Faculty Bibliography

The best clinical treatment for spinal metastases requires an integrated approach with input from an interdisciplinary cancer team. The principle goals of treatment are maintenance or improvement in neurologic function and ambulation, spinal stability, durable tumor control, and pain relief. The past decade has witnessed an explosion of new technologies that have impacted our ability to reach these goals, such as separation surgery and minimally invasive spinal procedures. The biggest advance, however, has been the evolution of stereotactic radiosurgery that has demonstrated durable tumor control both when delivered as definitive therapy and as a postoperative adjuvant even for tumors considered markedly resistant to conventional external beam radiation. In this paper, we perform an update on the management of spinal metastases demonstrating the integration of these new technologies into a decision framework NOMS that assesses four basic aspects of a patient's spine disease: Neurologic, Oncologic, Mechanical Instability and Systemic disease.

BACKGROUND CONTEXT/BACKGROUND:Pathologic vertebral compression fractures (VCFs) represent a major source of morbidity and diminished quality of life in the spinal oncology population. Procedures with low morbidity that effectively treat patients with pathologic fractures are especially important in the cancer population where life expectancy is limited. Vertebroplasty and kyphoplasty are often not effective for mechanically unstable pathologic fractures extending into the pedicle and facet joints. Combination of cement augmentation and percutaneous instrumented stabilization represents a minimally invasive treatment option that does not delay radiation and systemic therapy. PURPOSE/OBJECTIVE:The objective of the study was to evaluate the safety and efficacy of cement-augmented short-segment percutaneous posterolateral instrumentation for tumor-associated VCF with pedicle and joint involvement. METHODS:Forty-four consecutive patients underwent cement-augmented percutaneous spinal fixation for unstable tumors between 2011 and 2014. Retrospective analysis of prospectively collected data, including visual analog pain scale (VAS) response score and procedural complications, was performed. RESULTS:Patients with a median composite Spinal Instability Neoplastic Scale score of 10 (range=8-15) were treated with constructs spanning one to four disk spaces (median of two spaces, constituting 84% of all cases). The proportion of patients with severe pain decreased from 86% preoperatively to 0%; 65% of patients reported no referable instability pain postoperatively. There was one adjacent-level fracture responsive to kyphoplasty, and one case of asymptomatic screw pullout. Two patients subsequently required decompression in the setting of disease progression despite radiation; there was no perioperative morbidity. CONCLUSIONS:Percutaneous cement-augmented posterolateral spinal fixation is a safe and effective option for palliation of appropriately selected mechanically unstable VCF that extends into pedicle and/or joint.

BACKGROUND:Tumor-associated sacral insufficiency fractures (SIF) present a significant clinical challenge. As survival increases for many malignancies, sacral fractures associated with metastases, sacral or extended pelvic radiation, and paraneoplastic osteoporosis are increasingly common and yet remain difficult to treat in the setting of the potentially significant morbidity of open sacral surgery. OBJECTIVE:To describe our prospective experience with sacroplasty for tumor-associated lesions, including the largest series to date of radiation-induced SIF. METHODS:Twenty-five patients with symptomatic SIF underwent 31 percutaneous fluoroscopy-guided sacroplasties with a median 5.8 mL of polymethyl methacrylate or a ceramic-resin composite under fluoroscopic guidance and with concurrent biopsy acquisition. Eighteen patients had fractures related to previous sacral or pelvic radiation; 4 had viable lytic lesions; and 2 had oncology-related osteoporosis. Postoperative pain reduction, procedural morbidity, and functional outcomes were recorded. RESULTS:Twenty of 25 patients (80%) had reduction in their visual analog pain score at a median follow-up of 6.5 months; no patients worsened. The mean visual analog scale score decreased from 8.8 to 4.7 postprocedurally (P < .001), with significant reductions regardless of the underlying pathology (P < .001 to P < .05). Six of 13 patients with pretreatment ambulatory impairment required fewer ambulatory aids and 3 were newly ambulatory. Extravertebral cement migration was noted in 18 procedures; however, no instance was clinically relevant. Six repeat or contralateral procedures were performed. No morbidity was encountered. CONCLUSION/CONCLUSIONS:Sacroplasty is a safe and effective option for the palliation of sacral fractures in the oncologic population.

