The Tokuhashi Score: Significant Predictive Value for the Life Expectancy of Patients With Breast Cancer With Spinal Metastases

Transcription

1 The Tokuhashi Score: Significant Predictive Value for the Life Expectancy of Patients With Breast Cancer With Spinal Metastases SPINE Volume 30, Number 19, pp , Lippincott Williams & Wilkins, Inc. Benjamin Ulmar, MD,* Marcus Richter, MD, Balkan Cakir, MD,* Rainer Muche, PhD, Wolfhart Puhl, MD,* and Klaus Huch, MD* Study Design. Retrospective study of 55 consecutive patients with spinal metastases secondary to breast cancer who underwent surgery. Objective. To evaluate the predictive value of the Tokuhashi score for life expectancy in patients with breast cancer with spinal metastases. Summary of Background Data. The score, composed of 6 parameters each rated from 0 to 2, has been proposed by Tokuhashi and colleagues for the prognostic assessment of patients with spinal metastases. Methods. A total of 55 patients surgically treated for vertebral metastases secondary to breast cancer were studied. The score was calculated for each patient and, according to Tokuhashi, the patients were divided into 3 groups with different life expectancy according to their total number of scoring points. In a second step, the grouping for prognosis was modified to get a better correlation of the predicted and definitive survival. Results. Applying the Tokuhashi score for the estimation of life expectancy of patients with breast cancer with vertebral metastases provided very reliable results. However, the original analysis by Tokuhashi showed a limited correlation between predicted and real survival for each prognostic group. Therefore, our patients were divided into modified prognostic groups regarding their total number of scoring points, leading to a higher significance of the predicted prognosis in each group (P ), and a better correlation of the predicted and real survival. Conclusion. The modified Tokuhashi score assists in decision making based on reliable estimators of life expectancy in patients with spinal metastases secondary to breast cancer. Key words: prognostic predictors, spinal metastasis, breast cancer, life expectancy, outcome analysis. Spine 2005;30: From the Departments of *Orthopedic Surgery and Spinal Cord Injury, Biometry and Medical Documentation, University of Ulm, Ulm, Germany; and Spine Center, St. Josefs Hospital, Wiesbaden, Germany. Acknowledgment date: August 31, Acceptance date: October 25, The manuscript submitted does not contain information about medical device(s)/drug(s). No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. Address correspondence and reprint requests to Klaus Huch, MD, Department of Orthopedic Surgery and Spinal Cord Injury, University of Ulm, Oberer Eselsberg 45, Ulm, Germany; klaus.huch@rku.de Because advances in tumor diagnosis and treatment have improved the prognosis of patients with cancer, the incidence of clinically apparent metastatic disease has increased. 1 Spinal metastases are a very common manifestation of bony metastasis. 2,3 Of vertebral metastases, 75% originate from carcinomas of the breast, prostate, lung, thyroid gland, and kidney. 4 Metastases of unknown primary tumors account for only 3% to 4%. 5 Spinal metastases are most often located in the vertebral body and in the peridural space, a dorsal localization is a rare finding. 1 Surgery has to strive for the maximum palliative effect, including the reduction of pain and maintenance or restoration of function and stability of the spine, with a minimum of operative morbidity and mortality. For the surgeon involved in treating the individual patient, it is not easy to choose the optimal treatment. The decision should be strongly influenced by the predicted survival. Therefore, Tokuhashi et al 6 developed a simple scoring system for the preoperative evaluation of the prognosis for patients with spinal metastases. This score consists of the sum of 6 parameters that are used to measure the severity of the disorder: (1) general condition of the patient, (2) the number of extraspinal bone metastases, (3) the number of vertebral metastases, (4) the number of metastases to the major internal organs, (5) the primary site of cancer, and (6) the spinal cord palsy. Each parameter is rated between 0 and 2. Zero signifies the worst prognosis (Table 1). Tokuhashi et al 6 proposed a survival of 3 months for patients with a total score of 5, a survival of 12 months for those with 8 points, and a survival of 12 months in patients with a total score of 9 points. Therefore, these investigators recommend excisional surgery of metastases when the total score reaches 9 and palliative treatment when the score is 5. Recently, Tokuhashi et al 8 modified this score by a more differentiated grouping of the primary tumor site from 0 to 5. This modification leads to a total score from 0 to 15 (Table 1) (Y. Tokuhashi, written communication, June 2004). 