Tumors of the Thoracic Spine

Published on 11/04/2015 by admin

Filed under Orthopaedics

Last modified 11/04/2015

Print this page

This article have been viewed 4474 times

46 Tumors of the Thoracic Spine

Timothy F. Witham, Vivek A. Mehta, Ziya L. Gokaslan

KEY POINTS

• Surgical decompression followed by radiotherapy has replaced radiotherapy alone as the gold standard for treatment of spinal cord compression caused by metastatic cancer.

• Tumor histology, location, and anterior column compromise should dictate the surgical approach.

• Anterior, posterior, combined, or en bloc approaches should be considered in a disease-directed fashion for the treatment of thoracic spinal column neoplasms.

• Instability must be considered, particularly when there is anterior column compromise.

• The clinical goals of surgical intervention should be local disease control, preservation or restoration of neurological function, stabilization of the spinal column, and pain control.

Introduction

Nearly 1.4 million new cases of cancer are diagnosed in the United States per year.¹ In a majority of the approximately 724,000 cancer-related deaths per year, patients ultimately succumb to complications related to metastatic disease. Approximately 30% to 90% of cancer patients will have evidence of metastases to the spinal column at autopsy, and 5% to 10% of cancer patients develop symptomatic metastatic epidural spinal cord compression (MESCC), an oncologic emergency.¹ There are approximately 25,000 cases of symptomatic MESCC in the united states each year.¹ The thoracic spine is the most common site of metastatic disease involvement in the vertebral column and the most common site of primary vertebral column tumors. Because symptomatic degenerative disease is less common in the thoracic spine relative to the cervical or lumbar spine, thoracic region pain in a middle-aged or elderly patient should be considered a red flag symptom. Imaging should be considered expeditiously, particularly if the patient has a history of primary cancer. Nonetheless, delay in the diagnosis of thoracic region tumors is common. Surgical intervention for thoracic spinal cord and spinal column neoplasms continues to present significant challenges. Lesions in the thoracic spine can be subdivided into four major categories: metastatic disease, intradural extramedullary tumors, intramedullary spinal cord tumors, and primary neoplasms involving skeletal elements.

Metastatic Tumors

Approximately 70% of metastatic disease to the vertebral column involves the thoracic spine, making it the most common site of spinal metastases.² As increased survival due to improved treatment strategies for primary neoplasms becomes a trend, the incidence of bone metastases to the spine will also increase. The unique venous drainage of visceral organs through the Batson plexus is one explanation for this anatomical phenomenon, along with the close proximity of the thoracic spine to thoracic and abdominal viscera.² Eighty percent of metastases involve the vertebral body, while 20% involve the posterior elements.² Of the four subtypes of tumors of the thoracic spine, metastatic tumors are by far the most common. Cancers that have a propensity to spread to the thoracic spine include breast, lung, leukemia/lymphoma, prostate, and renal cell.² Surgical management requires consideration of the need for decompression of the neural elements, for restoration of spinal stability, and for pain relief. With tumors of breast, prostate, or renal origin, aggressive resection is indicated, as this has been shown to have a favorable impact on ambulatory function and may have a favorable impact on survival.³

Intradural Extramedullary Tumors

Intradural extramedullary (IDEM) tumors account for two thirds of intradural spinal tumors.⁴ The most common IDEM tumors of the thoracic spine are meningiomas, neurofibromas and schwannomas, which account for over 80% of tumors in this location.⁵ Meningiomas arise from arachnoid cap cells; 75% of spinal meningiomas are contained within the thoracic spine, and they show a strong predilection for women (80%).⁵ Meningiomas have a benign biological behavior in the spinal column and management should be directed toward gross total resection to achieve a cure. This may not be feasible with anteriorly situated, calcified tumors, and the risk of neurological injury must be weighed against the indolent growth rates usually observed after subtotal resection. Radiotherapy is thus reserved for recurrent tumors that cannot be completely resected.

Schwannomas arise from Schwann cells of the posterior nerve roots and are most common in the thoracic and upper lumbar spine. The location, in 13% of cases, may be both extradural and intradural. Spinal schwannomas have an increased frequency in certain genetic disorders, such as neurofibromatosis type 2 (NF2).