PURPOSE/OBJECTIVE:Sterilization of surgical margins for lesions involving the dura is complicated by the tolerance of the spinal cord and/or cauda equina, especially in the setting of prior radiation therapy (RT); use of intraoperative brachytherapy may allow local delivery of therapeutic dose without damaging sensitive structures. METHODS AND MATERIALS/METHODS:Patients with malignant lesions involving the dura received intraoperative brachytherapy with a (32)P plaque after maximal resection of the tumor. Local recurrence (LR) was analyzed using competing risks analysis; overall survival was analyzed using Kaplan-Meier statistics. RESULTS:Between September 2009 and April 2013, 68 patients with 69 lesions in the spine were treated with the (32)P plaque. Median followup was 10 months. Most patients (n=59, 85.5%) had previously been treated with at least one course of prior RT to the treated site. About 38 (55%) lesions received postoperative RT (median dose, 30 Gy; range, 18-30 Gy). The LR and overall survival at 12 months were 25.5% (95% confidence interval [CI]=15.5-37%) and 59.5% (95% CI=46-73%), respectively. For patients who received postoperative RT, LR at 12 months was 18.5% (95% CI=7.5-33%) compared with 34% (95% CI=18-51%) for those who were treated with the plaque alone (p=0.08 and 0.04 on univariate and multivariable analysis, respectively). There were no acute or long-term complications from treatment observed in this cohort. CONCLUSIONS:The (32)P intraoperative brachytherapy plaque is a useful adjunct to surgical intervention for primary recurrent and metastatic lesions of the spine involving the dura, and is not associated with additional toxicity.

The NOMS considerations provide a dynamic decision framework to determine the optimal combination of systemic and radiation therapies and surgery. Generally, NSCLC metastases to the spine require SRS because cEBRT usually fails to provide consistent long-term local control. Patients with spinal cord compression secondary to NSCLC require surgical decompression to safely undergo SRS and to reduce the risk of radiation-induced spinal cord injury. Separation surgery allows spinal cord decompression and spinal stabilization using the posterior approach and, in combination with SRS, has been shown to provide reliable local control with low risk of wound complication or spinal hardware fracture.

BACKGROUND CONTEXT/BACKGROUND:Lumbar metastases can result in spinal instability and mechanical radiculopathy, characterized by radicular pain produced by axial loading. This pain pattern represents a definitive symptom of neoplastic instability and may serve as a reliable indication for surgical stabilization. PURPOSE/OBJECTIVE:We examined the results of surgical decompression and fixation in the treatment of mechanical radiculopathy. STUDY DESIGN/SETTING/METHODS:A retrospective clinical study. PATIENT SAMPLE/METHODS:An internally maintained spine neurosurgery database was queried between February 2002 and April 2010. Patients were identified and deemed eligible for inclusion in this study based on the presence of all the following: metastatic tumor, lumbar surgery, and lumbar radiculopathy. OUTCOME MEASURES/METHODS:Visual analog scale (VAS) of pain and Eastern Cooperative Oncology Group (ECOG) status. METHODS:The Memorial Sloan-Kettering Cancer Center Department of Neurosurgery operative database was queried over an 8-year period to identify all patients with spinal metastases who underwent lumbar surgery. Only patients whose operative indication included mechanical radiculopathy were included. Pre- and postoperative pain was assessed with the VAS of pain, whereas pre- and postoperative performance status was evaluated using the ECOG. RESULTS:Fifty-five patients were included in the cohort. L2 and L3 were the most common levels involved, and most patients underwent multilevel posterior decompression and instrumented fusion. After surgery, 98% of patients reported pain relief. A significant difference between average pre- and postoperative pain scores was found (p<.01). Overall, 41.5% of patients experienced improvement in their ECOG score postoperatively. CONCLUSIONS:Mechanical radiculopathy in patients with spinal metastases represents a highly reliable surgical indication. Spinal decompression and fixation is an effective treatment for pain palliation in this patient population.

