J. Radiat. Res., 52, 641 645 (2011) Regular Paper Radiotherapy for Patients with Symptomatic Intramedullary Spinal Cord Metastasis Haruko HASHII 1,4 *, Masashi MIZUMOTO 1,4 *, Ayae KANEMOTO 1,4, Hideyuki HARADA 1, Hirofumi ASAKURA 1, Takayuki HASHIMOTO 1,4, Kazuhisa FURUTANI 1, Hirohisa KATAGIRI 2, Yoko NAKASU 3 and Tetsuo NISHIMURA 1 Radiotherapy/Intramedullary spinal cord metastases/iscm/vertebral metastasis/radiation therapy. The study was performed to evaluate radiotherapy for patients with intramedullary spinal cord metastasis (ISCM) and to identify the clinical features of ISCM. The subjects were 18 patients (8 men, 10 women) with ISCM who underwent radiotherapy between September 2002 and February 2008. The primary lesions were lung cancer in 8 patients (2 small cell, 6 non-small cell), breast cancer in 6, malignant melanoma in 2, renal cell carcinoma in 1, and rectal cancer in 1 patient. Diagnosis, symptoms and survival of these patients were compared with those for 544 patients with vertebral metastases who underwent radiotherapy at the same institute between September 2002 and November 2006. In the 18 patients with ISCM, the 6-month survival rate after radiotherapy was 36% and the median survival period was 4.0 months. Ten patients had neurological improvement or pain relief after radiotherapy. Brain metastases were six fold more frequent in the patients with ISCM than in those with vertebral metastasis [89% vs. 15%, p = 0.001]. At the time of radiotherapy, back pain in patients with vertebral metastasis was more frequent [97% vs. 33%, p = 0.001] but neurological deficits were less common [24% vs. 100%, p = 0.001]. Most ISCM cases were diagnosed by contrast-enhanced MRI, with detection by contrast-enhanced CT in only 3/18 cases (17%). ISCM has a poor prognosis and most patients have neurological deficits that impair quality of life. Early diagnosis by MRI is important for suspected ISCM to allow initiation of radiotherapy before development of neurological deficits. INTRODUCTION *Corresponding author: Phone: +81-29-853-7100, Fax: +81-29-853-7102, E-mail: mizumoto1717@hotmail.com Divisions of 1 Radiation Oncology, 2 Orthopaedic Oncology and 3 Neurosurgery Shizuoka Cancer Center Hospital, 1007 Shimonagakubo, Nagaizumi Town, Shizuoka, 411-8777, Japan; 4 Division of Radiation Oncology, University of Tsukuba, 2-1-1 Amakubo, Tsukuba, Ibaraki, 305-0005, Japan. doi:10.1269/jrr.10187 Intramedullary spinal cord metastasis (ISCM) is a rare condition. Chason reported only 10 cases (2%) among 200 patients with central nervous system metastases in a series of 1066 autopsy cases. 1) Lung and breast cancers are the most common primary tumor sites in ISCM, 2 8) but ISCM is seen in a wide variety of cancer. 9 16) The prognosis of patients with ISCM is very poor and many have neurological deficits and pain. Radiotherapy and surgery can be effective in selected cases 17 21) : for example, Lee et al. reported a transient improvement in motor function in 2 of 11 patients with ISCM who received radiotherapy at a total dose of 20 30 Gy. 17) However, there is only limited information on treatment, we performed a retrospective study of 18 patients with ISCM who received radiotherapy at our institute. MATERIALS AND METHODS Patients Eighteen patients with ISCM underwent radiotherapy at our hospital between September 2002 and February 2008. The patients comprised 8 men and 10 women, and had a median age of 55 years old (range: 37 to 76 years old). The primary lesions were lung cancer (n = 8, 44%, 2 small cell and 6 non-small cell lung cancer), breast cancer (n = 6, 33%), malignant melanoma (n = 2, 11%), renal cell carcinoma (n = 1, 6%), and rectal cancer (n = 1, 6%). Seventeen of the 18 patients had neurological deficits. Using the Frankel classification, 22) Two patient was classified in category A (complete motor loss and sensory loss), 2 in category B (complete motor loss and incomplete sensory loss), 8 in
