Br a i n metastases are the tumors most frequently. Safety and efficacy of Gamma Knife surgery for brain metastases in eloquent locations

Transcription

1 J Neurosurg 113:79 83, 2010 Safety and efficacy of Gamma Knife surgery for brain metastases in eloquent locations Clinical article Ni c o l a s De a, M.D., Mar t i n Bo r d u a s, Br e n d a n Ke n n y, M.D., F.R.C.S.I., Dav i d Fo r t i n, M.D., F.R.C.S.C., a n d Dav i d Ma t h i e u, M.D., F.R.C.S.C. Divisions of Neurosurgery and Neuro-oncology, Université de Sherbrooke, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Quebec, Canada Object. Brain metastases are the most frequently occurring cerebral tumors. Tumors that are located in eloquent cerebral parenchyma can cause considerable morbidity and may pose a significant challenge during surgery. Gamma Knife surgery (GKS) is a recognized treatment modality for brain metastases. This study was undertaken to assess the safety and efficacy of GKS, specifically for brain metastases in eloquent locations. Methods. Charts of patients harboring brain metastases that were treated by GKS at the Centre Hospitalier Universitaire de Sherbrooke between August 2004 and April 2008 were reviewed. Planning images were assessed by an independent neurosurgeon to assess tumor location. Eloquent locations included the primary motor, somatosensory, speech, and visual cortices; the basal ganglia; the thalamus; and the brainstem. Data on survival, tumor response, and complications were analyzed and compared with data published on surgical treatment of these lesions. Results. During the study period, 650 metastases in 295 patients were treated with GKS; of these, 164 metastases in 95 patients were located in eloquent areas. In this subgroup, the median age of patients was 59 years and women constituted 57.9% of the population. The median Karnofsky Performance Scale score was 80% (range 50% 100%). Patients were categorized according to their recursive partitioning analysis class: Class 1, 22.1%; Class 2, 70.5%; and Class 3, 7.4% of patients. Non small cell lung cancer was the most common primary tumor (63.2% of metastases), followed by small cell lung (8.4%), breast (7.4%), colorectal (5.3%), and renal cell (4.2%) cancers, as well as melanoma (4.2%). The median dose to the tumor margin was 18 Gy (range Gy). The median duration of survival after GKS was 8.2 months. The recursive partitioning analysis class was the most significant variable affecting survival (p < ). Immediate control was achieved in 92.9% of tumors, and 68.6% of tumors were still controlled at the last follow-up. The median time to tumor progression was 16 months. Higher margin dose (p = 0.002), the absence of edema (p = 0.009), and the non small cell lung cancer tissue type (p = 0.035) positively affected response rates. Steroid medications were no longer used in 46% of patients after GKS. New neurological deficits occurred in 5.7% of patients and seizures in 5.7%. All these deficits were transient and patients completely recovered in response to a temporary course of steroids. Imaging studies showed that new edema occurred in 8.6% of treated metastases and biopsy-proven radiation necrosis in 1.4%. Conclusions. Gamma Knife surgery is safe and effective for brain metastases located in eloquent areas. (DOI: / GKS10957) Ke y Wo r d s brain metastases Gamma Knife surgery eloquent brain location Br a i n metastases are the tumors most frequently encountered in neurosurgical practice. Up to 40% of patients with systemic cancer will develop CNS metastases over the course of their disease. 9,22 This percentage has been rising in recent years because of more aggressive treatment of the primary tumor with subsequent longer survival of patients and also because of more liberal and appropriate use of sensitive methods Abbreviations used in this paper: GKS = Gamma Knife surgery; KPS = Karnofsky Performance Scale; NSCLC = non small cell lung cancer; RPA = recursive partitioning analysis; WBRT = wholebrain radiation treatment. of cerebral imaging, leading to more lesions being diagnosed. 1 The optimal way to manage cases of metastases is often a matter of controversy. The necessity of WBRT for all cerebral metastases and the modalities used for adjuvant therapy are examples of unresolved issues. Surgery followed by radiation therapy has shown clear survival advantages over radiotherapy alone, improving the mean duration of survival from approximately 4 to months. 5,14,18,19 Radiosurgery has emerged as a minimally invasive alternative to microsurgical removal in a well-selected group of patients. 6,10,11,15,16 Nowadays, metastases constitute the most common indication for radiosurgery in many centers. A review of the recent lit- 79

