- Email: [email protected]
Osteosarcoma of the spine: experience in 26 patients treated at the Massachusetts General Hospital
The Spine Journal 10 (2010) 708–714
Clinical Study
Osteosarcoma of the spine: experience in 26 patients treated at the Massachusetts General Hospital Andrew J. Schoenfeld, MDa,*, Francis J. Hornicek, MD, PhDa, Francis X. Pedlow, MDa, Wendy Kobayashi, BSb, Ronald T. Garcia, BSa, Thomas F. DeLaney, MDb, Dempsey Springfield, MDa, Henry J. Mankin, MDa, Joseph H. Schwab, MD, MSa a
Department of Orthopaedic Surgery, Harvard Medical School, Massachusetts General Hospital, Boston, MA 02114, USA Department of Radiation Oncology, Harvard Medical School, Massachusetts General Hospital, Boston, MA 02114, USA
b
Received 6 February 2010; revised 21 April 2010; accepted 22 May 2010
Abstract
BACKGROUND CONTEXT: Because of the low incidence, treatment recommendations for spinal osteosarcoma are guided by the results of small series and case reports. Many include patients who presented for treatment over the course of three to four decades. PURPOSE: The goal of this investigation was to report the treatments, results, and overall survivorship of 26 patients treated for osteosarcoma of the spine at a single institution. STUDY DESIGN: Retrospective prognostic series (Level III evidence). PATIENT SAMPLE: Twenty-six patients treated at a single center for osteosarcoma of the spine over a 26-year period. OUTCOME MEASURES: Estimation of patient survival, local recurrence, and the presence of metastatic disease. METHODS: We performed a retrospective review of cases of osteosarcoma involving the spine treated at our institution between 1982 and 2008. Medical charts, radiology reports, pathology reports, and operative notes were reviewed for all patients. Available imaging studies were also reviewed. The log-rank test was used to compare baseline differences between groups. Survivorship analysis was performed using Kaplan–Meier methodology. The effect of Paget osteosarcoma, type of resection, presence of local recurrence, tumor size, surgical margins, and metastases on overall survival were also investigated using the log-rank test. RESULTS: Twenty-six patients were included for review in this study. Twenty individuals were treated surgically, and 24 were treated with radiation with a mean dose of 62.2 Gy (range 20–84.7 Gy). Twenty-five patients received chemotherapy. Of those treated surgically, seven received en bloc resection. The median overall survival for all patients in our series was 29.5 months (standard error 14.7, 95% confidence interval 0.6–58). Local recurrence developed in 7 patients (27%), and metastasis occurred in 16 individuals (62%). Patients with Paget osteosarcoma had worse overall survival (p!.001). CONCLUSIONS: Results presented here confirm a poor prognosis for patients with spinal osteosarcoma. Although combination therapies, including surgery, chemotherapy, and high-dose radiation, achieve adequate short-term survival, the 5-year mortality rate remains high. Published by Elsevier Inc.
Keywords:
Osteosarcoma; Spine; En bloc resection; Radiation; Outcomes
FDA device/drug status: not applicable. Author disclosures: FJH (consulting, Stryker Corporation; speaking and/or teaching arrangements, Stryker Corporation; trips/travel, Stryker Corporation; scientific advisory board, Chordoma Foundation, Desmoid Tumor Research Foundation, American Association Tissue Banks; research support: staff and/or materials, PharmaMar, Stephan L Harris Chordoma Center, Sarcoma Foundation, Centecor Corporation; grants, PharmaMar, Stephan L Harris Chordoma Center, Sarcoma Foundation, Centecor Corporation; fellowship support, Zimmer Corporation); TFD 1529-9430/$ - see front matter Published by Elsevier Inc. doi:10.1016/j.spinee.2010.05.017
(speaking and/or teaching arrangement, IBA Proton Therapy); DS (stock ownership, including options and warrants, Johnson & Johnson; research support: staff and/or materials, Stryker; fellowship support, Zimmer); JHS (research support: staff and/or materials, DePuy, Synthes Spine, Globus, Medtronic; fellowship support, AO Spine Fellowship). * Corresponding author. Department of Orthopaedic Surgery, Harvard Medical School, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA. Tel.: (617) 670-0304. E-mail address: [email protected] (A.J. Schoenfeld)
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
709
Introduction Osteosarcoma is the most common primary malignant tumor of bone; however, only 3% to 5% of cases occur in the spine [1–6]. Primary osteogenic sarcoma of the spine has been treated with multimodality therapy including surgery, chemotherapy, and radiation [7–9]. The most effective surgical intervention in the spine is presumed to be a wide resection, similar to the experience with osteosarcoma in the extremities [10–14]. However, because of its anatomic location and the proximity of vital structures, wide resection of spinal osteosarcoma is not always feasible. Furthermore, although the combination of surgical resection with radiation and chemotherapy for osteosarcoma of the spine has achieved encouraging short-term results, reports of long-term survival reported to date remain below 20% [5,7,9,12,13]. Because of the low incidence of spinal osteosarcoma, treatment recommendations are guided by small series and case reports that include patients who presented for treatment over the course of three to four decades [1,4,12,15]. Additionally, there are only a few studies that include patients treated with en bloc resections [5,9,12]. One recent series demonstrated that wide, or marginal, resections resulted in improved survival among patients with spinal osteosarcoma [9]. The goal of this investigation was to describe the outcomes of patients with osteosarcoma of the spine, treated at our institution, over a 26-year period. To the best of our knowledge, this report represents one of the largest series of patients treated for spinal osteosarcoma.
Materials and methods We performed a retrospective review of all surgical cases of spinal osteosarcoma treated at our institution from 1982 through 2008. Permission from the Massachusetts General Hospital Institutional Review Board was obtained before commencing this study. Three databases were queried to obtain patient information. The first was the Massachusetts General Hospital Orthopedic Oncology database [16]. This is a patient registry that includes all benign and malignant tumors treated by the Orthopedic Oncology service at our institution. The second was the Massachusetts General Hospital Cancer Center Tumor Registry database, and the third was the sarcoma database maintained by the Department of Radiation Oncology. Collected information was cross-referenced between the three databases for accuracy and to eliminate patient duplication. Patients’ medical records, pathology reports, clinic notes, operative reports, hospital discharge summaries, and radiology records were also reviewed. Only patients with a confirmed tissue diagnosis of osteosarcoma, those treated medically or surgically at our institution, and survivors with more than 1-year clinical follow-up were included for analysis in this investigation. Surgical reports were reviewed to determine whether the tumor was removed in one piece (en bloc resection) versus
Context Osteosarcoma of the spine is rare, and surgical resection strategies are limited by adjacent critical structures. The authors presented cases cared for at their center over a 26-year period. Contribution Despite technical advances in surgical, chemotherapeutic, and radiation-based therapies, the 5-year prognosis for survival remained poor. Implications Often, the available evidence for rare diseases is Level IV. In such cases, the coordinated establishment of registries can make this information ‘‘best’’ available, improving the likelihood that novel effective treatments might emerge and be recognized. —The Editors
a piecemeal (intralesional) excision. The tumor grade, histological subtype, and surgical margins were based on a review of pathology reports. Tumor stage was retrospectively classified based on the Enneking surgical staging system [17]. The methods of radiation treatment were recorded for each patient. Radiation was delivered as part of the definitive treatment or for palliative purposes in cases of metastatic disease. The total dose of radiation delivered was recorded and whether the radiation was given preoperatively, intraoperatively, or postoperatively. The type and duration of chemotherapy were also determined. The characterization of local recurrence was based on a review of patient notes, as well as radiology reports and images, when available. An identical method was used to document metastasis. For patients who were treated with an intralesional resection, or radiation alone, recurrence was determined to be an increase in the size of the tumor after treatment or an increase in tumor burden resulting in severe pain and/or neurologic compromise. The duration of follow-up was determined based on the date of surgery at our institution. Patients were followed up in the postoperative period and reevaluated at 3 months, 6 months, 1 year, and annually thereafter. Physical examination and plain film imaging studies were performed at every follow-up visit, and computed tomographic or magnetic resonance imaging was also ordered based on the patient symptoms or physical examination findings. The last date of follow-up represents the last documented contact with the patient or the date of their death. Disease status was recorded as no evidence of disease, or alive with disease, for patients still living at the time of final evaluation. Deaths were reported for those who died of any cause and those who died from complications related to
710
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
osteosarcoma. The national Social Security Index was cross-referenced against our mortality data to ensure accuracy. Discrepancies in medical records were resolved by the consensus of two researchers (AJS and JHS) who independently reviewed patients’ hospital and office charts. Kaplan–Meier methods were used to calculate diseasespecific survival for all patients with osteosarcoma. In addition, we used the log-rank test to determine the effect that type of osteosarcoma (ie, conventional vs. Paget, or radiation induced), surgery, type of resection, surgical margins, presence of local recurrence, metastasis, and tumor size had on disease-specific survival. We also sought to determine whether patients treated entirely at our institution had better survival than those initially treated at an outside facility. A p value of .05 or less was deemed significant. A b error was calculated for those comparisons that did not demonstrate statistical significance.
