Shots = number of isocenters.

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 75

The histologic types and grading were as follows: 10 pineocytomas (Grade I), 8 choroid plexus papillomas (CPP) (Grade I), 4 central neurocytomas (CN) (Grade II), 3 pleomorphic xanthoastrocytomas (PXA) (Grade II), 2 gangliogliomas (Grade I), 1 subependymoma (Grade I), 1 papillary glioneuronal tumor (Grade I), and 1 mixed oligoastrocytoma (Grade II). All tumors were histologically verified: 20/30 (67%) were surgical residuals or postoperative recurrences, while biopsy confirmed the diagnosis in the remaining 10/30 cases: 4 (13%) during neuroendoscopy approach and 6 (20%) with stereotactic brain intervention. There were 16 males and 14 females; the mean age of the clinical series was 34.7 years (7-70 years). Further pre-GKRS treatments were as follows: RT alone, 2/30 patients (7%), and RT followed by chemotherapy, 1/30 (3%). The TV was always less than 15 cc. In 24/30 patients, pre-GKRS neurologic deficits were documented. The mean and range values of the

> Dose Plan Mean

Overall series 16.5 (12-22.4) 50.2 (40-70) 104 (11.4-257.4) 9.6 (1-25) Pineocytoma 16.7 (13-20) 50.5 (50-55) 110.1 (18.9-257.4) 13.9 (5-25) CPP 15.1 (12-20.2) 48.7 (45-50) 72.9 (11.4.189.7) 7.1 (3-11) CN 16.4 (14-18) 47.5 (40-50) 151.6 (84.2-185.4) 7.0 (5-10) PXA 17.0 (13-20) 50.0 (50-50) 152.1 (69.6-248.5) 10.3 (2-25)

PD = prescription dose; Gy = gray; PI = prescription isodose; ID = integral dose; mJ = milli Joule;

University Hospital of Verona from February 1993 to February 2009.

Table 2. Summary of radiosurgical parameters of 30 unusual LGPNTs treated with GKRS at

The mean prescription dose (PD) was 16.5 Gy while the mean prescription isodose (PI) corresponded to 50.2%. The average value of integral dose (ID), parameter which describes the relationship between treated TV and delivered dose appropriately, was 104 mJ. The mean number of shots needed for dose planning was 9.6. Follow-up data, including causes of death, were obtained from hospital notes, imaging studies, and contact with relatives and family physicians. Medical records and MRI scans for all patients were carefully reviewed by neurosurgeons and neuroradiologists. This retrospective study was approved by the Ethical Commetee of our University Hospital. All patients identifiers were removed before

In the group of pineocytoma, 7 out of 10 patients were young adults, under 40-year-old. Diagnosis was confirmed by bioptic samples harvest during a neuroendoscopic procedure

Parameters (Range)

PD Gy PI % ID mJ # Shots

18.2 (12.5-22.4) 54.0 (50-70) 74.4 (19.8-191.09 6.8 (1-16)

radiosurgical dose planning parameters are summarized in Table 2.

use of GKRS in unusual LGPNTs, also, is that the highly conformal treatment planning possible with this technique allows delivery of citotoxic dose of radiation to the target field while minimazing the dose delivered to neighboring vital structures. Hence, radiosurgery may be a less morbid alternative for treating deep-seated or functional-located lesions surrounded by eloquent structures. It therefore becomes possible to avoid toxic systemic side effects due to less effective antiblastic drugs, to preserve neurological function, to prevent further tumor growth and/or reinterventions in residual or recurrent tumors with remarkably low risks of acute radiation injury or delayed radiation therapy (RT)-related sequelae. In this chapter, the results of a retrospective study on 30 unusual LGPNTs of the brain treated with GKRS and followed up for at least 2 years will be presented and discussed. The end-points of this study will regard the OS, overall local TGC, actuarial Survival and local PFS (on the whole series and evaluated on different histologies) rates, and GKRS-related permanent side effects. Furthermore, a statistical analysis concerning the identification of potential prognostic factors influencing post-GKRS OS and local PFS among some selected independent variables will be performed, as well.

## **2. Materials and methods**

## **2.1 Overall series**

Between February 1993 and February 2009, GKRS was performed on 30 patients (Table 1) with unusual LGPNTs of the brain (Grade I-II according to the WHO classification) (Louis et al., 2007). Inclusion criteria for GKRS included: lesions deep-seated in the brain or located in eloquent regions, histology confirmation, residual or recurrent tumors, tumor volume (TV) less than 15 cc, conditions of high operative risk, and patient refusal of microsurgery or reoperation.


N° Pt = number of treated patients; yr = year; TV = tumor volume; cc = cubic cm; Pre-GK Treats. = pregamma knife treatments; M = male; F = female; RT = radiotherapy; Ch = chemotherapy; Papillary Gl. Tum. = papillary glioneuronal tumor.

Table 1. Summary of clinical characteristics of 30 unusual LGPNTs treated with GKRS at University Hospital of Verona from February 1993 to February 2009.

use of GKRS in unusual LGPNTs, also, is that the highly conformal treatment planning possible with this technique allows delivery of citotoxic dose of radiation to the target field while minimazing the dose delivered to neighboring vital structures. Hence, radiosurgery may be a less morbid alternative for treating deep-seated or functional-located lesions surrounded by eloquent structures. It therefore becomes possible to avoid toxic systemic side effects due to less effective antiblastic drugs, to preserve neurological function, to prevent further tumor growth and/or reinterventions in residual or recurrent tumors with remarkably low risks of acute radiation injury or delayed radiation therapy (RT)-related sequelae. In this chapter, the results of a retrospective study on 30 unusual LGPNTs of the brain treated with GKRS and followed up for at least 2 years will be presented and discussed. The end-points of this study will regard the OS, overall local TGC, actuarial Survival and local PFS (on the whole series and evaluated on different histologies) rates, and GKRS-related permanent side effects. Furthermore, a statistical analysis concerning the identification of potential prognostic factors influencing post-GKRS OS and local PFS among

Between February 1993 and February 2009, GKRS was performed on 30 patients (Table 1) with unusual LGPNTs of the brain (Grade I-II according to the WHO classification) (Louis et al., 2007). Inclusion criteria for GKRS included: lesions deep-seated in the brain or located in eloquent regions, histology confirmation, residual or recurrent tumors, tumor volume (TV) less than 15 cc, conditions of high operative risk, and patient refusal of microsurgery or

Histology N° Pt Age yr Surgery TV cc Pre-GK

Overall series 16/14 38.4 (7-70) 20/10 4.9 (0.7-14.5) 2 RT

Pineocytoma 5/5 31.5 (7-58) 6/4 5.0 (0.8-11.7) 1 RT CPP 6/2 44.2 (22-70) 5/3 3.7 (0.7-9.9) 1 RT CN 1/3 33.2 (21-43) 4/0 6.5 (3.7-9.0) None PXA 1/2 48.0 (28-64) 2/1 7.3 (2.3-14.5) None

N° Pt = number of treated patients; yr = year; TV = tumor volume; cc = cubic cm; Pre-GK Treats. = pregamma knife treatments; M = male; F = female; RT = radiotherapy; Ch = chemotherapy; Papillary Gl.

Table 1. Summary of clinical characteristics of 30 unusual LGPNTs treated with GKRS at

University Hospital of Verona from February 1993 to February 2009.

M/F Mean (Range) /Biopsy Mean (Range) Treats.

2/3 19,2 (11-24) 3/2 3.7 (0.7-11.3) 1 RT+Ch

1 RT+Ch

some selected independent variables will be performed, as well.

**2. Materials and methods** 

**2.1 Overall series** 

reoperation.

Miscellaneous: Ganglioglioma Subependymoma Oligoastrocytoma Papillary Gl. Tum.

Tum. = papillary glioneuronal tumor.

The histologic types and grading were as follows: 10 pineocytomas (Grade I), 8 choroid plexus papillomas (CPP) (Grade I), 4 central neurocytomas (CN) (Grade II), 3 pleomorphic xanthoastrocytomas (PXA) (Grade II), 2 gangliogliomas (Grade I), 1 subependymoma (Grade I), 1 papillary glioneuronal tumor (Grade I), and 1 mixed oligoastrocytoma (Grade II). All tumors were histologically verified: 20/30 (67%) were surgical residuals or postoperative recurrences, while biopsy confirmed the diagnosis in the remaining 10/30 cases: 4 (13%) during neuroendoscopy approach and 6 (20%) with stereotactic brain intervention. There were 16 males and 14 females; the mean age of the clinical series was 34.7 years (7-70 years). Further pre-GKRS treatments were as follows: RT alone, 2/30 patients (7%), and RT followed by chemotherapy, 1/30 (3%). The TV was always less than 15 cc. In 24/30 patients, pre-GKRS neurologic deficits were documented. The mean and range values of the radiosurgical dose planning parameters are summarized in Table 2.


PD = prescription dose; Gy = gray; PI = prescription isodose; ID = integral dose; mJ = milli Joule; # Shots = number of isocenters.

Table 2. Summary of radiosurgical parameters of 30 unusual LGPNTs treated with GKRS at University Hospital of Verona from February 1993 to February 2009.

The mean prescription dose (PD) was 16.5 Gy while the mean prescription isodose (PI) corresponded to 50.2%. The average value of integral dose (ID), parameter which describes the relationship between treated TV and delivered dose appropriately, was 104 mJ. The mean number of shots needed for dose planning was 9.6. Follow-up data, including causes of death, were obtained from hospital notes, imaging studies, and contact with relatives and family physicians. Medical records and MRI scans for all patients were carefully reviewed by neurosurgeons and neuroradiologists. This retrospective study was approved by the Ethical Commetee of our University Hospital. All patients identifiers were removed before data analysis.

## **2.1.1 Pineocytoma**

In the group of pineocytoma, 7 out of 10 patients were young adults, under 40-year-old. Diagnosis was confirmed by bioptic samples harvest during a neuroendoscopic procedure

Gamma Knife Radiosurgery in the Management of

intervention.

**2.1.5 Others (Miscellaneous)** 

the other patient of our series.

**2.2 Radiosurgical technique and procedure**

A B

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 77

Fig. 1. A) 28-year old man with cystic temporo-occipital PXA. Calculation of stereotactic coordinates and trajectory for Ommaya reservoir insertion and tumor nodule biopsy. B) GKRS dose planning performed on shrinkaged tumor 10 days after double stereotactic

This group of patients included 2 gangliogliomas (Grade I), 1 subependymoma (Grade I), 1 papillary glioneuronal tumor (Grade I), and 1 mixed oligoastrocytoma (Grade II). The mean age (19.2 years) was lower than in all the other groups. There were 2 males and 3 females. The tumor sites were as follows: 2 deep frontal, 1 basal ganglia, 1 occipital, and 1 third ventricle. There were 3 pre-GKRS surgical interventions and 2 stereotactic brain biopsies. In 3 out of 5 patients, neurologic examination resulted negative on the day of radiosurgery. The mean TV treated with GKRS was 3.7 cc. A 24-year-old woman underwent stereotactic brain biopsy for a left frontal deep-seated tumor in another institution. Histologic diagnosis was grade II mixed oligoastrocytoma and a treatment protocol of RT and chemo-therapy was performed successively. In this group, the mean selected PI (54%) was higher than in all

Our radiosurgical technique was already described in detail in previous reports (Lunsford et al., 1989, Nicolato et al., 1997). In brief, after the application of the magnetic resonance imaging (MRI) compatible Leksell Model G stereotactic frame (Elekta Instruments AB, Stockholm, Sweden) to the patient's head, high resolution 1.0 Tesla MRI was performed in all cases. Millimetric volumetric images in axial and coronal orthogonal planes after gadolinium enhancement were used. First a sagittal MRI localizer or a 3-D scout survey (which included axial, coronal and sagittal images) were performed. Contrast enhanced Spoiled Gradient Recalled Acquisition in Steady State (SPGR) sequence was then used to image the tumor and surrounding brain*.* T2 weighted MR images using Fast Spin Echo technique also were acquired to assess the infiltrative tumor volume. The target volume included enhanced and non-enhanced tumor regions. Three-dimensional dose plans were developed by using commercially available software, i.e., Kula (Elekta Instruments) from February 1993 to February 1998 and Leksell Gamma Plan (versions 4.12, 5.34 and 8.3, Elekta Instruments) after February 1998. The neurological surgeon, radiation oncologist, and medical physicist created a highly conformal dose planning using multiple collimators and performed the dose selection. In all patients, the radiosurgery dose was prescribed to the entire tumor volume as defined using contrast enhanced imaging. GKRS was performed with either Model B, C 201-source Co60, or Perfexion Leksell Gamma Knife (Elekta Instruments). The procedure was performed under local anesthesia in 28/30 patients;

in 4 cases. In the other histologic groups, there were no patients in whom the histologic diagnosis was achieved following a neuroendoscopic intervention. The treated TV was always inferiour to 12 cc (mean, 5.0 cc). Only a 58-year-old woman, operated on in another institution, underwent RT following incomplete surgical removal. Neurologic deficits were present in 9/10 patients at the moment of GKRS. The 10 pineocytoma tumors were administered a radiosurgical dose comprised between 13 and 20 Gy with a mean PI of 50.5% and a mean ID of 110.1 mJ. The dose plannings for pineocytoma were particularly conformal as documented by the higher mean number of utilized isocenters (13.9) compared to the mean number of shots employed in the other histologic groups.

## **2.1.2 Choroid plexus papilloma (CPP)**

The mean age in this group of patients was 44.2 years and most cases were males. The tumors were located in the following sites: 2 cerebellum, 2 ponto-cerebellar angle, 1 basal ganglia, 1 brain stem, 1 pineal region, and 1 skull base. Diagnosis was achieved by surgical intervention in 5 out of 8 cases. The TV treated with GKRS was always small, mean 3.7 cc, inferiour to the means of all the other histologic types of our series. In a 25-year-old man, two operations where performed before GKRS: the last histological examination confirmed a grade I CPP. Only in one case, RT was performed after partial surgical resection and before GKRS. All the patients complaint neurologic symptoms at last evaluation prior to GKRS. From the radiosurgical point of view, the mean PD and ID were lower than those of all the other histologic groups of our series, probably due to the highly critical brain location of most of these tumors.

### **2.1.3 Central neurocytoma (CN)**

The mean age was 33.2 years. The tumor location was in the lateral ventricles in 2 cases, and one case in the basal ganglia and in the occipital lobe, respectively. Histologic diagnosis was defined by means of surgical intervention in all neoplasms. In 2 out of 4 patients, the pre-GKRS neurologic evaluation was negative. The mean TV treated with radiosurgery was 6.5 cc. None of the patients underwent to other treatments prior to GKRS. The mean PD and ID were 16.4 Gy and 151.6 mJ, respectively.