642 H. Hashii et al. category C (loss of motor function: no practical use), and 6 in category D (useful motor function: could move the lower limb and many could walk with or without aids). One patient had an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 1, 3 were PS 2, 9 were PS 3, and 5 were PS 4. Twelve of the 18 patients had a single tumor and 6 had multiple tumors along their spinal cord. Fourteen patients had a well-defined tumor and 4 had diffuse tumors. Brain metastases were found in 14 patients before radiotherapy and in 2 after radiotherapy. The patient and tumor characteristics are summarized in Table 1. Radiation therapy Irradiation was performed with 4 to 18 MV photons from a linear accelerator and was mainly delivered through a single posterior field. The radiation dose was prescribed to Table 1. Patient and tumor characteristics. Characteristics Number Age (years) 37 76 (range) 55 (median) Gender Male 8 Female 10 ECOG performance status 1 1 2 3 3 9 4 5 Primary tumor Breast cancer 6 Non-small cell lung cancer 6 Small cell lung cancer 2 Malignant melanoma 2 Renal cell carcinoma 1 Rectal cancer 1 Tumor characteristics Single 12 Multiple 6 Well-defined type 14 Diffuse type 4 Brain metastases Yes, before radiotherapy 14 Yes, after radiotherapy 2 No 1 No data available 1 ECOG, Eastern Cooperative Oncology Group the center of the spinal cord using a radiation port in a threedimensional treatment planning system. The treatment volume usually encompassed 1 normal vertebra above and below the metastatic lesion. The radiation schedule was determined mainly based on prognosis or histology at a conference among primary doctors and radiation oncologists. Eight patients received 30 Gy in 10 fractions, 5 received 20 Gy in 5 fractions, and 5 received schedules of 21 to 40 Gy in 7 to 20 fractions. Toxicities were evaluated using CTCAE ver. 3.0. 23) Informed consent was obtained from the patients for the procedures described in the study. Statistical analysis All statistical analyses were performed with the SPSS software package version 11.0 (SPSS Inc., Chicago, IL). Overall survival was defined as the interval between the start of radiotherapy and the last follow-up visit when the patient was alive. The Kaplan-Meier method was used for calculation of survival rate. A log-rank test was performed for evaluation of differences between two categories, with P < 0.05 considered to indicate a significance difference. RESULTS Treatment outcome Of the 18 patients, 14 were followed up until death. All intramedullary spinal cord metastases were clearly diagnosed by T1-weighted MRI with gadolinium enhancement. Only 3 were diagnosed by contrast-enhanced computed tomography (CT). The 3- and 6-month overall survival rates after radiotherapy were 54% and 36%, respectively, and the median overall survival period was 4.0 months (Fig. 1). After radiotherapy, 10 patients (56%) had improvement of neurological deficits or relief of pain. Neurological deficits improved from Frankel category B to D in 1 patient and improved slightly within the same category in 7. Four patients had relief of pain, but 8 (44%) showed no change in neurological deficits or pain. There were no exact data about the time of symptom-onset. The times to start of radiotherapy from the diagnosis were almost within 8 days. There was no significant correlation between the time from diagnosis and outcome, but in the only case in which neurological deficits improved from Frankel category B to D, radiotherapy was started in the same day of diagnosis. There were no acute or late toxicities of CTCAE grade 3 or greater. The details of the treatment outcomes are shown in Table 2. Representative case One of the patients was a 49-year-old woman who presented with breast cancer. She had a history of right mastectomy 6 years earlier. At the time of initial consultation, there were multiple metastases to the brain, liver, supraclavicular nodes, and bones. A neurologic examination revealed impaired motor function in the right lower extremities. Two
