Spinal Cord Compression in Metastatic Prostate Cancer

Transcription

1 European Urology European Urology 44 (2003) Spinal Cord Compression in Metastatic Prostate Cancer H. Tazi, A. Manunta *, A. Rodriguez, J.J. Patard, B. Lobel, F. Guillé Service d Urologie, Centre Hospitalier Universitaire Pontchaillou, Rue Henri Le, Guilloux, Rennes Cedex, France First published online 11 July 2003 Abstract Introduction: Spinal cord compression (SCC) in metastatic prostate cancer is not rare occurring in 1 to 12% of patients. We have analysed patients treated for this condition in our institution assessing outcome and prognostic factors. Material and Methods: Retrospective analysis of the notes of 24 patients hospitalised with SCC due to metastatic prostate cancer from 1987 to Results: At presentation 3 patients were ambulant with mild neurological deficit, 12 patients were paraparetic and 9 patients were paraplegic. Diagnosis was established by emergency myelogram, CT-scan or MRI of the spine. 8 patients had received no hormone treatment prior to diagnosis of SCC. 19 patients presented dorsal or lumbar pain requiring opioid treatment on average 60 days before onset of neurological symptoms (range ). All patients underwent steroid treatment; the 8 patients without prior hormone therapy were treated with bilateral orchidectomy, 1 out of these 8 patients had castration without other treatments; 12 patients underwent radiotherapy alone and 9 radiotherapy and laminectomy; 2 patients were in terminal conditions and receive only steroid treatment. Overall 15/24 patients were ambulant after treatment. 8 out of 9 patients treated by laminectomy and radiotherapy were ambulant after treatment versus 7 out of 12 patients treated by radiotherapy alone. 17 patients died during follow-up with a median survival of 4 months (2 weeks to 49 months). 7 patients were alive at the last control with a mean follow-up of 10 months (range 4 40). Conclusion: Outlook in patients with spinal cord compression from metastatic prostate cancer is poor. Efforts must be concentrated on prevention of spinal cord compression. Patients with hormone resistant prostate cancer who develop persistent back pain should undergo imaging studies (bone scan, spine CT-scan or MNR) and prophylactic local radiotherapy to the spine if bony metastases are identified. # 2003 Elsevier B.V. All rights reserved. Keywords: Prostatic neoplasms; Neoplasm metastasis; Spinal cord compression 1. Introduction Spinal cord compression is a common neurologic complication of advanced malignancies and should be considered an oncologic emergency [1]. Its frequency in patients with any malignancy is 5 to 10% [2,3]. Prostate cancer is second only to lung cancer as a cause of metastatic spinal cord compression in men. 1 to 12% of prostate cancer patients will develop this complication [3 8]. In 12 to 19% of cases spinal cord compression is the first presenting sign [6,8 10]. The slow natural history of prostate cancer justifies an aggressive treatment in order to preserve or restore * Corresponding author. Tel. þ ; Fax: þ address: andrea.manunta@chu-rennes.fr (A. Manunta). neurologic function, to relieve pain, and to improve the patient s quality of life [11]. However in spite of the improvement in imaging techniques little has changed over the years in term of neurologic recovery after treatment of established cord compression. We retrospectively reviewed our experience with prostate cancer patients suffering from spinal cord compression. 2. Patients and methods From 1987 to 2001, 24 patients were hospitalised for spinal cord compression due to metastatic prostate cancer. Median age was 64 years (range 42 75). In all patients spinal cord compression was /$ see front matter # 2003 Elsevier B.V. All rights reserved. doi: /s (03)

2 528 H. Tazi et al. / European Urology 44 (2003) Table 1 Patients characteristics confirmed by imaging (emergency myelogram in 6 cases, CT-scan in 3, MRI of the spine in 15 cases). Patients were classified according to their neurological status at the time of hospitalisation: 3 patients were ambulant with mild neurological deficit, 12 patients were paraparetic and 9 patients were paraplegic. The level of compression was cervical and dorsal in 1 patient, dorsal in 10 patients and thoraco-lumbar in 13 patients. Patients characteristics are outlined on Table 1. All patients received high dose steroids (Dexamethasone 10 mg IV initially and then 4 mg IV every 6 hours). Patients who had not undergone prior androgen deprivation therapy were treated with bilateral orchidectomy. Radiotherapy was administered in doses ranging from 20 Gy in 5 fractions to 30 Gy in over 10 fractions. Decision to perform laminectomy was undertaken in consultation with the neurosurgery department. 