#### **2.1.4 Pleomorphic xanthoastrocyroma (PXA)**

Mean age and TV treated with GKRS (48.0 years and 7.3 cc, respectively) in this group of patients were higher than in the other groups of our series. Tumor lacations were as follows: 1 brain stem, 1 temporal, and 1 temporo-occipital. In 2 cases, diagnosis was achieved by surgical intervention. In a 28-year old man affected with cystic right temporo-occipital PXA, a double stereotactic procedure was performed in the same neurosurgical session (Figure 1): first, an Ommaya reservoir was inserted for cyst aspiration; secondly, a biopsy of the tumor solid nodule for histopathological diagnosis definition was performed preceeded by changing of stereotactic coordinates and trajectory. All patients were symptomatic at the time of radiosurgery. No other treatment before GKRS was performed. Mean PD and ID delivered were higher than in the other groups of our series ( 17.0 Gy and 152.1 mJ, respectively).

A B

Fig. 1. A) 28-year old man with cystic temporo-occipital PXA. Calculation of stereotactic coordinates and trajectory for Ommaya reservoir insertion and tumor nodule biopsy. B) GKRS dose planning performed on shrinkaged tumor 10 days after double stereotactic intervention.

## **2.1.5 Others (Miscellaneous)**

76 Gamma Knife Radiosurgery

in 4 cases. In the other histologic groups, there were no patients in whom the histologic diagnosis was achieved following a neuroendoscopic intervention. The treated TV was always inferiour to 12 cc (mean, 5.0 cc). Only a 58-year-old woman, operated on in another institution, underwent RT following incomplete surgical removal. Neurologic deficits were present in 9/10 patients at the moment of GKRS. The 10 pineocytoma tumors were administered a radiosurgical dose comprised between 13 and 20 Gy with a mean PI of 50.5% and a mean ID of 110.1 mJ. The dose plannings for pineocytoma were particularly conformal as documented by the higher mean number of utilized isocenters (13.9) compared to the

The mean age in this group of patients was 44.2 years and most cases were males. The tumors were located in the following sites: 2 cerebellum, 2 ponto-cerebellar angle, 1 basal ganglia, 1 brain stem, 1 pineal region, and 1 skull base. Diagnosis was achieved by surgical intervention in 5 out of 8 cases. The TV treated with GKRS was always small, mean 3.7 cc, inferiour to the means of all the other histologic types of our series. In a 25-year-old man, two operations where performed before GKRS: the last histological examination confirmed a grade I CPP. Only in one case, RT was performed after partial surgical resection and before GKRS. All the patients complaint neurologic symptoms at last evaluation prior to GKRS. From the radiosurgical point of view, the mean PD and ID were lower than those of all the other histologic groups of our series, probably due to the highly critical brain location of

The mean age was 33.2 years. The tumor location was in the lateral ventricles in 2 cases, and one case in the basal ganglia and in the occipital lobe, respectively. Histologic diagnosis was defined by means of surgical intervention in all neoplasms. In 2 out of 4 patients, the pre-GKRS neurologic evaluation was negative. The mean TV treated with radiosurgery was 6.5 cc. None of the patients underwent to other treatments prior to GKRS. The mean PD and ID

Mean age and TV treated with GKRS (48.0 years and 7.3 cc, respectively) in this group of patients were higher than in the other groups of our series. Tumor lacations were as follows: 1 brain stem, 1 temporal, and 1 temporo-occipital. In 2 cases, diagnosis was achieved by surgical intervention. In a 28-year old man affected with cystic right temporo-occipital PXA, a double stereotactic procedure was performed in the same neurosurgical session (Figure 1): first, an Ommaya reservoir was inserted for cyst aspiration; secondly, a biopsy of the tumor solid nodule for histopathological diagnosis definition was performed preceeded by changing of stereotactic coordinates and trajectory. All patients were symptomatic at the time of radiosurgery. No other treatment before GKRS was performed. Mean PD and ID delivered were higher than in the other groups of our series ( 17.0 Gy and 152.1 mJ,

mean number of shots employed in the other histologic groups.

**2.1.2 Choroid plexus papilloma (CPP)** 

most of these tumors.

respectively).

**2.1.3 Central neurocytoma (CN)** 

were 16.4 Gy and 151.6 mJ, respectively.

**2.1.4 Pleomorphic xanthoastrocyroma (PXA)** 

This group of patients included 2 gangliogliomas (Grade I), 1 subependymoma (Grade I), 1 papillary glioneuronal tumor (Grade I), and 1 mixed oligoastrocytoma (Grade II). The mean age (19.2 years) was lower than in all the other groups. There were 2 males and 3 females. The tumor sites were as follows: 2 deep frontal, 1 basal ganglia, 1 occipital, and 1 third ventricle. There were 3 pre-GKRS surgical interventions and 2 stereotactic brain biopsies. In 3 out of 5 patients, neurologic examination resulted negative on the day of radiosurgery. The mean TV treated with GKRS was 3.7 cc. A 24-year-old woman underwent stereotactic brain biopsy for a left frontal deep-seated tumor in another institution. Histologic diagnosis was grade II mixed oligoastrocytoma and a treatment protocol of RT and chemo-therapy was performed successively. In this group, the mean selected PI (54%) was higher than in all the other patient of our series.

## **2.2 Radiosurgical technique and procedure**

Our radiosurgical technique was already described in detail in previous reports (Lunsford et al., 1989, Nicolato et al., 1997). In brief, after the application of the magnetic resonance imaging (MRI) compatible Leksell Model G stereotactic frame (Elekta Instruments AB, Stockholm, Sweden) to the patient's head, high resolution 1.0 Tesla MRI was performed in all cases. Millimetric volumetric images in axial and coronal orthogonal planes after gadolinium enhancement were used. First a sagittal MRI localizer or a 3-D scout survey (which included axial, coronal and sagittal images) were performed. Contrast enhanced Spoiled Gradient Recalled Acquisition in Steady State (SPGR) sequence was then used to image the tumor and surrounding brain*.* T2 weighted MR images using Fast Spin Echo technique also were acquired to assess the infiltrative tumor volume. The target volume included enhanced and non-enhanced tumor regions. Three-dimensional dose plans were developed by using commercially available software, i.e., Kula (Elekta Instruments) from February 1993 to February 1998 and Leksell Gamma Plan (versions 4.12, 5.34 and 8.3, Elekta Instruments) after February 1998. The neurological surgeon, radiation oncologist, and medical physicist created a highly conformal dose planning using multiple collimators and performed the dose selection. In all patients, the radiosurgery dose was prescribed to the entire tumor volume as defined using contrast enhanced imaging. GKRS was performed with either Model B, C 201-source Co60, or Perfexion Leksell Gamma Knife (Elekta Instruments). The procedure was performed under local anesthesia in 28/30 patients;

Gamma Knife Radiosurgery in the Management of

N°

Wors. = Neurological worsening; OS = Overall survival.

Hospital of Verona from February 1993 to February 2009.

Pt Me/Med

Histology

2 patients.

diffusion in the other 2 patients.

Histology Act. Surv. %

RT = Radiation therapy.

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 79

Overall series 30 85.7/66.8 16/30 8/30 6‡/30 26/4¶ (87.0) Pineocytoma 10 66.7/54.6 6/10 3/10 1/10 9/1 (90.0) CPP 8 92.6/71.8 2/8 2/8 4/8 6/2 (75.0) CN 4 172.0/180.6 4/4 – – 4/0 (100.0) PXA 3 45.6/43.7 1/3 1/3 1/3 2/1 (66.7) Miscellaneous 5 68.0/68.3 3/5 2/5 – 5/0 (100.0)

N° Pt = number of treated patients with al least 2-year-follow-up; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement;

¶: the cause of death was tumor distant progression in 2 cases and both local and distant tumor

Ov. TGC%

Overall series 90.0 87.0 90.0 91.0 91.0 0.0 0.0

‡: due to tumor distant progression in 4 cases and both to local and distant tumor diffusion in the other

Table 3. Summary of clinical results concerning 30 LGPNTs treated with GKRS at University

At 5y At 10y At 5y At 10y Mort. Perm

Pineocytoma 92.0 92.0 100.0 100.0 100.0 0.0 0.0 1 RT

CPP 90.0 80.0 87.5 90.0 90.0 0.0 0.0 1 surg.int.

CN 100.0 100.0 75.0 83.0 83.0 0.0 0.0 1 surg.int.

PXA 80.0 80.0 66.7 80.0 80.0 0.0 0.0 1 VPS Miscellaneous 100.0 100.0 100.0 100.0 100.0 0.0 0.0 1 surg.int.

Act. Surv. = Actuarial survival; Ov. TGC = Overall tumor growth control; Act. Local PFS = Actuarial local progression-free survival; Mort. = Mortality; Perm. Compl = Permanent Complications; Post-GK treats. = Post-GK treatments; surg.int. = surgical intervention; VPS = Ventricolo-peritoneal shunt;

Table 4. Summary of outcome results concerning 30 LGPNTs treated with GKRS at

University Hospital of Verona from February 1993 to February 2009.

Act. Local PFS%

GK-related sequelae

Compl

Clinical Results

FU mos. No def./Impr Stable Wors. (OS%)

Alive/ Dead

Post-GK treats.

3 surg.int. 1 VPS 3 RT

1 RT

1 RT

general anesthesia was administered in the two children younger than 14. All patients were discharged from the hospital within 24 hours after treatment, and they were evaluated clinically and with serial contrast-enhanced imaging using MRI at intervals of 6 months. The neurological status of the patient during the follow-up period was defined as no deficit pre-GKRS, if the patient were without deficits before radiosurgery and remained clinically negative, improved, in case of complaint amelioration, stable or worsened. For the evaluation of TV, the mass lesion was measured by using the Gamma Plan software or the OsiriX medical imaging program (version 4.19) developed at the University Hospital of Geneva, Switzerland. Local TGC was defined as complete disappearance of enhancing and nonenhancing neoformation, lesion shrinkage or stable disease.

### **2.3 Statistical analysis**

Survival and local PFS curves were calculated using the life-table system and the Kaplan– Meier method (Kaplan & Meier, 1958). Length of survival and local PFS were evaluated on an actuarial basis from the day of GKRS treatment to the time of patient death or tumor progression and/or last follow-up, respectively. Survival and local PFS curves were compared and evaluated using Breslaw's test (Breslaw,1974). We performed an univariate analysis using the log-rank test to detect variables that might influence the length of survival and local PFS. We matched the survival with four different variables, age, sex (male vs. female), histologic type (pineocytoma vs. choroid plexus papilloma vs. central neurocytoma vs. pleomorphic xanthoastrocytoma vs. miscellaneuos), prior surgical resection vs. biopsy, and the local PFS with seven different parameters, age, sex (male vs. female), histologic type (pineocytoma vs. choroid plexus papilloma vs. central neurocytoma vs. pleomorphic xanthoastrocytoma vs. miscellaneuos), prior surgical resection vs. biopsy, prescription dose, integral dose and length of follow-up. In Table 5, the statistical comparison between parameters in the different groups of patients were based on the Welch Modified Two-Sample t-Test. On the basis of internationally accepted criteria, values of P ≤ 0.05 were considered statistically significant. Statistical analysis was effected using the Stata version 8.2 (StataCorp, College Station, Texas) and MATLAB for graphics*.*

## **3. Results**

#### **3.1 Overall series**

The results of our series are summarized in Tables 3 and 4. The median duration of followup was 66.8 months (range, 24.7-195.97 months). At the end of the study, February 28th, 2009, the neurologic examination remained negative or showed an improvement of previous deficits in 16 patients, in 8 cases the complaints were stable and in the other 6 a worsening due to tumor progression was observed. Among these last patients, four were deceased. The cause of death was related to distant tumour progression in 2 cases and both to local and distant disease diffusion in the other 2 cases. The overall survival was 87%, and the actuarial survival rate at 5 and 10 years was 90% and 87%, respectively (Figure 2A). The median survival in the 4 patients who died was 50 months (range, 24.7-73 months) from GKRS. The local TGC was achieved in 27/30 (90%) tumors with a 91% local actuarial PFS both at 5 and 10 years (Figure 2B). The median time to progression for the three patients with uncontrolled tumor was 18.5 months (range, 9.8-26.7 months). No GKRS-related permanent side-effect was observed on the whole series. To date, neither malignant tumor

general anesthesia was administered in the two children younger than 14. All patients were discharged from the hospital within 24 hours after treatment, and they were evaluated clinically and with serial contrast-enhanced imaging using MRI at intervals of 6 months. The neurological status of the patient during the follow-up period was defined as no deficit pre-GKRS, if the patient were without deficits before radiosurgery and remained clinically negative, improved, in case of complaint amelioration, stable or worsened. For the evaluation of TV, the mass lesion was measured by using the Gamma Plan software or the OsiriX medical imaging program (version 4.19) developed at the University Hospital of Geneva, Switzerland. Local TGC was defined as complete disappearance of enhancing and

Survival and local PFS curves were calculated using the life-table system and the Kaplan– Meier method (Kaplan & Meier, 1958). Length of survival and local PFS were evaluated on an actuarial basis from the day of GKRS treatment to the time of patient death or tumor progression and/or last follow-up, respectively. Survival and local PFS curves were compared and evaluated using Breslaw's test (Breslaw,1974). We performed an univariate analysis using the log-rank test to detect variables that might influence the length of survival and local PFS. We matched the survival with four different variables, age, sex (male vs. female), histologic type (pineocytoma vs. choroid plexus papilloma vs. central neurocytoma vs. pleomorphic xanthoastrocytoma vs. miscellaneuos), prior surgical resection vs. biopsy, and the local PFS with seven different parameters, age, sex (male vs. female), histologic type (pineocytoma vs. choroid plexus papilloma vs. central neurocytoma vs. pleomorphic xanthoastrocytoma vs. miscellaneuos), prior surgical resection vs. biopsy, prescription dose, integral dose and length of follow-up. In Table 5, the statistical comparison between parameters in the different groups of patients were based on the Welch Modified Two-Sample t-Test. On the basis of internationally accepted criteria, values of P ≤ 0.05 were considered statistically significant. Statistical analysis was effected using the Stata version

The results of our series are summarized in Tables 3 and 4. The median duration of followup was 66.8 months (range, 24.7-195.97 months). At the end of the study, February 28th, 2009, the neurologic examination remained negative or showed an improvement of previous deficits in 16 patients, in 8 cases the complaints were stable and in the other 6 a worsening due to tumor progression was observed. Among these last patients, four were deceased. The cause of death was related to distant tumour progression in 2 cases and both to local and distant disease diffusion in the other 2 cases. The overall survival was 87%, and the actuarial survival rate at 5 and 10 years was 90% and 87%, respectively (Figure 2A). The median survival in the 4 patients who died was 50 months (range, 24.7-73 months) from GKRS. The local TGC was achieved in 27/30 (90%) tumors with a 91% local actuarial PFS both at 5 and 10 years (Figure 2B). The median time to progression for the three patients with uncontrolled tumor was 18.5 months (range, 9.8-26.7 months). No GKRS-related permanent side-effect was observed on the whole series. To date, neither malignant tumor

nonenhancing neoformation, lesion shrinkage or stable disease.