2 N. Dea et al. erature showed no difference in survival and local control rates when comparing radiosurgery alone with resection plus radiation therapy of solitary brain metastases suitable for radiosurgery. 15,16 Metastases located in eloquent cerebral areas can cause considerable morbidity and may pose a significant surgical challenge. Even though the feasibility and safety of resection of lesions in eloquent areas has been studied, 1,3,17,23 to date no study has specifically focused on the safety and efficacy of GKS for metastases in eloquent areas of the brain. This article was designed to provide a retrospective analysis of patient survival, local control rates, and complications that are encountered when treating patients with cerebral metastases in eloquent areas with GKS. Methods The charts of patients with brain metastases treated by GKS at the Centre Hospitalier Universitaire de Sherbrooke between August 2004 and April 2008 were retrospectively reviewed. Approval from the institutional research ethics committee was obtained for the study. All procedures were planned using thin-slice volumetric Gd-enhanced T1-weighted MR imaging sequences after application of a Leksell G stereotactic frame. The images were imported to a workstation running Leksell GammaPlan software. All treatments were performed using a Leksell Gamma Knife model 4C (Elekta AB). All planning MR images were assessed by an independent neurosurgeon to define the exact tumor location and identify those lesions located in eloquent brain areas. Eloquent locations included the primary motor, somatosensory, speech, and visual cortices; the basal ganglia; the thalamus; and the brainstem. Patients with metastases in noneloquent locations were excluded from further analyses. Patients in whom radiosurgery was performed on the tumor bed after resection of a metastasis in an eloquent location were also excluded. The treatment dose was chosen according to the location and size of the tumor. Thus, cortical and deep nuclei tumors smaller than 1 cm received a dose of Gy at the margin; tumors between 1 and 2 cm received Gy; tumors between 2 and 3 cm received Gy; and tumors larger than 3 cm received Gy. Brainstem tumors received a margin dose ranging between 14 and 16 Gy. The survival, response, and complication rates were analyzed using the Kaplan-Meier method. Cox regression analyses were performed to assess the impact of different variables on outcomes in patients. Variables related to patients and tumors that were analyzed included patient sex and age, primary cancer origin, KPS score, RPA class, symptoms at presentation, presence of brain edema, and use of corticosteroid medications. The treatment characteristics that were studied included the type of treatment (radiosurgery alone, boost after radiotherapy), the dosage, and the irradiated volume. All patients were followed up clinically and by MR imaging at 2, 4, and 6 months and thereafter every 3 months. Immediate tumor control was defined as the best imaging response during the course of follow-up and included disappearance, reduction, or stabilization of the tumor, whereas the final response corresponded to the last response documented on available MR images. The imaging response was assessed using the criteria proposed by Macdonald et al. 12 Patients who were not taking steroid medications at the time of radiosurgery did not receive new prescriptions for those medications after treatment. For patients who were already receiving steroids, the medication was progressively tapered as soon as possible depending on the presence of edema, the size of the lesion, and the clinical need of the patient. This was usually achieved within 1 or 2 months. Results During the study period, a total of 650 cerebral metastases were treated in 295 patients at our institution. Of these, we identified 164 metastases located in eloquent areas in 95 patients. The median patient age at radiosurgery was 59 years (range years), with 42.1% male and 57.9% female patients. The median KPS score was 80% (range 50% 100%). When stratified according to Radiation Therapy Oncology Group RPA criteria, 22.1% of cases were Class 1; 70.5% were Class 2; and 7.4% were Class 3. Most patients had NSCLC (63.2% of metastases). Other histological types included small cell lung (8.4%), breast (7.4%), colorectal (5.3%), and renal cell (4.2%) cancer, as well as melanoma (4.2%) and various other primary tumors (7.5% of metastases). Radiosurgery was performed as a single treatment in 72% of cases and as a boost to WBRT in 28%. The median margin treatment dose was 18 Gy (range Gy), whereas the median maximal dose was 36 Gy (range Gy). The median tumor volume treated was 2.6 cm 3 (range cm 3 ). After radiosurgery, the median overall survival duration in patients whose metastases were located in eloquent areas was 8.2 months (mean 11.5 months) (Fig. 1). The survival rate was 58% at 6 months, 34.6% at 1 year, and 10.9% at 2 years. Variables that significantly influenced