Results Between 1982 and 2008, 35 patients were evaluated for spinal osteosarcoma at the Massachusetts General Hospital. Two patients were excluded because of insufficient followup, and three did not receive their treatment at our institution. Four individuals lacked a tissue diagnosis confirming osteosarcoma histology and were also excluded from the series. Twenty-six patients were ultimately included for review. The cohort consisted of 15 men and 11 women, with an average age of 46 years at the time of initial presentation (range 5–82 years). Eight patients were treated between 1982 and 1998, and the remaining 18 were treated in the years 1999 to 2008. The cervical spine was involved in three instances, the thoracic in six (Fig. 1), and the lumbar spine in nine, and eight patients had osteosarcoma of the sacrum. There were nine cases of chondroblastic subtype, nine instances of osteoblastic subtype, two tumors with fibroblastic subtype, and one case each of small-cell- and
giant-cell–rich tumors. There were no cases with telangiectatic osteosarcoma. Four individuals had Paget osteosarcoma, and another four had radiation-associated osteosarcoma. At the time of initial presentation, all patients were considered to have high-grade tumors, classified as Enneking Stage IIB. Ten patients had tumors that were larger than 8 cm as determined on preoperative computed tomographic scans and/or magnetic resonance imaging. Eight patients were initially treated at outside facilities and only referred to the Massachusetts General Hospital after incomplete resection or tumor recurrence (Fig. 1 and Fig. 2). Eighteen individuals were treated entirely at our institution. Six patients did not receive surgery, whereas 20 underwent surgical intervention for their osteosarcoma. Of these, seven received a complete en bloc resection of their tumors using combined anterior-posterior procedures (Fig. 2). Among patients receiving en bloc resection, three achieved a wide or marginal excision and four had contaminated margins. Thirteen patients underwent incomplete, or intralesional, resections. No attempt at en bloc excision was made in these instances. Of the six patients who were not treated surgically, five received a combination of chemotherapy and radiation and one was treated with radiation alone. Twenty-four patients received radiation therapy. The average dose delivered was 62.2 Gy (range 20–84.7 Gy). Six patients received preoperative radiation, and six were treated with intraoperative radiation. Eighteen individuals received a combination of photon and proton therapy, and six were treated with photons alone. Of the two patients who did not receive radiation, one had achieved a wide surgical margin and was treated with neoadjuvant chemotherapy. The other individual was deemed to require palliative care after surgery, and radiation was withheld. Twenty-five patients were treated with chemotherapy. The sole individual who did not receive chemotherapy had a Paget associated osteosarcoma and was treated with radiation only for palliative pain relief. Twenty-one were treated with adriamycin, methotrexate, and cisplatin,
Fig. 1. (Left) Axial and (Middle) sagittal computed tomographic images and (Right) axial magnetic resonance scan of a 46-year-old woman with osteosarcoma at T6. The patient initially presented to an outside facility with progressive neurologic deficit in the lower extremities. An emergent posterior decompression was performed, and the patient was transferred to the Massachusetts General Hospital for definitive treatment. The patient received preoperative chemotherapy and radiation treatment before surgical intervention. All the images were obtained after preoperative chemotherapy and radiation but before the definitive surgery.
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
711
Fig. 2. Postoperative (Left) anteroposterior and (Right) lateral radiographic images of the same patient presented in Fig. 1. The patient received a staged en bloc excision of T6 with anterior-posterior reconstruction. A marginal margin was obtained at the time of surgery. The patient received postoperative chemotherapy and radiation but ultimately developed lung metastases and died 10 months after surgery.
whereas four patients received a regimen of etoposide and ifosfamide only. Local recurrence ultimately developed in seven patients (27%), with six of these having received surgery. Metastatic disease occurred in 16 patients (62%). Thirteen patients had metastases at one site, whereas three had metastases in multiple locations (Table 1). The lung was the most common site of metastatic spread. At the last follow-up, 19 patients (73%) were dead of disease. Five patients (19%) were alive with no evidence of disease, and two (8%) were alive with disease. Of these two individuals, one had developed local recurrence and the other was living with lung metastases. Two of the five individuals alive without disease had been treated with radiation and chemotherapy only (Table 2). Five of the seven patients who had received en bloc excision and two of the three individuals with wide, or marginal, margins were dead of disease. Three of the four patients treated with etoposide and ifosfamide only were also dead of disease. Median overall survival for patients in the cohort was 29.5 months (standard error [SE] 14.7, 95% confidence Table 1 Sites of metastatic spread in the 16 patients with metastatic osteosarcoma of the spine Site
Number of cases (%)*
Lung Bone Brain Liver Other
13 3 1 1 3
(81) (19) (6) (6) (19)
* Percentages do not add up to 100% because some patients had metastatic disease at more than one location.
interval [CI] 0.6–58) (Fig. 3). All eight patients treated before 1999 died of disease (average survival 29.2 months, range 3.4–64.5 months), whereas 11 of the 18 individuals treated after 1999 were dead of disease (average survival 26.5 months, range 2.9–78 months). The 1- and 5-year survival rates for all patients were 69% and 31%, respectively. Average follow-up for patients alive at final evaluation was 35.9 months (range 22–78 months). The average follow-up for those with no evidence of disease was 42.8 months (range 22–78 months) (Table 2). There were 10 treatment-related complications in eight patients (31%). Incidental durotomy during surgery was the most common, occurring in four instances. Postoperative infection presented in two cases, and neurologic deficit, myocardial infarction, and deep venous thrombosis all occurred in one case each. One patient died in the postoperative period from chemotherapy-related complications. This patient had previously received an en bloc resection with contaminated margins. The average overall survival for patients without Paget and radiation-associated osteosarcoma were 38 months (SE 7.3, 95% CI 23.4–52.5) and 29.5 months (SE 31.6, 95% CI 2.7–56.3), respectively. Median survival for patients with Paget osteosarcoma was 5.6 months (SE 1.4, 95% CI 2.9–8.2), whereas those with radiation-associated osteosarcoma had a median survival of 14.9 months (SE 13.1, 95% CI 0–44.4). Patients with Paget osteosarcoma were found to have a significantly increased risk of mortality relative to those with conventional osteosarcoma (p!.001). With the exception of the presence of Paget osteosarcoma, no other risk factors exhibited a statistically significant influence on mortality. There was a statistical trend for patients with metastases to have an increased risk of
712
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
Table 2 Details of five patients with spinal osteosarcoma and no evidence of disease Age (y)
Gender
Location
Year of first treatment
Enneking stage
Previous surgery
Margin
Radiation dose
Intraoperative radiation
Chemo
Follow-up (mo)
56 5 52 20 69
M F F M M
T L C S S
2007 2007 2005 2003 2006
IIB IIB IIB IIB IIB
Yes Yes No No No
Contaminated Intra Intra None None
70.2 50.4 64.8 70.2 74
Yes No No N/A N/A
Yes Yes Yes Yes Yes
28 22 51 78 35
M, male; F, female; C, cervical; T, thoracic; L, lumbar; S, sacrum; Intra, intralesional; None, patient did not receive surgery; N/A, not applicable.