8.2 (StataCorp, College Station, Texas) and MATLAB for graphics*.*

**2.3 Statistical analysis**

**3. Results**

**3.1 Overall series**


N° Pt = number of treated patients with al least 2-year-follow-up; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = Neurological worsening; OS = Overall survival.

‡: due to tumor distant progression in 4 cases and both to local and distant tumor diffusion in the other 2 patients.

¶: the cause of death was tumor distant progression in 2 cases and both local and distant tumor diffusion in the other 2 patients.

Table 3. Summary of clinical results concerning 30 LGPNTs treated with GKRS at University Hospital of Verona from February 1993 to February 2009.


Act. Surv. = Actuarial survival; Ov. TGC = Overall tumor growth control; Act. Local PFS = Actuarial local progression-free survival; Mort. = Mortality; Perm. Compl = Permanent Complications; Post-GK treats. = Post-GK treatments; surg.int. = surgical intervention; VPS = Ventricolo-peritoneal shunt; RT = Radiation therapy.

Table 4. Summary of outcome results concerning 30 LGPNTs treated with GKRS at University Hospital of Verona from February 1993 to February 2009.

Gamma Knife Radiosurgery in the Management of

analysis.

**3.2 Pineocytoma** 

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 81

(14.0 Gy) than that of the 27 tumors with local TGC (16.8 Gy), and the mean ID was higher (151.1 mJ) compared to that of the group under control (98.74 mJ). Because the ID is the result of volume multiplied by average dose, we evaluated the mean TV between the groups of patients with under control vs. progressed neoplasms, as well. The 3 patients with tumor progression presented a mean TV of 7.3 cc versus 4.6 cc of the tumors with favourable response. Even though the statistical analysis did not shaw significant results, the considerable difference of the two mean TV values authorizes to think that a possible correlation between TV and local TGC after GKRS could be proposed. Finally, it must be considered that the small number of observations might represent a bias for the statistical

In this group of patients, the median observation time was 54.6 months (range, 25.3-130.0 months). On last follow-up day, 9 out of 10 patients were neurologically stable or improved (Table 3); only a 16-year-old boy worsened because of distant tumor progression and underwent post-GKRS RT, but unfortunately he died at 42 months from radiosurgery (OS, 90%). All the other patients did not needed further treatments following GKRS. In the whole ten pineocytomas of our series, local TGC was achieved (overall TGC, 100%) (Figure 4). The 5 and 10 year actuarial survival and local TGC were 92% and 100%, respectively (Table 4).

Fig. 4. A) Posterior III ventricle pineocytoma on GKRS day. The diagnosis was achieved by means of stereotactic neuroendoscopy biopsy. B) MR imaging follow-up at 3 months and C)

The 8 patients of this group were followed-up for 71.8-month median time. A neurologic deterioration was observed in 4 patients: in 3 cases, due to tumor distant progression, and in 1 both to local and distant progression. During the follow-up period, 2 of these last patients died: 1 because of tumor distant progression and 1 due to local and distant progression (OS, 75%). The 5- and 10-year actuarial survival rates were 90% and 80%, respectively. Local TGC was reported in 7 patients (87.5%), with an actuarial local TGC of 90% both at 5- and 10-years. The 2 patients with distant progression refused further treatments: one of them is died at 57.7 months from GKRS, and the other is still alive with mild neurologic deficits 181.7 months after radiosurgery. The treatments performed after GKRS were surgical intervention for local tumor

The median follow-up period was 180.6 months (range, 142.7-183.9 months). At last observation, all patients were neurologically intact or clinically improved. The OS and the 5-

A B C

**3.3 Choroid plexus papilloma (CPP)** 

**3.4 Central neurocytoma (CN)** 

at 18 months from GKRS showing the gradual disappearance of the tumor.

progression in one case and RT in another patient with distant progression.

Fig. 2. Kaplan-Meier estimate of A) overall survival and B) local PFS curves in 30 LGPNTs.

transformation nor GKRS-related mortality were registered. Post-GKRS treatments were represented by: 3 surgical removals, 1 ventriculo-peritoneal shunt and 3 fractionated irradiations. No chemotherapy treatment was performed on the whole series after GKRS. All variables tested at univariate analysis on the whole series did not show a statistically significant influence either as concerns survival or local PFS rates. In particular, the different histological type analized did not seem to have prognostic significance, as well (Figures 3A and 3B). Therefore, multivariate analysis was not performed. Nevertheless, a positive trend for PD and ID seems to emerge from statistical analysis (Table 5): in the 3 patients with local progression, the mean dose delivered to the periphery of the neoplasms was lower

Fig. 3. Kaplan-Meier estimate of A) overall survival and B) local PFS curves after GKRS for histologic types (P = NS).


N° Pt = number of treated patients; TV = tumor volume; cc = cubic cm; PD = prescription dose; Gy = gray; ID = integral dose; mJ = milli Joule.

Table 5. statistical comparison between radiosurgical outcome and TV, PD, and ID in 30 LGPNTs treated with GKRS.

(14.0 Gy) than that of the 27 tumors with local TGC (16.8 Gy), and the mean ID was higher (151.1 mJ) compared to that of the group under control (98.74 mJ). Because the ID is the result of volume multiplied by average dose, we evaluated the mean TV between the groups of patients with under control vs. progressed neoplasms, as well. The 3 patients with tumor progression presented a mean TV of 7.3 cc versus 4.6 cc of the tumors with favourable response. Even though the statistical analysis did not shaw significant results, the considerable difference of the two mean TV values authorizes to think that a possible correlation between TV and local TGC after GKRS could be proposed. Finally, it must be considered that the small number of observations might represent a bias for the statistical analysis.

## **3.2 Pineocytoma**

80 Gamma Knife Radiosurgery

Fig. 2. Kaplan-Meier estimate of A) overall survival and B) local PFS curves in 30 LGPNTs.

Fig. 3. Kaplan-Meier estimate of A) overall survival and B) local PFS curves after GKRS for

P value Mean PD

N° Pt = number of treated patients; TV = tumor volume; cc = cubic cm; PD = prescription dose; Gy =

Table 5. statistical comparison between radiosurgical outcome and TV, PD, and ID in 30

(Gy)

<0.55 = 0.138 = 0.141

P value Mean ID (mJ) P value

transformation nor GKRS-related mortality were registered. Post-GKRS treatments were represented by: 3 surgical removals, 1 ventriculo-peritoneal shunt and 3 fractionated irradiations. No chemotherapy treatment was performed on the whole series after GKRS. All variables tested at univariate analysis on the whole series did not show a statistically significant influence either as concerns survival or local PFS rates. In particular, the different histological type analized did not seem to have prognostic significance, as well (Figures 3A and 3B). Therefore, multivariate analysis was not performed. Nevertheless, a positive trend for PD and ID seems to emerge from statistical analysis (Table 5): in the 3 patients with local progression, the mean dose delivered to the periphery of the neoplasms was lower

A B

A B

(cc)

Local TGC 27 4.61 16.8 98.74

Local Progr. 3 7.30 14.0 151.1

histologic types (P = NS).

GKRS outcome N° Pt Mean TV

gray; ID = integral dose; mJ = milli Joule.

LGPNTs treated with GKRS.

In this group of patients, the median observation time was 54.6 months (range, 25.3-130.0 months). On last follow-up day, 9 out of 10 patients were neurologically stable or improved (Table 3); only a 16-year-old boy worsened because of distant tumor progression and underwent post-GKRS RT, but unfortunately he died at 42 months from radiosurgery (OS, 90%). All the other patients did not needed further treatments following GKRS. In the whole ten pineocytomas of our series, local TGC was achieved (overall TGC, 100%) (Figure 4). The 5 and 10 year actuarial survival and local TGC were 92% and 100%, respectively (Table 4).

Fig. 4. A) Posterior III ventricle pineocytoma on GKRS day. The diagnosis was achieved by means of stereotactic neuroendoscopy biopsy. B) MR imaging follow-up at 3 months and C) at 18 months from GKRS showing the gradual disappearance of the tumor.

## **3.3 Choroid plexus papilloma (CPP)**

The 8 patients of this group were followed-up for 71.8-month median time. A neurologic deterioration was observed in 4 patients: in 3 cases, due to tumor distant progression, and in 1 both to local and distant progression. During the follow-up period, 2 of these last patients died: 1 because of tumor distant progression and 1 due to local and distant progression (OS, 75%). The 5- and 10-year actuarial survival rates were 90% and 80%, respectively. Local TGC was reported in 7 patients (87.5%), with an actuarial local TGC of 90% both at 5- and 10-years. The 2 patients with distant progression refused further treatments: one of them is died at 57.7 months from GKRS, and the other is still alive with mild neurologic deficits 181.7 months after radiosurgery. The treatments performed after GKRS were surgical intervention for local tumor progression in one case and RT in another patient with distant progression.

## **3.4 Central neurocytoma (CN)**

The median follow-up period was 180.6 months (range, 142.7-183.9 months). At last observation, all patients were neurologically intact or clinically improved. The OS and the 5-

Gamma Knife Radiosurgery in the Management of

**4. Discussion 4.1 Overall series**

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 83

Several series describing the results of stereotactic radiosurgery on LGPNTs of the brain including a large number of patients have been reported. However, they always dealt with low-grade gliomas. Kida (Kida et al., 2000) and Heppner (Heppner et al., 2005) presented their experience with GKRS on 51 and 49 patients with grade I and II astrocytomas, respectively. Hadjipanayis (Hadjipanayis et al., 2002a) treated 37 pilocytic astrocytomas of every age, and, more recently, Kano (Kano et al., 2009b) described the outcome following GKRS in 50 pilocytic astrocytomas in pediatric age, exclusively. Simonova (Simonova et al., 2005) reported a series of 70 cases including different types of grade I and II gliomas in whom stereotactic radiosurgery was applied. Sarkar (Sarkar et al., 2002) and Kano (Kano et al., 2009c) reported two GKRS studies on 18 and 30 patients affected with oligodendrogliomas and mixed oligodendroastrocytomas, respectively, but both series included low- and high-grade tumors. As concerns the unusual primitive neuroepithelial tumors of the brain, the largest published series always comprised low- and high-grade tumors (Kano et al., 2009d, Mori et al., 2009, Lekovic et al., 2007, Hasegawa et al., 2002, Kobayashi et al., 2001), while the studies with exclusively unusual LGPNTs were always represented by a few cases (usually, less than 15). To our knowledge, the work of our team represent the largest series of unusual LGPNTs treated with stereotactic radiosurgery. In a study performed on 10 patients (8 pineocytoamas, 1 CPP, and 1 CN), Lekovic (Lekovic et al., 2007) reported an overall survival of 90%, a 100% local TGC, with neither GKRS-related mortality nor permanent morbidity. But, the patients were observed for a median period of less than 2 years (20.5 months). Our results obtained on 30 patients followed up for at least 2 years (median, 66.8 months; range, 24.7-195.97 months) confirmed this excellent outcome. In brief, the 5- and 10-year actuarial survival rates were 90% and 87%, respectively, with a 91% actuarial local PFS both at 5- and 10-year. The median time to progression for the three patients with uncontrolled tumor was 18.5 months and the median survival in the 4 patients who died was 50 months. In 2 of them, the cause of death was exclusively due to tumor distant progression. Also in our series, no complications were attributable to GKRS. The only study on a series of LGPNTs in which was performed an analysis of potential prognostic factors influencing GKRS outcome is referred to Sarkar (Sarkar et al., 2002). The authors studied a series of 18 patients with 21 tumors – 10 oligodendrogliomas and 11 mixed oligoastrocytomas – and found that factors associated with an improved survival rate included younger age and smaller tumors. In our series, univariate analysis on selected independent variables did not show any factors of significance. Nevertheless, we observed that patients with local TGC presented a smaller mean TV (4.61 cc vs. 7.0 cc), a higher mean PD (16.8 Gy vs. 14.0 Gy), and a lower mean ID (98.74 mJ vs. 151.1 mJ) than those with tumor progression. This trend could suggest that multisession GKRS with the Extend System (Elekta Instrument) should be better in those patients affected with larger tumors, thus achieving an increased dose delivery to the target volume maintaining a negligible risk of permanent side effects on the surrounding normal brain tissue. Some authors (Kobayashi et al., 2001, Kim et al., 2007) propose GKRS as primary treatment without histological diagnosis when a deep-seated or critically located brain tumor showing the imaging characteristics of an unusual LGPNTs is identified. On the contrary, other collegues state that obtaining a histological diagnosis remains the main aim for rational treatment planning (Matsunaga et al., 2010; Martin et al., 2003; Yen et al., 2007;

and 10-year actuarial survival rates were 100% (Table 4). The local TGC was achieved in 3 out of 4 cases (75%) and the actuarial local TGC was 83% both at 5- and 10-years. In a 26 year-old woman with a TV at radiosurgery of 9.0 cc was decided to associate RT after GKRS. She is alive and well at 177.3 months from radiosurgery. A surgical intervention was needed in a 21-year-old man because of local progression of an intraventricular residual tumor at 9.8 months after GKRS. He underwent radiosurgery for a 5.9 cc neoplasm and the administered PD was 14.0 Gy, the lowest value than in all the other patients treated for central neurocytoma.

## **3.5 Pleomorphic xanthoastrocyroma (PXA)**

The median follow-up time in this group of patients was 43.7 months (range, 25.1-70.9 months). At the end of the study, one patient showed an improvement of the neulogic complaints, the second was clinically stable and the third presented a gradual deterioration due to a brain stem tumor progression and died 25.1 months after GKRS. This patient was already operated prior to radiosurgery. She refused further RT and the only possible treatment was VPS. In this group, the OS and the local TGC rates were 66.7% and the actuarial survival and local TGC rates were 80% both at 5- and 10-years (Figure 5).

Fig. 5. Same case than in figures 1. MRI follow up at 32 months from GKRS showing a significant tumor reduction.