Radiotherapy for Spinal Cord Metastases 643 weeks after the appearance of this symptom, T1-weighted MRI revealed an intramedullary mass that was markedly enhanced by gadolinium at the Th 9 level (Fig. 2a). Radiotherapy was administered at a dose of 30 Gy in 10 fractions. Fig. 1. Kaplan-Meier estimate of overall survival for patients with intramedullary spinal cord metastasis (ISCM) or spine metastasis. The 1-, 3- and 6-month survival rates after radiotherapy were 77%, 54% and 36%, respectively, for patients with ISCM, and 93%, 68% and 50%, respectively, for patients with spine metastases. Two weeks after completion of the spinal cord radiotherapy, there was a significant improvement in the motor function of the lower extremities and the patient was able to walk with aid. One month after radiotherapy, MRI showed significant improvement of ISCM (Fig. 2b). Comparison with patients with vertebral metastasis A total of 544 patients with vertebral metastasis were treated by radiotherapy at our hospital between September 2002 and November 2006. 24) During the same period 9 patients with ISCM received radiotherapy; thus, the frequency of vertebral metastasis (544 patients) was 60 times higher than that of ISCM (9 patients) in our hospital. At the time of radiotherapy, the frequency of back pain was higher in patients with vertebral metastasis compared to patients with ISCM [526/544 (97%) vs. 3/9 (33%), p = 0.001], but significantly fewer patients with vertebral metastases had neurological deficits [133/544 (24%) vs. 9/9(100%), p = 0.001]. Most patients with vertebral metastasis were diagnosed by CT, whereas 7 of 9 patients with ISCM could not be diagnosed by contrast-enhanced CT. In contrast, a definite diagnosis of ISCM was made by MRI in all 9 patients. At the time of radiotherapy, the incidence of brain metastases was significantly higher in patients with ISCM compared to those with vertebral metastasis [at least 8/9 (89%) vs. 81/ 544 patients (15%), p = 0.001]; that is, brain metastases was about six times more frequent in patients with ISCM. The characteristics of the patients with ISCM or vertebral metastasis are shown in Table 3 and the poorer prognosis of Table 2. Treatment outcome No. Primary tumor Age (years) Gender Total dose (Gy) Fraction size (Gy) Frankel Before RT Frankel After RT Neurological deficit Survival (Weeks) Follow-up and status 1. Non-small cell lung cancer 49 M 20 4 C C Improved 8 Death from cancer 2. Non-small cell lung cancer 55 M 35 2.5 D D Improved 26 Death from cancer 3. Non-small cell lung cancer 57 M 30 3 D D No change 9 Death from cancer 4. Non-small cell lung cancer 64 F 30 3 D D Improved 17 Death from cancer 5. Non-small cell lung cancer 65 M 30 3 C C No change 4 Alive, lost to follow up 6. Non-small cell lung cancer 76 M 30 3 D D Improved 20 Alive 7. Small cell lung cancer 67 M 21 3 D D No change (pain relieved) 20 Death from cancer 8. Small cell lung cancer 72 F 30 3 C C No change (pain relieved) 28 Alive 9. Breast cancer 41 F 40 2 D D Improved 33 Death from cancer 10. Breast cancer 49 F 30 3 C C Improved 44 Death from cancer 11. Breast cancer 50 F 20 4 C C No change 4 Death from cancer 12. Breast cancer 51 F 20 4 A A No change 2 Death from cancer 13. Breast cancer 53 F 30 3 D D Improved 12 Death from cancer 14. Breast cancer 57 F 20 4 C C No change 17 Death from cancer 15. Malignant melanoma 54 F 39 3 D D No change 12 Death from cancer 16. Malignant melanoma 56 M 20 4 D D No change 4 Death from cancer 17. Renal cell carcinoma 65 M 30 3 A A No change 3 Death from cancer 18. Rectal cancer 37 F 28 3.5 B D Improved 33 Alive
644 H. Hashii et al. (a) (b) Fig. 2. (a) T1-weighted MRI with gadolinium enhancement revealed an intramedullary mass (arrow). (b) The intramedullary mass decreased in size 1 month after external beam irradiation (arrow). patients with ISCM is illustrated in Fig. 1. DISCUSSION We found a very poor outcome of patients with ISCM (median survival time: 4.0 months), as in previous reports. 