Neurofibromas of spinal nerve root origin may be seen sporadically or in patients with neurofibromatosis. These tumors will have an extradural component in 30% of cases.⁵ The surgical goal in the treatment of thoracic schwannomas and neurofibromas is complete surgical resection. If the tumor cannot be separated from the nerve root of origin, then sacrifice of that root may be necessary. Fortunately, this is usually of little consequence in the thoracic spine.

IDEM tumors of the thoracic spine most often manifest with long tract signs, and the corticospinal tracts are particularly vulnerable. Stiffness and muscle fatigue are often primary signs with spasticity frequently occurring secondarily. In patients presenting with myelopathy, surgical indications are straightforward. It is the incidentally found meningioma or nerve sheath tumor that may present a dilemma. Patient age, medical comorbidities, and tumor growth observed with serial imaging help direct the necessity and timing of surgical intervention.

Intramedullary Spinal Cord Tumors

Intramedullary spinal cord tumors (IMSCTs) make up 2% to 8.5% of all central nervous system(CNS) tumors and approximately one third of primary spinal column tumors.⁶ Approximately 90% are of glial origin.⁴ The major types observed in the thoracic spine are astrocytomas and ependymomas, which are seen with almost equal frequency in adults. Hemangioblastomas, which are less commonly observed, may be sporadic or a part of von Hippel-Lindau syndrome. Astrocytomas are most often seen at the cervicothoracic junction and in the lower thoracic spinal cord. In adults, ependymomas are the most common IMSCT. However, the majority of these occur in the cervical spine or arise from the filum terminale (myxopapillary ependymoma). Treatment of ependymomas is focused on gross total resection. Patients who have incompletely-removed tumors should be considered for re-resection or followed closely with serial imaging and considered for reoperation if growth is documented. Radiation therapy is often reserved for incomplete resection after reoperation. Astrocytomas in adults tend to be infiltrative and blend imperceptibly with the spinal cord at the margins of the tumor. For this reason, total removal may not be possible. Most commonly, infiltrative tumors treated with subtotal resection will undergo radiation therapy. For high-grade infiltrative lesions, biopsy or subtotal resection is also often followed by radiation therapy. For some low-grade lesions treated with subtotal resection, postoperative radiation therapy remains controversial.⁶

Pain is the most common presentation of IMSCT in the thoracic spine; it usually localizes to the level of the tumor and is either regional back pain or radicular.⁴ Up to one third of patients may also experience sensory or motor complaints and spasticity. The most common sign of an IMSCT within the thoracic spine is a mild scoliosis with spasticity and sensory disturbance.⁶ Until recently, it was believed that posterior decompression and radiation therapy were the limits of therapy, but aggressive gross total resection, with the aid of microsurgical tools and intraoperative monitoring, has been shown to be safe and to improve functional recovery and reduce tumor recurrence.

Primary Vertebral Column Tumors

Primary tumors of the vertebral column are rare, making up less than 10% of all tumors involving the spinal column.⁷ It should be emphasized that when patients present with a vertebral column mass in the absence of impending neurological compromise, minimally invasive, image-guided biopsy should be performed prior to definitive treatment planning. Many of the primary neoplasms of the thoracic spinal column are best treated via radical en bloc surgical resection. These types of surgical approaches require careful and detailed preoperative surgical planning by an experienced surgical team. Therefore, knowing the tumor histology up front is critical to the surgical plan.

Primary osseous tumors can be divided into three categories: benign, benign but locally aggressive, and malignant. Benign tumors include hemangiomas, osteoid osteomas/osteoblastomas, chondroma/osteochondromas, aneurysmal bone cysts, and eosinophilic granulomas. Hemangiomas are the most common primary neoplasm affecting the thoracic spine and have been seen in approximately 11% of all postmortem examinations.⁷ Infrequently, hemangiomas can behave in an atypical or locally aggressive fashion and can cause spinal cord compression. In this situation, aggressive surgical resection is advocated. Giant cell tumors have a variable biological behavior and may act in a locally aggressive fashion. Because of this, the authors often recommend radical en bloc excision of tumors exhibiting this pathology and will strongly consider adjuvant radiation therapy postoperatively. Chordomas occurring in the thoracic spine are almost uniformly locally aggressive and recurrences are common after intralesional resection, including gross total resection. Similarly, we recommend radical en bloc resection techniques for thoracic chordomas. Chemotherapy protocols for chordomas have not been promising to date, nor have conventional radiation therapy modalities. Proton beam irradiation therapy has been shown to be the most promising radiation modality and patients are evaluated postoperatively for this modality on an individual basis. However, it is our belief that the most effective treatment modality for achieving long term local disease control is aggressive en bloc resection with negative margins, including resection of the biopsy tract if possible. Proton beam irradiation should not be relied on for cases where gross total resection cannot be achieved. However, it may be employed even in cases of en bloc resection with negative margins.