## **3.6 Others (Miscellaneous)**

The median period of follow-up was 68.3 months (range, 27.5-109 months). All patients in this heterogeneous group of unusual LGPNTs showed a negative, improved or stable neurological status at last observation (OS 100%), with an actuarial survival rate at 5- and 10-years of 100% and 100%, respectively. All tumors treated with GKRS were under control (Table 4). A 24-year-old woman affected with a grade II mixed oligoastrocytoma and treated prior to radiosurgery with RT and chemotharapy elsewhere developed an expansive lesion with mass effect in left frontal region, far from the target volume treated with GKRS. Surgical removal was needed and histologic examination showed that it deals with radionecrosis, secondary to the previous RT. Now, she is well at 68.3 months from radiosurgery. She was the only patient in this group who needed a further treatment after GKRS.

## **4. Discussion**

82 Gamma Knife Radiosurgery

and 10-year actuarial survival rates were 100% (Table 4). The local TGC was achieved in 3 out of 4 cases (75%) and the actuarial local TGC was 83% both at 5- and 10-years. In a 26 year-old woman with a TV at radiosurgery of 9.0 cc was decided to associate RT after GKRS. She is alive and well at 177.3 months from radiosurgery. A surgical intervention was needed in a 21-year-old man because of local progression of an intraventricular residual tumor at 9.8 months after GKRS. He underwent radiosurgery for a 5.9 cc neoplasm and the administered PD was 14.0 Gy, the lowest value than in all the other patients treated for central

The median follow-up time in this group of patients was 43.7 months (range, 25.1-70.9 months). At the end of the study, one patient showed an improvement of the neulogic complaints, the second was clinically stable and the third presented a gradual deterioration due to a brain stem tumor progression and died 25.1 months after GKRS. This patient was already operated prior to radiosurgery. She refused further RT and the only possible treatment was VPS. In this group, the OS and the local TGC rates were 66.7% and the

actuarial survival and local TGC rates were 80% both at 5- and 10-years (Figure 5).

Fig. 5. Same case than in figures 1. MRI follow up at 32 months from GKRS showing a

The median period of follow-up was 68.3 months (range, 27.5-109 months). All patients in this heterogeneous group of unusual LGPNTs showed a negative, improved or stable neurological status at last observation (OS 100%), with an actuarial survival rate at 5- and 10-years of 100% and 100%, respectively. All tumors treated with GKRS were under control (Table 4). A 24-year-old woman affected with a grade II mixed oligoastrocytoma and treated prior to radiosurgery with RT and chemotharapy elsewhere developed an expansive lesion with mass effect in left frontal region, far from the target volume treated with GKRS. Surgical removal was needed and histologic examination showed that it deals with radionecrosis, secondary to the previous RT. Now, she is well at 68.3 months from radiosurgery. She was the only patient in this group who needed a further treatment after

neurocytoma.

significant tumor reduction.

**3.6 Others (Miscellaneous)** 

GKRS.

**3.5 Pleomorphic xanthoastrocyroma (PXA)** 

## **4.1 Overall series**

Several series describing the results of stereotactic radiosurgery on LGPNTs of the brain including a large number of patients have been reported. However, they always dealt with low-grade gliomas. Kida (Kida et al., 2000) and Heppner (Heppner et al., 2005) presented their experience with GKRS on 51 and 49 patients with grade I and II astrocytomas, respectively. Hadjipanayis (Hadjipanayis et al., 2002a) treated 37 pilocytic astrocytomas of every age, and, more recently, Kano (Kano et al., 2009b) described the outcome following GKRS in 50 pilocytic astrocytomas in pediatric age, exclusively. Simonova (Simonova et al., 2005) reported a series of 70 cases including different types of grade I and II gliomas in whom stereotactic radiosurgery was applied. Sarkar (Sarkar et al., 2002) and Kano (Kano et al., 2009c) reported two GKRS studies on 18 and 30 patients affected with oligodendrogliomas and mixed oligodendroastrocytomas, respectively, but both series included low- and high-grade tumors. As concerns the unusual primitive neuroepithelial tumors of the brain, the largest published series always comprised low- and high-grade tumors (Kano et al., 2009d, Mori et al., 2009, Lekovic et al., 2007, Hasegawa et al., 2002, Kobayashi et al., 2001), while the studies with exclusively unusual LGPNTs were always represented by a few cases (usually, less than 15). To our knowledge, the work of our team represent the largest series of unusual LGPNTs treated with stereotactic radiosurgery. In a study performed on 10 patients (8 pineocytoamas, 1 CPP, and 1 CN), Lekovic (Lekovic et al., 2007) reported an overall survival of 90%, a 100% local TGC, with neither GKRS-related mortality nor permanent morbidity. But, the patients were observed for a median period of less than 2 years (20.5 months). Our results obtained on 30 patients followed up for at least 2 years (median, 66.8 months; range, 24.7-195.97 months) confirmed this excellent outcome. In brief, the 5- and 10-year actuarial survival rates were 90% and 87%, respectively, with a 91% actuarial local PFS both at 5- and 10-year. The median time to progression for the three patients with uncontrolled tumor was 18.5 months and the median survival in the 4 patients who died was 50 months. In 2 of them, the cause of death was exclusively due to tumor distant progression. Also in our series, no complications were attributable to GKRS. The only study on a series of LGPNTs in which was performed an analysis of potential prognostic factors influencing GKRS outcome is referred to Sarkar (Sarkar et al., 2002). The authors studied a series of 18 patients with 21 tumors – 10 oligodendrogliomas and 11 mixed oligoastrocytomas – and found that factors associated with an improved survival rate included younger age and smaller tumors. In our series, univariate analysis on selected independent variables did not show any factors of significance. Nevertheless, we observed that patients with local TGC presented a smaller mean TV (4.61 cc vs. 7.0 cc), a higher mean PD (16.8 Gy vs. 14.0 Gy), and a lower mean ID (98.74 mJ vs. 151.1 mJ) than those with tumor progression. This trend could suggest that multisession GKRS with the Extend System (Elekta Instrument) should be better in those patients affected with larger tumors, thus achieving an increased dose delivery to the target volume maintaining a negligible risk of permanent side effects on the surrounding normal brain tissue. Some authors (Kobayashi et al., 2001, Kim et al., 2007) propose GKRS as primary treatment without histological diagnosis when a deep-seated or critically located brain tumor showing the imaging characteristics of an unusual LGPNTs is identified. On the contrary, other collegues state that obtaining a histological diagnosis remains the main aim for rational treatment planning (Matsunaga et al., 2010; Martin et al., 2003; Yen et al., 2007;

Gamma Knife Radiosurgery in the Management of

Author RS

Desmukh

Hasegawa

Kobayashi

dev

Perm. = Permanent; GK = Gamma Knife.

Hospital, Komaki, Japan.

Center, Phoenix, Arizona, USA. ‡: due to tumor distant progression.

N° Pt

Me/Med FU mos.

No def.

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 85

remains a challange because of their deep location and associated critical structures. The surgical major morbidity rate associated with pineocytomas is comprised between 3% and 6.8%, and permanent minor morbidity rate was 3-28%. The risk of surgical mortality associated with pineocytoma removal has been reduced in these last 20 years, but it remains less than 2% (Friedman et al., 2001, Bruce & Ogden, 2004). The role of adjuvant RT has not been clearly delineated. Furthermore, pineacytomas are traditionally considered radioresistant tumors. For all these reasons, the use of radiosurgery in the treatment of pineocytoma is more and more growing and diffusing worldwide and studies reporting the experience of different institutions are now available (Table 6). The first and more relevant published series always regard treatment performed with GK device. The median follow-up period varies between 11.0 and 73.5 months. Neither neurological worsening nor GKRSrelated permanent complications were reported. In our studies, 10 patients were followed up for a median period of 54.6 months; we observed a clinical worsening in 1 case, only, due to distant tumor progression. As concerns outcome results (Table 7), the overall survival

/Impr

Kano 2009† GK 13 54.1 Me 66.7 – – 23.0 0.0 Mori 2009\* GK 5 49.0/33.5 – – – 0.0 0.0 Lekovic 2007¶ GK 8 17.4/11.0 – – – 0.0 0.0 Reyns 2006 GK 7 31.7/37.0 – – – – –

2004¶ GK 3 19.3/12.0 – – – – –

2002† GK 10 69.1/73.5 – – – 10.0 0.0

2001\* GK 3 21.7 Me – – – – – Subach 1998† GK 8 – – – – – – Backlund 1974 GK 2 – – – – – – Present series GK 10 66.7/54.6 6/10 3/10 1/10‡ 10.0 0.0

RS dev = Radiosurgical device; N° Pt = number of treated patients; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = neurological worsening; Sympt. Compl. = Symptomatic complications; Trans. = transient;

†: Departments of Neurological Surgery and Radiation Oncology, The University of Pittsburgh, and Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, Pa., USA. \*: Nagoya Radiosurgery Center, Nagoya Kioritsu Hospital, and Gamma Knife Center, Komaki City

¶: Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph's Hospital and Medical

Table 6. Clinical data in previous and present series of pineocytomas treated with GKRS.

Clinical

Results Sympt. Compl.

Stable Wors. Trans. Perm.

% %

Tyler-Kabara et al., 2001; Kano et al., 2009d; Lekovic et al., 2007; Reyns et al., 2007; Dershmukh et al., 2004). In our series,all tumors were histologically verified before GKRS. We do not believe that empirical treatment of brain tumors with radiosurgery is justified, as the treatment paradigm is critically dependent upon tumor histological grade. Therefore, it is of paramount importance to achieve the histological diagnosis for decision making prior to GKRS. We suggest the use of radiosurgery as a primary treatment modality for those patients in whom an adeguate tissue diagnosis of an unusual LGPNTs has been obtained with endoscopy or stereotactic biopsy.

## **4.2 Pineocytoma**

Pineal region tumors are rare and account for 0.4 to 1% of intracranial tumors in Western countries and for 2.2 to 8% of intracranial tumors in northeastern Asian countries (Deshmukh et al., 2004). They are 10 times more common in children than in adults. According to the statistics of the Brain Tumor Registry of Japan, pineal parenchimal tumors such as pineocytoma or pineoblastoma account for 7% of all pineal region tumors (Kobayashi et al., 2001). Individual clinical experience of these tumors is thus limited. Pineocytoma is tipically localized to the pineal area and compresses adjacent structures, including the cerebral acqueduct, brain stem and cerebellum. Their growth may extend into the third ventricle. The majority of patients exhibit neuro-ophtalmologic findings, particularly Parinaud syndromes (Grimoldi et al., 1998). MRI findings such as enhancement, calcification, and welldefined margins are suggestive of a pineocytoma but are by no means diagnostic (Chiechi et al., 1995). Therefore, tissue diagnosis is imperative. Pineocytoma is a slowly growing tumor with a relatively favorable prognosis in most cases. According to the 2007 classification, pineocytomas correspond histologically to WHO grade I. From the pathological point of view, pineocytomas tend to recapitulate the normal pineal gland (Shild et al., 1996) (Figure 6). Despite improved microsurgical technique, resection of pineocytoma

Fig. 6. Pineocytoma. X200 H-E. Classical morphological appearance of uniform cells with rosette and trabeculae.

Tyler-Kabara et al., 2001; Kano et al., 2009d; Lekovic et al., 2007; Reyns et al., 2007; Dershmukh et al., 2004). In our series,all tumors were histologically verified before GKRS. We do not believe that empirical treatment of brain tumors with radiosurgery is justified, as the treatment paradigm is critically dependent upon tumor histological grade. Therefore, it is of paramount importance to achieve the histological diagnosis for decision making prior to GKRS. We suggest the use of radiosurgery as a primary treatment modality for those patients in whom an adeguate tissue diagnosis of an unusual LGPNTs

Pineal region tumors are rare and account for 0.4 to 1% of intracranial tumors in Western countries and for 2.2 to 8% of intracranial tumors in northeastern Asian countries (Deshmukh et al., 2004). They are 10 times more common in children than in adults. According to the statistics of the Brain Tumor Registry of Japan, pineal parenchimal tumors such as pineocytoma or pineoblastoma account for 7% of all pineal region tumors (Kobayashi et al., 2001). Individual clinical experience of these tumors is thus limited. Pineocytoma is tipically localized to the pineal area and compresses adjacent structures, including the cerebral acqueduct, brain stem and cerebellum. Their growth may extend into the third ventricle. The majority of patients exhibit neuro-ophtalmologic findings, particularly Parinaud syndromes (Grimoldi et al., 1998). MRI findings such as enhancement, calcification, and welldefined margins are suggestive of a pineocytoma but are by no means diagnostic (Chiechi et al., 1995). Therefore, tissue diagnosis is imperative. Pineocytoma is a slowly growing tumor with a relatively favorable prognosis in most cases. According to the 2007 classification, pineocytomas correspond histologically to WHO grade I. From the pathological point of view, pineocytomas tend to recapitulate the normal pineal gland (Shild et al., 1996) (Figure 6). Despite improved microsurgical technique, resection of pineocytoma

Fig. 6. Pineocytoma. X200 H-E. Classical morphological appearance of uniform cells with

has been obtained with endoscopy or stereotactic biopsy.

**4.2 Pineocytoma** 

rosette and trabeculae.

remains a challange because of their deep location and associated critical structures. The surgical major morbidity rate associated with pineocytomas is comprised between 3% and 6.8%, and permanent minor morbidity rate was 3-28%. The risk of surgical mortality associated with pineocytoma removal has been reduced in these last 20 years, but it remains less than 2% (Friedman et al., 2001, Bruce & Ogden, 2004). The role of adjuvant RT has not been clearly delineated. Furthermore, pineacytomas are traditionally considered radioresistant tumors. For all these reasons, the use of radiosurgery in the treatment of pineocytoma is more and more growing and diffusing worldwide and studies reporting the experience of different institutions are now available (Table 6). The first and more relevant published series always regard treatment performed with GK device. The median follow-up period varies between 11.0 and 73.5 months. Neither neurological worsening nor GKRSrelated permanent complications were reported. In our studies, 10 patients were followed up for a median period of 54.6 months; we observed a clinical worsening in 1 case, only, due to distant tumor progression. As concerns outcome results (Table 7), the overall survival


RS dev = Radiosurgical device; N° Pt = number of treated patients; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = neurological worsening; Sympt. Compl. = Symptomatic complications; Trans. = transient; Perm. = Permanent; GK = Gamma Knife.

†: Departments of Neurological Surgery and Radiation Oncology, The University of Pittsburgh, and Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, Pa., USA. \*: Nagoya Radiosurgery Center, Nagoya Kioritsu Hospital, and Gamma Knife Center, Komaki City Hospital, Komaki, Japan.

¶: Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, Phoenix, Arizona, USA.