mortality, but this did not reach significance because of a lack of power (p5.27, b50.29). Additional risks of mortality for the remaining factors and the resultant p values are encountered in Table 3. Discussion Spinal osteosarcoma is a rare condition, representing 3% to 5% of all cases of osteosarcoma and approximately 10% of primary spinal tumors [2,3,5,6,9,12]. Wide resection and neoadjuvant chemotherapy have revolutionized the treatment of osteosarcoma in the extremities, significantly improving patient outcomes [10–14]. However, no similar benefit has been evinced for spinal osteosarcoma over the last 30 years, despite significant advances in diagnostic and treatment modalities. Advanced surgical techniques, enabling complete removal of the tumor, are presently available [5,12,13,15, 18–20], and a number of studies have suggested that these procedures, combined with chemotherapy or radiation, may
impart long-term survival [5,8,9,12,13,21,22]. Nonetheless, anatomic constraints of the spine can limit the ability to achieve wide surgical margins, and en bloc resection is not always practical. Traditionally, adjuvant radiotherapy was limited by the dose constraints of the spinal cord and other critical normal tissues, but recent techniques such as intensitymodulated radiation therapy [23] and proton radiation therapy [8] now allow radiation doses to be delivered that may be effective for microscopic and even gross residual tumor. Nonetheless, for most patients treated to date, long-term survival for conventional spinal osteosarcoma remains poor, and the outlook is even worse for those with Paget osteosarcoma [24–26]. One of the major obstacles in studying the impact of treatment regimens on spinal osteosarcoma is that the condition is rare, and it is difficult for any one center to evaluate a number of patients treated in similar fashion over a short period of time. Thus, treatment recommendations are informed by a small number of series encompassing patients treated over the course of decades [1,4,7,9] or singular case reports with limited long-term follow-up [19,22,27–30]. The present study documented outcomes for patients with spinal osteosarcoma treated over the course of a 26year period. Results confirm that, despite advances in diagnostic and treatment regimens, the long-term outcome for individuals with osteosarcoma of the spine is poor. Table 3 Risks of mortality for patients with osteosarcoma of the spine Risk factor
Fig. 3. Kaplan–Meier curve illustrating the disease-specific survival estimates for all patients. Cum survival, cumulative survival; f/u months, follow-up in months.
Paget osteosarcoma Radiation-induced osteosarcoma Tumor size O8 cm Tumor size !8 cm Wide/marginal excision No wide/marginal excision En bloc resection No en bloc resection Surgical intervention No surgery Local recurrence No local recurrence Metastases No metastases All treatment at the Massachusetts General Hospital Initial treatment elsewhere
Mortality rate, % (median survival, mo) 100 (5.6) 100 (14.9) 80 69 66 74 71 74 80 50 100 63 94 40 72 75
p Value (b error) !.001 .91 (0.99) .45 (0.97) .58 (0.93) .47 (0.95) .25 (0.93) .48 (0.86) .27 (0.29) .94 (0.95)
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
Although those with Paget osteosarcoma and metastases may be at greater risk for early mortality, en bloc excision, negative surgical margins, and the use of adjuvant chemotherapy and radiation were not found to positively influence outcome. We recognize that this may be because of the low number of patients in our study. However, when compared to other series available in the literature, this report represents one of the largest series of patients with spinal osteosarcoma. Early reports regarding spinal osteosarcoma documented uniformly poor outcomes with patient mortality occurring shortly after presentation. Shives et al. [4] published a series of 27 patients with osteosarcoma of the spine, collected over the course of 70 years. Ninety-six percent of patients in this cohort died of disease, with a median survival of only 10 months. Similarly, Barwick et al. [1] documented the cases of 10 patients with spinal osteosarcoma treated at a single center over 60 years. In this series, there was a 90% patient mortality and median survival was 6 months. Sundaresan et al. [13] reviewed the experience of 11 patients treated from 1978 to 1984 with multimodal therapy, including aggressive surgical resection, chemotherapy, and radiation. Although there was a 55% mortality in this series, only one individual developed metastases while receiving multimodal treatment. More recently, Ozaki et al. [9] reported results for 22 patients with spinal osteosarcoma treated from 1979 to 1998. Twelve patients underwent surgery, and 10 were treated nonoperatively with radiation and/or the Cooperative Osteosarcoma Study Group chemotherapy protocol. Survival at 1 year was 86%, but only 36% were alive after more than 2 years. These authors reported that the presence of metastases and tumors larger than 10 cm adversely impacted outcome, whereas surgical resection with wide margins improved survival. In terms of patient age, the population in the present study was older than that presented in the work of Ozaki et al. [9]. The rate of metastases was also substantially higher (62% vs. 27%), although median survival and overall mortality approximated each other [9]. The high rate of metastases in the present study may be attributed to the size of the cohort, but such a figure can also reflect the aggressive nature of spinal osteosarcoma. It is interesting to note that in the study of Ozaki et al. [9], the sacrum was the most frequent site of osteosarcoma (68% of patients in the series). In the present investigation, the lumbar region was most commonly involved and sacral tumors comprised only 31% of the total. This is important, as Ozaki et al. [9] reported that sacral osteosarcoma was associated with a significantly increased risk of mortality. Additionally, the cohort presented by Ozaki et al. [9] included no patients with Paget osteosarcoma. Sarcomatous transformation in the setting of Paget disease has long been known to carry a poor prognosis [3,5,24–26,31]. For example, a report of 13 cases of Paget osteosarcoma of the spine
713
from the Scottish Tumor Registry documented an average survival of 4.2 months [25]. If patients with Paget sarcoma are excluded from the present investigation, although median overall survival remains the same, the 1-year survival rate approaches the figure of 86% presented in the work of Ozaki et al. [9]. This investigation also differs from previous works in that most patients in the series were treated aggressively with radiation, including proton beam therapy. Several reports have documented satisfactory outcomes for patients with spinal tumors treated using similar protocols [5,8,21,32]. Wagner et al. [32] encountered an overall survival of 65%, and a continuous disease-free survival of 54%, in patients with spinal, or pelvic, tumors treated with surgical resection and a combination of pre- and postoperative radiation, although most of these patients did not have osteosarcoma. DeLaney et al. [8,21] have also reported similar results, with local control rates for spinal sarcomas approaching 80% after varying degrees of resection and high-dose photon/proton radiation. A dose-response relationship with regard to local control of osteosarcomas was reported in one series with poor local control after 30 Gy and effective control when more than 90 Gy was delivered [21]. However, the delivery of conventional radiation to the spine must be balanced with the risk of local toxicity to the spinal cord. For this reason, most large series of osteogenic sarcoma involving the spine have traditionally not used doses greater than 50 Gy. Presently, at our institution, en bloc resection with a wide surgical margin is considered for patients with osteosarcoma of the spine who have at least one pedicle free of tumor and no evidence of metastatic disease [18]. If there is significant concern that a negative margin can be achieved only by sacrificing nerve roots, the sensory and motor functions of those roots are taken into account and included in the preoperative discussion with the patient, so that an informed decision regarding treatment can be made. There are instances where, short of spinal cord transection, an en bloc resection with negative margins is not possible. If the morbidity associated with en bloc resection with negative margins is unacceptable to the patient, then we recommend preoperative radiation followed by gross total resection, followed again with postoperative radiation. Our institution delivers more than 70 Gy of radiation in this setting. This protocol is in addition to standard neoadjuvant and adjuvant chemotherapy [8,21,32]. Quality of life, projected longevity, and the patient’s symptoms figure strongly in decisions regarding surgery for those who have advanced local disease and/or metastases. Patient wishes are paramount in these situations after an extensive preoperative discussion. Although en bloc resection may not be suitable or acceptable in all cases, surgical intervention can still have a positive impact on a patient’s quality of life, especially if the tumor is causing progressive neurologic deficits and resulting in loss of the ability to ambulate.