8 Table 1 also includes the revised evaluation system for the prognosis of metastatic spine tumors with a modification of the number of points regarding the primary site of tumor. This new score was used for the prognosis evaluation in 234 operative and nonoperative patients (Y. Tokuhashi, written communication, June 2004). Patients with spinal metastases and an intermediate or good prognosis seem to benefit from a more radical (excisional) removal of metastatic spine lesions Although all surgical procedures remain palliative by definition, evaluation of the prognosis is necessary to make the decision for surgical procedures (i.e., dorsoventral 2222

2 The Tokuhashi Score Ulmar et al 2223 Table 1. Original Evaluation Score for the Prognosis of Metastatic Spine Tumors Parameters Scoring Points No. Patients General conditions (performance status)* Poor (PS 10% 40%) 0 8 Moderate (PS 50% 70%) 1 22 Good (PS 80% 100%) 2 25 No. extraspinal bone metastases foci No. metastases in the vertebral bodies Metastases to the major internal organs Unremovable 0 22 Removable 1 No metastases 2 33 Primary site of the cancer Lung, stomach 0 Kidney, liver, uterus, unidentified, other 1 Thyroid, prostate, breast, rectum 2 55 Primary site of the cancer Pancreas, esophagus, stomach, bladder, 0 osteosarcoma, lung Liver, gallbladder, unidentified 1 Others 2 Uterus, kidney 3 Rectum 4 Thyroid, prostate, breast 5 55 Spinal cord palsy Complete 0 1 Incomplete 1 8 None 2 46 *Adapted with permission from Gann Monogr Cancer Res 1967;22: Adapted with permission from Spine 1990;15: Y. Tokuhashi, written communication, June PS parameter sum. approach vs. isolated posterior instrumentation). Stabilization and restoration of the anterior spinal column after surgical removal of the diseased vertebra is, as in acute traumatic spinal lesions with a collapse of the vertebra, also recommended in patients with spinal instabilities caused by metastases with a good long-term prognosis. Therefore, calculation of the survival is not only useful to improve the prognosis with a more radical surgical procedure 9 11 but also to prevent mechanical complications of an isolated dorsal instrumentation in the case of predicted long-term survival. 4,12 16 In these patients, reconstruction of the vertebral body following vertebral removal should provide sufficient long-term stability. In times of decreasing finances, the indication for surgery, especially when using costly implants in patients with tumor with a significantly reduced life expectancy, should be based on proved algorithms and scores. The present study was performed to compare the survival predicted by the Tokuhashi score, 6 with the definitive survival of patients with spinal metastases caused by breast cancer because the origin of the primary tumor has substantial influence on the survival. 8 Including only patients with the same type of cancer reduces the number of variable parameters in the Tokuhashi score from6to5. Patients and Methods In this retrospective study, 55 patients with spinal metastases secondary to breast cancer underwent surgery for spine fusion for reduction of pain or for neurologic deficits between September 1984 and September Histologic evaluation verified breast cancer as the primary tumor in all cases. Before surgery, patients were examined with plain anterior-posterior and lateral radiographs of the affected spinal segment, computerized tomography with sagittal reconstructions of the involved vertebra, and magnetic resonance imaging focused on the affected spinal segment to establish the extent of the osseous and spinal canal involvement. A systemic search for other metastases was performed in cooperation with the gynecologists. This process included clinical evaluation, mammography of the affected and contralateral mamma and