Survival Following CyberKnife Radiosurgery and Hypofractionated Radiotherapy for Newly Diagnosed Glioblastoma Multiforme

Transcription

1 Technology in Cancer Research and Treatment ISSN Volume 7, Number 3, June 2008 Adenine Press (2008) Survival Following CyberKnife Radiosurgery and Hypofractionated Radiotherapy for Newly Diagnosed Glioblastoma Multiforme Current therapeutic goals for treatment of Glioblastoma Multiforme (GBM) involve gross total resection followed by multifractionated focal external beam radiation therapy (EBRT). Patients treated with optimal therapy have a median survival of approximately months. In the present study, we sought to determine whether a hypofractionated dosing schedule using CyberKnife is at least as effective as multifractionated focal EBRT. A retrospective analysis was conducted on 20 histopathologically confirmed GBM patients treated with CyberKnife at Okayama Kyokuto Hospital in Japan after gross total resection (n=11), subtotal resection (n=8), or biopsy (n=1). Eight patients also received adjuvant ACNU and Vincrisitine chemotherapy according to local protocol; however, no patient received any other form of radiation besides post surgical/biopsy CyberKnife treatment. The treated tumor volumes ranged from 9.62 cm cm 3 (mean: cm 3 ). The marginal dose (D90) ranged from Gy Gy (mean: Gy) with a maximum mean dose of Gy (range: Gy Gy). The prescribed isodose line ranged from 50.38%-85.68% with a mean of 79.25%. Treatment was delivered in 1-8 fractions (mean: 5.65). Patients were followed from 2-36 months (mean: months). Overall median survival was 16 months with 55% of patients alive at 12 months and 34% of patients alive at 24 months. Median survival of patients in Recursive Partitioning Analysis (RPA) classes III or IV was 32 months versus 12 months for those in RPA class V. Median survival for patients who received gross total resection was 36 months versus 8 months for those who underwent subtotal resection or biopsy. The results of this study using CyberKnife stereotactic radiosurgery (SRS) and hypofractionated radiotherapy compared favorably to historic data using focal EBRT in newly diagnosed post surgical GBM patients. A larger prospective analysis that compares CyberKnife SRS and hypofractionated radiotherapy to focal EBRT is warranted. John D. Lipani, M.D., Ph.D. 1,4* Paul S. Jackson, M.D., Ph.D. 1 Scott G. Soltys, M.D. 2 Kengo Sato, M.D. 3 John R. Adler, M.D. 1 1 Department of Neurosurgery 2 Radiation Oncology Stanford University Medical Center Stanford, California 3 Yokohama CyberKnife Center Yokohama, Japan 4 Institute for Neurosciences Capital Health System Trenton, NJ Key words: Stereotactic radiosurgery; Hypofractionated radiotherapy; CyberKnife; GBM. Introduction Glioblastoma multiforme (GBM) is the most common and aggressive of the malignant primary brain neoplasms in adults, accounting for approximately 12,000 new cases annually in the US (1). The overall prognosis for GBM has changed little in the past 30 years despite major improvements in neuroimaging, neurosurgery, radiation treatment techniques, and supportive care (2). Without * Corresponding Author: John D. Lipani, M.D., Ph.D. jdlipani@aol.com Abbreviations: ACNU, 1-(4-amino-2-methyl-5-pyrimidinyl) methyl-3-(2-chloroethyl)-3-nitrosourea hydrochloride; CT, Computed tomography; CK, CyberKnife; DRRs, Digitally reconstructed radiographs; EBRT, External beam radiation therapy; GBM, Glioblastoma multiforme; KPS, Karnofsky performance scale; MRI, Magnetic resonance imaging; MRS, Magnetic resonance spectroscopy; RPA, Recursive partitioning analysis; SRS, Stereotactic radiosurgery; WBRT, Whole brain radiation therapy. 249