‡: due to tumor distant progression.

Table 6. Clinical data in previous and present series of pineocytomas treated with GKRS.

Gamma Knife Radiosurgery in the Management of

histopathological diagnosis is assured.

**4.3 Choroid plexus papilloma (CPP)** 

neuropsychological sequelae associated with fractionated RT.

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 87

radiosurgical outcome is still present in long-term studies, as well. Therefore, we conclude that GKRS must be taken into account when considering the treatment management of pineocytoma, also as a primary choice in selected patients providing that a previous

CPPs are epithelial tumors of the choroid plexus that account for <1% of adult brain tumors (Kim et al., 2008). The majority occur within the ventricular system, the lateral ventricle being the most frequent location in children and the IVth ventricle in adults. The usual MRI findings are characterized by large, circumscribed, contrast-enhanced intraventricular tumor with occasional cyst formation and associated hydrocephalus. Tumoral calcification may be seen on plain radiographs. CPPs are slow-growing, epithelial tumors of the choroid plexus defined as grade I according to the WHO classifacation. Microsurgical resection is the preferred management for these tumors. Gross total resection is expected to be curative, with infrequent recurrence (Krishnan et al., 2004). However, complete resection is not always possible because CPPs have a deepseated location, close proximity to critical structures (e.g. brain-stem), florid vascularity, and capacity for local invasion into underlying brain parenchyma (Krishnan et al., 2004). Furthermore, microsurgery-related permanent complications are not negligible with morbidity and mortality rates up to 25% and 16.7%, respectively (Talacchi et al., 1999). Acute postoperative complications were frequent, most notably a 22% incidence of temporary swallowing dysfunction. This condition often led to placement of a percutaneous endoscopic gastrostomy tube or tracheostomy for aspiration, or both (Krishnan et al., 2004). Therefore, additional options for treatment resistant residual or recurrent tumors are needed. Additional management options in grade I CPP include repeat surgery and RT. Krishman (Krishman et al., 2004) noted that irradiation after subtotal resection was associated with a failure of local tumor control in one half of patients. They concluded that conventional fractionated RT after initial subtotal resection did not improve outcomes. The use of stereotactic radiosurgery for CPP has been described rarely. To date, there are 9 cases reported, only (Tables 8 and 9). Duke (Duke et al., 1997) published the first case of a third-ventricle CPP who underwent radiosurgery with an excellent neurologic and imaging result. Eder (Eder et al., 2001) reported one case of CPP in pediatric age with TV shrinkage following GKRS. Also Lekovic (Lekovic et al., 2007) reported one case with partial response at 95 months from radiosurgery. To date, the most numerous published series goes back to Kim (Kim et al., 2008), who treated with GKRS 11 locally or distant recurrent intracranial CPPs in 6 patients. He described rates of overall survival and local TGC of 66.7% and 36.4%, respectively, but the study included some aggressive tumors, also. To our knownledge, the present is the largest series ever reported before regarding 8 patients affected with intracranial CPP and treated with radiosurgery. They were followed up for a median period of 71.8 months; the 5-year actuarial survival and local PFS rates were 90% and 90%, respectively. Even though the number of reported patients is limited, the long lasting TGC in most cases without GKRSrelated permanent side effects strongly suggests that radiosurgery could represent a valid alternative for the treatment of intracranial CPP, especially in resistant tumors, thus avoiding invasive and high risk repeated microsurgery and the potential long term


N° Pt = number of treated patients; Ov. Surv. = Overall survival; Act. Surv. = Actuarial survival; y = years; Ov. Local TGC = Overall local tumor growth control; Act. Local PFS = Actuarial local progression-free survival.

†: Departments of Neurological Surgery and Radiation Oncology, The University of Pittsburgh, and Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, Pa., USA. \*: Nagoya Radiosurgery Center, Nagoya Kioritsu Hospital, and Gamma Knife Center, Komaki City Hospital, Komaki, Japan.

¶: Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, Phoenix, Arizona, USA.

Table 7. Radiosurgical outcome in previous and present series of pineocytomas.

ranged between 87.5% and 100%. Mori (Mori et al., 2009) is the only author who reported a 10-year actuarial survival rate of 67%. In our series, the 10-year actuarial survival rate was 92%. The local TGC and PFS were excellent in all published studies. Among 5 cases with a median follow-up of 33.5 months, Mori (Mori et al., 2009) reported that the tumors were controlled in 4 cases. Only one patient developed cerebrospinal fluid dissemination after GKRS. Lekovic (Lekovic et al., 2007) described only 1 death in a series of 8 pineocytomas. The patient died 2 months after GKRS; however, MR images obtained immediately before the patient's death showed a 75% reduction in the size of the lesion. Kano (Kano et al., 2009d) reported a 5-year local PFS rate of 100% in 13 pineocytomas observed for a median period of 54.1 months. Reyns (Reyns et al., 2006) treated 8 pineocytomas and in 7 of them a follow-up period was available (median, 37.0 months). The author reported a 100% rate of overall survival and local TGC and he suggested that GKRS may represent a useful therapeutic modality in selected cases of pineal parenchymal tumours as part of a multidisciplinary approach. Our data achieved on a series with long term observation period strengthen the extremely encouraging results already published. Hence, GKRS has showed to be a valid, effective and safe treatment modality in benign tumors with critical surgical approach and considered relatively radioresistant to conventional fractionated RT, such as the pineocytoma. The excellent radiosurgical outcome is still present in long-term studies, as well. Therefore, we conclude that GKRS must be taken into account when considering the treatment management of pineocytoma, also as a primary choice in selected patients providing that a previous histopathological diagnosis is assured.

## **4.3 Choroid plexus papilloma (CPP)**

86 Gamma Knife Radiosurgery

Pt % At 5y At 10y TGC% At 3y At 5y At 10y

Author N° Ov. Surv. Act. Surv. % Ov. Local Act. Local PFS%

Kano 2009† 13 – 92.3 – 100.0 100.0 100.0 – Mori 2009\* 5 – 100.0 67.0 80.0 – 85.0 – Lekovic 07¶ 8 87.5 – – 100.0 – – – Reyns 2006 7 100.0 – – 100.0 – – – Desmukh 04¶ 3 – – – 100.0 – – –

2002† 10 90.0 – – 100.0 – – – Kobayashi 2001\* 3 100.0 – – 100.0 – – – Subach 1998† 8 – – – 100.0 – – – Backlund 1974 2 – – – 100.0 – – – Present series 10 90.0 92.0 92.0 100.0 – 100.0 100.0

N° Pt = number of treated patients; Ov. Surv. = Overall survival; Act. Surv. = Actuarial survival; y = years; Ov. Local TGC = Overall local tumor growth control; Act. Local PFS = Actuarial local

†: Departments of Neurological Surgery and Radiation Oncology, The University of Pittsburgh, and Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, Pa., USA. \*: Nagoya Radiosurgery Center, Nagoya Kioritsu Hospital, and Gamma Knife Center, Komaki City

¶: Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph's Hospital and Medical

ranged between 87.5% and 100%. Mori (Mori et al., 2009) is the only author who reported a 10-year actuarial survival rate of 67%. In our series, the 10-year actuarial survival rate was 92%. The local TGC and PFS were excellent in all published studies. Among 5 cases with a median follow-up of 33.5 months, Mori (Mori et al., 2009) reported that the tumors were controlled in 4 cases. Only one patient developed cerebrospinal fluid dissemination after GKRS. Lekovic (Lekovic et al., 2007) described only 1 death in a series of 8 pineocytomas. The patient died 2 months after GKRS; however, MR images obtained immediately before the patient's death showed a 75% reduction in the size of the lesion. Kano (Kano et al., 2009d) reported a 5-year local PFS rate of 100% in 13 pineocytomas observed for a median period of 54.1 months. Reyns (Reyns et al., 2006) treated 8 pineocytomas and in 7 of them a follow-up period was available (median, 37.0 months). The author reported a 100% rate of overall survival and local TGC and he suggested that GKRS may represent a useful therapeutic modality in selected cases of pineal parenchymal tumours as part of a multidisciplinary approach. Our data achieved on a series with long term observation period strengthen the extremely encouraging results already published. Hence, GKRS has showed to be a valid, effective and safe treatment modality in benign tumors with critical surgical approach and considered relatively radioresistant to conventional fractionated RT, such as the pineocytoma. The excellent

Table 7. Radiosurgical outcome in previous and present series of pineocytomas.

Hasegawa

progression-free survival.

Hospital, Komaki, Japan.

Center, Phoenix, Arizona, USA.

CPPs are epithelial tumors of the choroid plexus that account for <1% of adult brain tumors (Kim et al., 2008). The majority occur within the ventricular system, the lateral ventricle being the most frequent location in children and the IVth ventricle in adults. The usual MRI findings are characterized by large, circumscribed, contrast-enhanced intraventricular tumor with occasional cyst formation and associated hydrocephalus. Tumoral calcification may be seen on plain radiographs. CPPs are slow-growing, epithelial tumors of the choroid plexus defined as grade I according to the WHO classifacation. Microsurgical resection is the preferred management for these tumors. Gross total resection is expected to be curative, with infrequent recurrence (Krishnan et al., 2004). However, complete resection is not always possible because CPPs have a deepseated location, close proximity to critical structures (e.g. brain-stem), florid vascularity, and capacity for local invasion into underlying brain parenchyma (Krishnan et al., 2004). Furthermore, microsurgery-related permanent complications are not negligible with morbidity and mortality rates up to 25% and 16.7%, respectively (Talacchi et al., 1999). Acute postoperative complications were frequent, most notably a 22% incidence of temporary swallowing dysfunction. This condition often led to placement of a percutaneous endoscopic gastrostomy tube or tracheostomy for aspiration, or both (Krishnan et al., 2004). Therefore, additional options for treatment resistant residual or recurrent tumors are needed. Additional management options in grade I CPP include repeat surgery and RT. Krishman (Krishman et al., 2004) noted that irradiation after subtotal resection was associated with a failure of local tumor control in one half of patients. They concluded that conventional fractionated RT after initial subtotal resection did not improve outcomes. The use of stereotactic radiosurgery for CPP has been described rarely. To date, there are 9 cases reported, only (Tables 8 and 9). Duke (Duke et al., 1997) published the first case of a third-ventricle CPP who underwent radiosurgery with an excellent neurologic and imaging result. Eder (Eder et al., 2001) reported one case of CPP in pediatric age with TV shrinkage following GKRS. Also Lekovic (Lekovic et al., 2007) reported one case with partial response at 95 months from radiosurgery. To date, the most numerous published series goes back to Kim (Kim et al., 2008), who treated with GKRS 11 locally or distant recurrent intracranial CPPs in 6 patients. He described rates of overall survival and local TGC of 66.7% and 36.4%, respectively, but the study included some aggressive tumors, also. To our knownledge, the present is the largest series ever reported before regarding 8 patients affected with intracranial CPP and treated with radiosurgery. They were followed up for a median period of 71.8 months; the 5-year actuarial survival and local PFS rates were 90% and 90%, respectively. Even though the number of reported patients is limited, the long lasting TGC in most cases without GKRSrelated permanent side effects strongly suggests that radiosurgery could represent a valid alternative for the treatment of intracranial CPP, especially in resistant tumors, thus avoiding invasive and high risk repeated microsurgery and the potential long term neuropsychological sequelae associated with fractionated RT.

Gamma Knife Radiosurgery in the Management of

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 89

usually arise from the neuronal cells of the septum pellucidum, fornix, or subependymal plate of the lateral and third ventricle, so these tumors are surrounded by cerebrospinal fluid and occur as a small tumor attached to normal structures. CN is defined as grade II according to the WHO classification. This neoplasm is generally considered a benign, slowgrowing tumor. Histological study shows a pattern composed of uniform small round cells, usually with clear cytoplasm and round nuclei (Figure 7). Its features match those of oligodendroglioma or ependymoma, leading to frequent misdiagnosis when further investigations are not performed. Immunohistochemistry is very useful for the diagnosis of neurocytoma, which shows a dot-like positivity for synaptophysin, negativity for glial fibrillary acidic protein and constant expression of neuron-specific enolase (Martin et al., 2003). Typical neurocytomas are characterized by a MIB-1 labeling index ≤3% and the absence of histologic atypia (Rades & Schild, 2006). Standard initial treatment for CNs is a total resection whenever possible. The prognosis is usually favorable after gross total resection, generally leading to cure and long-term survival. Furthermore, tumor resection not only provides a histological diagnosis but also restores the intracranial cerebrospinal fluid circulation. But, tumors in more than half of patients with CN cannot be completely resected. Furthermore, recurrences, even after complete resection, and tumor progression after subtotal resection have been reported up to 33% of cases (Matsunaga et al., 2010, Kim et al., 2007, Yen et al., 2007, Martin et al., 2003, Cobery et al., 2001). Finally, malignant transformation is known to occur in a few cases, resulting in tumor progression, intracerebral hemorrhage, or craniospinal dissemination (Matsunaga et al., 2010). For treatment of residual or recurrent CNs, fractionated RT has been advocated because these tumors tend to have high radiosensitivity due to their high vascularity. The results showed effective local tumor control. However, because of the benign clinical course of CN, the young mean age of the affected patients and the well-known long-term adverse effects of conventional RT, such as cognitive dysfunction and secondary tumor formation, routine use of conventional RT for residual or recurrent CN has been critized by several authors (Kim et al., 2007, Yen et al., 2007, Martin et al., 2003). The advantage of a focused radiation is that a

Fig. 7. CN: X100 H-E. Olicodendroglial like cells with round nuclei and clear haloes.


RS dev = Radiosurgical device; N° Pt = number of treated patients; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = Neurological worsening; Perm. Sympt. Compl. = permanent symptomatic complications; GK = Gamma Knife.

‡: 3/4 due to tumor distant progression and 1/4 due to local and distant progression.

Table 8. Clinical data in previous and present series of CPP treated with GKRS.


N° Pt = number of treated patients; Ov. Surv. = overall survival; Act. Surv. = actuarial survival; Ov. Local TGC = overall local tumor growth control; Act. Local PFS = Actuarial local progression-free survival; Ped. Pt. = Pediatric patient.

‡: 2 deaths: 1 due to tumor distant progression and 1 due to local and distant progression.

Table 9. Radiosurgical outcome in previous and present series of CPP.