transvaginal, and abdominal sonography. The staging was completed by plain radiographs of the thorax to detect pulmonary metastases and bone scan to detect other osseous metastases, especially in the spine. Computerized tomography of the chest, abdomen, and brain was performed in each patient with suspicious clinical findings. In addition, the preoperative Tokuhashi score 6 was calculated for all patients (Table 1). The survival dates were censored at February 3, All findings were entered into an Excel database (Microsoft, Corp., Redmond, WA) to ensure that all vertebral, bony, and visceral metastases would be listed and analyzed in the same way, and enable comparisons to be made wherever possible. The statistical unit of analysis was the patient, and statistical analysis was performed on the 55 patients available for the follow-up study. To test the prognostic value of the Tokuhashi score, its total number of points was evaluated in all patients. The possibility to evaluate this preoperative score retrospectively was described by Enkaoua et al. 17 Statistical analyses were performed using the log-rank test for univariate analysis. Survival curves were created using the Kaplan-Meier life table analyses. A P value 0.05 (2-tailed) was considered significant. Results The mean survival of all study patients was months (range ; median 16.2). At the end of the study, 46 patients (83.6%) had died. In 9 surviving patients (16.4%), mean observation period was months (range ; median 24.9). Only 1 patient has been observed less than 12 months. The Tokuhashi score 6 was calculated for all patients (Table 1), and the number of patients with the same score is depicted in Figure 1. For the palliative group (n 7; 12.7%), the survival was predicted according to Tokuhashi et al 6 to be less than 3 months. In our evaluation (Table 2), these patients revealed a mean survival of months (range ; median 5.0). The indifferent group (n 26; 47.3%) had a predicted survival of 12 months. However, our group showed a mean survival of months (range ; median 14.8). The excisional group (n 22; 40%) had a predictive survival of 12 months and revealed in our study group a mean

3 2224 Spine Volume 30 Number Figure 1. Total number of score points of the Tokuhashi evaluation score 6 for our study patients (n 55). survival of months (range ; median 25.6). The prognostication of the survival interval using the original evaluation of Tokuhashi et al 6 was significant (P ) (Figure 2A). Because the original evaluation of the Tokuhashi score has shown only a limited correlation between predicted and real survival, classification of the Tokuhashi score was modified. There were 4 patients (7.3%) included in the first group (0 4 points instead of 0 5 points), with a mean survival of months (range ; median 2.9). The second group (5 8 points instead of 6 8 points) included 29 patients (52.7%) with a mean survival of months (range ; median 14.4). The third group ( 9 points) included 22 patients (40%), with a mean survival of months (range ; median 25.6). The modified score showed a higher significance for the prognosticated survival (P ) than the original one (Figure 2B). The differences among the original, revised, and modified evaluations of the Tokuhashi score are listed in Table 2. Table 2. Original and Modified Evaluation of the Tokuhashi Score for the Study Group (n 55) in Comparison to the Original Prediction by Tokuhashi et al and According to the Revisions by Tokuhashi Original evaluation score by Tokuhashi et al 6 and revised evaluation score by Tokuhashi (written communication, June 2004) Original evaluation by Ulmar et al (present study) Modified evaluation by Ulmar et al (present study) Group I Group II Group III Figure 2. A, Predictive survival of the Tokuhashi score regarding the prognostic groups according to Tokuhashi. 