2 250 Lipani et al. therapy, patients with GBM uniformly die within three months (3, 4). Patients treated with optimal therapy, including surgical resection, radiation therapy, and chemotherapy, have a median survival of approximately months, with 27% of patients surviving at 2 years and 13% of patients surviving up to 4 years (5-7). The pathogenesis of GBM is complex and multiple factors such as extent of surgical resection, tumor size, performance status, and age, play a role in the ultimate outcome of treatment (8-11). Current therapeutic goals include gross total resection followed by concurrent focal external beam radiation therapy (EBRT) and chemotherapy with temozolomide (6, 7, 12). Early studies showed that postoperative whole brain radiation therapy (WBRT) prolongs median survival from 14 weeks (surgery alone) to 36 weeks (surgery followed by WBRT) (13-15). However, an increased emphasis on local tumor control has been recommended due to the fact that up to 90% of patients develop local recurrences within 2 cm of the initial tumor location (16-19). Compared to focal EBRT, the use of WBRT has not improved outcomes and leads to greater toxicity (20). With focal EBRT, the dose is applied to a limited field that includes the enhancing volume on computed tomography (CT) or magnetic resonance imaging (MRI) scans with a 2-3 cm margin or a 1-2 cm margin beyond the peri-tumoral edema as defined by T2-weighted MRI images. The standard dose of EBRT is 60 Gy in single daily fractions of 2 Gy, five times a week for six weeks. Although a dose response up to 60 Gy exists for EBRT (21), there is no benefit to further dose escalation with the addition of a SRS boost (22) or brachytherapy (23) to conventional EBRT. Stereotactic radiosurgery (SRS) and hypofractionated radiotherapy allows for the delivery of high dose radiation to surgically inaccessible or previously irradiated areas with extreme precision. Prior studies using Gamma Knife SRS have demonstrated some success in achieving local control with recurrent GBM s in patients who have already undergone traditional EBRT (24, 25). However, the use of SRS or hypofractionated radiotherapy in place of focal EBRT after surgery has yet to be investigated. In this study, we sought to determine whether a high ablative radiation dose using CyberKnife SRS in 1-5 fractions or hypofractionated radiotherapy in 6-8 fractions is at least as effective as focal multifractionated EBRT. In addition to limiting exposure to critical structures, using a hypofractionated *D, Dead; A, Alive Male 6/D subtotal 4 temporal 2 81 Female 10/D subtotal 2 frontal 3 69 Female 27/A gross total 2 frontal 4 57 Male 12/D subtotal 4 caudate head 5 46 Female 2/D subtotal 4 caudate head 6 68 Male 6/D gross total 4 temporal 7 60 Female 12/D subtotal 4 temporal 8 72 Male 24/A gross total 4 frontal 9 70 Female 15/D gross total 68 frontal Male 24/D gross total 4 vermis Female 5/D biopsy 2 temporal Female 36/D gross total 4 temporal Male 16/A gross total 4 temporal Male 7/D subtotal 2 temporal Male 18/D gross total 20 occipital Male 16/D gross total 12 frontal Female 32/D subtotal 8 temporal Male 28A gross total 8 frontal Male 8/D subtotal 12 frontal Male 25/A gross total 12 occipital Patient Number Age (y) Gender Follow up (mo)/endpoint KPS RPA (D/A) * Number of fractions Volume (cm 3 ) D90 (cgy) Dmax (cgy) Resection Surgery/CK Interval (Weeks) Location Table I Demographic and clinical patient data.