3. Results No. patients Neurological status Mild neurological deficit 3 Paraparesis 12 Paraplegia 9 Site of spinal cord compression Cervical and dorsal 1 Dorsal 10 Thoraco-lumbar 13 Prior androgen deprivation 16/24 Dorsal and/or loin pain 19 (average 60 days before onset SCC) Bladder function Normal 12 Urinary retention patients were known to have prostate cancer before onset of neurological symptoms. Initial treatment of prostate cancer was radical prostatectomy in 1 patient and androgen deprivation in 16 patients for a median period of 35 months (range 4 92). In 2 patients spinal cord compression developed soon after diagnosis of prostate cancer before any treatment had been started. The median time interval between diagnosis of prostate cancer and the development of neurological symptoms was 32 months (range 2 92). 19 patients presented dorsal or lumbar pain requiring opioid treatment on average 60 days before onset of neurological symptoms (range ). All patients underwent steroid treatment, 12 patients underwent radiotherapy alone, 9 radiotherapy and laminectomy, 1 castration alone. 2 patients did not receive specific treatment for spinal cord compression other than steroids; they were in terminal conditions and died few days after onset of spinal cord compression. Median delay between onset of neurological symptoms and treatment of spinal cord compression was 60 hours (min 8 hours max 27 days). Overall 15/24 patients were ambulant after treatment (9 full neurological recovery, 6 could walk with aids). 12 patients presented with urinary retention, normal voiding was restored in 5 patients within a median period of 8 days (range 3 18), in the remaining 7 patients chronic urinary retention was managed by indwelling urethral catheter, intermittent self catheterisation was not undertaken given their general conditions and poor short term outlook. Out of 9 patients with paraplegia on admission 3 were ambulant after treatment and 6 remained wheelchair bound (2 were in terminal conditions and were not treated). In the group of 12 patients with paraparesis 9 were ambulant after treatment and 3 deteriorated and become wheelchair bound. All patients with focal neurological deficit were ambulant after treatment. Outcome in relation to pretreatment neurological status is resumed in Table 2. Table 2 Functional outcome in relation to pre-treatment functional status Pretreatment neurological status Treatment received Full neurologic recovery Improvement Stabilization Failure 3 Focal neurological deficit 2 Rt RtþL 12 Paraparesis 6 RtþL Rt 9 Paraplegia 4 Rt RtþL 1 Castr 2 No ttt Rt: Radiotherapy alone; L: Laminectomy; Castr: Castration alone; No ttt: No treatment.

3 H. Tazi et al. / European Urology 44 (2003) Out of 12 patients undergoing radiotherapy alone 7 were ambulant after treatment and 5 were paraplegic. Out of these 12 patients 2 were ambulatory at presentation and remained so after treatment, 6/12 were paraparetic: 4 became ambulatory after treatment. Only 1 out of 4 paraplegic patients treated by radiotherapy alone regained function. 9 patients underwent radiotherapy and laminectomy, 8 (88.8%) were ambulant after treatment and 1 was paraplegic. 8 patients were not on androgen deprivation at the time of cord compression: 5 underwent laminectomy, radiotherapy and castration; 2 radiotherapy and castration; and 1 with established paraplegia underwent castration alone. After treatment 5 were ambulant and 3 paraplegic. 17 patients died during follow-up with a median survival of 4 months (2 weeks to 49 months). In patients without prior androgen deprivation therapy, the median survival was 15.5 months. 7 patients were alive at the last control with a mean follow-up of 10 months (range 4 40). 4. Discussion Prostate cancer metastases to the spine occur frequently and 1 to 12% of prostate cancer patients will develop spinal cord compression [3 8]. Functional outlook is related to the degree of neurological involvement at diagnosis and treatment should be undertaken without delay. Involvement of the spinal cord is secondary to local extension of a vertebral metastasis. Tumor compression leads to venous obstruction and development of vasogenic edema. At this early stage steroids can decrease local edema and resolve most neurological symptoms. However if compression is not relieved ischemia and eventually spinal cord infarction will ensue. The time laps between appearance of the first neurological symptoms due to reversible edema and permanent neurological damage may range from days to weeks and consitutes a window of opportunity which should not be missed [11,12]. Back pain is the most common presenting symptom occurring in 75 to 100% of the patients, usually gradual in onset and progressive within days to months, and it may be focal, radicular or reffered [10 14]. In our series 79% of the patients had developed back pain for an average of 60 days before onset of neurological symptoms. Almost all patients who experience pain complain of local pain, usually close to the site of the lesion subsequently identified during diagnostic evaluation. This is thought to be due to periosteal stretching as the vertebral metastasis enlarges [3]. Normal neurologic examination