## **4.4 Central neurocytoma (CN)**

CN was characterized by Hassoun (Hassoun et al., 1982) in the 1980's as a distinct histological entity with a typical immunohistochemical profile and ultrastructural features of neuronal differentiation. CN accounts for approximately 0.1% of all the primary CNS tumors and it is a typically disease of young adulthood, occurring in the second and third decades of life (Matsunaga et al., 2010; Tyler-Kabara et al., 2001). This unusual LGPNT

Kim 2008 GK 6 57.3/55.5 – – – –

Lekovic 2007 GK 1 96.0 – – – 0.0

Eder 2001 GK 1 – – 1 – 0.0

Duke 1997 GK 1 17.0 Excellent – – 0.0

Present series GK 8 92.6/71.8 2/8 2/8 4/8‡ 0.0

RS dev = Radiosurgical device; N° Pt = number of treated patients; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = Neurological worsening; Perm. Sympt. Compl. = permanent symptomatic complications;

‡: 3/4 due to tumor distant progression and 1/4 due to local and distant progression. Table 8. Clinical data in previous and present series of CPP treated with GKRS.

Kim 2008 6 66.7 – – 36.4 (with some aggressive

Lekovic 2007 1 100.0 – – 100.0 – – Eder 2001 1 100.0 – – 100.0 (Ped. Pt.) – – Duke 1997 1 100.0 – – 100.0 – –

series 8 75.0‡ 90.0 80.0 87.5 90.0 90.0

CN was characterized by Hassoun (Hassoun et al., 1982) in the 1980's as a distinct histological entity with a typical immunohistochemical profile and ultrastructural features of neuronal differentiation. CN accounts for approximately 0.1% of all the primary CNS tumors and it is a typically disease of young adulthood, occurring in the second and third decades of life (Matsunaga et al., 2010; Tyler-Kabara et al., 2001). This unusual LGPNT

N° Pt = number of treated patients; Ov. Surv. = overall survival; Act. Surv. = actuarial survival; Ov. Local TGC = overall local tumor growth control; Act. Local PFS = Actuarial local progression-free

‡: 2 deaths: 1 due to tumor distant progression and 1 due to local and distant progression.

Table 9. Radiosurgical outcome in previous and present series of CPP.

No def. /Impr Stable Wors. Compl.

Surv. Act. Surv. % Ov. Local Act. Local PFS%

tumors) – –

% At 5y At 10y TGC% At 5y At 10y

Clinical Results

Perm. Sympt.

Me/Med FU mos.

Author RS

GK = Gamma Knife.

Author N°

Present

Pt

survival; Ped. Pt. = Pediatric patient.

**4.4 Central neurocytoma (CN)** 

Ov.

dev

N° Pt

usually arise from the neuronal cells of the septum pellucidum, fornix, or subependymal plate of the lateral and third ventricle, so these tumors are surrounded by cerebrospinal fluid and occur as a small tumor attached to normal structures. CN is defined as grade II according to the WHO classification. This neoplasm is generally considered a benign, slowgrowing tumor. Histological study shows a pattern composed of uniform small round cells, usually with clear cytoplasm and round nuclei (Figure 7). Its features match those of oligodendroglioma or ependymoma, leading to frequent misdiagnosis when further investigations are not performed. Immunohistochemistry is very useful for the diagnosis of neurocytoma, which shows a dot-like positivity for synaptophysin, negativity for glial fibrillary acidic protein and constant expression of neuron-specific enolase (Martin et al., 2003). Typical neurocytomas are characterized by a MIB-1 labeling index ≤3% and the absence of histologic atypia (Rades & Schild, 2006). Standard initial treatment for CNs is a total resection whenever possible. The prognosis is usually favorable after gross total resection, generally leading to cure and long-term survival. Furthermore, tumor resection not only provides a histological diagnosis but also restores the intracranial cerebrospinal fluid circulation. But, tumors in more than half of patients with CN cannot be completely resected. Furthermore, recurrences, even after complete resection, and tumor progression after subtotal resection have been reported up to 33% of cases (Matsunaga et al., 2010, Kim et al., 2007, Yen et al., 2007, Martin et al., 2003, Cobery et al., 2001). Finally, malignant transformation is known to occur in a few cases, resulting in tumor progression, intracerebral hemorrhage, or craniospinal dissemination (Matsunaga et al., 2010). For treatment of residual or recurrent CNs, fractionated RT has been advocated because these tumors tend to have high radiosensitivity due to their high vascularity. The results showed effective local tumor control. However, because of the benign clinical course of CN, the young mean age of the affected patients and the well-known long-term adverse effects of conventional RT, such as cognitive dysfunction and secondary tumor formation, routine use of conventional RT for residual or recurrent CN has been critized by several authors (Kim et al., 2007, Yen et al., 2007, Martin et al., 2003). The advantage of a focused radiation is that a

Fig. 7. CN: X100 H-E. Olicodendroglial like cells with round nuclei and clear haloes.

Gamma Knife Radiosurgery in the Management of

RS dev N° Pt

GK

Author

Matsunaga

Rades 2006 (reviews)

Tyler-Kabara

Anderson

Bertalanffy

Knife; Linac = Linear accelerator.

**4.5 Pleomorphic xanthoastrocyroma (PXA)** 

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 91

<sup>2010</sup>GK 7 63.6 Me – – – 0.0 Kim 2007 GK 13 53.7/61.0 – 100.0 – 0.0 Yen 2007 GK 7 60.0 Me 100.0 – – 0.0 Lekovic 2007 GK 1 54.0 – 1/1 – 0.0

Martin 2003 Linac 4 33.0/37.5 75.0 – 25.0 25.0 Javedan 2003 GK 1 25.0 1/1 – – 0.0 Kim 2003 Linac 1 51.0 – – – 0.0 Hara 2003 GK 1 12.0 – – – 0.0

<sup>2001</sup>GK 4 45.7/46.0 – 100.0 – 0.0

<sup>2001</sup>GK 4 16.5/13.0 100.0 – – 0.0

<sup>2001</sup>GK 3 32.0/24.0 67.0 33.0 – 0.0 Cobery 2001 GK 4 44.0/32.5 100.0 – – 0.0 Pollock 2001 GK 1 34.0 1/1 – – 0.0 Present series GK 4 172.0/180.6 4/4 – – 0.0 RS dev = Radiosurgical device; N° Pt = number of treated patients; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. =

Neurological worsening; Perm. Sympt. Compl. = permanent symptomatic complications; GK = Gamma

PXA is a rare tumor accounting for less than 1% of all astrocytic neoplasms. At MRI, PXA does not show peculiar imaging features. This unusual LGPNT of the brain belongs to grade II according to WHO classification. Histopathological patterns includes cellular pleomorphism, focal eosinophilic protein droplets, and regions with an interstitial reticulin fiber network (Figure 8). Maximum surgical removal is considered the first treatment of choice. The efficacy of adjuvant radiotherapy has not yet been established, largely because of the relative rarity of this disease. Chemotherapy for PXA has been generally considered ineffective. The only described case of PXA treated with GKRS showed anaplastic features (Koga et al., 2009). Nevertheless, the authors chose to perform radiosurgery on 8 distinct intracranial tumor nodules in six different GKRS sessions during a 50 month follow up

Table 10. Clinical data in previous and present series of CN treated with GKRS.

/Impr

Linac 15 6 42 Med – – – –

Clinical Results

Perm. Sympt.

Stable Wors. Compl.

Me/Med FU mos.

No def.

high dose with a steep fall off in radiation can be delivered precisely in one single treatment session. In addition, the residual or recurrent CNs are usually characterised by highly vascularized and well demarkated small volumes with generally intraventricular growth; therefore, they tend to be surrounded by CSF, and only a small part of the tumor has contact with the brain tissue. This makes them ideal targets for GKRS, as most of the dose surplus hits the CSF and the radiation burden to the neighbouring brain tissue can be kept at a minimum. In this way, the side effects of radiosurgery are minimised. All these advantages have led to the emergence of GKRS as an attractive alternative to conventional RT in the treatment of residual or recurrent CNs. The first CN patient treated with radiosurgery was described by Schild (Schild et al., 1997) in 1997. Since 1997, several studies of CNs treated with radiosurgery have demonstrated good response and high TGC. The mechanism of tumor response to GKRS is not yet clear. Several hypothesis have been formulated to explain the TV shrinkage: 1) cytotoxic effect; 2) accelerated programmed cell death; or 3) obliteration of nutritive vessels (Yen et al., 2007). The latter was reported by Kulkarni (Kulkarni et al., 2002) after they observed a decrease in contrast enhancement in the tumors after RT. The shrinkage of the residual neurocytoma began approximately 1 year after fractionated RT; this timing corresponds to the endothelial damage caused by irradiation which induces the proliferation of smooth muscle cells and the production of extracellular collagen by these cells, which leads to progressive stenosis and obliteration of the cAVM nidus. In addition, the contractile activity of these gamma ray–activated, spindle–shaped smooth muscle cells and the transformation of the resting cells into an activated form after irradiation may be relevant to the shrinking process and eventual occlusion of AVMs after radiosurgery. The activation of this obliteration mechanism is earlier in pediatric patients and young adults than in adults due to the greater number of nonresting cells found in young patients' vessels, as reported in an our previous study on cerebral arteriovenous malformations (Nicolato et al., 2005). To date, more than 50 patients treated with radiosurgery for CN has been described (Table 10). From the neurological point of view, the outcome was favourable in most cases. Only in one patient, a radiosurgery-related clinical worsening occurred: she was treated with a 6 MV LINAC system for a 8.09 cc left lateral ventricle CN. Radiation necrosis and brain edema were developed and ventriculo-peritoneal shunt was needed (Martin et al., 2003). The overall survival reported by the different authors varies between 67% and 100%, but none of the patients died due to the tumor progression. Rades (Rades & Schild, 2006) reviewed the data of all the CN patients reported since 1997. There were 21 cases treated with incomplete tumor resection followed by stereotactic radiosurgery, GK in 15 and Linac in 6 CNs. The median follow up period was 42 months. The 5-year actuarial survival and local PFS rates was 100% and 100%, respectively. The 4 patients of our series were observed for a median period of 180.6 months. The 10-year survival rate was 100%. Local tumor control was achieved in 75% of cases: in only one patient, tumor progression was documented at 9.3 months from GKRS. He was operated and he was alive and well at last follow-up. In conclusion, we suggest GKRS for residual CNs after incomplete resection or early detection of tumor recurrence with relatively small volume, which will reduce the long-term risk of radiation injury to the surrounding normal brain tissue compared with conventional RT. Because CN tends to show local recurrence leading to clinical malignant sequelae such as tumor progression, intracranial hemorrhage, or craniospinal dissemination and rarely malignant transformation, we strongly recommend GKRS for small residual or recurrent tumors rather than conservative follow up to obtain good tumor growth control.

high dose with a steep fall off in radiation can be delivered precisely in one single treatment session. In addition, the residual or recurrent CNs are usually characterised by highly vascularized and well demarkated small volumes with generally intraventricular growth; therefore, they tend to be surrounded by CSF, and only a small part of the tumor has contact with the brain tissue. This makes them ideal targets for GKRS, as most of the dose surplus hits the CSF and the radiation burden to the neighbouring brain tissue can be kept at a minimum. In this way, the side effects of radiosurgery are minimised. All these advantages have led to the emergence of GKRS as an attractive alternative to conventional RT in the treatment of residual or recurrent CNs. The first CN patient treated with radiosurgery was described by Schild (Schild et al., 1997) in 1997. Since 1997, several studies of CNs treated with radiosurgery have demonstrated good response and high TGC. The mechanism of tumor response to GKRS is not yet clear. Several hypothesis have been formulated to explain the TV shrinkage: 1) cytotoxic effect; 2) accelerated programmed cell death; or 3) obliteration of nutritive vessels (Yen et al., 2007). The latter was reported by Kulkarni (Kulkarni et al., 2002) after they observed a decrease in contrast enhancement in the tumors after RT. The shrinkage of the residual neurocytoma began approximately 1 year after fractionated RT; this timing corresponds to the endothelial damage caused by irradiation which induces the proliferation of smooth muscle cells and the production of extracellular collagen by these cells, which leads to progressive stenosis and obliteration of the cAVM nidus. In addition, the contractile activity of these gamma ray–activated, spindle–shaped smooth muscle cells and the transformation of the resting cells into an activated form after irradiation may be relevant to the shrinking process and eventual occlusion of AVMs after radiosurgery. The activation of this obliteration mechanism is earlier in pediatric patients and young adults than in adults due to the greater number of nonresting cells found in young patients' vessels, as reported in an our previous study on cerebral arteriovenous malformations (Nicolato et al., 2005). To date, more than 50 patients treated with radiosurgery for CN has been described (Table 10). From the neurological point of view, the outcome was favourable in most cases. Only in one patient, a radiosurgery-related clinical worsening occurred: she was treated with a 6 MV LINAC system for a 8.09 cc left lateral ventricle CN. Radiation necrosis and brain edema were developed and ventriculo-peritoneal shunt was needed (Martin et al., 2003). The overall survival reported by the different authors varies between 67% and 100%, but none of the patients died due to the tumor progression. Rades (Rades & Schild, 2006) reviewed the data of all the CN patients reported since 1997. There were 21 cases treated with incomplete tumor resection followed by stereotactic radiosurgery, GK in 15 and Linac in 6 CNs. The median follow up period was 42 months. The 5-year actuarial survival and local PFS rates was 100% and 100%, respectively. The 4 patients of our series were observed for a median period of 180.6 months. The 10-year survival rate was 100%. Local tumor control was achieved in 75% of cases: in only one patient, tumor progression was documented at 9.3 months from GKRS. He was operated and he was alive and well at last follow-up. In conclusion, we suggest GKRS for residual CNs after incomplete resection or early detection of tumor recurrence with relatively small volume, which will reduce the long-term risk of radiation injury to the surrounding normal brain tissue compared with conventional RT. Because CN tends to show local recurrence leading to clinical malignant sequelae such as tumor progression, intracranial hemorrhage, or craniospinal dissemination and rarely malignant transformation, we strongly recommend GKRS for small residual or recurrent tumors rather than conservative follow up to obtain

good tumor growth control.


RS dev = Radiosurgical device; N° Pt = number of treated patients; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = Neurological worsening; Perm. Sympt. Compl. = permanent symptomatic complications; GK = Gamma Knife; Linac = Linear accelerator.

Table 10. Clinical data in previous and present series of CN treated with GKRS.

## **4.5 Pleomorphic xanthoastrocyroma (PXA)**

PXA is a rare tumor accounting for less than 1% of all astrocytic neoplasms. At MRI, PXA does not show peculiar imaging features. This unusual LGPNT of the brain belongs to grade II according to WHO classification. Histopathological patterns includes cellular pleomorphism, focal eosinophilic protein droplets, and regions with an interstitial reticulin fiber network (Figure 8). Maximum surgical removal is considered the first treatment of choice. The efficacy of adjuvant radiotherapy has not yet been established, largely because of the relative rarity of this disease. Chemotherapy for PXA has been generally considered ineffective. The only described case of PXA treated with GKRS showed anaplastic features (Koga et al., 2009). Nevertheless, the authors chose to perform radiosurgery on 8 distinct intracranial tumor nodules in six different GKRS sessions during a 50 month follow up

Gamma Knife Radiosurgery in the Management of

N° Pt

No def.

