grade, location of disease, and extent of residual disease. For patients with positive CSF cytology or disease throughout the CNS axis, the entire craniospinal axis is treated to a dose of 30–36 Gy at 1.5–1.8 Gy/fraction and the primary tumor site is boosted to a total dose of 50–55.8 Gy. Higher doses to the craniospinal axis (ie, 39.6–45 Gy at 1.8 Gy/fraction) can be considered in patients with gross leptomeningeal disease.

Given the complexity and length of field irradiated, traditional craniospinal techniques involved multiple abutting photon fields (Fig. 5A) which were carefully positioned to match at specific junctions (shifted during the course of treatment to prevent high-dose overlap in the spinal cord), which resulted in significant unnecessary dose to the anterior thoracic and abdominal organs. Newer techniques involve using multiple partially overlapping fields to reduce the exposure to nearby organs-at-risk. One of these techniques uses a helical delivery platform to modulate radiation dose in the craniospinal axis (Fig. 5B). Intensity-modulated PBT allows dose deposition precisely in the target volume with minimal exposure to the surrounding structures (Fig. 5C). Dosimetric studies have demonstrated clear advantages of PBT,⁶² and clinical studies have revealed a reduction in acute treatment-related morbidity, such as hematologic and gastrointestinal toxicity.^63,64 This treatment delivery will also have the potential to reduce the risk of late treatment-related toxicities, such as radiation-induced malignancies.^65,66

Fig. 5.

Sagittal planning CT scans of (A) a patient undergoing craniospinal irradiation using multiple abutting conventional radiotherapy fields, (B) helical arc delivery technique, or (C) pencil beam scanning proton therapy with radiotherapy isodose lines illustrated for each treatment type. Note the improved dose distribution, with reduced extraneous dose of radiation to anteriorly situated organs, with the more advanced radiotherapy techniques.

Hemangioblastoma

Hemangioblastomas (WHO grade I) represent the third most common intramedullary spinal cord tumor and are often found in the cervical or lumbar region.⁶⁷ Approximately half are located in the spinal cord, one-third are completely intramedullary, while the remainder have an extramedullary component of disease.⁶⁸ Approximately 80% of hemangioblastomas develop in the posterior fossa, while 20% occur in the cervical or lumbar spine.⁶⁷ About one-quarter of patients with hemangioblastoma have von Hippel–Lindau (VHL) disease, characterized by the VHL mutation. One study reported that spinal hemangioblastomas were strongly associated with the VHL syndrome (in 88% of cases) and often were associated with significant VHL expression in multilevel disease.⁶⁹ Overall, the understanding of the role of mutated genes other than VHL in spinal hemangioblastoma remains limited.

Patients with spinal hemangioblastomas should be screened for other manifestations of VHL with an MRI of the brain (cranial hemangioblastoma), neuro-ophthalmologic examination (retinal angioma), and CT imaging of the body (renal cell carcinomas, pheochromocytoma). Management options for patients with spinal hemangioblastoma include active surveillance, resection, high-dose (>50 Gy) fractionated external beam radiotherapy, and SRS. Primary resection with coagulation of feeding arteries is the standard treatment for large, symptomatic lesions, and those with an associated syrinx.

Prior to the development of SRS, patients unable to undergo resection were treated with conventionally fractionated external beam radiotherapy to low doses with poor outcomes. Dose escalation to 50 Gy or higher led to more encouraging results⁷⁰ and helped to reestablish the role of radiotherapy. Subsequent studies have demonstrated modest outcomes in patients who have extensive spinal disease, require adjuvant therapy for residual disease, or develop recurrent disease after prior resection (Fig. 6).⁷¹ In patients with disseminated disease, radiotherapy is the dominant therapy for disease control. Rapid technology developments in treatment planning, upgrades to treatment delivery systems, and advances in on-board imaging capabilities have led to the introduction of a high-precision SRS alternative in select patients with spinal hemangioblastomas who are unable to undergo resection, have VHL disease, or have multiple tumors. Spine SRS series have been very encouraging given the high rates of disease control (often exceeding 90%), improvements in disease-related neurologic symptoms, and lack of significant adverse toxicity, although patient selection is key to appropriate triage of patients for SRS and a discussion with respect to radiation injury to the spinal cord associated with the high dose per fraction inherent to spine SRS.^72,73 It is important to note that in the radiotherapy experience for intracranial hemangioblastomas, local control rates appear to be higher in VHL-associated disease compared with sporadic cases and likely the same would be true in the spine.^71,74,75