survival in the multivariate analyses were RPA class (p < ) and repeat treatment with GKS (p = 0.002) or chemotherapy (p = 0.001). The median survival duration was 23.6 months for patients in RPA Class 1, 7.4 months for those in RPA Class 2, and only 1.8 months for those in RPA Class 3 (Fig. 2). Of note, in this study, having a solitary brain metastasis did not significantly impact survival compared with having multiple cerebral lesions. After GKS, follow-up neuroimages were available for 115 metastases (70.1% of lesions) in 70 patients. Immediate tumor control (disappearance, reduction in size, and stabilization) was obtained in 92.9% of tumors, and 68.6% of the treated metastases continued to demonstrate control at the time of last available follow-up (Table 1). The median length of the available follow-up was 6 months. The median time to tumor progression was 16 months, as depicted by the Kaplan-Meier curve in Fig. 3. Local control rates were 83% at 6 months post-gks, 64% at 12 months post-gks, and 42% at 2 years post-gks. In the multivariate models, the presence of edema around 80

3 Radiosurgery for brain metastases in eloquent locations TABLE 1: Local control rates in brain metastases in eloquent areas Response to GKS Best (initial) Response Final Response complete response 14.3% 12.9% partial response 61.4% 42.9% stable disease 17.1% 12.9% progressive disease 7.1% 31.4% Fig. 1. Kaplan-Meier curve depicting overall survival after radiosurgery. Vertical lines indicate patient deaths. the tumor at the time of GKS was predictive of a worse outcome (p = 0.009), whereas increased margin dose (p = 0.002) and NSCLC tissue type (p = 0.035) were associated with a better tumor response. During follow-up, 42% of the cohort developed new cerebral metastases. As shown in the survival analysis, the development of new metastases did not significantly influence overall patient survival. After GKS, radiation-induced imaging changes (a transient increase in the T2 signal with increased Gd uptake on T1 MR images) were seen in 8.6% of treated tumors, indicating the presence of new edema. Radiation necrosis proven by pathology tests occurred in only 1 (1.4%) of 70 patients. This patient was a 37-year-old woman with a solitary metastasis from a sarcoma in the somatosensory cortex. Progressive enlargement of the tumor was seen on imaging studies 6 months after radiosurgery. Although the patient remained asymptomatic, a biopsy of the lesion was performed to guide management. Pathological findings revealed radiation necrosis, which completely resolved at 18 months without further intervention. Clinically, posttreatment seizures occurred in 4 (5.7%) of 70 patients and new or worsening neurological deficit occurred in another 4 patients (5.7%). All these deficits were transient, and the patients completely recovered with the aid of a temporary course of steroid medications. No permanent deficit was recorded. A summary of the new and worsening neurological deficits encountered in our series is shown in Table 2. Overall, compared with the patients pretreatment medication statuses, steroid medications were tapered and stopped in 46% of patients. Discussion Management of brain metastases can be challenging. Treatment modalities include microsurgical removal, WBRT, stereotactic radiosurgery, or a combination of these. The final decision is a very individualized one and is often made by a multidisciplinary panel of experts including neurosurgeons, medical oncologists, and radiation oncologists. There can also be great variation in management between different centers, based on the available technology and on patient and physician preferences. The indispensable role of surgery for tissue diagno- Fig. 2. Kaplan-Meier curves of survival after radiosurgery according to RPA classification. Vertical lines show time of a censored observation (censuré). 81

4 N. Dea et al. Fig. 3. Kaplan-Meier depicting local control after radiosurgery. Vertical lines indicate tumor progression. sis, treatment of intracranial hypertension or symptomatic mass effect, and treatment of lesions with a volume greater than 15 cm 3 is generally well accepted. 