Me/Med FU mos.

RS dev

Author

Present

survival.

radiosurgical treatment.

**4.6 Others (Miscellaneous)** 

GK = Gamma Knife.

‡: due to tumor distant progression.

‡: dead due to tumor local and distant progression.

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 93

Koga 2009 GK 1 50.0 – – – 0.0

series GK 3 45.6/43.7 1/3 1/3 1/3‡ 0.0

RS dev = Radiosurgical device; N° Pt = number of treated patients; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = Neurological worsening; Perm. Sympt. Compl. = permanent symptomatic complications;

Table 12. Clinical data in previous and present series of PXA treated with GKRS.

Author N° Ov. Surv Act. Surv % Ov. Local Act. Local PFS%

period achieving a successful local control of all treated lesions without radiosurgeryrelated permanent side effects. The patient died 66 months after the disease onset; the cause of death was identified with a distant craniospinal axis tumor nodule dissemination. In this chapter, we report 3 patients affected with grade II PXA who underwent GKRS and followed up for a median period of 43.7 months. To our knowledge, this is the only reported "small" series of low-grade PXA treated with GKRS. One patient affected with a brain stem PXA died at 22.1 months from GKRS due to local and distant tumor progression. In the other 2 cases, a long term PFS was achieved and they are well and alive at last follow up performed at 43.7 months and 70.9 months from

This heterogeneous group of patients is represented by different histological types of extremely unusual LGPNT of the brain: gangliolgioma, mixed oligoastrocytoma, subependymoma, and papillary glioneuronal tumor. The results of our experience with GKRS is excellent: negative or stable neurologic conditions in all cases, no radiosurgeryrelated complications, 100% survival and local TGC rate. Obviously, other experiences with GKRS on such rare tumors are limited or even absent (Tables 14 and 15). In the few reported cases, the authors described encouraging responses to radiosurgery, as well. The results

Koga 2009 1 0.0 – – 100.0 – – Present series 3 66.7‡ 80.0 80.0 66.7 80.0 80.0

N° Pt = number of treated patients; Ov. Surv. = overall survival; Act. Surv. = actuarial survival; Ov. Local TGC = Overall local tumor growth control; Act. Local PFS = Actuarial local progression-free

Table 13. Radiosurgical outcome in previous and present series of PXA.

Pt % At 5y At 10y TGC% At 5y At 10y

/Impr

Clinical Results

Perm. Sympt.

Stable Wors. Compl.


N° Pt = number of treated patients; Ov. Surv. = overall survival; Act. Surv. = actuarial survival; Ov. Local TGC = Overall local tumor growth control; Act. Local PFS = Actuarial local progression-free survival.

Table 11. Radiosurgical outcome in previous and present series of CN.

Fig. 8. PXA. X200 H-E. Presence of pleomorphic bizarre xanthomized cells in a fibrillary background.


RS dev = Radiosurgical device; N° Pt = number of treated patients; Me = Mean; Med = Median; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = Neurological worsening; Perm. Sympt. Compl. = permanent symptomatic complications; GK = Gamma Knife.

‡: due to tumor distant progression.

92 Gamma Knife Radiosurgery

Author N° Ov. Surv. Act. Surv. % Ov.Local Act. Local PFS%

Matsunaga 2010 7 – – – 87.5 – – Kim 2007 13 – – – 84.6 – – Yen 2007 7 85.7 – – 88.9 – – Lekovic 07 1 – – – 1/1 – – Rades 2006 (rews.) 156 100.0 100.0 – 95.2 100.0 – Martin 2003 4 100.0 – – 100.0 – – Javedan 2003 1 1/1 – – 1/1 – – Kim 2003 1 1/1 – – 1/1 – – Hara 2003 1 1/1 – – 1/1 – – Tyler-Kabara 2001 4 100.0 – – 100.0 – – Anderson 2001 4 100.0 – – 100.0 – – Bertalanffy 2001 3 67.0 – – 100.0 – – Cobery 2001 4 100.0 – – 100.0 – – Pollock 2001 1 1/1 – – 1/1 – – Present series 4 100.0 100.0 100.0 75.0 83.0 83.0

N° Pt = number of treated patients; Ov. Surv. = overall survival; Act. Surv. = actuarial survival; Ov. Local TGC = Overall local tumor growth control; Act. Local PFS = Actuarial local progression-free

Fig. 8. PXA. X200 H-E. Presence of pleomorphic bizarre xanthomized cells in a fibrillary

Table 11. Radiosurgical outcome in previous and present series of CN.

survival.

background.

Pt % At 5y At 10y TGC% At 5y At 10y

Table 12. Clinical data in previous and present series of PXA treated with GKRS.


N° Pt = number of treated patients; Ov. Surv. = overall survival; Act. Surv. = actuarial survival; Ov. Local TGC = Overall local tumor growth control; Act. Local PFS = Actuarial local progression-free survival.

‡: dead due to tumor local and distant progression.

Table 13. Radiosurgical outcome in previous and present series of PXA.

period achieving a successful local control of all treated lesions without radiosurgeryrelated permanent side effects. The patient died 66 months after the disease onset; the cause of death was identified with a distant craniospinal axis tumor nodule dissemination. In this chapter, we report 3 patients affected with grade II PXA who underwent GKRS and followed up for a median period of 43.7 months. To our knowledge, this is the only reported "small" series of low-grade PXA treated with GKRS. One patient affected with a brain stem PXA died at 22.1 months from GKRS due to local and distant tumor progression. In the other 2 cases, a long term PFS was achieved and they are well and alive at last follow up performed at 43.7 months and 70.9 months from radiosurgical treatment.

## **4.6 Others (Miscellaneous)**

This heterogeneous group of patients is represented by different histological types of extremely unusual LGPNT of the brain: gangliolgioma, mixed oligoastrocytoma, subependymoma, and papillary glioneuronal tumor. The results of our experience with GKRS is excellent: negative or stable neurologic conditions in all cases, no radiosurgeryrelated complications, 100% survival and local TGC rate. Obviously, other experiences with GKRS on such rare tumors are limited or even absent (Tables 14 and 15). In the few reported cases, the authors described encouraging responses to radiosurgery, as well. The results

Gamma Knife Radiosurgery in the Management of

Histology Author RS dev N° Pt

Schröttner

Present

Present

Present

Present

LGPNTs, particularly as concerns pineocytoma and CN.

**4.7 Future research/perspectives** 

Papillary Gl. Tumour

Ganglioglioma Kim 1999 GK 1 – 1/1 solid

<sup>2002</sup>GK 3

Subependymoma Ecker 2004 GK 1 1/1 1/1 – –

Oligoastrocytoma Sarkar 2002 GK 11 les – – 42.0 Younger age

RS dev = Radiosurgical device; N° Pt = number of treated patients; Ov. Surv. = Overall survival; Ov. Local TGC = Overall local tumor growth control; Act. Local PFS = Actuarial local progression-free survival; GK = Gamma Knife; compon. = component; Lin = Linear accelerator; TVs = Tumor volumes. Table 15. Radiosurgical outcome in previous and present series of rare unusual LGPNTs.

The results described with the GKRS treatment in such unusual LGPNTs are very intersting. Nevertheless, the need for further future perspectives and researches emerge from all these studies. First of all, the application of multisession GKRS with the Extend system (Elekta instruments, AB) on larger TVs with the aim of increasing the local TGC without exposing the patients to higher risk of radiation toxicity on the surrounding normal brain tissue need to be studied. Second, it should be interesting to investigate if there is any potential correlation between biomarker expression (Ki67, MGMT, PCNA, p53, etc.) in such unusual LGPNTs and GKRS outcome. Finally, basing on the literature data and our experience results, it should be suitable that GKRS is included in the international guidelines for good clinical practice as part of the therapeutic armamentarium for the management of unusual

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 95

Ov. Surv.

series GK 2 2/2 2/2 – None

Seol 2003 Lin 1 1/1 0/1 – – Im 2003 GKLin 2 2/2 0/2 – – Roos 2000 Lin 1 1/1 1/1 – –

series GK 1 1/1 1/1 – None

series GK 1 1/1 1 100.0 None

series GK 1 1/1 1/1 100.0 None

Ov. Local TGC%

Act. Local PFS%

At 5y

compon. – –

Prognostic Factors

Smaller TVs

seems to be particularly encouraging when GKRS is employed. To our knowledge, there are no other patients affected with papillary glioneuronal tumor who underwent GKRS reported in the literature; therefore, the case of the present series represents the first papillary glioneuronal tumor treated with radiosurgery.


RS dev = Radiosurgical device; N° Pt = number of treated patients; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = Neurological worsening; Perm. Sympt. Compl. = permanent symptomatic complications; GK = Gamma Knife; Lin = Linear accelerator; les = lesions; Papillary Gl. Tumour = Papillary Glioneuronal Tumour.

Table 14. Clinical data in previous and present series of rare unusual LGPNTs treated with GKRS.

seems to be particularly encouraging when GKRS is employed. To our knowledge, there are no other patients affected with papillary glioneuronal tumor who underwent GKRS reported in the literature; therefore, the case of the present series represents the first

> FU mos.

> > /Impr

<sup>1999</sup>GK 1 14.0 – – – 0.0

<sup>2002</sup>GK 3 – – – – –

series GK 2 68.4 – 2/2 – 0.0

<sup>2004</sup>GK 1 54.0 – 1/1 – 0.0

<sup>2003</sup>Lin 1 24 – – – 0.0

<sup>2000</sup>Lin 1 16 1/1 – – 0.0

series GK 1 31.2 1/1 – – 0.0

<sup>2002</sup>GK 11 les – – – – 0.0

series GK 1 68.3 1 – – 0.0

series GK 1 104.0 1/1 – – 0.0

RS dev = Radiosurgical device; N° Pt = number of treated patients; mos. = months; No def./Impr = No pre-GK neurological deficit/Neurological improvement; Wors. = Neurological worsening; Perm. Sympt. Compl. = permanent symptomatic complications; GK = Gamma Knife; Lin = Linear accelerator;

Table 14. Clinical data in previous and present series of rare unusual LGPNTs treated with

Lin 2 22.5 – – – 0.0

dev Pt No def.

Clinical Results

Perm. Sympt.

Stable Wors. Compl.

papillary glioneuronal tumor treated with radiosurgery.

Histology Author RS N°

Schröttner

Im 2003 GK

Present

Seol

Roos

Present

Present

Present

les = lesions; Papillary Gl. Tumour = Papillary Glioneuronal Tumour.

Ganglioglioma Kim

Subependymoma Ecker

Oligoastrocytoma Sarkar

Papillary Gl. Tumour

GKRS.


RS dev = Radiosurgical device; N° Pt = number of treated patients; Ov. Surv. = Overall survival; Ov. Local TGC = Overall local tumor growth control; Act. Local PFS = Actuarial local progression-free survival; GK = Gamma Knife; compon. = component; Lin = Linear accelerator; TVs = Tumor volumes.

Table 15. Radiosurgical outcome in previous and present series of rare unusual LGPNTs.

## **4.7 Future research/perspectives**

The results described with the GKRS treatment in such unusual LGPNTs are very intersting. Nevertheless, the need for further future perspectives and researches emerge from all these studies. First of all, the application of multisession GKRS with the Extend system (Elekta instruments, AB) on larger TVs with the aim of increasing the local TGC without exposing the patients to higher risk of radiation toxicity on the surrounding normal brain tissue need to be studied. Second, it should be interesting to investigate if there is any potential correlation between biomarker expression (Ki67, MGMT, PCNA, p53, etc.) in such unusual LGPNTs and GKRS outcome. Finally, basing on the literature data and our experience results, it should be suitable that GKRS is included in the international guidelines for good clinical practice as part of the therapeutic armamentarium for the management of unusual LGPNTs, particularly as concerns pineocytoma and CN.

Gamma Knife Radiosurgery in the Management of

Vol.48, pp 518-523.

13-24.

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 97

Friedman, J.A., Lynch, J.J., Buckner, J.C., Scheithauer, B.W., Raffel, C. (2001). Managment of

Grimoldi, N., Tomei, G., Stankov, B., Lucini, V., Masini, B., Caputo, V., Repetti, M.L.,

Hadjipanayis, C.G., Niranjan, A., Tyler-Kabara, E., Kondziolka, D., Flickinger, J.C.,

Hara, M., Aoyagi, M., Yamamoto, M., Maehara, T., Takada, Y., Nojiri, T., Ohno, K. (2003).

Hasegawa, T., Kondziolka, D., Hadjipanayis, C.G., Flickinger, J.C., Lunsford, L.D. (2002).

Hassoun, J., Gambarelli, D., Grisoli, F., Pellet, W., Salamon, G., Pellissier, J.F., Toga, M.

Heppner, P.A., Sheehan, J.P., Steiner, L.E. (2005). Gamma knife surgery for low-grade

Im, S.-H., , Paek, S.H., Choi, Y.-L., Chi, J.G., Kim, D.G., Jung, H.W., Cho, B.-K. (2003).

Javedan, S.P., Manwaring, K., Smith, K.A. (2003). Treatment of posterior third ventricular

and stereotactic radiosurgery. *Minim Invasive Neurosurg,* Vol.46, pp 165–168. Kano, K., Kondziolka, D., Niranjan, A., Flickinger, J.C., Lunsford, L.D. (2009). Stereotactic

Kano, K., Niranjan, A., Kondziolka, D., Flickinger, J.C., Pollack, I., Jakacki, R., Lunsford, L.D.

Kano, H., Niranjan, A., Khan, A., Flickinger, J.C., Kondziolka, D., Lieberman, F., Lunsford,

Kano, H., Niranjan, A., Kondziolka, D., Flickinger, J.C., Kondziolka, D., Lunsford, L.D.

multimodal therapy is necessary. *J Neurosurg*, Vol.97, pp 56-64.

a case report. *J Neurooncol,* Vol.62, pp 269–273.

gliomas. Neurosurgery, Vol.57, No.6, pp 1132-1139.

pediatric patients. *J Neurooncol*, Vol.95, pp 219-229.

oligodendrogliomas? *J Neurosurg*, Vol.110, pp 564–571.

subependymoma. *Journal of Neuro-Oncology*, Vol.61, pp 57–67.

*Neurosurgery*, Vol.51, pp 880-889.