does not exclude the possibility of impending spinal cord compression. Rodichok reported that in thirty-six percent of patients who presented with back pain but who had normal neurologic findings, there was evidence of epidural metastases on myelography [15]. Weakness is common by the time the diagnosis established. The weakness is usually bilateral and involves the lower extremities. The degree of weakness is associated with the likelihood of success in preserving or restoring ambulation and is therefore a useful prognostic tool [9,11,16]. Barcena et al. reported post treatment ambulation rates for pretreatment ambulatory, paraparetic and paraplegic patients as 70%, 44% and 22%, respectively [17]. Smith et al. reported that 92% of ambulatory patients, 83% of paraparetic patients and 20% of paraplegic patients were ambulatory after appropriate treatment [9]. In our study, all 3 patients ambulatory at presentation remained so after treatment. Of 12 paraparetic patients 9 (75%) became ambulatory after treatment. Of 9 paraplegic patients, only 3 (33%) were ambulant after treatment. Loss of sensation is rarely the first symptom, but approximately 50% of patients are affected by the time the diagnosis is established. The precise level of spinal cord compression can be determined from the level of sensory loss. Less frequently autonomic dysfunction may occur. 50% of our patients were in urinary retention at initial presentation and only 41% of this patients regained normal voiding after treatment. Gilbert et al. considered autonomic dysfunction as an unfavorable prognostic sign. Of 65 of their patients with urinary incontinence or retention at the time of diagnosis 44 (66%) either were or became non-ambulatory despite treatment [3]. The delay in the diagnosis of spinal cord compression may result in significant neurological deterioration, as rapid progression toward paraplegia will occur in all patients [18]. Huddart suggested that the time of treatment from the onset of more severe neurological deterioration could be more crucial than the time from the first neurological symptoms. 70 to 75% of patients would be improved by current treatments as long as critical damage has not occurred [10]. Plain radiographs of the entire spine usually are helpful in the diagnostic evaluation of suspected spinal metastases. The plain film of the spine can correctly predict the presence or absence of spinal cord compression in over 80% of symptomatic patients [15]. The CT-scan may be useful if surgery is contemplated

4 530 H. Tazi et al. / European Urology 44 (2003) because it provides good anatomy bony details and can not only determine the extent of bony destruction at the level of primary lesion, but also evaluates bone quality above and below the lesion, in order to determine the feasibility of stabilizing the spine after a surgical decompression [12]. However CT-scan explores a limited area in the spine, and many authors recommend imaging of the entire spine in order to detect cord compression at sites distant from the symptomatic lesion [9,11,19]. Smith et al. reported that 27% of their patients had multiples compressive lesions. Flynn et al. found 23% of patients with 2 sites of compression. Only 1 patient in our series had double site of compression. Spinal cord involvement at multiple levels was correlated with poor prognosis in Huddart s study [10]. Magnetic resonnance imaging (MRI) is now the radiologic procedure of choice in the diagnosis and localization of spinal cord compression [1,11]. The sensitivity and specificity of MRI in the diagnosis of spinal cord compression du to extradural masses are similar to those of myelography [20]. Furthermore MRI was more sensitive for diagnosing intramedullary metastasis, paravertebral masses and bony metastasis. Myelography is now reserved for patients with claustrophobia, pacemakers and metal near the area of interest [1]. Aggressive investigation in prostate cancer patients with back pain will decrease delay in diagnosis and initiation of palliative therapy for spinal cord compression. The goal in the treatment of spinal cord compression secondary to metastatic prostate carcinoma is palliation. Pain relief and prevention of neurologic deterioration through local tumor control are of primary importance. Corticosteroides are usually instituted immediately after the diagnosis is made. Greenberg reported that patients receiving high dose bolus of dexamethasone did have rapide and complete relief of pain prior to undergoing radiation [21]. Androgen deprivation has an important role in the treatment of spinal cord compression. Its greatest benefit is seen in patients with prostate cancer who have not previously undergone hormonal therapy. The median survival in those patients was 42 and 16 months, in the studies of Huddart and Flynn respectively compared to 7.6 and 6 months for the patients who had received androgen deprivation prior to the diagnosis of spinal cord compression [10,11]. In our study, eight