6 Small dashes (----) indicate patients with a total number of 5 score points. Long dashes ( ) show patients with a total number of 6 8 score points. Solid line ( ) illustrates patients with a total number of 9 points. B, Predictive survival of the Tokuhashi score with modified restrictions of the prognostic groups. Small dashes ( ) indicate patients with a total number of 4 score points. Long dashes ( ) show patients with a total number of 5 8 score points. Solid line ( ) indicates patients with a total number of 9 score points. Discussion This study confirms the value of the Tokuhashi score to predict the time of survival in females with spinal metastases caused by breast carcinoma. However, our modification could significantly improve the predictive value of the score to estimate life expectancy. Therefore, the score can facilitate surgical planning in spinal metastases in breast carcinoma. Survival is influenced by various isolated parameters. According to Tokuhashi et al, 6 these parameters should be combined in a significant but easy, usable score to evaluate the survival prognosis. Therefore, these investigators proposed a prognostic scoring system, taking 6 variables into account, for the preoperative evaluation of patients with spinal metastases. The influence of the general medical condition for the prediction of the survival is controversially discussed. For nonneurologic cases, the Karnofsky index has proved to be effective. 18 If the Karnofsky index of patients with spinal metastases is more than 70%, the survival period will increase. 19 When the general state is not good (i.e., Karnofsky index less than 40%), palliative treatments other than surgery should be considered for patients without neurologic deficits. 18

4 The Tokuhashi Score Ulmar et al 2225 The survival is significantly lower for patients with bone metastases combined with additional visceral spread in contrast to isolated bony metastases, and significantly lower for patients with multiple bony in contrast to solitary bony metastases. 20 Therefore, the bony and visceral spread, and the extent and number of bony metastases have been accepted as a predictive factor for the patient s survival. 4,6,17,20 22 Variable survival rates have been reported by several investigators dependent on the primary site of tumor. 18,23 29 Tatsui et al 27 evaluated the 1-year survival of 425 patients with spinal metastases of different origin after detection by bone scan. They reported a 1-year survival rate of 83.3% for patients with prostatic cancer, 77.7% for those with breast cancer, 51.2% for those with renal cancer, 44.6% for patients with uterine cancer, 21.7% for patients with lung cancer, and 0% of those with gastric cancer. Therefore, Tokuhashi et al 6 included the site of tumor as one major prognostic factor influencing the survival. Enkaoua et al 17 confirmed the prognostic value of the Tokuhashi score but recommended assigning 0 points for carcinomas of unknown primary. Oberndorfer and Grisold 30 followed 38 patients with different primary tumors and could show that the Tokuhashi score decreased 3 months after various therapies from 0.8 to 2.4 score points. Riegel et al 31 also showed a significant correlation between score and survival for 139 patients. Paraplegia is also a controversially discussed prognostic factor in patients with spinal metastases. It has been reported that the rapidity of the onset of muscular weakness has a considerable bearing on the ultimate prognosis. A delay of less than 24 hours between the onset of the symptoms and appearance of a full-blown neurologic syndrome leads to a poor prognosis no matter what treatment is offered. 32 In contrast, Spiegel et al 33 reported that the presence of a neurologic deficit did not significantly influence the survival among patients with melanoma. Enkaoua et al 17 also reported that epidural metastases were not significantly associated with the length of survival. Patients with paraplegia seem to die sooner because of progression of their cancer, but not for paraplegia itself. Tomita et al 34 did not accept the neurologic state as a prognostic factor for their system of survival prognostication in spinal metastases. They also proposed the surgical classification of the spinal tumors based only on 3 factors: (1) grade of malignancy of the primary tumor; (2) visceral metastases to vital organs (lung, liver, kidneys, and brain); and (3) bone metastases. 11,34,35 The score ranges between 2 and 10, and the lower the total numbers of points, the better the predicted survival, but the primary cancer type, generalized metastases, neurologic impairment, or general state of the patient were not considered. 34 Tokuhashi 6 and Enkaoua 17 et al indicate surgery without regarding biomechanical aspects in a predicted long-time-survival of the patient. However, to avoid the problems of isolated dorsal instrumentations during further follow-up, stabilization and restoration of the spine in patients with a good prognosis is necessary. 