714
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
There are several limitations to this study. First, this was a retrospective investigation, conducted using information compiled from institutional databases. Therefore, this study suffers from the limitations inherent to such a research design. Second, the patients under investigation were treated over the course of 26 years by different physicians, and the care of these individuals evolved with advances in the understanding of osteosarcoma, its response to chemotherapy and radiation, and the technical improvements in the field of spine surgery. The impact of evolving treatment modalities on disease course and outcome cannot be quantified but must be considered as potential confounders. In addition, our study lacked sufficient power to detect differences between treatments rendered and presenting characteristics thought to be important in survival, such as metastasis. While recognizing these limitations, it is hoped that this study may serve as a baseline to which future prospective investigations may be compared. At present, all patients treated for spinal osteosarcoma at our center are enrolled in a prospective database. In conclusion, the results of this study confirm a continued poor prognosis for patients with spinal osteosarcoma. Although combination therapies, including surgery, chemotherapy, and high-dose radiation, may enhance short-term survival, the mortality rate at 5 years remains high. Paget associated osteosarcoma of the spine has a particularly poor prognosis.
References [1] Barwick KW, Huvos AG, Smith J. Primary osteogenic sarcoma of the vertebral column: a clinicopathologic correlation of ten patients. Cancer 1980;46:595–604. [2] Dreghorn CR, Newman RJ, Hardy GJ, Dickson RA. Primary tumors of the axial skeleton. Experience of the Leeds Regional Bone Tumor Registry. Spine 1990;15:137–40. [3] Kelley SP, Ashford RU, Rao AS, Dickson RA. Primary bone tumours of the spine: a 42-year survey from the Leeds Regional Bone Tumour Registry. Eur Spine J 2007;16:405–9. [4] Shives TC, Dahlin DC, Sim FH, et al. Osteosarcoma of the spine. J Bone Joint Surg Am 1986;68:660–8. [5] Sundaresan N, Rosen G, Huvos AG, Krol G. Combined treatment of osteosarcoma of the spine. Neurosurgery 1988;23:714–9. [6] Weinstein JN, McLain RF. Primary tumors of the spine. Spine 1987;12:843–51. [7] Bielack SS, Wulff B, Delling G, et al. Osteosarcoma of the trunk treated by multimodal therapy: experience of the Cooperative Osteosarcoma Study Group (COSS). Med Pediatr Oncol 1995;24: 6–12. [8] DeLaney TF, Park L, Goldberg SI, et al. Radiotherapy for local control of osteosarcoma. Int J Radiat Oncol Biol Phys 2005;61: 492–8. [9] Ozaki T, Flege S, Liljenqvist U, et al. Osteosarcoma of the spine: experience of the Cooperative Osteosarcoma Study Group. Cancer 2002;94:1069–77.
[10] Mankin HJ, Hornicek FJ, Rosenberg AE, et al. Survival data for 648 patients with osteosarcoma treated at one institution. Clin Orthop Relat Res 2004;429:286–91. [11] Schwab JH, Antonescu CR, Athanasian EA, et al. A comparison of intramedullary and juxtacortical low-grade osteogenic sarcoma. Clin Orthop Relat Res 2008;466:1318–22. [12] Sciubba DM, Okuno SH, Dekutoski MB, Gokaslan ZL. Ewing and osteogenic sarcoma: evidence for multidisciplinary management. Spine 2009;34:S58–68. [13] Sundaresan N, Rosen G, Boriani S. Primary malignant tumors of the spine. Orthop Clin North Am 2009;40:21–36. [14] Weber K, Damron TA, Frassica FJ, Sim FH. Malignant bone tumors. Instr Course Lect 2008;57:673–88. [15] Tasdemiroglu E, Bagatur E, Ayan I, et al. Primary spinal column sarcomas. Acta Neurochir (Wien) 1996;138:1261–6. [16] Refaat Y, Gunnoe J, Hornicek FJ, Mankin HJ. Comparison of quality of life after amputation or limb salvage. Clin Orthop Relat Res 2002;397:298–305. [17] Enneking WF, Spanier SS, Goodman MA. A system for the surgical staging of musculoskeletal sarcoma. Clin Orthop Relat Res 1980;153: 106–20. [18] Boriani S, Weinstein JN, Biagini R. Primary bone tumors of the spine. Terminology and surgical staging. Spine 1997;22:1036–44. [19] Kawahara N, Tomita K, Fujita T, et al. Osteosarcoma of the thoracolumbar spine: total en bloc spondylectomy. A case report. J Bone Joint Surg Am 1997;79:453–8. [20] Tomita K, Kawahara N, Baba H, et al. Total en bloc spondylectomy. A new surgical technique for primary malignant vertebral tumors. Spine 1997;22:324–33. [21] DeLaney TF, Liebsch NJ, Pedlow FX, et al. Phase II study of high-dose photon/proton radiotherapy in the management of spine sarcomas. Int J Radiat Oncol Biol Phys 2009;74:732–9. [22] Nishida J, Abe M, Shiraishi H, et al. Familial occurrence of telangiectatic osteosarcoma: cousin cases. J Pediatr Orthop 1994;14: 119–22. [23] Yamada Y, Lovelock DM, Yenice KM, et al. Multifractionated image-guided and stereotactic intensity-modulated radiotherapy of paraspinal tumors: a preliminary report. Int J Radiat Oncol Biol Phys 2005;62:53–61. [24] Huang TL, Cohen NJ, Sahgal S, Tseng CH. Osteosarcoma complicating Paget’s disease of the spine with neurologic complications. Clin Orthop Relat Res 1979;141:260–5. [25] Scottish Bone Tumor Registry, Sharma H, Mehdi SA, MacDuff E, et al. Paget sarcoma of the spine: Scottish Bone Tumor Registry experience. Spine 2006;31:1344–50. [26] Sofka CM, Ciavarra G, Saboeiro G, Ghelman B. Paget’s disease of the spine and secondary osteosarcoma. HSS J 2006;2:188–90. [27] Amritanand R, Venkatesh K, Cherian R, et al. Telangiectatic osteosarcoma of the spine: a case report. Eur Spine J 2008;17:S342–6. [28] Bauernhofer T, Sto¨ger H, Kasparek AK, et al. Combined treatment of metastatic osteosarcoma of the spine. Oncology 1999;57:265–8. [29] Sar C, Eralp L. Transoral resection and reconstruction for primary osteogenic sarcoma of the second cervical vertebra. Spine 2001;26: 1936–41. [30] Simon RG, Irwin RB. An unusual presentation of telangiectatic osteosarcoma. Am J Orthop 1996;25:375–9. [31] Mankin HJ, Hornicek FJ. Paget’s sarcoma: a historical and outcome review. Clin Orthop Relat Res 2005;438:97–102. [32] Wagner TD, Kobayashi W, Dean S, et al. Combination short-course preoperative irradiation, surgical resection, and reduced-field highdose postoperative irradiation in the treatment of tumors involving the bone. Int J Radiat Oncol Biol Phys 2009;73:259–66.