3 Radiosurgery and Hypofractionated Radiotherapy as Primary Treatment for GBM 251 treatment schedule with CyberKnife would result in a much shorter treatment time and may provide a better quality of life for GBM patients. Materials and Methods Patients A retrospective analysis was conducted on 20 histopathologically confirmed GBM patients (11 males, 9 females, age range: 38-88, mean: years) treated with CyberKnife after gross total resection (n=11), subtotal resection (n=8), or biopsy (n=1). After granting informed consent, all patients received CyberKnife treatment by co-author Dr. Kengo Sato at Okayama Kyokuto Hospital in Japan between November 2000 and June All patients chose CyberKnife treatment over current standards as supported by evidence-based medicine. Temozolomide was not approved for use in Japan until September Therefore, eight patients also received adjuvant ACNU and Vincristine chemotherapy according to local standards; however, no patient received any other form of radiation besides post surgical/biopsy CyberKnife treatment. Timing of radiosurgery treatment was determined based on patients clinical status, personal preferences, or recommendations of the referring oncologists. The time from diagnosis of GBM to CyberKnife treatment ranged from 2-68 weeks (median time interval: 4 weeks). Karnofsky Performance Scale (KPS) ranged from 50 to 100 (mean: 81.5). Patients were also classified according to Radiation Oncology Group (RTOG) Recursive Partitioning Analysis (RPA) classification (26) (Class III, 20%; IV, 10%; and V, 70%). Demographic and clinical patient data are presented in Tables I and II. CyberKnife Treatment A non-isocentric treatment plan was generated for each patient using 1.25 mm contrast-enhanced axial CT images. These images were used to create digitally reconstructed radiographs (DRRs) to facilitate skull tracking during treatment. In addition, 1 mm slice T2-weighted MRI (n=2), FLAIR sequence MRI (n=2), or MR spectroscopy (n=16) images were fused to the planning CT for enhanced tumor delineation. Images were then transferred to the CyberKnife workstation where the treatment target volume and surrounding critical structures were defined. For T2- weighted and FLAIR images, tumor margins were defined as 1-2 cm beyond the hyper-intense tumor margins. For MR spectroscopy images, tumor margins were defined where the choline/n-acetylaspartate ratio was greater than 1 (Figure 1). An inverse treatment planning method was used to limit the radiation dose exposure to critical struc- Demographics Number treated 20 Age Table IIA Summary of demographic and clinical patient data. Average (Y) Minimum (Y) 38 Maximum (Y) 88 Sex (% female) 40.00% KPS Average 81.5 Stdev RPA Class Follow up 5 70% (14) 4 10% (2) 3 20% (4) Length of follow up (months) Mean Median 15.5 Minimum 2 Maximum 36 Treatment parameters Interval surgery to CK (weeks) Median 4 Minimum 2 Maximum 68 Treatment volume Average (cm 3 ) Minimum 9.62 Maximum Marginal Dose (D90) Average (cgy) Minimum 1999 Maximum 4147 Max Dose Average (cgy) Minimum 2333 Maximum Isodose Average (cgy) 79.25% Minimum 50.38% Maximum 85.68% Fractions Mean 5.65 Median 5 Minimum 1 Maximum 8 Table IIB Summary of demographic and clinical patient data. Median survivals (months) All patients 16 Percent alive at 12 months 55.00% 24 months 33.75% Median Survival pairwise comparisons RPA 5 12 vs RPA 3 or 4 32 Gross total 36 vs Subtotal+biopsy 8 p-value for pairwise comparisons RPA class Extent of resection Surgery and Chemotherapy Surgical resections Gross total 55% Subtotal 40% Biopsy 5% Adjunct chemo 45.00%

4 252 Lipani et al. target size relative to a desired prescription dose and on the target s proximity to critical structures. Patients who were on prophylactic corticosteroids prior to their treatment continued their medication as prescribed. Those who were not taking steroid medication were administered dexamethasone 4 mg by mouth immediately following each treatment session to relieve or prevent potential soft tissue irritation. Statistical Analysis Retrospective data was recorded on patient demographics, radiosurgical treatment parameters, extent of resection, adjuvant chemotherapy, and survival at follow-up. These data were logged in a database and a descriptive statistical analysis was generated (Microsoft Excel). Survival data and pairwise statistical comparisons were analyzed using Kaplan- Meier survival curves (Graphpad Prism). Results Figure 1: 39 year old male with right frontal GBM. tures such as the optic chiasm, optic nerves, or brainstem. The marginal or prescription dose was defined at the dose of 90% volume covered on the dose-volume histogram. The CyberKnife system s treatment planning process was previously described (27, 28). During treatment the patient was placed supine on the CyberKnife treatment couch wearing a custom aquaplast immobilization mask (WFR Corp, Wyckoff, NJ). Orthogonal X-ray images were acquired and registered to the DRRs previously obtained from the pretreatment CT. Skull tracking was performed throughout treatment delivery to ensure maintenance of optimal position. On average, more than 100 non-isocentric beams were used to irradiate a single target stereotactically, with relative homogeneity within the target volume (29). The treated tumor volumes ranged from 9.62 cm cm 3 (mean: cm 3 ). The marginal dose (D90) ranged from Gy Gy (mean: Gy) with a mean maximum dose of Gy (range: Gy Gy). The prescribed isodose line ranged from 50.38%-85.68% with a mean of 79.25%. Treatment was delivered in 1-8 fractions (mean: 5.65). The number of fractions was determined based on the Patients were followed up from 2-36 months (mean: months). Overall median survival was 16 months with 55% of patients alive at 12 months and 34% of patients alive at 24 months (Figure 2). Median survival of patients in RPA classes III or IV was 32 months versus 12 months for those in RPA class V (Figure 3). Median survival for patients who received gross total resection was 36 months versus 8 months for those who underwent subtotal resection or biopsy (Figure 4). Discussion Considering the unfavorable outcomes of standard multimodality treatment for GBM, SRS or hypofractionated radio- Percent survival Survival (Months) Figure 2: Overall survival following CyberKnife treatment.

5 Radiosurgery and Hypofractionated Radiotherapy as Primary Treatment for GBM 253 RPA Class Extent of Resection RPA 5 RPA 3/ gross total subtotal Survival (%) Survival (%) Time (Months) Time (Months) Figure 3: RPA class. Figure 4: Extent of Resection. therapy in place of a prolonged focal EBRT treatment regimen deserves to be explored. The median survival time of patients in this study (16 months) was comparable to that reported by Stupp et al. (6) using conventional radiation therapy with concomitant temozolomide (14.6 months) and without temozolomide (12.1 months). The 2-year survival (33.8%) of patients in this study was also favorably comparable to the study reported by Stupp et al. (6) (26.5%). The effect of concomitant ACNU and Vincristine on 45% of patients in this study is difficult to assess given the small sample size and the existence of other variables such as RPA class, and extent of surgical resection. However, the authors question whether SRS or hypofractionated radiotherapy can be offered as a suitable option to all GBM patients. Especially to those who are not candidates for extensive surgical resection or who cannot tolerate a six week course of radiation. Comparable survival matched with significantly shorter treatment time is especially important with regard to quality of life in these terminally ill patients. Since local recurrence of GBM inevitably occurs despite aggressive multimodal treatment regimens, numerous infiltrative tumor cells are clearly left untreated (30-34). This raises the question as to whether it is possible to improve outcome by optimizing targeting methods used to guide treatment planning. CT and MRI are often utilized in conjunction with one another and are the most common imaging techniques used for tumor delineation and targeting. However, the border of contrast enhancement on both CT and MRI is not the histologic border of the tumor. Similarly, contrast enhancement cannot distinguish tumor necrosis from edema. Targeting based on T2-weighted or FLAIR MRI may be of benefit as tumor cells are often found beyond enhancing borders (35); however, these methods still do not overcome the difficulty discerning tumor from edema or radiation necrosis. We, therefore, used MR Spectroscopy in the latter majority of our patients since we believe that it helps to distinguish tumor from necrosis or edema and more accurately guides treatment planning. In support of this, a previous study by Chan et al. found that the incorporation of metabolic imaging into Gamma Knife SRS treatment planning was beneficial to patient survival (36). The extent to which MR Spectroscopy targeting has influenced the outcome in our study has yet to be determined. There are several prognostic variables that are known to have an effect on survival in GBM patients which led to the development of the Recursive Partitioning Classification by Curran et al. (26). Consequently, survival times are often reported for patient groups in terms of these prognostic classes (37-40). When we analyzed patient survival according to RPA class the number of patients with RPA class 3 or 4 was too small to calculate meaningful median survival independently so they were grouped. We then compared patients with class 5 versus those with class 3 or 4 and found, as expected, there was a relation between a lower class and longer survival time. In a pairwise comparison, this association approached significance (Table IIB). Our results compare favorably to the RPA classification by Mirimanoff et al. (41), updated for use in the Temozolomide era. That study reported a median survival for class 5 of 10 months while we found ours to be 12 months. Our combined class 3 and 4 group was small but had a median survival of 32 months, much better than the 17 or 15 month median survival reported for classes 3 and 4, respectively, in the Temodar study. The extent of surgical resection also had its predicable effects on prolonging survival in our study, which had been previously demonstrated in GBM patients (42, 43). However, the extent to which these factors and others have influenced the results of our study is uncertain. Despite this, patients with GBM require a multimodality approach that should be based on rationally and individually designed treatment strategies. Conclusion In this retrospective analysis, survival rates using primary SRS or hypofractionated radiotherapy were comparable to historic controls using conventional fractionated external

6 254 Lipani et al. beam radiotherapy. Whether SRS or hypofractionated radiotherapy may be used as a suitable alternative to focal EBRT in newly diagnosed GBM patients has yet to be determined. A larger prospective analysis that compares CyberKnife SRS or hypofractionated radiotherapy to focal EBRT in post surgical GBM patients is warranted. References Bruner, J. M. Neuropathology of malignant gliomas. Semin Oncol 21, (1994). Deorah, S., Lynch, C. F., Sibenaller, Z. A., Ryken, T. C. Trends in brain cancer incidence and survival in the United States: Surveillance, Epidemiology, and End Results Program, 1973 to Neurosurg Focus 20, E1 (2006). Burger, P. C., Vogel, F. S., Green, S. B. Glioblastoma multiforme and anaplastic astrocytoma. Pathologic criteria and prognostic implications. Cancer 56, (1985). Coffey, R. J., Lunsford, L. D., Taylor, F. H. Survival after stereotactic biopsy of malignant gliomas. Neurosurgery 22, (1988). Nelson, D. F., Nelson, J. S., Davis, D. R., et al. Survival and prognosis of patients with astrocytomas with atypical or anaplastic features. J Neuro-Oncol 3, (1985). Stupp, R., Mason, W. P., van den Bent, M. J., et al. Radiotherapy plus Concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352, 10 (2005). Mirimanoff, R., Mason, W. P., van den Bent, M. J., Kortmann, R., et al. Is Long-Term Survival in Glioblastoma Possible? Updated Results of the EORTC/NCIC Phase III Randomized Trial on Radiotherapy (RT) and Concomitant and Adjuvant Temozolomide (TMZ) versus RT Alone, Int J Radiat Oncol Biol Phys 69, Issue 3, Supplement 1, Proceedings of the American Society for Therapeutic Radiology and Oncology 49th Annual Meeting, 49th Annual Meeting of the American Society for Therapeutic Radiology and Oncology 1, Page S2. November (2007). Keles, G. E., Anderson, B., Berger, M. S. The Effect of Extent of Resection on Time to Tumor Progression and Survival in Patients Glioblastoma Multiforme of the Cerebral Hemisphere. Surg Neurol 52, (1999). Wood, J. R., Green, S. B., Shapiro, W. R. The prognostic importance of tumor size in malignant gliomas: A computed tomographic scan study by the brain tumor cooperative group. J Clin Oncology 6, (1988). Byar, D. P., Green, S. B., Strike, T. A. Prognostic Factors for malignant glioma, In Oncology of the nervous system, pp Ed., Walk, M. D. Boston, MA, Marinus Nijhoff (1983). Silbergeld, D. L., Rostomily, R. C., Alvord, E. C., Jr. The cause of death in patients with glioblastoma is multifactorial: clinical factors and autopsy findings in 117 cases of supratentorial glioblastoma in adults. J Neurooncol 10, (1991). Gilbert, M. R., Loghin, M. The Treatment of Malignant Gliomas. Curr Treat Options Neurol 7, (2005). Walker, M. D., Alexander, E., Hunt W. E., et al. Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic gliomas. J Neurosurg 49, (1978). Walker, M. D., Strike, T. A., Shelino, G. E. An analysis of dose-effect relationship in the radiotherapy of malignant gliomas. Int J Radiat Oncol Biol Phys 5, (1979). Kristiansen, K., Hagen, S., Kollevoid, T., et al. Combined modality therapy of operated astrocytomas grade III and IV. Confirmation of the value of postoperative irradiation and lack of potentiation of bleomycin on survival time. A prospective multicenter trial of the Scandinavian Glioblastoma Study Group. Cancer 47, (1981) Ammirati, M., Galicich, J. H., Arbit, E., Liao, Y. Reoperation in the treatment of recurrent intracranial malignant gliomas. Neurosurgery 21, (1987). Hochberg, F. H., Pruitt, A. Assumptions in the radiotherapy of glioblastoma. Neurology 30, (1980). Masciopinto, J. E., Levin, A. B., Mehta, M. P., Rhode, B. S. Stereotactic radiosurgery for glioblastoma: a final report of 31 patients. J Neurosurg 82, (1995). Wallner, K. E., Galicich, J. H., Krol, G., Arbit, E., Malkin, M. G. Patterns of failure following treatment for glioblastoma multiforme and anaplastic astrocytoma. Int J Radiat Oncol Biol Phys 16, (1989). Shapiro, W. R., Green, S. B., Burger, P. C., et al. Randomized trial of three chemotherapy regimens and two radiotherapy regimens in postoperative treatment of malignant glioma: Brain Tumor Cooperative Group Trial J Neurosurg 71, 1-9 (1989). Souhami, L., Seiferheld, W., Brachman, D., Podgorsak, E. B., et al. Randomized comparison of stereotactic radiosurgery followed by conventional radiotherapy with carmustine to conventional radiotherapy with carmustine for patients with glioblastoma multiforme: report of Radiation Therapy Oncology Group protocol. Int J Radiat Oncol Biol Phys 60, (2004). Laperriere, N. J., Leung, P. M., McKenzie, S., Milosevic, M., et al. Randomized study of brachytherapy in the initial management of patients with malignant astrocytoma. Int J Radiat Oncol Biol Phys 41, (1998). Selker, R. G., Shapiro, W. R., Burger, P., Blackwood, M. S., Arena, V. C., et al. Brain Tumor Cooperative Group.The Brain Tumor Cooperative Group NIH Trial 87-01: a randomized comparison of surgery, external radiotherapy, and carmustine versus surgery, interstitial radiotherapy boost, external radiation therapy, and carmustine. Neurosurgery 51, ; discussion (2002). Fiveash, J. B., Spencer, S. A. Role of radiation therapy and radiosurgery in glioblastoma multiforme. Cancer J 9, (2003). Roberge, D., Souhami, L. Stereotactic radiosurgery in the management of intracranial gliomas. Technol Cancer Res Treat 2, (2003). Curran, W. J., Jr., Scott, C. B., Horton, J., Nelson, J. S., Weinstein, A. S., Fischbach, A. J., et al. Recursive partitioning analysis of prognostic factors in three Radiation Therapy Oncology Group malignant glioma trials. J Natl Cancer Inst 85, (1993). Collins, S. P., Coppa, N. D., Zhang, Y., Collins, B. T., McRae, D. A., Jean, W. C. CyberKnife(R) radiosurgery in the treatment of complex skull base tumors: analysis of treatment planning parameters. Radiat Oncol 1, 46 (2006). Adler, J. R., Jr., Murphy, M. J., Chang, S. D. et al. Image-guided robotic radiosurgery. Neurosurgery 44, ; discussion (1999). Yu, C., Jozsef, G., Apuzzo, M. L., Petrovich, Z. Dosimetric comparison of CyberKnife with other radiosurgical modalities for an ellipsoidal target. Neurosurgery 53, ; discussion (2003). Burger, P. C., Dubois, P. J., Schold, S. C., Smith, K. R., Odom, G. L., Crafts, D. C., et al. Computerized tomographic and pathologic studies of the untreated, quiescent, and recurrent glioblastoma multiforme. J Neurosurg 58, (1983). Burger, P. C., Heinz, E. R., Shibata, T., Kleihues, P. Topographic anatomy and CT correlations in the untreated glioblastoma multiforme. J Neurosurg 68, (1988). Halperin, E. C., Bentel, G., Heinz, E. R., Burger, P. C. Radiation therapy treatment planning in supratentorial glioblastoma multiforme: an analysis based on post mortem topographic anatomy with CT correlations. Int J Radiat Oncol Biol Phys 17, (1989). Kelly, P. J., Daumas-Duport, C., Scheithauer, B. W., Kall, B. A., Kispert, D. B. Stereotactic histologic correlations of computed tomography- and magnetic resonance imaging-defined abnormalities in patients with glial neoplasms. Mayo Clin Proc 62, (1987).

7 Radiosurgery and Hypofractionated Radiotherapy as Primary Treatment for GBM Lunsford, L. D., Martinez, A. J., Latchaw, R. E. Magnetic resonance imaging does not define tumor boundaries. Acta Radiol Suppl 369, (1986). Miwa, K., Shinoda, J., Yano, H., Okumura, A., Iwama, T., Nakashima, T., et al. Discrepancy between lesion distributions on methionine PET and MR images in patients with glioblastoma multiforme: insight from a PET and MR fusion image study. J Neurol Neurosurg Psychiatry 75, (2004) Chan, A. A., Lau, A., Pirzkall, A., Chang, S. M., Verhey, L. J., Larson, D., et al. Proton magnetic resonance spectroscopy imaging in the evaluation of patients undergoing gamma knife surgery for Grade IV glioma. J Neurosurg 101, (2004) Hsieh, P. C., Chandler, J. P., Bhangoo, S., Panagiotopoulos, K., Kalapurakal, J. A., Marymont, M. H., et al. Adjuvant gamma knife stereotactic radiosurgery at the time of tumor progression potentially improves survival for patients with glioblastoma multiforme. Neurosurgery 57, ; discussion (2005). Kondziolka, D., Flickinger, J. C., Bissonette, D. J., Bozik, M., Lunsford, L. D. Survival benefit of stereotactic radiosurgery for patients with malignant glial neoplasms. Neurosurgery 41, ; discussion (1997) Mahajan, A., McCutcheon, I. E., Suki, D., Chang, E. L., Hassenbusch, S. J., Weinberg, J. S., et al. Case-control study of stereotactic radiosurgery for recurrent glioblastoma multiforme. J Neurosurg 103, (2005) Sarkaria, J. N., Mehta, M. P., Loeffler, J. S., Buatti, J. M., Chappell, R. J., Levin, A. B., et al. Radiosurgery in the initial management of malignant gliomas: survival comparison with the RTOG recursive partitioning analysis. Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 32, (1995) Mirimanoff, R. O., Gorlia, T., Martin, W., Van den Bent. M. J., et al. Radiotherapy and Temozolomide for Newly Diagnosed Glioblastoma: Recursive Partitioning Analysis of the EORTC 26981/22981 NCIC CE3 Phase III Randomized Trial. J Clin Onc 24, (2006). Ciric, I., Vick, N. A., Mikhael, M. A., Cozzens, J., Eller, T., Walsh, A. Aggressive surgery for malignant supratentorial gliomas. Clin Neurosurg 36, (1990) Hentschel, S. J., Sawaya, R. Optimizing outcomes with maximal surgical resection of malignant gliomas. Cancer Control 10, (2003). Received: January 17, 2008; Revised: March 24, 2008; Accepted: April 4, 2008