2 8) Ten of the 18 patients with ISCM obtained relief of symptoms after radiotherapy, but none achieved complete remission. Six patients showed no change in symptoms and another two patients had exacerbated symptoms at 2 and 4 months, respectively, after radiotherapy. The frequency of ISCM was less than 2% among all cases of vertebral metastases treated by radiotherapy in our hospital. Therefore, most cases of back pain and neurological deficits caused by a metastatic lesion were due to vertebral metastases. However, patients with ISCM presented Table 3. Characteristics of patients with ISCM or spine metastases Characteristics ISCM (n = 9) Spine metastases (n = 544) P value Primary tumor Lung 4 (44%) 151 (28%) Breast 4 (44%) 115 (21%) Gastrointestinal 0 (0%) 94 (17%) Others 1 (11%) 184 (34%) Symptoms and signs Back pain 3 (33%) 526 (97%) 0.001 Paralytic symptom 9 (100%) 133 (24%) 0.001 Diagnosed by CT scan 2 (22%) Almost all Brain metastases 8 (89%) 81 (15%) 0.001 with rapidly progressing and serious neurological deficits that reduced their motor activity, and in our most effective case radiotherapy was started at the same day of diagnosis, which probably indicates that early diagnosis of ISCM and early treatment are important. The majority of primary tumors in cases of ISCM are lung or breast cancers characterized by a high frequency of hematogenous metastasis. However, lung and breast cancers are also the two most common primary tumors in patients with vertebral metastasis, which makes differentiation of ISCM difficult based on the primary tumor. Two clues may help in differential diagnosis of ISCM and vertebral metastases. First, brain metastases in patients with ISCM were six times more frequent than in patients with vertebral metastases, which may be explained by the widespread hematogenous metastasis to the central nervous system in ISCM, including to the spinal cord. Second, symptom patterns at the time of diagnosis differed between the two groups of patients: 100% of patients with ISCM presented with focal neurological deficits and 33% had back pain, whereas 97% of those with vertebral metastases presented with back pain and 24% had focal neurological deficits. In this study, all cases of ISCM were diagnosed by MRI with gadolinium enhancement, whereas only three were identifiable on a CT scan. Therefore, prompt use of gadoliniumenhanced MRI is recommended after exclusion of vertebral metastasis for patients with progressive paralytic symptoms (rather than back pain) and brain metastases. A few reports have shown the effectiveness of radiotherapy for ISCM, 2,17) but the numbers of patients in these studies were insufficient to develop an optimal radiotherapy schedule. Radiation schedules of 20 Gy in 5 fractions to 40 Gy in 20 fractions were used in Lee et al., 17) Conill et al. 2) and the current study, based on the dose tolerated by the spinal cord. Since patients with ISCM have a very poor prognosis, we believe that long-course radiotherapy (such as
Radiotherapy for Spinal Cord Metastases 645 40 Gy in 20 fractions) is unsuitable, but we note that longcourse radiotherapy used for patients with vertebral metastases results in longer survival. 24,25) A favorable tumor type (breast or prostate cancer), no visceral metastasis, and a good performance status were found to be prognostic factors for survival in these reports, and therefore long-course radiotherapy may be useful for selected patients with ISCM with expected longer survival. However, the majority of patients with ISCM have poor life expectancy due to brain metastases and a poor performance status caused by neurological deficits. Furthermore, in-field recurrence may occur after radiotherapy, and therefore short-term radiotherapy at a total dose up to the tolerance dose of the spinal cord is suitable in most patients with ISCM. Most patients with ISCM in the current study had a poor performance status and impaired quality of life due to neurological deficits, and about half showed no improvement of neurological deficits after radiotherapy. This suggests that prompt radiotherapy is required to maintain QOL in patients with ISCM. Our findings also emphasize the importance of early diagnosis by gadolinium-enhanced MRI when ISCM is suspected. REFERENCES 1. Chason JL, Walker FB and Landers JN (1963) Metastatic carcinoma in the central nervous system and dorsal root ganglia. A prospective autopsy study. Cancer 16: 781 787. 