Clinical Study
Osteosarcoma of the spine: experience in 26 patients treated at the Massachusetts General Hospital Andrew J. Schoenfeld, MDa,*, Francis J. Hornicek, MD, PhDa, Francis X. Pedlow, MDa, Wendy Kobayashi, BSb, Ronald T. Garcia, BSa, Thomas F. DeLaney, MDb, Dempsey Springfield, MDa, Henry J. Mankin, MDa, Joseph H. Schwab, MD, MSa a
Department of Orthopaedic Surgery, Harvard Medical School, Massachusetts General Hospital, Boston, MA 02114, USA Department of Radiation Oncology, Harvard Medical School, Massachusetts General Hospital, Boston, MA 02114, USA
b
Received 6 February 2010; revised 21 April 2010; accepted 22 May 2010
Abstract
BACKGROUND CONTEXT: Because of the low incidence, treatment recommendations for spinal osteosarcoma are guided by the results of small series and case reports. Many include patients who presented for treatment over the course of three to four decades. PURPOSE: The goal of this investigation was to report the treatments, results, and overall survivorship of 26 patients treated for osteosarcoma of the spine at a single institution. STUDY DESIGN: Retrospective prognostic series (Level III evidence). PATIENT SAMPLE: Twenty-six patients treated at a single center for osteosarcoma of the spine over a 26-year period. OUTCOME MEASURES: Estimation of patient survival, local recurrence, and the presence of metastatic disease. METHODS: We performed a retrospective review of cases of osteosarcoma involving the spine treated at our institution between 1982 and 2008. Medical charts, radiology reports, pathology reports, and operative notes were reviewed for all patients. Available imaging studies were also reviewed. The log-rank test was used to compare baseline differences between groups. Survivorship analysis was performed using Kaplan–Meier methodology. The effect of Paget osteosarcoma, type of resection, presence of local recurrence, tumor size, surgical margins, and metastases on overall survival were also investigated using the log-rank test. RESULTS: Twenty-six patients were included for review in this study. Twenty individuals were treated surgically, and 24 were treated with radiation with a mean dose of 62.2 Gy (range 20–84.7 Gy). Twenty-five patients received chemotherapy. Of those treated surgically, seven received en bloc resection. The median overall survival for all patients in our series was 29.5 months (standard error 14.7, 95% confidence interval 0.6–58). Local recurrence developed in 7 patients (27%), and metastasis occurred in 16 individuals (62%). Patients with Paget osteosarcoma had worse overall survival (p!.001). CONCLUSIONS: Results presented here confirm a poor prognosis for patients with spinal osteosarcoma. Although combination therapies, including surgery, chemotherapy, and high-dose radiation, achieve adequate short-term survival, the 5-year mortality rate remains high. Published by Elsevier Inc.
Keywords:
Osteosarcoma; Spine; En bloc resection; Radiation; Outcomes
FDA device/drug status: not applicable. Author disclosures: FJH (consulting, Stryker Corporation; speaking and/or teaching arrangements, Stryker Corporation; trips/travel, Stryker Corporation; scientific advisory board, Chordoma Foundation, Desmoid Tumor Research Foundation, American Association Tissue Banks; research support: staff and/or materials, PharmaMar, Stephan L Harris Chordoma Center, Sarcoma Foundation, Centecor Corporation; grants, PharmaMar, Stephan L Harris Chordoma Center, Sarcoma Foundation, Centecor Corporation; fellowship support, Zimmer Corporation); TFD 1529-9430/$ - see front matter Published by Elsevier Inc. doi:10.1016/j.spinee.2010.05.017
(speaking and/or teaching arrangement, IBA Proton Therapy); DS (stock ownership, including options and warrants, Johnson & Johnson; research support: staff and/or materials, Stryker; fellowship support, Zimmer); JHS (research support: staff and/or materials, DePuy, Synthes Spine, Globus, Medtronic; fellowship support, AO Spine Fellowship). * Corresponding author. Department of Orthopaedic Surgery, Harvard Medical School, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA. Tel.: (617) 670-0304. E-mail address: [email protected] (A.J. Schoenfeld)
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
709
Introduction Osteosarcoma is the most common primary malignant tumor of bone; however, only 3% to 5% of cases occur in the spine [1–6]. Primary osteogenic sarcoma of the spine has been treated with multimodality therapy including surgery, chemotherapy, and radiation [7–9]. The most effective surgical intervention in the spine is presumed to be a wide resection, similar to the experience with osteosarcoma in the extremities [10–14]. However, because of its anatomic location and the proximity of vital structures, wide resection of spinal osteosarcoma is not always feasible. Furthermore, although the combination of surgical resection with radiation and chemotherapy for osteosarcoma of the spine has achieved encouraging short-term results, reports of long-term survival reported to date remain below 20% [5,7,9,12,13]. Because of the low incidence of spinal osteosarcoma, treatment recommendations are guided by small series and case reports that include patients who presented for treatment over the course of three to four decades [1,4,12,15]. Additionally, there are only a few studies that include patients treated with en bloc resections [5,9,12]. One recent series demonstrated that wide, or marginal, resections resulted in improved survival among patients with spinal osteosarcoma [9]. The goal of this investigation was to describe the outcomes of patients with osteosarcoma of the spine, treated at our institution, over a 26-year period. To the best of our knowledge, this report represents one of the largest series of patients treated for spinal osteosarcoma.
Materials and methods We performed a retrospective review of all surgical cases of spinal osteosarcoma treated at our institution from 1982 through 2008. Permission from the Massachusetts General Hospital Institutional Review Board was obtained before commencing this study. Three databases were queried to obtain patient information. The first was the Massachusetts General Hospital Orthopedic Oncology database [16]. This is a patient registry that includes all benign and malignant tumors treated by the Orthopedic Oncology service at our institution. The second was the Massachusetts General Hospital Cancer Center Tumor Registry database, and the third was the sarcoma database maintained by the Department of Radiation Oncology. Collected information was cross-referenced between the three databases for accuracy and to eliminate patient duplication. Patients’ medical records, pathology reports, clinic notes, operative reports, hospital discharge summaries, and radiology records were also reviewed. Only patients with a confirmed tissue diagnosis of osteosarcoma, those treated medically or surgically at our institution, and survivors with more than 1-year clinical follow-up were included for analysis in this investigation. Surgical reports were reviewed to determine whether the tumor was removed in one piece (en bloc resection) versus
Context Osteosarcoma of the spine is rare, and surgical resection strategies are limited by adjacent critical structures. The authors presented cases cared for at their center over a 26-year period. Contribution Despite technical advances in surgical, chemotherapeutic, and radiation-based therapies, the 5-year prognosis for survival remained poor. Implications Often, the available evidence for rare diseases is Level IV. In such cases, the coordinated establishment of registries can make this information ‘‘best’’ available, improving the likelihood that novel effective treatments might emerge and be recognized. —The Editors
a piecemeal (intralesional) excision. The tumor grade, histological subtype, and surgical margins were based on a review of pathology reports. Tumor stage was retrospectively classified based on the Enneking surgical staging system [17]. The methods of radiation treatment were recorded for each patient. Radiation was delivered as part of the definitive treatment or for palliative purposes in cases of metastatic disease. The total dose of radiation delivered was recorded and whether the radiation was given preoperatively, intraoperatively, or postoperatively. The type and duration of chemotherapy were also determined. The characterization of local recurrence was based on a review of patient notes, as well as radiology reports and images, when available. An identical method was used to document metastasis. For patients who were treated with an intralesional resection, or radiation alone, recurrence was determined to be an increase in the size of the tumor after treatment or an increase in tumor burden resulting in severe pain and/or neurologic compromise. The duration of follow-up was determined based on the date of surgery at our institution. Patients were followed up in the postoperative period and reevaluated at 3 months, 6 months, 1 year, and annually thereafter. Physical examination and plain film imaging studies were performed at every follow-up visit, and computed tomographic or magnetic resonance imaging was also ordered based on the patient symptoms or physical examination findings. The last date of follow-up represents the last documented contact with the patient or the date of their death. Disease status was recorded as no evidence of disease, or alive with disease, for patients still living at the time of final evaluation. Deaths were reported for those who died of any cause and those who died from complications related to
710
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
osteosarcoma. The national Social Security Index was cross-referenced against our mortality data to ensure accuracy. Discrepancies in medical records were resolved by the consensus of two researchers (AJS and JHS) who independently reviewed patients’ hospital and office charts. Kaplan–Meier methods were used to calculate diseasespecific survival for all patients with osteosarcoma. In addition, we used the log-rank test to determine the effect that type of osteosarcoma (ie, conventional vs. Paget, or radiation induced), surgery, type of resection, surgical margins, presence of local recurrence, metastasis, and tumor size had on disease-specific survival. We also sought to determine whether patients treated entirely at our institution had better survival than those initially treated at an outside facility. A p value of .05 or less was deemed significant. A b error was calculated for those comparisons that did not demonstrate statistical significance.