4,12 16 Harrington 32 designed a 5-category classification for spinal metastases to evaluate patients for surgical and nonsurgical indications: The classification regards only bone destruction and neurologic compromise: (1) no significant neurologic involvement, (2) involvement of bone without collapse or instability, (3) major neurologic impairment (sensory or motor) without significant involvement of the bone, (4) vertebral collapse with pain resulting from mechanical causes or instability but with no significant neurologic involvement, and (5) vertebral collapse or instability combined with major neurologic impairment. He recommended that patients in categories 1 3 be treated with radiotherapy, chemotherapy, or hormonal manipulation, and patients in categories 4 or 5 should consider undergoing surgery. Based on a 2-column concept, Kostuik et al 36 make their decision for surgical intervention in spinal instability. The anterior column includes the vertebral body with the cortex and is further divided into 4 quadrants, an anterior and posterior, and a left and right part of the vertebral body. The posterior column consists of pedicles, laminae, and spinal cord, and is divided into a left and right part. If no more than 2 parts of the 6 parts are destroyed, spinal stability is expected. McLain and Weinstein 37 suggested that the optimal surgical approach should be based on the extent of bony involvement. They divided the vertebra into 4 zones to select adequate surgery: (1) process spinous to pars interarticularis and the inferior facets; (2) the superior articular facet, the transverse process, and the pedicle; (3) anterior three fourths of the vertebral body; and (4) the posterior one fourth of the vertebral body. Tumor extension was classified as A C for intraosseous, extraosseous, and distant tumor spread. According to Taneichi et al, 38 surgery should be considered when the risk of impending collapse is significant: more than 50% to 60% involvement of the vertebral body with no destruction of adjacent structures, or 25% to 30% involvement with costovertebral joint destruction in the thoracic spine and 35% to 40% involvement of the vertebral body, or 20% to 25% involvement with destruction of the posterior elements in the thoracolumbar and lumbar spine. Conclusions In our study, patients with spinal metastasis secondary to breast cancer survived longer (mean survival months; range months; median 16.2 months) than estimated by the Tokuhashi score and shown by Coleman 39 (median survival 20 months). Using a modified grouping of the Tokuhashi score, we could identify patient groups with a mean survival of 3 months (group 1), 21.7 months (group 2), and 38.9 months (group 3). In connection with biomechanical aspects, this score allows adaptation of the therapy to the expected survival. No surgery in group 1 with a very limited life

5 2226 Spine Volume 30 Number expectancy, a dorsal approach in group 2, and a dorsoventral approach in group 3. 40,41 The revisions performed by Tokuhashi (written communication, June 2004) do not change the results of this study. However, a prospective study will ultimately determine the true value of this score. Key Points Prognostic factors in patients with spinal metastases strongly influence the surgical pathway. Spinal metastases are a very common manifestation of bony metastases, 75% of them originate from carcinomas of the breast, prostate, lung, thyroid, and kidney. Breast cancer is the most common female cancer with frequent vertebral metastases. Life expectancy is a key factor for surgical planning. Outcome analysis in spinal metastases assigns a value to prognostic scores. References 1. Harrington KD. Orthopedic surgical management of skeletal complications of malignancy. Cancer 1997;80: Boland PJ, Lane JM, Sundaresan N. Metastatic disease of the spine. Clin Orthop 1982;169: Enneking WF. Musculoskeletal Tumor Surgery. Vol. 2. New York, NY: Churchill Livingston; 1983: Harrington KD. The use of methylmethacrylate for vertebral body replacement and anterior stabilization of pathologic fracture dislocation of the spine due to metastatic malignant disease. J Bone Joint Surg 1981;63A: Holmes FF, Fouts TL. Metastatic cancer of unknown primary site. Cancer 1970;26: Tokuhashi