patients (33%) were not on hormone therapy at the time of diagnosis of spinal cord compression. After treatment 5 patients (62.5%) were ambulant. The median survival of these patients was 15.5 months compared with 6 months in patients with prior androgen deprivation therapy. Radiation therapy is the mainstay of treatment of patients with a metastatic prostate carcinoma to the spine [12]. The decision between radiation alone and surgery followed by radiation therapy can be a more difficult one. Flynn and Shipley reported that ambulatory and moderately paretic patients had the most successful outcome if treated with radiation alone. All of their ambulatory, 52% of their paraparetic and 38% of their paraplegic patients were ambulatory following radiation [10]. Smith et al. found that radiation alone was effective in all of their ambulatory patients and in 83% of their paraparetic patients. However, only 1 of 5 paraplegic patients (20%) regained function [9]. In our study, we found nearly the same results with radiotherapy alone effective in all of our ambulatory patients, in 66.6% of our paraparetic patients and only in 25% of paraplegic patients. For decreasing local recurrences, Kaminski et al. recommended that the radiation port encompass at least two vertebral bodies above and below the level of compression. The risk of further episodes of cord compression amounted to 50% at 3 years [22]. Immediate surgical decompression should be used in patients with an expected life span of at least 6 months who deteriorate during radiation, who have had previous radiation in the involved site, who have a potentially correctable unstable spine or who had spine compression of unknown histology [11]. Finally, looking only at paraplegic or severely paraparetic patients, surgery followed by radiation therapy appeared to be superior to radiation alone in improving neurologic function. In Flynn s study, only 2 of 6 paraplegic patients treated with radiation alone demonstrated neurologic improvement compared to 6 of 8 treated with radiation plus surgery. Our study confirms that radiotherapy and surgery were more effective; 88.8% of patients who underwent this combined treatment were ambulant after. It would however be difficult to draw conclusion from our small series of patients presenting with different neurological status and undergoing different treatments. Patients undergoing laminectomy and radiotherapy may have benefited from a selection bias as surgery would not be offered to patients with fully established paraplegia. In Smith s study, the average survival for all patients was 18 months. The mean survival of patients without paraplegia was 18.9 months. Survival for patients with paraplegia was significantly shorter averaging only 3.9 months from the onset of paraplegia [9]. The median survival in the Massachusetts hospital patients was

5 H. Tazi et al. / European Urology 44 (2003) months, with 17 patients (31%) alive at 1 year, 7 patients (13%) alive at 2 years and 4 patients (7%) alive at 5 years after diagnosis of cord compression. Significant long term survival is possible when hormonal manipulation is initiated at the time of cord compression, compared with patients already in relapse after hormonal therapy when they present with cord compression [11]. It is widely accepted that the neurologic status prior to treatment is a major determinant influencing the outcome. This underscores the importance of immediate treatment before further neurologic deterioration and before the damage to the spinal cord becomes permanent. Zelefsky reported that pretreatment ambulatory status and response documented by repeat myelography had a significant correlation with survival. In their study, the median survival was 9.5 months among patients who obtained a complete response compared with 2 months for those who did not respond [23]. Outlook in patients with spinal cord compression from metastatic prostate cancer is poor. The main prognostic factor for functional recovery is the degree of neurological damage before treatment. Early detection and treatment is mandatory in order to prevent irreversible spinal cord damage. MRI can be used currently to screen patients judged to be at high risk of developing spinal cord compression, identifying asymptomatic or incipient areas of cord compression and possible multiple sites of compression at different levels of the spinal cord. It would be then possible to treat patients identified prophylactically to prevent neurological damage. Given the high morbidity of spinal cord compression and the expense of caring for paraparetic and paraplegic patients, this could be an efficient and cost effective approach [10]. Biphosphonate treatment may be a useful adjunctive measure. In Saad s study [24] zoledronic acid, a third generation biphosphonate, given intravenously every three weeks significantly reduced skeletal-related events compared to placebo. In spite of the diagnostic improvement represented by the introduction of MRI little has been changed in the prognosis of spinal cord compression over the years; the rate of neurological recovery of recent studies being comparable to older series reported before the introduction of this diagnosis technique [1,8,9,13]. Effort must especially be concentrated on prevention of spinal cord compression in patients with hormon resistant prostate cancer who develop persistent back pain. In the present series 19 patients presented dorsal or lumbar pain requiring opioid treatment on average 60 days before onset of neurological symptoms. MRI and local radiotherapy to the spine might have prevented cord compression in most patients as well as palliating more effectively pain from bony metastasis. 