Results Between 1982 and 2008, 35 patients were evaluated for spinal osteosarcoma at the Massachusetts General Hospital. Two patients were excluded because of insufficient followup, and three did not receive their treatment at our institution. Four individuals lacked a tissue diagnosis confirming osteosarcoma histology and were also excluded from the series. Twenty-six patients were ultimately included for review. The cohort consisted of 15 men and 11 women, with an average age of 46 years at the time of initial presentation (range 5–82 years). Eight patients were treated between 1982 and 1998, and the remaining 18 were treated in the years 1999 to 2008. The cervical spine was involved in three instances, the thoracic in six (Fig. 1), and the lumbar spine in nine, and eight patients had osteosarcoma of the sacrum. There were nine cases of chondroblastic subtype, nine instances of osteoblastic subtype, two tumors with fibroblastic subtype, and one case each of small-cell- and
giant-cell–rich tumors. There were no cases with telangiectatic osteosarcoma. Four individuals had Paget osteosarcoma, and another four had radiation-associated osteosarcoma. At the time of initial presentation, all patients were considered to have high-grade tumors, classified as Enneking Stage IIB. Ten patients had tumors that were larger than 8 cm as determined on preoperative computed tomographic scans and/or magnetic resonance imaging. Eight patients were initially treated at outside facilities and only referred to the Massachusetts General Hospital after incomplete resection or tumor recurrence (Fig. 1 and Fig. 2). Eighteen individuals were treated entirely at our institution. Six patients did not receive surgery, whereas 20 underwent surgical intervention for their osteosarcoma. Of these, seven received a complete en bloc resection of their tumors using combined anterior-posterior procedures (Fig. 2). Among patients receiving en bloc resection, three achieved a wide or marginal excision and four had contaminated margins. Thirteen patients underwent incomplete, or intralesional, resections. No attempt at en bloc excision was made in these instances. Of the six patients who were not treated surgically, five received a combination of chemotherapy and radiation and one was treated with radiation alone. Twenty-four patients received radiation therapy. The average dose delivered was 62.2 Gy (range 20–84.7 Gy). Six patients received preoperative radiation, and six were treated with intraoperative radiation. Eighteen individuals received a combination of photon and proton therapy, and six were treated with photons alone. Of the two patients who did not receive radiation, one had achieved a wide surgical margin and was treated with neoadjuvant chemotherapy. The other individual was deemed to require palliative care after surgery, and radiation was withheld. Twenty-five patients were treated with chemotherapy. The sole individual who did not receive chemotherapy had a Paget associated osteosarcoma and was treated with radiation only for palliative pain relief. Twenty-one were treated with adriamycin, methotrexate, and cisplatin,
Fig. 1. (Left) Axial and (Middle) sagittal computed tomographic images and (Right) axial magnetic resonance scan of a 46-year-old woman with osteosarcoma at T6. The patient initially presented to an outside facility with progressive neurologic deficit in the lower extremities. An emergent posterior decompression was performed, and the patient was transferred to the Massachusetts General Hospital for definitive treatment. The patient received preoperative chemotherapy and radiation treatment before surgical intervention. All the images were obtained after preoperative chemotherapy and radiation but before the definitive surgery.
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
711
Fig. 2. Postoperative (Left) anteroposterior and (Right) lateral radiographic images of the same patient presented in Fig. 1. The patient received a staged en bloc excision of T6 with anterior-posterior reconstruction. A marginal margin was obtained at the time of surgery. The patient received postoperative chemotherapy and radiation but ultimately developed lung metastases and died 10 months after surgery.
whereas four patients received a regimen of etoposide and ifosfamide only. Local recurrence ultimately developed in seven patients (27%), with six of these having received surgery. Metastatic disease occurred in 16 patients (62%). Thirteen patients had metastases at one site, whereas three had metastases in multiple locations (Table 1). The lung was the most common site of metastatic spread. At the last follow-up, 19 patients (73%) were dead of disease. Five patients (19%) were alive with no evidence of disease, and two (8%) were alive with disease. Of these two individuals, one had developed local recurrence and the other was living with lung metastases. Two of the five individuals alive without disease had been treated with radiation and chemotherapy only (Table 2). Five of the seven patients who had received en bloc excision and two of the three individuals with wide, or marginal, margins were dead of disease. Three of the four patients treated with etoposide and ifosfamide only were also dead of disease. Median overall survival for patients in the cohort was 29.5 months (standard error [SE] 14.7, 95% confidence Table 1 Sites of metastatic spread in the 16 patients with metastatic osteosarcoma of the spine Site
Number of cases (%)*
Lung Bone Brain Liver Other
13 3 1 1 3
(81) (19) (6) (6) (19)
* Percentages do not add up to 100% because some patients had metastatic disease at more than one location.
interval [CI] 0.6–58) (Fig. 3). All eight patients treated before 1999 died of disease (average survival 29.2 months, range 3.4–64.5 months), whereas 11 of the 18 individuals treated after 1999 were dead of disease (average survival 26.5 months, range 2.9–78 months). The 1- and 5-year survival rates for all patients were 69% and 31%, respectively. Average follow-up for patients alive at final evaluation was 35.9 months (range 22–78 months). The average follow-up for those with no evidence of disease was 42.8 months (range 22–78 months) (Table 2). There were 10 treatment-related complications in eight patients (31%). Incidental durotomy during surgery was the most common, occurring in four instances. Postoperative infection presented in two cases, and neurologic deficit, myocardial infarction, and deep venous thrombosis all occurred in one case each. One patient died in the postoperative period from chemotherapy-related complications. This patient had previously received an en bloc resection with contaminated margins. The average overall survival for patients without Paget and radiation-associated osteosarcoma were 38 months (SE 7.3, 95% CI 23.4–52.5) and 29.5 months (SE 31.6, 95% CI 2.7–56.3), respectively. Median survival for patients with Paget osteosarcoma was 5.6 months (SE 1.4, 95% CI 2.9–8.2), whereas those with radiation-associated osteosarcoma had a median survival of 14.9 months (SE 13.1, 95% CI 0–44.4). Patients with Paget osteosarcoma were found to have a significantly increased risk of mortality relative to those with conventional osteosarcoma (p!.001). With the exception of the presence of Paget osteosarcoma, no other risk factors exhibited a statistically significant influence on mortality. There was a statistical trend for patients with metastases to have an increased risk of
712
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
Table 2 Details of five patients with spinal osteosarcoma and no evidence of disease Age (y)
Gender
Location
Year of first treatment
Enneking stage
Previous surgery
Margin
Radiation dose
Intraoperative radiation
Chemo
Follow-up (mo)
56 5 52 20 69
M F F M M
T L C S S
2007 2007 2005 2003 2006
IIB IIB IIB IIB IIB
Yes Yes No No No
Contaminated Intra Intra None None
70.2 50.4 64.8 70.2 74
Yes No No N/A N/A
Yes Yes Yes Yes Yes
28 22 51 78 35
M, male; F, female; C, cervical; T, thoracic; L, lumbar; S, sacrum; Intra, intralesional; None, patient did not receive surgery; N/A, not applicable.