2. Conill C, et al (2007) Clinical outcome in patients with intramedullary spinal cord metastases from lung cancer. Clin Transl Oncol 9: 172 176. 3. Kosmas C, et al (2005) Intramedullary spinal cord metastases in breast cancer: report of four cases and review of the literature. J Neurooncol 71: 67 72. 4. Potti A, et al (2001) Intramedullary spinal cord metastases (ISCM) and non-small cell lung carcinoma (NSCLC): clinical patterns, diagnosis and therapeutic considerations. Lung Cancer 31: 319 323. 5. Wada H, et al (2001) Intramedullary spinal cord metastasis of lung adenocarcinoma causing Brown-Séquard Syndrome. Nihon Kokyuki Gakkai Zasshi 39: 590 594. 6. Schiff D and O Neill BP (1996) Intramedullary spinal cord metastases: clinical features and treatment outcome. Neurology 47: 906 912. 7. Holoye P, et al (1984) Spinal cord metastasis in small cell carcinoma of the lung. Int J Radiat Oncol Biol Phys 10: 349 356. 8. Murphy KC, et al (1983) Intramedullary spinal cord metastases from small cell carcinoma of the lung. J Clin Oncol 1: 99 106. 9. Törnwall J, Snäll J and Mesimäki K (2008) A rare case of spinal cord metastases from oral SCC. Br J Oral Maxillofac Surg 46: 594 595. 10. Bakshi A, et al (2006) Successful complete regression of isolated intramedullary spinal cord metastases from epithelial ovarian carcinoma with chemotherapy and radiotherapy. Indian J Cancer 43: 136 138. 11. Kastritis E, et al (2006) Intramedullary spinal cord and leptomeningeal metastases in a patient with carcinoma of the uterine cervix. Gynecol Oncol 102: 124 127. 12. Gazzeri R, et al (2006) Pure intramedullary spinal cord metastasis secondary to gastric cancer. Neurosurg Rev 29: 173 177. 13. Jalali K, et al (2004) Spinal cord metastasis from prostate cancer. Prog Urol 14: 554 557. 14. Poggi MM, et al (2001) Intramedullary spinal cord metastasis from renal cell carcinoma: detection by positron emission tomography. Clin Nucl Med 26: 837 839. 15. Honma Y, Kawakita K and Nagao S (1996) Intramedullary spinal cord and brain metastases from thyroid carcinoma detected 11 years after initial diagnosis--case report. Neurol Med Chir 36: 593 597. 16. Lyding JM, et al (1987) Intramedullary spinal cord metastasis in Hodgkin s disease. Rapid diagnosis and treatment resulting in neurologic recovery. Cancer 60: 1741 1744. 17. Lee SS, et al (2007) Intramedullary spinal cord metastases: a single-institution experience. J Neurooncol 84: 85 89. 18. Watanabe M, et al (2006) Intramedullary spinal cord metastasis: a clinical and imaging study of seven patients. J Spinal Disord Tech 19: 43 47. 19. Gasser T, et al (2005) Surgical treatment of intramedullary spinal cord metastases of systemic cancer: functional outcome and prognosis. J Neurooncol 73: 163 168. 20. Kalayci M, et al (2004) Intramedullary spinal cord metastases: diagnosis and treatment - an illustrated review. Acta Neurochir 146: 1347 1354. 21. Connolly ES Jr, et al (1996) Intramedullary spinal cord metastasis: report of three cases and review of the literature. Surg Neurol 46: 337 338. 22. Frankel HL, et al (1969) The value of postural reduction in the initial management of closed injuries of the spine with paraplegia and tetraplegia. I Paraplegia 7: 179 192. 23. National Cancer Institute. National Cancer Institute common toxicity criteria. http://ctep.cancer.gov/protocoldevelopment/ electronic applications/docs/ctcaev3.pdf [accessed October 9, 2009] 24. Mizumoto M, et al (2008) Prognostic factors of survival after radiotherapy for metastases to the spinal column and a scoring system of survival: review of 544 patients in Shizuoka Cancer Center Hospital. Cancer 113: 2816 2822. 25. Rades D, Dunst J and Schild SE (2008) The first score predicting overall survival in patients with metastatic spinal cord compression. Cancer 112: 157 161. 26. Mizumoto M, et al (2011) Radiotherapy for patients with metastases to the spinal column: a review of 603 patients at Shizuoka Cancer Center Hospital. Int J Radiat Oncol Biol Phys 79(1): 208 213. Received on December 27, 2010 Revision received on April 5, 2011 Accepted on April 11, 2011 J-STAGE Advance Publication Date: July 15, 2011