BACKGROUND CONTEXT/BACKGROUND:Spine metastases occur frequently in patients with cancer. A variety of surgical approaches, including anterior transcavitary, lateral extracavitary, posterolateral, and/or combined techniques are used for spinal cord decompression and restoration of spinal stability. The incidence of symptomatic hardware failure is unknown for the majority of these approaches. PURPOSE/OBJECTIVE:The purpose of this study was to determine the incidence of symptomatic hardware failure and the associated risk factors in patients with metastatic epidural spinal cord compression (MESCC). STUDY DESIGN/SETTING/METHODS:This was a retrospective study. PATIENT SAMPLE/METHODS:The current series analyzes a cohort of 318 patients who underwent separation surgery, which involves single-stage posterolateral decompression and posterior segmental instrumentation for MESCC. OUTCOME MEASURES/METHODS:The event of interest was hardware failure; the competing event was death resulting from any cause. All patients were monitored for survival analysis. A competing risk analysis was conducted to examine univariately a number of potential risk factors associated with hardware failure, including junctional level, gender, construct length, and the presence or absence of prior chest wall resection. METHODS:A retrospective analysis and chart review were performed for 318 consecutive patients who underwent posterolateral decompression and posterior screw-rod fixation without supplemental anterior fixation from March 2004 to June 2011 at our institution. The median follow-up time for survivors without hardware failure was 399 days (range, 9-2,828), with a mean operative time of 3 hours. A total of 78% of patients died during the 7-year study period. RESULTS:Of the 318 patients, nine (2.8%) exhibited signs and symptoms of hardware failure and required revision of the instrumentation. Patients with chest wall resection and those with initial construct length greater than six contiguous spinal levels exhibited a statistically significantly higher risk of symptomatic hardware failure than their counterparts. We observed a trend toward an increased risk of failure in women compared with men (p=.09). CONCLUSIONS:The incidence of hardware failure is low in patients with MESCC who undergo posterolateral decompression and posterior screw-rod instrumentation. Moreover, the short operative time and low morbidity profile associated with this approach make it a reliable and acceptable method for the surgical treatment of MESCC. Patients with constructs spanning six or more levels or those with prior chest wall resection are at higher risk for instrumentation failure.

Significant evidence emerging in the spinal oncology literature recommends radiosurgery as a primary modality of treatment of spinal metastasis. Improvements in the methods of delivering radiation have increased the ability to provide a higher and more exacting dose of radiation to a tumor bed than previously. Using treatment-planning software, radiation is contoured around a specific lesion with the intent of administering a tumoricidal dose. Combined with a minimally invasive, tumor-load reducing surgery, this advanced form of radiation therapy can provide better local control of the tumor compared with conventional external beam radiation.