1,3 However, metastases located in eloquent areas of the brain can pose a considerable surgical challenge, and the resection of those tumors can lead to significant neurological morbidity. Few articles in the literature have specifically addressed the topic of metastases in eloquent brain locations. Weil and Lonser 23 reviewed the cases of 17 consecutive patients with metastases within the primary motor cortex, who underwent complete microsurgical tumor removal without local recurrence. This cohort had a mean survival of 10.6 months. Three patients (17.6%) had transient or reversible neurological deficits. No permanent incidence of morbidity was seen. That study was designed to specifically study metastases in the primary motor area. No other eloquent locations were included. Paek et al. 17 reported a series of 208 patients with single or multiple metastases treated by surgical removal. For a subgroup of 27 patients harboring lesions within eloquent parts of the brain, neurological deterioration occurred in 19% of cases. The median survival time for the whole group was 8 months. In yet another series report, Sawaya 20 showed a 7% major neurological complication rate for tumors in eloquent areas and a local recurrence rate of 8% 12%. In our study, clinically significant complications (either seizures or new or worsening neurological deficits) occurred in 11.4% of patients. This compares favorably with data in the surgical literature but contrasts with data from a review article focusing on the complications of radiosurgery for metastatic brain disease published by Williams and colleagues in Those authors reported a global complication rate of 40%, with a 32% risk of new neurological deficits after radiosurgery. However, the patients in that report were treated between 1993 and 2004 using a linear accelerator, and dose planning was done using CT scans. In contrast, all our patients were treated with a Gamma Knife model 4C using standard volumetric MR imaging. The better definition afforded by MR imaging for planning and dosimetry may, in part, explain the difference between our results and those reported by Williams and colleagues. Nevertheless, we believe our results more accurately represent the expected posttreatment course in patients after radiosurgery for eloquent brain metastases using current planning techniques. The median survival of patients in our study was 8 months, which is comparable to what is reported in the literature after radiosurgery for brain metastases. 2,4,21 This suggests that having metastases in eloquent brain locations does not confer a more negative prognosis if those metastases are treated adequately. In this study, the only variables that influenced survival were RPA class, subsequent GKS treatment, and subsequent systemic chemotherapy. The RPA class is consistently associated with an impact on survival in the literature. 7,8 Patients with a good functional status tend to benefit from more aggressive subsequent treatment. This could explain why subsequent GKS and chemotherapy improved survival in our patients. Most of our patients were initially treated by GKS alone in an effort to delay WBRT for as long as possible, and as a consequence, almost one-half developed new remote brain metastases. As demonstrated in other studies, the occurrence of new lesions does not, in itself, impact overall survival if salvage therapy is offered. 2,13,21 The initial withholding of WBRT also provides the option of administering it later in the course of the disease if needed. Our study showed that radiosurgery is an effective and safe therapeutic modality for metastases in eloquent brain. This study is the first, to our knowledge, to specifically address this question. An article by Yen et al. 25 reported the results of brainstem metastases treated by GKS but did not include metastatic tumors in other eloquent areas. We acknowledge the fact that our study has several limitations. It was retrospectively designed, with the inherent statistical biases associated with this type of study. Also, in the de- TABLE 2: Summary of clinical complications encountered in this series* Case No. Age (yrs), Sex Metastasis Location Primary Cancer Edema at Treatment Margin Dose (Gy) Isodose (%) Volume Treated (cm 3 ) Complication 1 63, M Wernicke area SCLC no new speech deficit 2 74, M primary motor cortex NSCLC no worsening motor deficit 3 50, F somatosensory cortex colon yes worsening limb apraxia 4 64, M primary motor cortex NSCLC yes worsening motor deficit * SCLC = small cell lung carcinoma. 82

5 Radiosurgery for brain metastases in eloquent locations sign, we did not differentiate cortical eloquence from deep eloquent areas such as the brainstem, which may differ in radiation sensitivity. However, this omission was made purposefully in an effort to analyze all the complications encountered when treating metastases in all eloquent areas; in our opinion, this did not create any significant bias in the study. Conclusions Gamma Knife surgery is an effective, safe, and minimally invasive treatment for brain metastases located in eloquent areas. The results of GKS compare favorably with surgery in this setting, as shown in the available surgical literature. Accordingly, GKS should be considered in the management algorithm as a primary treatment modality for adequately sized brain metastases in eloquent locations. Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript preparation include the following. Conception and design: Mathieu. Acquisition of data: Dea, Borduas. Analysis and interpretation of data: Mathieu, Dea. Drafting the article: Dea. Critically revising the article: Mathieu, Fortin. Reviewed final version of the manuscript and approved it for submission: all authors. Statistical analysis: Dea. Administrative/ technical/material support: Kenny. Study supervision: Mathieu. Acknowledgment The authors thank Nathalie Carrier, M.Sc., for assistance with the statistical analyses. References 1. Al-Shamy G, Sawaya R: Management of brain metastases: the indispensable role of surgery. J Neurooncol 92: , Aoyama H, Shirato H, Tago M, Nakagawa K, Toyoda T, Hatano K, et al: Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA 295: , Black PM, Johnson MD: Surgical resection for patients with solid brain metastases: current status. J Neurooncol 69: , Chen JC, Bugoci DM, Girvigian MR, Miller MJ, Arellano A, Rahimian J: Control of brain metastases using frameless image-guided radiosurgery. Neurosurg Focus 27(6):E6, DeAngelis LM, Mandell LR, Thaler HT, Kimmel DW, Galicich JH, Fuks Z, et al: The role of postoperative radiotherapy after resection of single brain metastases. Neurosurgery 24: , Flickinger JC, Lunsford LD, Somaza S, Kondziolka D: Radiosurgery: its role in brain metastasis management. Neurosurg Clin N Am 7: , Gaspar L, Scott C, Rotman M, Asbell S, Phillips T, Wasserman T, et al: Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys 37: , Gaspar LE, Scott C, Murray K, Curran W: Validation of the RTOG recursive partitioning analysis (RPA) classification for brain metastases. Int J Radiat Oncol Biol Phys 47: , Gavrilovic IT, Posner JB: Brain metastases: epidemiology and pathophysiology. J Neurooncol 75:5 14, Gerosa M, Nicolato A, Foroni R: The role of gamma knife radiosurgery in the treatment of primary and metastatic brain tumors. Curr Opin Oncol 15: , Gerosa M, Nicolato A, Foroni R, Zanotti B, Tomazzoli L, Miscusi M, et al: Gamma knife radiosurgery for brain metastases: a primary therapeutic option. J Neurosurg 97 (5 Suppl): , Macdonald DR, Cascino TL, Schold SC Jr, Cairncross JG: Response criteria for phase II studies of supratentorial malignant glioma. J Clin Oncol 8: , Mathieu D, Kondziolka D, Flickinger JC, Fortin D, Kenny B, Michaud K, et al: Tumor bed radiosurgery after resection of cerebral metastases. Neurosurgery 62: , Mintz AH, Kestle J, Rathbone MP, Gaspar L, Hugenholtz H, Fisher B, et al: A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer 78: , Muacevic A, Kreth FW, Horstmann GA, Schmid-Elsaesser R, Wowra B, Steiger HJ, et al: Surgery and radiotherapy compared with gamma knife radiosurgery in the treatment of solitary cerebral metastases of small diameter. J Neurosurg 91: 35 43, Muacevic A, Wowra B, Siefert A, Tonn JC, Steiger HJ, Kreth FW: Microsurgery plus whole brain irradiation versus Gamma Knife surgery alone for treatment of single metastases to the brain: a randomized controlled multicentre phase III trial. J Neurooncol 87: , Paek SH, Audu PB, Sperling MR, Cho J, Andrews DW: Reevaluation of surgery for the treatment of brain metastases: review of 208 patients with single or multiple brain metastases treated at one institution with modern neurosurgical techniques. Neurosurgery 56: , Patchell RA, Tibbs PA, Regine WF, Dempsey RJ, Mohiuddin M, Kryscio RJ, et al: Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 280: , Patchell RA, Tibbs PA, Walsh JW, Dempsey RJ, Maruyama Y, Kryscio RJ, et al: A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 322: , Sawaya R: Surgical treatment of brain metastases. Clin Neurosurg 45:41 47, Sneed PK, Suh JH, Goetsch SJ, Sanghavi SN, Chappell R, Buatti JM, et al: A multi-institutional review of radiosurgery alone vs. radiosurgery with whole brain radiotherapy as the initial management of brain metastases. Int J Radiat Oncol Biol Phys 53: , Soffietti R, Rudā R, Mutani R: Management of brain metastases. J Neurol 249: , Weil RJ, Lonser RR: Selective excision of metastatic brain tumors originating in the motor cortex with preservation of function. J Clin Oncol 23: , Williams BJ, Suki D, Fox BD, Pelloski CE, Maldaun MV, Sawaya RE, et al: Stereotactic radiosurgery for metastatic brain tumors: a comprehensive review of complications. Clinical article. J Neurosurg 111: , Yen CP, Sheehan J, Patterson G, Steiner L: Gamma knife surgery for metastatic brainstem tumors. J Neurosurg 105: , 2006 Manuscript submitted June 12, Accepted August 19, Portions of this work were presented in poster form at the following conferences: Annual Meeting of the Congress of Neurological Surgeons, New Orleans, Louisiana, October 24 29, 2009, and the Leksell Gamma Knife Society 15th International Meeting, Athens, Greece, May 16 20, Address correspondence to: David Mathieu, M.D., F.R.C.S.C., Service de Neurochirurgie, Centre Hospitalier Universitaire de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, Quebec, Canada J1H 5N4. david.mathieu@usherbrooke.ca. 83