mortality, but this did not reach significance because of a lack of power (p5.27, b50.29). Additional risks of mortality for the remaining factors and the resultant p values are encountered in Table 3. Discussion Spinal osteosarcoma is a rare condition, representing 3% to 5% of all cases of osteosarcoma and approximately 10% of primary spinal tumors [2,3,5,6,9,12]. Wide resection and neoadjuvant chemotherapy have revolutionized the treatment of osteosarcoma in the extremities, significantly improving patient outcomes [10–14]. However, no similar benefit has been evinced for spinal osteosarcoma over the last 30 years, despite significant advances in diagnostic and treatment modalities. Advanced surgical techniques, enabling complete removal of the tumor, are presently available [5,12,13,15, 18–20], and a number of studies have suggested that these procedures, combined with chemotherapy or radiation, may
impart long-term survival [5,8,9,12,13,21,22]. Nonetheless, anatomic constraints of the spine can limit the ability to achieve wide surgical margins, and en bloc resection is not always practical. Traditionally, adjuvant radiotherapy was limited by the dose constraints of the spinal cord and other critical normal tissues, but recent techniques such as intensitymodulated radiation therapy [23] and proton radiation therapy [8] now allow radiation doses to be delivered that may be effective for microscopic and even gross residual tumor. Nonetheless, for most patients treated to date, long-term survival for conventional spinal osteosarcoma remains poor, and the outlook is even worse for those with Paget osteosarcoma [24–26]. One of the major obstacles in studying the impact of treatment regimens on spinal osteosarcoma is that the condition is rare, and it is difficult for any one center to evaluate a number of patients treated in similar fashion over a short period of time. Thus, treatment recommendations are informed by a small number of series encompassing patients treated over the course of decades [1,4,7,9] or singular case reports with limited long-term follow-up [19,22,27–30]. The present study documented outcomes for patients with spinal osteosarcoma treated over the course of a 26year period. Results confirm that, despite advances in diagnostic and treatment regimens, the long-term outcome for individuals with osteosarcoma of the spine is poor. Table 3 Risks of mortality for patients with osteosarcoma of the spine Risk factor
Fig. 3. Kaplan–Meier curve illustrating the disease-specific survival estimates for all patients. Cum survival, cumulative survival; f/u months, follow-up in months.
Paget osteosarcoma Radiation-induced osteosarcoma Tumor size O8 cm Tumor size !8 cm Wide/marginal excision No wide/marginal excision En bloc resection No en bloc resection Surgical intervention No surgery Local recurrence No local recurrence Metastases No metastases All treatment at the Massachusetts General Hospital Initial treatment elsewhere
Mortality rate, % (median survival, mo) 100 (5.6) 100 (14.9) 80 69 66 74 71 74 80 50 100 63 94 40 72 75
p Value (b error) !.001 .91 (0.99) .45 (0.97) .58 (0.93) .47 (0.95) .25 (0.93) .48 (0.86) .27 (0.29) .94 (0.95)
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
Although those with Paget osteosarcoma and metastases may be at greater risk for early mortality, en bloc excision, negative surgical margins, and the use of adjuvant chemotherapy and radiation were not found to positively influence outcome. We recognize that this may be because of the low number of patients in our study. However, when compared to other series available in the literature, this report represents one of the largest series of patients with spinal osteosarcoma. Early reports regarding spinal osteosarcoma documented uniformly poor outcomes with patient mortality occurring shortly after presentation. Shives et al. [4] published a series of 27 patients with osteosarcoma of the spine, collected over the course of 70 years. Ninety-six percent of patients in this cohort died of disease, with a median survival of only 10 months. Similarly, Barwick et al. [1] documented the cases of 10 patients with spinal osteosarcoma treated at a single center over 60 years. In this series, there was a 90% patient mortality and median survival was 6 months. Sundaresan et al. [13] reviewed the experience of 11 patients treated from 1978 to 1984 with multimodal therapy, including aggressive surgical resection, chemotherapy, and radiation. Although there was a 55% mortality in this series, only one individual developed metastases while receiving multimodal treatment. More recently, Ozaki et al. [9] reported results for 22 patients with spinal osteosarcoma treated from 1979 to 1998. Twelve patients underwent surgery, and 10 were treated nonoperatively with radiation and/or the Cooperative Osteosarcoma Study Group chemotherapy protocol. Survival at 1 year was 86%, but only 36% were alive after more than 2 years. These authors reported that the presence of metastases and tumors larger than 10 cm adversely impacted outcome, whereas surgical resection with wide margins improved survival. In terms of patient age, the population in the present study was older than that presented in the work of Ozaki et al. [9]. The rate of metastases was also substantially higher (62% vs. 27%), although median survival and overall mortality approximated each other [9]. The high rate of metastases in the present study may be attributed to the size of the cohort, but such a figure can also reflect the aggressive nature of spinal osteosarcoma. It is interesting to note that in the study of Ozaki et al. [9], the sacrum was the most frequent site of osteosarcoma (68% of patients in the series). In the present investigation, the lumbar region was most commonly involved and sacral tumors comprised only 31% of the total. This is important, as Ozaki et al. [9] reported that sacral osteosarcoma was associated with a significantly increased risk of mortality. Additionally, the cohort presented by Ozaki et al. [9] included no patients with Paget osteosarcoma. Sarcomatous transformation in the setting of Paget disease has long been known to carry a poor prognosis [3,5,24–26,31]. For example, a report of 13 cases of Paget osteosarcoma of the spine
713
from the Scottish Tumor Registry documented an average survival of 4.2 months [25]. If patients with Paget sarcoma are excluded from the present investigation, although median overall survival remains the same, the 1-year survival rate approaches the figure of 86% presented in the work of Ozaki et al. [9]. This investigation also differs from previous works in that most patients in the series were treated aggressively with radiation, including proton beam therapy. Several reports have documented satisfactory outcomes for patients with spinal tumors treated using similar protocols [5,8,21,32]. Wagner et al. [32] encountered an overall survival of 65%, and a continuous disease-free survival of 54%, in patients with spinal, or pelvic, tumors treated with surgical resection and a combination of pre- and postoperative radiation, although most of these patients did not have osteosarcoma. DeLaney et al. [8,21] have also reported similar results, with local control rates for spinal sarcomas approaching 80% after varying degrees of resection and high-dose photon/proton radiation. A dose-response relationship with regard to local control of osteosarcomas was reported in one series with poor local control after 30 Gy and effective control when more than 90 Gy was delivered [21]. However, the delivery of conventional radiation to the spine must be balanced with the risk of local toxicity to the spinal cord. For this reason, most large series of osteogenic sarcoma involving the spine have traditionally not used doses greater than 50 Gy. Presently, at our institution, en bloc resection with a wide surgical margin is considered for patients with osteosarcoma of the spine who have at least one pedicle free of tumor and no evidence of metastatic disease [18]. If there is significant concern that a negative margin can be achieved only by sacrificing nerve roots, the sensory and motor functions of those roots are taken into account and included in the preoperative discussion with the patient, so that an informed decision regarding treatment can be made. There are instances where, short of spinal cord transection, an en bloc resection with negative margins is not possible. If the morbidity associated with en bloc resection with negative margins is unacceptable to the patient, then we recommend preoperative radiation followed by gross total resection, followed again with postoperative radiation. Our institution delivers more than 70 Gy of radiation in this setting. This protocol is in addition to standard neoadjuvant and adjuvant chemotherapy [8,21,32]. Quality of life, projected longevity, and the patient’s symptoms figure strongly in decisions regarding surgery for those who have advanced local disease and/or metastases. Patient wishes are paramount in these situations after an extensive preoperative discussion. Although en bloc resection may not be suitable or acceptable in all cases, surgical intervention can still have a positive impact on a patient’s quality of life, especially if the tumor is causing progressive neurologic deficits and resulting in loss of the ability to ambulate.