Y, Matsuzaki H, Toriyama S, et al. Scoring system for the preoperative evaluation of metastatic spine tumor prognosis. Spine 1990;15: Karnofsky DA. Clinical evaluation of anticancer drugs: Cancer chemotherapy. Gann Monogr Cancer Res 1967;22: Tokuhashi Y. Letter to the editor. Spine 2000;25: Boriani S, Biagini R, De Lure F, et al. En bloc resections of bone tumors of the thoracolumbar spine. A preliminary report on 29 patients. Spine 1996;21: Sundaresan N, Steinberger AA, Moore F, et al. Indications and results of combined anterior-posterior approaches for spine tumor surgery. J Neurosurg 1996;85: Tomita K, Kawahara N, Baba H, et al. Total en bloc spondylectomy for solitary spinal metastases. Int Orthop 1994;18: Dick W. Fixateur interne. State of the art reviews. Spine 1992;6: Gertzbein SD. Neurologic deterioration in patients with thoracic and lumbar fractures after admission to the hospital. Spine 1994;19: Gertzbein SD, Court-Brown CM, Marks P, et al. The neurological outcome following surgery for spinal fractures. Spine 1988;13: Hertlein H, Hartl WH, Dienemann H, et al. Thoracoscopic repair of thoracic spine trauma. Eur Spine J 1995;1: Lord CF, Herndon JH. Spinal cord compression secondary to kyphosis associated with radiation therapy for metastatic disease. Clin Orthop Relat Res 1986;210: Enkaoua EA, Doursounian L, Chatellier G, et al. Vertebral metastases. A critical appreciation of the prognostic Tokuhashi score in a series of 71 cases. Spine 1997;22: Chataigner H, Onimus M. Surgery in spinal metastasis without spinal cord compression: Indications and strategy related to the risk of recurrence. Eur Spine J 2000;9: Nazarian S, Guigui P, Gouvernet J. Place de la chirurgie dans le traitement des métastses du rachis. Résultats globaux. Rev Chir Orthop 1997;83(suppl 3): Dürr HR, Müller PE, Lenz T, et al. Surgical treatment of bone metastases in patients with breast cancer. Clin Orthop 2002;396: Swenerton KD, Legha SS, Smith T, et al. Prognostic factors in metastatic breast cancer treated with combination chemotherapy. Cancer Res 1979;39: Yamashita K, Yonenobu S, Fuji T. Staging of metastatic spinal tumor. Seikei Geka 1986;21: Brice J, McKissock W. Surgical treatment of malignant extradural spinal tumors. BMJ 1965;1: Constans JP, de Divitiis E, Donzelli R, et al. Spinal metastases with neurological manifestations. J Neurosurg 1983;59: Hall AJ, McKay NS. The result of laminectomy in compression of the cord or cauda equina by extradural malignant tumor. J Bone Joint Surg 1973;55-B: Onimus M, Schraub S, Bertin D, et al. Surgical treatment of vertebral metastases. Spine 1986;11: Tatsui H, Onomura T, Morishita S, et al. Survival rates of patients with metastatic spinal cancer after scintigraphic detection of abnormal radioactive accumulation. Spine 1996;18: White WA, Patterson RH, Bergland RM. Role of surgery in the treatment of spinal cord compression by metastatic neoplasm. Cancer 1971;27: Young RF, Feldmann RA. Metastatic tumor of the spine. J Neurosurg 1979; 50: Oberndorfer S, Grisold W. Letter to the editor. Spine 2000;25: Riegel T, Schilling T, Sitter H, et al. Analysis of factors affecting the prognosis of vertebral metastases [in German]. Zentralbl Neurochir 2002;63: Harrington KD. Metastatic disease of the spine. J Bone Joint Surg 1986;68A: Spiegel DA, Sampson JH, Richardson WJ, et al. Metastatic melanoma to the spine. Diagnosis, risk factors and prognosis in 114 patients. Spine 1995;20: Tomita K, Kawahara N, Kobayashi T, et al. Surgical strategy for spinal metastases. Spine 2001;3: Tomita K, Kawahara N, Baba H, et al. Total en bloc spondylectomy: A new surgical technique for primary malignant vertebral tumors. Spine 1997;22: Kostuik JP, Errico TJ, Gleason TF, et al. Spinal stabilisation of vertebral column tumors. Spine 1988;13: McLain RF, Weinstein JN. Tumors of the spine. Semin Spine Surg 1990;2: Taneichi H, Kaneda K, Takeda N, et al. Risk factors and probability of the vertebral collapse in metastases of thoracic and lumbar spine. Spine 1997; 22: Coleman RE. Skeletal complications of malignancy. Cancer 1997;80(suppl): Huch K, Cakir B, Dreinhöfer KE, et al. A new dorsal modular fixation device allows a modified approach in cervical and cervicothoracic neoplastic lesions. Eur Spine J 2004;13: Kluger P, Korge A, Scharf H P. Strategy for the treatment of patients with spinal neoplasms. Spinal Cord 1997;35: