The Role of Postoperative Radiation Therapy in the Treatment of Meningeal Hemangiopericytoma—Experience From the SEER Database

The Role of Postoperative Radiation Therapy in the Treatment of Meningeal Hemangiopericytoma—Experience From the SEER Database

International Journal of Radiation Oncology biology physics www.redjournal.org Clinical Investigation: Sarcoma The Role of Postoperative Radiatio...

342KB Sizes 69 Downloads 71 Views

International Journal of

Radiation Oncology biology

physics

www.redjournal.org

Clinical Investigation: Sarcoma

The Role of Postoperative Radiation Therapy in the Treatment of Meningeal HemangiopericytomadExperience From the SEER Database Alexander M. Stessin, MD, PhD,* Cristina Sison, PhD,y Jaime Nieto, MD,z Muri Raifu, MD,z and Baoqing Li, MD, PhD* *Department of Radiation Oncology, Weill Medical College of Cornell University, New York, New York; yDepartment of Biostatistics, the Feinstein Institute for Medical Research, Manhasset, New York; and zDepartment of Neurosurgery, New York Hospital Queens, Flushing, New York Received Apr 2, 2012, and in revised form May 25, 2012. Accepted for publication May 30, 2012

Summary Because of its rare incidence and relatively recent reclassification as a unique disease, there is a paucity of large prospective studies characterizing the role of postoperative radiation therapy for meningeal hemangiopericytoma. Our analysis of 76 cases from the Surveillance, Epidemiology, and End Results database showed a significant cause-specific survival benefit to postoperative radiation therapy. We recommend that postoperative radiation be considered after subtotal resection for patients who can tolerate it.

Purpose: The aim of this study was to examine the effect of postoperative radiation therapy (RT) on cause-specific survival in patients with meningeal hemangiopericytomas. Methods and Materials: The Surveillance, Epidemiology, and End Results database from 19902008 was queried for cases of surgically resected central nervous system hemangiopericytoma. Patient demographics, tumor location, and extent of resection were included in the analysis as covariates. The Kaplan-Meier product-limit method was used to analyze cause-specific survival. A Cox proportional hazards regression analysis was conducted to determine which factors were associated with cause-specific survival. Results: The mean follow-up time is 7.9 years (95 months). There were 76 patients included in the analysis, of these, 38 (50%) underwent gross total resection (GTR), whereas the other half underwent subtotal resection (STR). Postoperative RT was administered to 42% (16/38) of the patients in the GTR group and 50% (19/38) in the STR group. The 1-year, 10-year, and 20-year cause-specific survival rates were 99%, 75%, and 43%, respectively. On multivariate analysis, postoperative RT was associated with significantly better survival (HR Z 0.269, 95% CI 0.084-0.862; PZ.027), in particular for patients who underwent STR (HR Z 0.088, 95% CI: 0.015-0.528; P<.008). Conclusions: In the absence of large prospective trials, the current clinical decision-making of hemangiopericytoma is mostly based on retrospective data. We recommend that postoperative RT be considered after subtotal resection for patients who could tolerate it. Based on the current literature, the practical approach is to deliver limited field RT to doses of 50-60 Gy while respecting the normal tissue tolerance. Further investigations are clearly needed to determine the optimal therapeutic strategy. Ó 2013 Elsevier Inc.

Reprint requests to: Baoqing Li, MD, PhD, Weil Medical College of Cornell University, Stich Radiation Oncology, 525 East 68th St, New York, NY 10065. Tel: (212) 746-3600; E-mail: [email protected] Int J Radiation Oncol Biol Phys, Vol. 85, No. 3, pp. 784e790, 2013 0360-3016/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ijrobp.2012.05.042

Conflict of interest: none.

Volume 85  Number 3  2013

Introduction Previously considered to be a variant of meningioma, hemangiopericytoma (HPC) has been recognized as a distinct clinicopathological entity in the World Health Organization classification since 1993 (1). Arising from Zimmerman pericytes within the capillary wall architecture, HPC accounts for 2%-3% of primary meningeal tumors and is characterized by a unique set of genetic and histopathological features as well as by a propensity to recur and metastasize extraneurally. These tumors typically follow an aggressive clinical course with reported recurrence rates of up to 90% within 7 months of initial therapy. Among patients who survive >10 years, long-term follow-up reveals a persistent risk of developing distant metastases as well as multiple local recurrences after the initial course of treatment (2-5). Gross total resection (GTR) has been a mainstay of treatment but can be rather difficult to achieve in the setting of neural involvement, dural sinus extension, hypervascularity, and anatomical inaccessibility of the tumor. Postoperative radiation therapy (RT), though often recommended and supported by multiple single-institute, small retrospective studies (4-12), remains controversial with a recent meta-analysis of published literature putting into question the effectiveness of RT in improving survival outcomes (13). The rarity of hemangiopericytomas has precluded the possibility of conducting large-scale prospective studies, and clinical decision-making has been based predominantly on single institution retrospective series with small patient cohorts. Therefore, to investigate the role of postoperative radiation in a large patient population, this study used data from the Surveillance, Epidemiology, and End Results (SEER) Program, which collects and publishes cancer incidence and survival data from 20 populationbased cancer registries, covering more than 25% of the population in the United States. We hypothesized that postoperative radiation would be efficacious in improving cause-specific survival in hemangiopericytoma patients.

SEER analysis of postoperative RT for HCP

785

“censored” for survival. Cause-specific survival was analyzed separately according to each of the patient characteristics (gender, age group, race) and according to treatment modality (surgery type, radiation). The effects of radiation on survival outcomes were assessed separately for patients who underwent GTR and those who underwent subtotal resection (STR). A Cox proportional hazards regression analysis was then carried out with surgery type and radiation, and their interaction included in the model. The median follow-up time was calculated using the reverse Kaplan-Meier method. The threshold for significance used in the study was .05. All analyses were performed using SAS 9.2 software (SAS Institute Inc., Cary, NC, USA).

Results Patient and tumor characteristics A query of the SEER database for patients with meningeal hemangiopericytoma yielded 134 cases. After eliminating all cases with missing data (21 cases), unresectable disease (6 cases), multiple primaries (12 cases), diagnosis before 1990 (15 cases), or patients who died within 6 months after surgery (5 cases), there were a total of 76 patients included in the study. Of these, 38 (50%) underwent GTR and the other half underwent STR. Postoperative RT was administered to 42% (16/38) of the patients in the GTR group and 50% (19/38) in the STR group. Demographic characteristics across different treatment groups are reported in Table 1. Demographic characteristics were not significantly different between the 4 treatment groups. In the overall sample, the female-to-male ratio was 1.37:1. Patient age at the time of diagnosis ranged from 23-80 years, with median Z 50 years of age. In 65.8% (50/76) of the examined cases, the tumor was located in the supratentorial region.

Survival analysis

Methods and Materials The SEER database (1990-2008) was queried to identify patients with surgically resected hemangiopericytoma. We chose 1990 as the earliest year of diagnosis to be included in the study because querying the database for “hemangiopericytoma” or “angioblastic meningioma” in the earlier years yielded only several cases as the rest were likely categorized together with other meningiomas (eg, as “meningioma, not otherwise specified”). Patient demographics (race/ethnicity, gender, age at presentation, year of diagnosis), tumor location (supratentorial region, spinal area, posterior fossa, ventricle, and extension outside the central nervous system), and treatment modalities (type of surgery, postoperative radiation) were included as covariates in multivariate analysis. Patients with unresectable disease, multiple primary malignancies, patients who died within 6 months after surgery, or incomplete case information were excluded from analysis of the data. The Kaplan-Meier product-limit method was used to estimate survival, and the log-rank test was used to compare survival between groups. Ninety-five percent confidence intervals for survival rates were calculated using Greenwood’s formula for computing the standard error. Subjects who were alive or dead from other causes as of the last follow-up were considered

The mean follow-up time was 7.9 years (95 months). Seventeen patients died from the disease during the study period. The 1-, 5-, 10-, and 20-year survival rates were 99%, 91%, 75%, and 43%, respectively. On univariate analysis, there were no significant cause-specific survival differences associated with any of the examined patient characteristics (gender, race, age) or tumor location. When the types of surgery were studied, there was also no significant difference in survival between patients who underwent GTR (þ/RT) and those who underwent STR (þ/- RT), with PZ.48 (Fig. 1). However, an exploratory pairwise-analysis suggested that patients who underwent GTR alone had better survival than those underwent STR alone (PZ.03; Fig 3). Regarding the role of postoperative radiation for all patients (GTR and STR), there was a trend toward improved survival outcomes with the addition of postoperative RT, with PZ.051 (Fig. 2). In Fig. 3, further subgroup analysis showed that the addition of postoperative RT imparted a significant survival benefit to patients who underwent STR (PZ.017) but not to those who underwent GTR (PZ.661). In addition, patients who underwent STR with postoperative RT had similar survival as those underwent GTR alone (PZ.31). We performed multivariate analysis using the Cox proportional hazard model for the extent of surgery, postoperative

786

International Journal of Radiation Oncology  Biology  Physics

Stessin et al.

Table 1

Patient and tumor characteristics by treatment received

n Mean age, y (SD) Gender Male Female Race White Black Other or unknown Tumor location Supratentorial Ventricle Posterior fossa Spinal Extension outside CNS

GTR Alone

GTR þ RT

STR þ RT

STR Alone

22 55.8  (16.3)

16 47.0  (14.8)

19 46.8  (12.6)

19 53.9  (18.1)

7 (31.8%) 15 (68.2%)

7 (43.8%) 9 (56.2%)

10 (52.6%) 9 (47.4%)

8 (42.1%) 11 (57.9%)

20 (90.9%) 0 (0%) 2 (9.1%)

14 (87.5%) 1 (6.25%) 1 (6.25%)

15 (79.0%) 2 (10.5%) 2 (10.5%)

16 (88.9%) 0 (0%) 2 (11.1%)

13 2 2 1 1

16 0 0 2 1

P .184 .604

.767

.611 12 4 3 1 2

(54.6%) (18.2%) (13.6%) (4.6%) (9.1%)

9 3 2 1 1

(56.2%) (18.8%) (12.5%) (6.25%) (6.25%)

(64.4%) (10.5%) (10.5%) (5.3%) (5.3%)

(84.2%) (0%) (0%) (10.5%) (5.3%)

Abbreviations: CNS Z central nervous system; GTR Z gross total resection; RT Z radiation therapy; SD Z standard deviation; STR Z subtotal resection. Data are presented as mean  (SD) for continuous variables, and as n (%) for categorical variables.

radiation, and their interaction. No other variables (demographic or clinical) were included in the model because none of them were significantly associated with survival in the univariate screening. Results of the Cox regression are reported in Table 2. Postoperative RT was associated with significantly higher survival (HR Z 0.269, 95% CI 0.084-0.862; PZ.027). There was a moderate but not statistically significant interaction effect between surgery type and radiation (PZ.068), indicating that the effect of radiation varies depending on the type of surgery. For patients who underwent STR, postoperative RT was associated with significantly better survival (HR Z 0.088, 95% CI 0.0150.528; P<.008).

Discussion Because of its rare incidence and relatively recent reclassification as a unique disease, there is a paucity of large prospective studies

characterizing the role of postoperative RT and surgical treatment for hemangiopericytoma (Table 3). To our knowledge, this is the first population-based clinical outcome study of patients with HPC. The present study demonstrated that postoperative RT is associated with a statistically significant survival benefit, in particular for patients who underwent STR, corroborating the findings from multiple small single-institute retrospective studies (Table 3). Although the specific underlying biological mechanism of survival benefit associated with postoperative radiation in HPC has not been clearly established, it was speculated that solitary cytoreductive surgery alone was inferior to surgery with adjuvant RT because STR alone did not alter tumor biology per se, but merely reduced tumor burden. Even in the case of GTR, microscopic pockets of malignant tumor cells likely remained present, inevitably leading to recurrence (10). However, a recent meta-analysis of a total of 194 patients with treatment information from 97 published articles put into question the

Fig. 1. Kaplan-Meier survival curve showing the comparing of cause-specific survival between gross total resection (GTR) and subtotal resection (STR).

Volume 85  Number 3  2013

SEER analysis of postoperative RT for HCP

787

Fig. 2. Kaplan-Meier survival curve showing the comparing of cause-specific survival between postoperative radiation therapy (RAD) and no postoperative radiation therapy (NO RAD). effectiveness of RT in improving survival outcomes (13). It must be noted that the meta-analysis included reports dating as far back as 1954. The inclusion of data across such a protracted period presents 2 potential problems. First, radiation techniques have evolved dramatically over the past 2 decades, and the results of RT given today are likely very different from those before the advent of 3-dimensional conformal and image guided RT. Second, it is difficult to control for the quality of reported data and to ascertain consistency in pathologic categorization, especially considering the relatively recent (World Health Organization, 1993) recognition of hemangiopericytomas as a distinct clinicopathological entity. In fact, the same group of investigators has found postoperative radiation to show promise in hindering tumor progression in 2 recent institutional reviews of patients treated at University of California, San Francisco (UCSF) (3, 10). The role of surgery is more established than that of postoperative radiation. With reported operative mortality between 0% and 15%, attempted complete resection has remained the standard primary

treatment for HPC. Our pairwise comparison result showed that patients who underwent GTR alone had better cause-specific survival than those underwent STR alone. This is in line with the widely accepted finding that GTR is the best approach for maximizing recurrence-free survival (2-5, 10, 11, 13). However, GTR in many series was not a common surgical outcome, ranging from about 40% in the UCSF series and the London series (5, 10), to 50% in the present SEER database, and to 55% in the Mayo Clinic series and the MD Anderson series (4, 11). In the setting of STR, especially when GTR was considered impossible or highly risky because of dural sinus invasion, neural involvement, dense tumor vascularity, or anatomic inaccessibility, the addition of radiation therapy to STR appears to be a reasonable alternative to GTR alone. This hypothesis was supported by our finding of equivalent survival outcomes between those who underwent STR with postoperative RT and those underwent GTR alone. Similar results were reported in the UCSF series, which showed that STR followed by adjuvant radiation had a tendency toward an increased recurrence-free interval as compared with GTR alone (10).

Fig. 3. Kaplan-Meier survival curve showing the comparing of cause-specific survival between gross total resection (GTR) þ no postoperative radiation therapy (RT), GTR þ postoperative RT, subtotal resection (STR) þ postoperative RT, and STR þ no postoperative RT.

788

International Journal of Radiation Oncology  Biology  Physics

Stessin et al.

Table 2 Multivariate Cox proportional hazard analysis for hazard or cause-specific survival HR Surgery STR GTR Radiation No RT RT

95% CI

P

0.431

Referent 0.141-1.317

.139

0.269

Referent 0.084-0.862

.027

Abbreviations: CI Z confidence interval; GTR Z gross total resection; HR Z hazard ratio; RT Z radiation therapy; STR Z subtotal resection.

Regarding details of surgery and the potential association with clinical outcome, our SEER-based study was limited by missing information. This is a drawback inherent to the SEER database that broadly categorizes surgical resection as gross total or subtotal but does not provide more detailed information, such as Simpson grade. The amount of STR is not specified in the SEER coding system. This is potentially confounding, because a 50% resection is very different from a 90% resection; yet, both cases would be categorized as STR in SEER database. Furthermore, all GTR reported in SEER may not be equal. This was suggested by a recent literature review that emphasized the variability in what has been reported as GTR of meningeal tumors (14). The second limitation of the SEER database is lack of information on details of radiation treatment, including radiation intent, dose, target volume, and technique. It has been shown that salvage RT with stereotactic radiosurgery, including gamma knife, played an important role for small recurrent HPC (15-17). Given that the SEER registries only record information on firstcourse therapy, one could reasonably assume that the survival benefit associated with postoperative radiation in the present study was observed in the setting of initial management rather than salvage therapy for recurrence. Furthermore, our study could not address the question of dose-dependent effects from a lack of radiation dose information. Although HPC of the central nervous system has been considered to respond relatively well to radiation, previous investigations have presented conflicting findings regarding the dose-response relationship for HPC (4, 6, 12, 13). In the meta-analysis discussed above, Rutkowski et al reported increased mortality rates in patients receiving >50 Gy when compared with those receiving 50 Gy. It was suggested that the observation was due to either long-term radiation toxicities or selection bias where patients with poor prognostic factors such as STR, larger size, or higher grade were treated with higher radiation doses (13). In contrast to the metaanalysis findings, most HPC studies have reported decreased recurrent rates associated with radiation dose of >50 Gy. Guthrie et al proposed the existence of a radiation dose-response relationship based on their finding that 88% (7/8) of patients who received doses <45 Gy developed local recurrence, compared with 22% (2/9) of those receiving >45 Gy. None of the 3 patients who received >51 Gy developed recurrences (4). Similar results were shown in a recent French study in which 2 of 2 patients who had received RT to 30 Gy subsequently developed local recurrences while in the high-dose group (>50 Gy) local recurrence was seen in only 1 of 8 patients (6). The radiation dose-response

relationship was also confirmed in an extensive analysis of published series and case reports by Mehta et al (12). Although the threshold response dose is likely about 50 Gy, the maximum dose and correlative toxicities have not been fully delineated. In their retrospective studies, Guthrie et al used doses of up to 56 Gy, Dufour et al used 64 Gy for a completely resected intraventricular HCP, and Combs et al delivered a maximum dose of 65 Gy using fractionated stereotactic RT (18). No radiation toxicities were reported in any of the three studies. Furthermore, our SEER study could not address the questions of radiation target volume and patterns of failure. In a modern series of HCP patients treated with fractionated stereotactic RT or intensity modulated radiation therapy, Combs et al described some practical details: stereotactic image fusion was used for target delineation and planning target volume was the area of contrast enhancement adding a safety margin of 5-20 mm, depending on tumor location and grade (18). In other studies, techniques have been described in less detail but a common practice has been to use limited-field radiation covering the tumor bed with a 1 cm margin (6) or otherwise unspecified margin (4, 11). However, the optimal margin still remains unclear, largely because of the lack of published information on in-field vs out-offield recurrences. In addition, the SEER database does not provide information on such important variables as the use of adjuvant chemotherapy, tumor size, or histopathological grade. Although the role of adjuvant chemotherapy for HPC is still considered controversial, it has been shown that larger tumor size (>6 cm) and higher grade were associated with increased rates of local recurrence and a higher likelihood of distant metastases (5, 7, 10). Because of the lack of information, however, it remains unclear whether or not postoperative radiation improves cause-specific survival in patients with or without the above factors although one would expect that many of these issues would be balanced in the present study. Last, the lack of prognostic significance associated with tumor location in the present study is in contrast to the findings from prior studies that showed that tumors of posterior fossa or non-skull base location were associated with increased recurrence rates and higher grade (3, 10, 19). Our study may have been underpowered to detect outcome differences according to anatomic location given that the majority of hemangiopericytomas in our data set were localized to the supratentorial region. Future studies, evaluating the role of RT, must take into consideration the recent strides made in the field of radiation oncology. Although a detailed review of the literature on radiosurgery and stereotactic fractionated RT is beyond the scope of this discussion, a growing body of evidence supports the use of this modality for the treatment of hemangiopericytomas (15-18). Furthermore, investigation of anti-vascular endothelial growth factor receptor agent bevacizumab or vascular endothelial growth factor receptor tyrosine kinase inhibitor sorafenib as part of the adjuvant regimen may be warranted in light of a few promising results from patients with hemangiopericytoma/solitary fibrous tumor treated with these agents (20). In conclusion, through this first population-based clinical study of hemangiopericytoma, we showed a significantly better survival associated with the addition of postoperative RT after STR. In the absence of large prospective trials, the current clinical decision-making of hemangiopericytoma is mostly based

Volume 85  Number 3  2013 Table 3

SEER analysis of postoperative RT for HCP

789

Review of selected studies of patients with meningeal hemangiopericytoma Type of study

Period

No. of Mean Survival cases follow-up, y at 5 y, %

Present study

SEER analysis, retrospective

1990-2008

76

7.9

Bastin and Mehta (12)

Literature review 1970-1991

80

NA

1954-2009

194

NA

Rutkowski (10), Single institution, 1989-2010 UCSF retrospective

35

NA

Schiariti (5), NHNN, London Combs (18), Heidelberg

Single institution, 1984-2008 retrospective

31

10.3

Single institution, 1984-2004 retrospective

25

3

Dufour (6), Marseille

Single institution, 1965-1999 retrospective

21

5

Kim (9), Seoul

Single institution, 1982-1999 retrospective

31

6.5

Soyour (11), MDACC

Single institution, 1979-1999 retrospective

25

9.3

Gutherie (4), Mayo Clinic

Single institution 1938-1987 retrospective

44

NA

Rutkowski (13) Literature meta-analysis

Recommendation on extent of surgery and use of postoperative RT

91 (CSS) GTR had better CSS than STR alone. RT was recommended for better CSS in STR and had a trend for better CSS in GTR. NA RT had better disease-free survival. A doseresponse relationship was noted. Recurrence risk was 14% for dose >50 Gy, whereas it was 78% for dose <45 Gy. 82 (OS) GTR had the greatest survival benefit, regardless of RT. RT was not recommended for survival benefit in either STR or GTR. RT dose >50 Gy had worse survival. 92 (OS) GTR had survival benefit. RT was recommended for better recurrence-free survival, but not overall survival benefit in STR and GTR. 93 (OS) GTR followed by RT provided the best survival. RT was recommended for better recurrence rate and better survival in STR and GTR. 100 (OS) High-precision RT was considered an effective and safe modality. FSRT/IMRT and stereotactic image fusion were used in all cases. PTV included the area of contrast enhancement adding a safety margin of 5-20 mm, depending on tumor location and grade. The median dose was 57.6 Gy (range 40-65 Gy), with 1.8-2.0 Gy per fraction. NA RT (dose >50 Gy) was recommended for better recurrence rate. Dose between 50 and 64 Gy was associated with 12% recurrence. RT field was limited to tumor bed with 1-cm margin. 96 (OS) GTR had better recurrence rate. RT (dose >50 Gy) was recommended in the setting of GTR for better recurrence-free survival, in spite of insufficient statistics power. 85 (OS) GTR had better recurrence rate, but not survival benefit. RT was recommended in spite of the lack of local control benefit secondary to insufficient statistics power. Cranial limited-field RT was used, with median dose of 54 Gy and maximum dose 61.2 Gy. 67 (OS) GTR had better recurrence rate and better survival. RT was recommended for better recurrence rate and better survival in STR and GTR. A doseresponse relationship was proposed. No recurrence occurred in patients treated with 51-56 Gy, whereas it was 88% recurrence in <45 Gy. RT field was limited to tumors with an unspecified margin.

Abbreviations: CSS: cause-specific survival; FSRT: fractionated stereotactic radiation therapy; GTR: gross total resection; IMRT: intensity modulated radiation therapy; MDACC: M D Anderson Cancer Center; NA: not available; NHNN: National Hospital for Neurology and Neurosurgery; OS: overall survival; PTV: planning target volume; STR: subtotal resection; UCSF: the University of California at San Francisco.

on retrospective data and we recommend that postoperative RT be considered in STR patients who could tolerate. Limited-field RT in the range of 50-60 Gy while respecting the normal tissue tolerance appears to represent a practical approach. Further investigations are clearly needed to determine the optimal therapeutic strategy.

References 1. Bailey P, Cushing H, Eisenhardt L. Angioblastic meningioma. Arch Pathol Lab Med 1928;6:453-490. 2. Galanis E, Buckner JC, Scheithauer BW, et al. Management of recurrent meningeal hemangiopericytoma. Cancer 1998;82(10):1915-1920.

790

Stessin et al.

3. Rutkowski MJ, Bloch O, Jian BJ, et al. Management of recurrent intracranial hemangiopericytoma. J Clin Neurosci 2011;18(11):1500-1504. 4. Guthrie BL, Ebersold MJ, Scheithauer BW, et al. Meningeal hemangiopericytoma: histopathological features, treatment, and long-term follow-up of 44 cases. Neurosurgery 1989;25(4):514-522. 5. Schiariti M, Goetz P, El-Maghraby H, et al. Hemangiopericytoma: long-term outcome revisited. J Neurosurg 2011;114(3):747-755. 6. Dufour H, Me´tellus P, Fuentes S, et al. Meningeal hemangiopericytoma: a retrospective study of 21 patients with special review of postoperative external radiotherapy. Neurosurgery 2001;48(4):756-762. 7. Ecker RD, Marsh WR, Pollock BE, et al. Hemangiopericytoma in the central nervous system: treatment, pathological features, and longterm follow up in 38 patients. J Neurosurg 2003;98(6):1182-1187. 8. Jha N, McNeese M, Barkley HT Jr., et al. Does radiotherapy have a role in hemangiopericytoma management? Report of 14 new cases and a review of the literature. Int J Radiat Oncol Biol Phys 1987; 13(9):1399-1402. 9. Kim JH, Jung HW, Kim YS, et al. Meningeal hemangiopericytomas: long-term outcome and biological behavior. Surg Neurol 2003;59(1): 47-53. 10. Rutkowski MJ, Jian BJ, Bloch O, et al. Intracranial hemangiopericytoma: Clinical experience and treatment considerations in a modern series of 40 adult patients. Cancer 2011. http://dx.doi.org/10.1002/cncr.26411. 11. Soyuer S, Chang EL, Selek U, et al. Intracranial meningeal hemangiopericytoma: the role of radiotherapy: report of 29 cases and review of the literature. Cancer 2004;100(7):1491-1497.

International Journal of Radiation Oncology  Biology  Physics 12. Bastin KT, Mehta MP. Meningeal hemangiopericytoma: defining the role for radiation therapy. J Neurooncol 1992;14(3):277-287. 13. Rutkowski MJ, Sughrue ME, Kane AJ, et al. Predictors of mortality following treatment of intracranial hemangiopericytoma. J Neurosurg 2010;113(2):333-339. 14. Wiemels J, Wrensch M, Claus EB. Epidemiology and etiology of meningioma. J Neurooncol 2010;99(3):307-314. 15. Sheehan J, Kondziolka D, Flickinger J, et al. Radiosurgery for treatment of recurrent intracranial hemangiopericytomas. Neurosurgery 2002;51(4):905-910. 16. Kano H, Niranjan A, Kondziolka D, et al. Postoperative stereotactic radiosurgery after resection of intracranial hemangiopericytomas. Int J Radiat Oncol Biol Phys 2008;72(5):1333-1339. 17. Veeravagu A, Jiang B, Patil CG, et al. CyberKnife stereotactic radiosurgery for recurrent, metastatic, and residual hemangiopericytomas. J Hematol Oncol 2011;4:26. 18. Combs SE, Thilmann C, Debus J, et al. Precision radiotherapy for hemangiopericytomas of the central nervous system. Cancer 2005; 104(11):2457-2465. 19. Kane AJ, Sughrue ME, Rutkowski MJ, et al. Anatomic location is a risk factor for atypical and malignant meningiomas. Cancer 2011; 117(6):1272-1278. 20. Parks MS, Ravi MS, Araujo DM. Inhibiting the VEGF-VEGFR pathway in angiosarcoma, epithelioid hemangioendothelioma, and hemangiopericytoma/solitary fibrous tumor. Curr Opin Oncol 2010; 22(4):351-355.