714
A.J. Schoenfeld et al. / The Spine Journal 10 (2010) 708–714
There are several limitations to this study. First, this was a retrospective investigation, conducted using information compiled from institutional databases. Therefore, this study suffers from the limitations inherent to such a research design. Second, the patients under investigation were treated over the course of 26 years by different physicians, and the care of these individuals evolved with advances in the understanding of osteosarcoma, its response to chemotherapy and radiation, and the technical improvements in the field of spine surgery. The impact of evolving treatment modalities on disease course and outcome cannot be quantified but must be considered as potential confounders. In addition, our study lacked sufficient power to detect differences between treatments rendered and presenting characteristics thought to be important in survival, such as metastasis. While recognizing these limitations, it is hoped that this study may serve as a baseline to which future prospective investigations may be compared. At present, all patients treated for spinal osteosarcoma at our center are enrolled in a prospective database. In conclusion, the results of this study confirm a continued poor prognosis for patients with spinal osteosarcoma. Although combination therapies, including surgery, chemotherapy, and high-dose radiation, may enhance short-term survival, the mortality rate at 5 years remains high. Paget associated osteosarcoma of the spine has a particularly poor prognosis.
References [1] Barwick KW, Huvos AG, Smith J. Primary osteogenic sarcoma of the vertebral column: a clinicopathologic correlation of ten patients. Cancer 1980;46:595–604. [2] Dreghorn CR, Newman RJ, Hardy GJ, Dickson RA. Primary tumors of the axial skeleton. Experience of the Leeds Regional Bone Tumor Registry. Spine 1990;15:137–40. [3] Kelley SP, Ashford RU, Rao AS, Dickson RA. Primary bone tumours of the spine: a 42-year survey from the Leeds Regional Bone Tumour Registry. Eur Spine J 2007;16:405–9. [4] Shives TC, Dahlin DC, Sim FH, et al. Osteosarcoma of the spine. J Bone Joint Surg Am 1986;68:660–8. [5] Sundaresan N, Rosen G, Huvos AG, Krol G. Combined treatment of osteosarcoma of the spine. Neurosurgery 1988;23:714–9. [6] Weinstein JN, McLain RF. Primary tumors of the spine. Spine 1987;12:843–51. [7] Bielack SS, Wulff B, Delling G, et al. Osteosarcoma of the trunk treated by multimodal therapy: experience of the Cooperative Osteosarcoma Study Group (COSS). Med Pediatr Oncol 1995;24: 6–12. [8] DeLaney TF, Park L, Goldberg SI, et al. Radiotherapy for local control of osteosarcoma. Int J Radiat Oncol Biol Phys 2005;61: 492–8. [9] Ozaki T, Flege S, Liljenqvist U, et al. Osteosarcoma of the spine: experience of the Cooperative Osteosarcoma Study Group. Cancer 2002;94:1069–77.
[10] Mankin HJ, Hornicek FJ, Rosenberg AE, et al. Survival data for 648 patients with osteosarcoma treated at one institution. Clin Orthop Relat Res 2004;429:286–91. [11] Schwab JH, Antonescu CR, Athanasian EA, et al. A comparison of intramedullary and juxtacortical low-grade osteogenic sarcoma. Clin Orthop Relat Res 2008;466:1318–22. [12] Sciubba DM, Okuno SH, Dekutoski MB, Gokaslan ZL. Ewing and osteogenic sarcoma: evidence for multidisciplinary management. Spine 2009;34:S58–68. [13] Sundaresan N, Rosen G, Boriani S. Primary malignant tumors of the spine. Orthop Clin North Am 2009;40:21–36. [14] Weber K, Damron TA, Frassica FJ, Sim FH. Malignant bone tumors. Instr Course Lect 2008;57:673–88. [15] Tasdemiroglu E, Bagatur E, Ayan I, et al. Primary spinal column sarcomas. Acta Neurochir (Wien) 1996;138:1261–6. [16] Refaat Y, Gunnoe J, Hornicek FJ, Mankin HJ. Comparison of quality of life after amputation or limb salvage. Clin Orthop Relat Res 2002;397:298–305. [17] Enneking WF, Spanier SS, Goodman MA. A system for the surgical staging of musculoskeletal sarcoma. Clin Orthop Relat Res 1980;153: 106–20. [18] Boriani S, Weinstein JN, Biagini R. Primary bone tumors of the spine. Terminology and surgical staging. Spine 1997;22:1036–44. [19] Kawahara N, Tomita K, Fujita T, et al. Osteosarcoma of the thoracolumbar spine: total en bloc spondylectomy. A case report. J Bone Joint Surg Am 1997;79:453–8. [20] Tomita K, Kawahara N, Baba H, et al. Total en bloc spondylectomy. A new surgical technique for primary malignant vertebral tumors. Spine 1997;22:324–33. [21] DeLaney TF, Liebsch NJ, Pedlow FX, et al. Phase II study of high-dose photon/proton radiotherapy in the management of spine sarcomas. Int J Radiat Oncol Biol Phys 2009;74:732–9. [22] Nishida J, Abe M, Shiraishi H, et al. Familial occurrence of telangiectatic osteosarcoma: cousin cases. J Pediatr Orthop 1994;14: 119–22. [23] Yamada Y, Lovelock DM, Yenice KM, et al. Multifractionated image-guided and stereotactic intensity-modulated radiotherapy of paraspinal tumors: a preliminary report. Int J Radiat Oncol Biol Phys 2005;62:53–61. [24] Huang TL, Cohen NJ, Sahgal S, Tseng CH. Osteosarcoma complicating Paget’s disease of the spine with neurologic complications. Clin Orthop Relat Res 1979;141:260–5. [25] Scottish Bone Tumor Registry, Sharma H, Mehdi SA, MacDuff E, et al. Paget sarcoma of the spine: Scottish Bone Tumor Registry experience. Spine 2006;31:1344–50. [26] Sofka CM, Ciavarra G, Saboeiro G, Ghelman B. Paget’s disease of the spine and secondary osteosarcoma. HSS J 2006;2:188–90. [27] Amritanand R, Venkatesh K, Cherian R, et al. Telangiectatic osteosarcoma of the spine: a case report. Eur Spine J 2008;17:S342–6. [28] Bauernhofer T, Sto¨ger H, Kasparek AK, et al. Combined treatment of metastatic osteosarcoma of the spine. Oncology 1999;57:265–8. [29] Sar C, Eralp L. Transoral resection and reconstruction for primary osteogenic sarcoma of the second cervical vertebra. Spine 2001;26: 1936–41. [30] Simon RG, Irwin RB. An unusual presentation of telangiectatic osteosarcoma. Am J Orthop 1996;25:375–9. [31] Mankin HJ, Hornicek FJ. Paget’s sarcoma: a historical and outcome review. Clin Orthop Relat Res 2005;438:97–102. [32] Wagner TD, Kobayashi W, Dean S, et al. Combination short-course preoperative irradiation, surgical resection, and reduced-field highdose postoperative irradiation in the treatment of tumors involving the bone. Int J Radiat Oncol Biol Phys 2009;73:259–66.