Fig. 6.

Seventy-seven year old patient who initially underwent resection of a solitary T4 intradural extramedullary hemangioblastoma and 5 years later was found on MR imaging to have not only recurrence at the T4 level but also gross disease at the level of C3 and sub-centimeter nodules extending down to the level of T8. (A) Sagittal post-contrast MRI indicating the disease extent. He was paraparetic from prior surgery and was neurologically asymptomatic. Partial spine radiation was performed to a dose of 54 Gy in 30 fractions using an IMRT technique encompassing C1 to T10. (B) Sagittal representation of the radiotherapy plan with representative isodose lines. Three years following treatment, he remains controlled.

Very robust patient immobilization and accurate target volume visualization and delineation are critical to planning spine SRS for patients with hemangioblastoma, and MR images are co-registered to the treatment planning CT scans to delineate the disease (the enhancing nodule, without the associated cyst), spinal cord, and other nearby organs-at-risk. Patients can be treated with a variety of radiosurgery platforms and the prescription dose in the literature ranges 16–20 Gy based on spinal cord tolerance.^72,73,76,77 Adverse neurologic events after radiosurgery have been reported, but serious toxicities occur very rarely following treatment.^71,73,77

Intradural Extramedullary Spinal Cord Tumors

Meningioma

Spinal meningiomas are the most common spinal tumors in adults, accounting for up to 38% of intradural spinal tumors.⁷⁸ The WHO classification of meningiomas includes 15 different histologic subtypes grouped into grade I, grade II and grade III categories.⁷ Psammomatous, meningothelial, and transitional histologies are the most common subtypes of meningiomas of the spine, and both psammomatous and clear cell meningiomas are almost exclusively found in the spine. Spinal meningiomas appear to have a lower risk for recurrence and malignant transformation than their intracranial counterparts.^79–81

Several studies have reported deletion of chromosome 22q and of its associated gene NF2 in spinal meningioma.^82,83 A comparative DNA microarray study of 7 spinal and 11 intracranial meningiomas found that spinal meningiomas were more likely to arise from a single-cell clone rather than from a collection of cells.⁸² Interestingly, the study also reported significant differences in gene expression between spinal and intracranial meningiomas. Mutations in the chromosome secondary structure regulator, SMARCE1, have been reported to be important in the formation of multiple spinal meningiomas, in the absence of NF2 mutation.⁸⁴ Molecular testing is not routinely performed in a clinical setting and has not been integrated into meningioma diagnosis.⁷

The primary treatment approach for meningiomas is gross total resection (Simpson grades I–II) and is associated with a high rate of disease control. Subtotal resection (Simpson grades III–IV) results in a recurrence rate of 20–100%, especially with late relapses which may be difficult to salvage.^85–88 Resection status, graded using the Simpson grading system for intracranial tumors, is useful for determining the extent of residual disease.⁸⁹ Radiotherapy remains the standard treatment for patients who are medically inoperable, for tumors that are unresectable, for patients who undergo subtotal resection (Simpson grades III–IV), and for WHO grades II and III tumors, extrapolating from lessons learned from intracranial series. It is important to note that there exists a controversy regarding the management of intracranial WHO grade II meningiomas after gross total resection (Simpson grades I–II)^90,91; however, given the potential neurologic morbidity from disease relapse and the relative rarity of these higher-grade tumors in the spine, often these patients undergo adjuvant radiotherapy.