Malignant types include plasmacytomas, chondrosarcomas, and osteosarcomas. Plasmacytomas make up nearly 30% of all primary tumors and have a propensity to occur in the thoracic spine.⁸ While these tumors are radiation-sensitive lesions, surgery may be necessary for failure of radiation therapy, acute neurological decline from spinal cord compression, and overt instability of the spinal column. Sarcomas present a specific treatment challenge. Biopsy is recommended to assess the specific histology and grade. Systemic workup for metastases is important prior to embarking on treatment. High-grade lesions that are large or those tumors associated with metastatic disease may be treated with neoadjuvant chemotherapy protocols to shrink the local disease and address metatstatic disease prior to surgical resection. En bloc resection is the technique that gives the patient the best opportunity for a longer-term survival with these tumor types, particularly when metastatic disease is not present.

Thoracic spinal column neoplasms present a host of treatment challenges. Knowing the tumor histology through a biopsy is ideal prior to embarking on a formal surgical plan. In the setting of impending neurological decline, a biopsy may not be possible and urgent decompression may be necessary. However, surgical planning is best determined when the histology, systemic disease status, patient age, and medical comorbidities are known.

Basic Science

Research related to the molecular biology and genetics of specific tumor types that affect the spinal column either as primary or metastatic lesions is beyond the scope of this text. However, several animal models of spinal column neoplasia and the translational research related to these animal models are worth discussion. Recently, animal models for metastatic disease to the spine and intramedullary spinal cord gliomas have been developed that will allow for more rigorous preclinical analysis of novel therapies. Mantha et al have established a reproducible model of metastatic breast adenocarcinoma to the vertebral column in rats.⁹ The establishment of this model represents the first of its kind and has allowed for the study of multiple treatment modalities focused on treating spinal metastatic disease, including surgery, radiation therapy, and the novel use of locally delivered chemotherapeutic agents.

Using the metastatic model developed by Mantha et al, Bagley et al have shown that radiation reliably delays paraparesis and death in this model, as does local delivery of paclitaxel.^10,¹¹ The combination of these two modalities has been shown to be more effective than either treatment alone. ¹¹ Additionally, Gok et al. have shown that the combination of decompressive surgery, local chemotherapy, and radiation has the most profound results in terms of preservation of neurological function and prolongation of survival in this animal model with of metastatic epidural spinal cord compression.¹² The hope is that this experimental data may translate into useful clinical trials designed to improve neurological morbidity and, potentially, survival in patients with MESCC due to metastatic breast cancer and other tumor histologies.

Pennant et al. have examined the efficacy of microsurgical excision in a rat model of intramedullary spinal cord tumor. The established model mimics the behavior of IMSCTS, both functionally and histopathologically.¹³ Following tumor implantation, animals were randomized into a treatment group (microsurgical resection) or to no treatment. The animals that underwent resection had a significant delay in the onset of functional paraplegia as compared to the controls and these results were highly reproducible. This new model allows for the study of new treatment options for high-grade intramedullary tumors.

Schuster et al have developed a novel model of spine metastasis using human osteoblasts implanted in immune incompetent (SCID) mice that has shown to be highly reproducible.¹⁴ This model allows for the study of bone-tumor interaction in metastatic disease, as well as the basic biology of bone metastases. Chordoma models have been difficult to establish, but through human cell culture, nude animal models have been conceptualized. Laboratory study of this extremely aggressive local tumor is much needed given the limited options for treatment outside of aggressive radical resection.