Chapter 190 Management of Primary Malignant Tumors of the Osseous Spine
Primary tumors of the spine are exceedingly rare. Owing to their rarity, few surgeons had gained enough experience and insight into their management until the 1970s when B. Stener first applied oncologic criteria in resecting spine tumors, previously submitted only to intralesional excisions (curettage). Stener1 was the first to plan and perform en bloc resections as they were already being performed in tumors of the gastrointestinal tract. His works are still unsurpassed examples of adapting surgery to tumor expansion and anatomic constraints to achieve a tumor-free margin by en bloc resection.
Later, R. Roy-Camille2 popularized a technique to standardize en bloc resection in the thoracic spine by the posterior approach and in the lumbar spine by a combined posterior and anterior approach. Some years later, K. Tomita3 proposed a similar technique, characterized by the use of a saw thinner than the Gigli saw, proposing to remove en bloc also the posterior arch. However, what is missing in these capital contributions—which represent the foundations upon which other surgeons have further advanced—is any consideration of the margins to be achieved, thus reducing the oncologic interest. These concepts were first applied to bone tumor by Enneking4 and more specifically to bone tumors of the spine by Campanacci and colleagues5 and by Talac and colleagues.6
Many examples are reported in the most recent literature of highly technically demanding surgical procedures performed either in tumors of the cervical or cervicothoracic spine,7–11 including in the resection functionally relevant structures with the purpose of achieving tumor-free margins.12–13
Classification
Less than 5% of the 2500 primary malignant bone tumors that are found each year in the United States occur in the spine.14 For that reason, few centers can achieve a critical level of experience in the management of these tumors. The terminology used to describe these tumors has tended to vary considerably from one region to the next. In addition, there were no staging systems that helped one decide which type of surgery to perform. All of these reasons entered into our decision to help develop the Weinstein, Boriani, Biagini (WBB) surgical staging system.15 This system is designed to unify the ways tumors are described in order to facilitate communication between physicians. Once a common descriptive language is accepted, it also helps to facilitate research efforts. The WBB system helps guide the surgeon with regard to what type of resection is possible.
The WBB system divides the axial presentation of the vertebrae involved with tumor into 12 zones similar to a clock face (Fig. 190-1). Position number 1 begins at the left half of the spinous process and position 12 ends at the right half of the spinous process. Zones 4 and 9 are particularly important to know because they define respectively the left and the right pedicle. Vertebrectomy with adequate surgical margins depends upon one of these two zones to be free of tumor. The vertebra is further divided into radial zones. The radial zones define the depth of tumor invasion. For instance, zone A represents a soft tissue mass extending beyond the confines of the bony cortex. Zone B describes tumor within the superifical bony vertebrae, and zone C defines tumor within the deep bony vertebrae. Zone D describes epidural tumor involvement and zone E is intradural. It is also important to describe the longitudinal extend of the tumor.
This system has been submitted to intra- and interobserver reliability assessment16 by a group of spine tumor experts
Indications for En Bloc Resection
The goal of en bloc surgery is to remove the whole tumor with a continuous shell of normal tissue, the margin. Often the tissues surrounding spine tumors are functionally very important and even unresectable. Sometimes nerve root resection leads to significant motor deficit; sometimes a situation arises where a meaningful surgical margin is not possible without transection of the dura or even the spinal cord. These options should be first weighted considering the long-term outcome of oncologically appropriate surgery versus intralesional surgery in each specific conditions.6,17–20 A 2009 systematic review of the literature22 offered evidence (low quality due to the small number of cases reported) that for thoracic and lumbar spine aggressive osteoblastoma, giant cell tumor (Enneking 3), en bloc resection—when anatomically feasible—can be recommended to minimize the risk of local recurrence. En bloc resection should be undertaken for chordoma and chondrosarcoma of the spine, provided wide or marginal margins can be achieved, because it is related to a reduced rate of local control and mortality. In osteosarcomas and Ewing sarcomas of the spine, en bloc resection is associated with better local control and overall survival if associated with a full course of chemotherapy.
The margins are a valid predictor of local and systemic prognosis,6,12,13,17–21 but the possibility to compensate intentional margin transgression with multimodal treatments should be carefully considered. The option of relevant functional sacrifices to target better systemic prognosis must be discussed with the patient. For some patients, the idea of paralysis is not worth considering. Conversely, others might wish to accept paralysis in exchange for possible cure. This is clearly a decision that only the patient should make in conjunction with the counsel of the surgeon. En bloc resections are sometimes proposed also in the treatment of bone metastases in the spine. This indication is open to discussion because the criterion to plan the treatment of metastatic disease is palliation, and risk-to-benefit ratio should always be carefully considered. However, when local control is the primary issue—in solitary metastases from clear cell tumor or colon carcinoma—en bloc resection, if feasible with acceptable risk, may be an option. Bilsky et al.,23 however, in a 2009 systematic literature review, concluded that stereotactic radiosurgery is better than en bloc resection in the treatment of solitary metastases from clear cell carcinoma without epidural extension.
Finally, the specific morbidity of en bloc resections should always be carefully considered.24
Preoperative Planning
• A posterior resection (Fig. 190-2A) for posteriorly occurring tumor (Fig. 190-2B)
• A posterior-only approach (Fig. 190-3A) for vertebral body tumors not expanding in layer A (Fig. 190-3B)
• A staged approach (see Fig. 190-5A) for vertebral body tumors expanding anteriorly in layer A (see Figure 190-5B).
• A posterior-only or staged sagittal resection (see Fig. 190-6A) for eccentrically located tumors (see Fig. 190-6B).