The location of the tumor, size, extent of disease, nearby critical structures, length of abutment of the spinal cord, prior surgery, age, and performance status of the patient are some of the factors used to determine the appropriate radiotherapy technique. For example, patients with small tumors can undergo single fraction radiosurgery, SBRT, or conventionally fractionated radiotherapy (Fig. 7). Single fraction radiosurgery or 2–5 fraction SBRT is typically used for patients with smaller well-defined tumors and has resulted in excellent rates of disease control.^92–95 Single fraction doses range 12–16 Gy^92,96 and common hypofractionated schedules include 18–21 Gy in 3 fractions⁹⁷ or 25 Gy in 5 fractions. Patients treated postoperatively often undergo treatment approximately 4–8 weeks after resection, and pre- and postoperative MR images are co-registered to the treatment planning CT scan to delineate the tumor, residual disease, and postoperative cavity, as well as identify any surfaces with tumor extension at initial diagnosis. For patients who undergo conventionally fractionated radiotherapy, in either the intact or postoperative setting, the dose ranges 50.4–54 Gy at 1.8–2 Gy/fraction,^89,98 but higher doses (54–60 Gy) can be considered for patients with higher-grade disease. Adverse events following radiotherapy for spinal meningiomas are rare, primarily observed in the setting of prior surgery, which may influence spinal cord tolerance to radiotherapy.^94,99,100

Fig. 7.

(A) Sagittal T1-weighted post-contrast MRI of a patient with a meningioma of the foramen magnum/upper cervical spine in whom resection was felt to be too high-risk. Isodose distributions of potential radiotherapy options for this patient with high expected rates of tumor control including (B) 14 Gy SRS plan, (C) 50.4 Gy in 28 fraction conventionally fractionated radiotherapy approach using pencil beam scanning proton therapy, or (D) a 25 Gy in 5 fraction stereotactic approach, all sparing the adjacent spinal cord (orange).

Conclusion

Primary spinal cord malignancies are rare tumors of the CNS, and multidisciplinary management is crucial to determining the appropriate treatment approach. In general, maximum safe resection is essential to provide an accurate diagnosis and relieve neurologic symptoms, and in most cases correlates with local control and overall disease outcomes. Discoveries in the field of molecular biology and whole-genome sequencing have identified fundamental differences between intracranial and spinal variants of the same tumors with the potential to identify tumor-specific therapeutic targets. Advances in radiation oncology have led to an array of treatment options for patients in the definitive, adjuvant, or salvage setting. Development of advanced treatment delivery platforms, such as proton radiotherapy as well as SRS, provide high-precision non-invasive treatments with favorable rates of tumor control. The rapid development of these non-invasive technologies that can deliver high doses of radiation rapidly with unprecedented precision has opened the door for new options for patients. Nonetheless, the management of recurrent and refractory tumors remains challenging. Systemic therapy options are limited, but recent advances in our understanding of molecular biology hold promise for novel effective therapies. Unfortunately, almost none of the data for these tumors are obtained from prospective trials, which remains a significant concern and deficit. Consequently, continued prospective research and long-term follow-up are needed to further define the roles of surgery, radiotherapy, and systemic therapy in the modern era to optimize the outcome for every patient diagnosed with an intradural spinal cord tumor.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest statement:

R. Kotecha: Elekta AB; Accuray Inc.; Chrysalis Biotherapeutics Inc.; Elsevier.

M. Mehta: Consultant: Insys, Remedy, IBA, Varian, Oncoceutics, Astra-Zeneca; DSMB: Monteris.

E. Chang: Brainlab.

P. Brown: Novella Clinical (DSMB), UpToDate (Contributor).

J. Suh: Chrysalis Biotherapeutics Inc.

S. Lo: Elekta AB.

A. Sahgal: Elekta AB; Accuray Inc.; Varian Medical Systems Inc.