PURPOSE/OBJECTIVE:Conventional radiation treatment (20-40 Gy in 5-20 fractions, 2-5 Gy per fraction) for sarcoma metastatic to the spine provides subtherapeutic doses, resulting in poor durable local control (LC) (50%-77% at 1 year). Hypofractionated (HF) and/or single-fraction (SF) image-guided stereotactic radiosurgery (IG-SRS) may provide a more effective means of managing these lesions. METHODS AND MATERIALS/METHODS:Patients with pathologically proven high-grade sarcoma metastatic to the spine treated with HF and SF IG-SRS were included. LC and overall survival (OS) were analyzed by the use of Kaplan-Meier statistics. Univariate and multivariate analyses were performed by the use of Cox regression with competing-risks analysis; all confidence intervals are 95%. Toxicities were assessed according to Common Terminology Criteria for Adverse Events, version 4.0. RESULTS:From May 2005 to November 11, 2012, 88 patients with 120 discrete metastases received HF (3-6 fractions; median dose, 28.5 Gy; n=52, 43.3%) or SF IG-SRS (median dose, 24 Gy; n=68, 56.7%). The median follow-up time was 12.3 months. At 12 months, LC was 87.9% (confidence interval [CI], 81.3%-94.5%), OS was 60.6% (CI, 49.6%-71.6%), and median survival was 16.9 months. SF IG-SRS demonstrated superior LC to HF IG-SRS (12-month LC of 90.8% [CI, 83%-98.6%] vs 84.1% [CI, 72.9%-95.3%] P=.007) and retained significance on multivariate analysis (P=.030, hazard ratio 0.345; CI, 0.132-0.901]. Treatment was well tolerated, with 1% acute grade 3 toxicity, 4.5% chronic grade 3 toxicity, and no grade >3 toxicities. CONCLUSIONS:In the largest series of metastatic sarcoma to the spine to date, IG-SRS provides excellent LC in the setting of an aggressive disease with low radiation sensitivity and poor prognosis. Single-fraction IG-SRS is associated with the highest rates of LC with minimal toxicity.

OBJECT/OBJECTIVE:Radiation therapy is known to impair wound healing. Higher dose per fraction is believed to increase this risk. This study sought to quantify rates of wound complication in patients receiving preoperative conventionally fractionated radiotherapy (XRT) or high-dose hypofractionated image-guided radiation therapy (IGRT) for spinal metastasis, and to identify predictors of wound complication. METHODS:The records of 165 consecutive patients who underwent spine surgery for metastasis at Memorial Sloan-Kettering Cancer Center between 1999 and 2010, with a history of prior radiation therapy, were reviewed. Patients with primary spine tumors, 2 courses of prior radiation therapy to the surgical site, total dose < 9 Gy, or radiation therapy adjacent to or partially overlapping the surgical site, were excluded. One hundred thirty patients received XRT (≤ 3 Gy/fraction) and 35 received IGRT (> 3 Gy/fraction). The total dose prescribed to the 100% isodose line to treat the planning target volume was 18-30 Gy in 1-5 fractions. Clinical factors evaluated included age, Karnofsky Performance Scale score, body mass index, presence of diabetes, smoking, ambulatory status, prior surgery at same spinal site, preoperative laboratory results (hemoglobin, lymphocyte count, and albumin), perioperative chemotherapy or steroids, estimated blood loss, extent of stabilization hardware, time between radiation therapy and surgery, number of vertebral bodies irradiated, total radiation dose, and dose per fraction of radiation therapy. Wound complication was defined as poor healing, dehiscence, or infection. Potential predictors of wound complication were assessed by univariate analyses using competing-risk methods to adjust for risk of death. results: For XRT patients, median dose was 30 Gy (range 11.5-70 Gy) with 72% of them receiving 3 Gy × 10 fractions. For IGRT patients, 66% received 18-24 Gy × 1 fraction and 23% received 6 Gy × 5 fractions. Groups differed only by the mean number of vertebral bodies treated (4.6 XRT and 1.8 IGRT, p < 0.0001). Wound complications occurred at a median of 0.95 months (range 0.4-3.9 months). A total of 22 wound events occurred in the XRT group and 2 in the IGRT group. The 6-month cumulative incidence of wound complications for XRT was 17% and for IGRT was 6%. There was no significant difference in wound complications between groups (IGRT vs XRT: hazard ratio 0.31, 95% CI 0.08-1.3; p = 0.11). Higher dose per fraction appeared to be associated with a lower risk of wound complication (hazard ratio 0.27, 95% CI 0.06-1.15; p = 0.08), which trended toward significance. Univariate analyses did not reveal any significant predictors of wound complications. CONCLUSIONS:Patients who underwent XRT or IGRT did not have significantly different rates of postoperative wound complications. This finding may be explained by the treatment of fewer vertebral bodies in IGRT patients, or by the low overall number of total events. With a wound complication rate of 6%, preoperative IGRT, a highly conformal treatment, resulted in a very low rate of surgical wound complication.