5. Conclusion Spinal cord compression occurs in up to 12% of prostate cancer patients and results in devastating morbidity. Combined treatment by laminectomy and radiotherapy seem to be associated with improved neurological function and longer survival especially in patients without prior androgen deprivation. Improvement in diagnostic and imaging techniques over the past decades has had little impact on functional outcome and neurologic recovery after treatment. Early recognition and treatment are essential if neurologic function is to be preserved but more emphasis should be placed on prevention of cord compression. Persistent back pain in patients with known prostate cancer is all too often treated by opioids while it should be investigated and treated aggressively with local radiotherapy before the onset of neurological complications. References [1] Osborn JL, Getzenberg RH, Trump DL. Spinal cord compression in prostate cancer. J Neurooncol 1995;23: [2] Barron KD, Hirano A, Araki S. Experiences with metastatic neoplasms involving the spinal cord. Neurology 1959;9: [3] Gilbert RW, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: diagnosis and treatment. Ann Neurol 1978;3: [4] Rubin H, Lome LG, Presman D. Neurological manifestations of metastatic prostate carcinoma. J Urol 1974;111: [5] Constans JP, de Divitiis E, Donzelli R. Spinal metastases with neurological manifestations. J Neurosurg 1983;59: [6] Kuban DA, El-Mahdi AM, Sigfred SV. Characteristics of spinal cord compression in adenocarcinoma of prostate. Urology 1986;18: [7] Liskow A, Chang CH, De Sanctis P. Epidural cord compression in association with genitourinary neoplasms. Cancer 1986;58: [8] Rosenthal MA, Rosen D, Raghavan D, Leicester J, Duval P, Besser B, et al. Spinal cord compression in prostate cancer. A 10 years experience. Br J Urol 1992;69: [9] Smith EM, Hampel N, Ruff RL, Bodner DR, Resnick MI. Spinal cord compression secondary to prostate carcinoma: treatment and prognosis. J Urol 1993;149: [10] Huddart RA, Rajan B, Law M. Spinal cord compression in prostate cancer: treatment outcome and prognostic factors. Radioth Oncol 1997;44: [11] Flynn DF, Shipley WU. Management of spinal cord compression secondary to metastatic prostatic carcinoma. Urol Clin North Am 1991;18: [12] Chen TC. Prostate cancer and spinal cord compression. Oncology 2001;15(7):

6 532 H. Tazi et al. / European Urology 44 (2003) [13] Shoskes DA, Perrin RG. The role of surgical management for symptomatic spinal cord compression in patients with metastatic prostate cancer. J Urol 1989;142: [14] Iacovou JW, Marks JC, Abrams PH, Gingell JC, Ball AJ. Cord compression and carcinoma of the prostate: is laminectomy justified? Br J Urol 1985;57: [15] Rodichok LD, Harper GR, Ruckdeschel JC, Price A, Roberson G, Barron KD, et al. Early diagnosis of spinal epidural metastases. Am J Med 1981;70: [16] Sorensen PS, Borgesen SE, Rohde K, Rasmusson B, Bach F, Boge- Rasmussen T, et al. Metastatic epidural spinal cord compression: Results of treatment and survival. Cancer 1990;65: [17] Barcena A, Lobato RD, Rivas JJ. Spinal metastatic disease: An analysis of factors determining functional prognosis and the choice of treatment. Neurosurgery 1984;15: [18] Maurice-Williams RS, Richardson PL. Spinal cord compression: delay in diagnosis and referral of a common neuosurgical emergency. Br J Neurosurg 1988;2: [19] Bonner JA, Lichter AS. A caution about the use of MRI to diagnose spinal cord compression. New Eng J Med 1990;332: [20] Carmody RF, Yang PJ, Seeley GW, Seeger JF, Unger EC, Johnson JE. Spinal cord compression due to metastatic disease: Diagnosis with MR imaging versus myelography. Radiology 1989;173: [21] Greenberg HS, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: Results from a new protocol. Ann Neurol 1980;8: [22] Kaminski HJ, Diwan VG, Ruff RL. Second occurrence of spinal epidural metastasis. Neurology 1991;41: [23] Zelefsky MJ, Scher HI, Krol G, Portenoy RK, Leibel SA, Fuks ZY. Spinal epidural tumor in patients with prostate cancer. Cancer 1992; 70: [24] Saad F, Gleason DM, Murray R, Tchekmedyian S, Venner P, Lacombe L, et al. A randomized, placebo-controlled trial of zoledronic acid in patients with hormone-refractory metastatic prostate carcinoma. J Natl Cancer Inst 2002;94(19):