*Neuropathol,* Vol.56, pp 151–156.

*Neurooncol*, Vol.95, pp 211-218.

Basel, Switzerland.

malignant pineal germ cell tumors with residual mature teratome. *Neurosurgery*,

Lazzarini, G., Gaini, S.M. (1998). Neuroendocrine , himmunohistochemical, and ultrastructural study of pineal region tumors. *J Pineal Res,* Vol.25, pp 147-158. Hadjipanayis, C.D., Kondziolka, D., Gardner, P., Niranjan, A., Dagam, S., Flickinger, J.C.,

Lunsford, L.D. (2002). Stereotactic radiosurgery for pilocytic astrocytomas when

Lunsford, L.D. (2002). Stereotactic radiosurgery for well-circumscribed fibrillary grade II astrocytomas: an initial experience. *Stereotactic Funct Neurosurg*, Vol.79, pp

Rapid shrinkage of remnant central neurocytoma after gamma knife radiosurgery:

The role of radiosurgery for the treatment of pineal parenchymal tumors.

(1982). Central neurocytoma. An electron-microscopic study of two cases. *Acta* 

Clinicopathological study of seven cases of symptomatic supratentorial

central neurocytoma with endoscopic biopsy, endoscopic third ventriculostomy

radiosurgery for pilocytic astrocytomas part 1: outcome in adult patients. *J* 

(2009) Stereotactic radiosurgery for pilocytic astrocytomas part 2: outcome in

L.D. (2009). Does radiosurgery have a role in the management of

(2009). Role of stereotactic radiosurgery in the managment of pineal parenchymal tumors, In: *Pineal Region Tumors. Diagnosis and treatment options,* T. Kobayashi & L.D. Lunsford, (Eds.), 44-58, ISSN 0079-6492, S. Karger AG, ISBN 978-3-8055-9077-8,

## **5. Conclusion**

GKRS may be already considered an effective and safe treatment alternative in multimodality approach for selected cases with pineocytoma and CNs, thus eliminating the need for reoperation of residual or recurrent tumors and avoiding the potential long-term side effects of conventional RT in these young adult patients. In the unusual LGPNTs with a limited number of treated patients – CPP, PXA, and other rare tumors – radiosurgery seems to be a valid complementary treatment tool in these rarer tumors, also. Nevertheless, multidisciplinary studies on large series of patients and long follow-up period with statistical analysis are needed to convincingly demonstrate the efficacy and safety of GKRS on such primary low-grade brain tumors.

## **6. Acknowledgment**

The authors thank Giovanni Nicolato, for his assistance in the preparation of this manuscript.

## **7. References**


GKRS may be already considered an effective and safe treatment alternative in multimodality approach for selected cases with pineocytoma and CNs, thus eliminating the need for reoperation of residual or recurrent tumors and avoiding the potential long-term side effects of conventional RT in these young adult patients. In the unusual LGPNTs with a limited number of treated patients – CPP, PXA, and other rare tumors – radiosurgery seems to be a valid complementary treatment tool in these rarer tumors, also. Nevertheless, multidisciplinary studies on large series of patients and long follow-up period with statistical analysis are needed to convincingly demonstrate the efficacy and safety of GKRS

The authors thank Giovanni Nicolato, for his assistance in the preparation of this

Anderson, R.C., Elder, J.B., Parsa, A.T., Issacson, S.R., Sisti, M.B. (2001). Radiosurgery for the treatment of recurrent central neurocytomas. *Neurosurgery,* Vol.48, pp 1231–1238. Backlund, E.O., Rahn, T., Sarby, B. (1974). Treatment of pinealomas by stereotaxic radiation

Bertalanffy, A., Roessler, K., Dietrich, W., Aichholzer, M., Prayer, D., Ertl, A., Kitz, K. (2001).

Boethius, J., E Ulfarsson, T Rahn, B Lippitz. (2002). Gamma Knife radiosurgery for pilocytic

Breslaw, N. (1974). Covariance analysis of censored survival date. *Biometrics,* Vol.30, pp 89-

Bruce, J.N., Ogden, A.T. (2004). Surgical strategies for treating patients with pineal region

Chiechi, M.V., Smirniotopoulos, J.G., Mena, H. (1995). Pineal parenchymal tumors: CT and

Cobery, S.T., Noren, G., Friehs, G.M., Chougule, P., Zheng, Z., Epstein, M.H., Taylor, W.

Deshmukh, V.R., Smith, K.A., Rekate, H.L., Coons, S., Spetzler, R.F. (2004). Diagnosis and

Duke, B.J., Kindt, G.W., Breeze, R.E. (1997). Pineal region choroid plexus papilloma treated with stereotactic radiosurgery: a case study. *Comput Aided Surg,* Vol.2, pp 135–138. Ecker, R.D., Pollock, B.E. (2004). Recurrent Subependymoma Treated with Radiosurgery.

Eder, H.G., Leber, K.A., Eustacchio, S., Pendl, G. (2001). The role of gamma knife

management of pineocytomas. *Neurosurgery*, Vol.55, pp 349–357.

radiosurgery in children. *Childs Nerv Syst,* Vol.17, pp 341–346.

(2001). Gamma knife surgery for treatment of central neurocytomas. Report of four

Gamma knife radiosurgery of recurrent central neurocytomas: a preliminary

surgery. *Acta Radiol Ther Phys Biol,* Vol.13, pp 368–376.

report. *J Neurol Neurosurg Psychiatry,* Vol.70, pp 489–493

astrocytomas. *J Neurosurg*, Vol.(Suppl 5)97, pp 677-680.

MR features. *J Comput Assist Tomogr,* Vol.19, pp 509–517.

tumors. *J Neuro-Oncol*, Vol.69, pp 221-236.

cases. *J Neurosurg,* Vol.94, pp 327–330.

*Stereotact Funct Neurosurg*, Vol.82, pp 58–60.

**5. Conclusion** 

on such primary low-grade brain tumors.

**6. Acknowledgment** 

manuscript.

**7. References**

99.


Gamma Knife Radiosurgery in the Management of

Vol.106, pp 1140–1143.

Vol.44:, pp 85–192.

*Neurochir (Wien)* 2006;148:5–11.

AG, ISBN 978-3-8055-9077-8, Basel, Switzerland.

childhood. *J Neurosurg Sci*, Vol.41, pp 359–371.

factors. *Int J Radiat Oncol Biol Phys*, Vol.64, pp 914-921.

neurocytoma: case report. *Neurosurgery*, Vol.48, pp 441–443.

Unusual Grade I/II Primitive Neuroepithelial Tumours of the Brain 99

Nicolato, A., Gerosa, M., Ferraresi, P., Piovan, E., Pasoli, A., Perini, S., Mazza, C. (1997).

Nicolato, A., F Lupidi, MF Sandri, R Foroni, P Zampieri, C Mazza, A Pasqualin , A

Pollock, B.E., Stafford, S.L. (2001). Stereotactic radiosurgery for recurrent central

Rades, D., Schild, S.E. (2006). Value of postoperative stereotactic radiosurgery and

Reyns, N., M Hayashi, O Chinot, L Manera, J-C Peragut, S Blond, J Regis. The role of

Roos, D.E., Brophy, B.P., Zavgorodni, S.F., Francis, J.W. (2000). Radiosurgery at the Royal

Sarkar, A., Pollock, B.E., Brown, P.D., Gorman, D.A. (2002). Evaluation of gamma knife

Schild, S.E., Scheithauer, B.W., Haddock, M.G., Wong, W.W., Lyons, M.K., Marks, L.B.,

Schröttner, O., Unger, F., Eder, H.G., Feichtinger, M., Pendl, G. (2002). Gamma-Knife

Seol, H.J., Hwang, S.-K., Choi, Y.L., Chi, J.G., Jung, H.W. (2003). A case of recurrent

Simonova, G., J Novotny, R Lisckac. (2005). Low-grade gliomas treated by fractionated

Subach, B.R., Lunsford, L.D., Kondziolka, D. (1998). Stereotactic radiosurgery in the treatment of pineal region tumors. *Prog Neurol Surg,* Vol.14, pp 175–194. Talacchi, A., De Micheli, E., Lombardo, C., Turazzi, S., Bricolo, A. (1999). Choroid plexus

Tyler-Kabara, E., Kondziolka, D., Flickinger, J.C., Lunsford, L.D. (2001). Stereotactic

oligodendroastrocytomas. *J Neurosurg,* Vol.97(5 Suppl), pp 653–656. Schild, S.E., Scheithauer, B.W., Haddock, M.G., Schiff, D., Burger, P.C., Wong, W.W., Lyons,

other germ cell tumors of the brain. *Cancer,* Vol.78, pp 2564–2571.

M.K. (1997). Central neurocytomas. *Cancer,* Vol.79, 790–795.

gamma knife surgery. *J Neurosurg,* Vol.(Suppl)102, pp 19-24.

results. *Acta Neurochir Suppl.*, Vol.84, pp 49-55.

*Oncology,* Vol.62 pp 315–320.

pp 621–629.

879–882.

*options,* T. Kobayashi & L.D. Lunsford, (Eds.), 107-118, ISSN 0079-6492, S. Karger

Stereotactic radiosurgery for the treatment of arteriovenous malformations in

Beltramello, M Gerosa (2006). Gamma knife radiosurgery for cerebral arteriovenous malformations in children/adolescents and adults. Part II: differences in obliteration rates, treatment-obliteration intervals, and prognostic

conventional radiotherapy for incompletely resected typical neurocytomas. *Cancer*,

Gamma Knife radiosurgery in the treatment of pineal parenchymal tumours. *Acta* 

Adelaide Hospital: The first 4 1/2 years' clinical experience. *Australas Radiol*,

radiosurgery in the treatment of oligodendrogliomas and mixed

Norman, M.G., Burger, P.C. (1996). Histologically confirmed pineal tumors and

radiosurgery of mesiotemporal tumour epilepsy observations and long-term

subependymoma with subependymal seeding: case report. *Journal of Neuro-*

papilloma of the cerebellopontine angle: a twelve patient series. *Surg Neurol,* Vol.51,

radiosurgery for residual neurocytoma. Report of four cases. *J Neurosurg,* Vol.95, pp


Kaplan, E.L.; Meier, P. (1958). Nonparametric estimation from incomplete observations. *J* 

Kida, Y., T Kobayashi, Y Mori. (2000). Gamma knife radiosurgery for low-grade

Kim, M.S., Lee, S.I., Sim, J.H. (1999). Brain tumors with cysts treated with gamma knife

Kim, C.Y., Paek, S.H., Kim, D.G. (2003). Linear accelerator radiosurgery for central

Kim, C.Y., Paek, S.H., Jeong, S.S., Chung, H.T., Han, J.H., Park, C.K., Jung, H.W., Kim, D.G.

Kim, I-Y., Niranjan, A., Kondziolka, D., Flickinger, J.C., Lunsford, L.D. (2008). Gamma knife

Kobayashi, T., Kida, Y., Mori, Y. (2001). Stereotactic gamma radiosurgery for pineal and

Koga, T., Morita, A., Maruyama, K., Tanaka, M., Ino, J., Shibahara, J., Louis, D.N.,

Krishnan, S., Brown, P.D., Scheithauer, B.W., Ebersold, M.J., Hammack, J.E., Buckner, J.C.

Kulkarni, V., Rajshekhar, V., Haran, R.P., Chandi, S.M. (2002). Long-term outcome in

Lekovic, G.P., Gonzales, L.F., Shetter, A.G., Porter, R.W., Smith, K.A., Brachman, D.,

Louis, D.N., Ohgaki H., Wiestler, O.D., Cavenee, W.K., Burger, P.C., Jouvet, A., Scheithauer,

Lunsford, L.D., Flickinger, J., Lindner, G., Maitz, A. (1989). Stereotactic radiosurgery of the

Martin, J.M., Katati, M., Lopez, E., Bullejos, J.A., Arregui, G., Busquier, H., Minguez, A.,

Mori, Y., Kobayashi, T., Hasegawa, T., Yoshida, K., Kida, Y. (2009). Stereotactic radiosurgery

neurocytoma: a case report. *J Neurooncol,* Vol.61, pp 249–254.

stereotactic irradiation. *Neuro-Oncology,* Vol.11, pp 446–451.

Secondary Treatment. *Cancer*, Vol.110, pp 2276–84.

related tumors. *J Neurooncol*, Vol.54, pp 301–309.

therapy. *Br J Neurosurg,* Vol.16, pp 126–132.

Neurosurgery, Vol.24, pp 151–159.

region tumors. *Neurosurg Focus*, Vol.23, No.6, PP E12.

Nervous System. Acta Neuropathol, Vol.114, pp 97–109.

astrocytomas: results of long term follow-up. J Neurosurg, Vol.(Suppl 3)93, pp 42-

radiosurgery: is microsurgery indicated ? *Stereotact Funct Neurosurg*, Vol. 72

(2007). Gamma Knife Radiosurgery for Central Neurocytoma. Primary and

radiosurgery for treatment resistant choroid plexus papillomas. *J Neurooncol*,

Reifenberger, G., Itami, J., Hara, R., Saito, N., Todo, T. (2009). Long-term control of disseminated pleomorphic xanthoastrocytoma with anaplastic features by means of

(2004). Choroid plexus papillomas: a single institutional experience*. J Neurooncol,*

patients with central neurocytoma following stereotactic biopsy and radiation

Spetzler, R.F. (2007). Role of Gamma Knife surgery in the management of pineal

B.W., Kleihues, P. (2007). The 2007 WHO Classi**f**cation of Tumours of the Central

brain using the first United States 201 cobalt-60 source gamma knife.

Olivares, G:; Hernandez, V., Arjona, V. (2003). Linear accelerator radiosurgery in treatment of central neurocytomas. *Acta Neurochir (Wien),* Vol.145, pp 749–754. Matsunaga, S., Shuto, T., Suenaga, J., Inomori, S., Fujino H. (2010). Gamma Knife

Radiosurgery for Central Neurocytomas. *Neurol Med Chir (Tokio)*, Vol.50, pp 107-

for pineal and related tumors, In: *Pineal Region Tumors. Diagnosis and treatment* 

*Am Stat Assoc*, Vol.53, pp 457-481.

46.

(Suppl.), pp 38-44.

Vol.90, pp 105–110.

Vol.68, pp 49–55.

113.

*options,* T. Kobayashi & L.D. Lunsford, (Eds.), 107-118, ISSN 0079-6492, S. Karger AG, ISBN 978-3-8055-9077-8, Basel, Switzerland.


**Functional and Vascular Disorders** 

Yen, C.P., Sheehan, J., Patterson, G., Steiner, L. (2007). Gamma Knife surgery for neurocytoma. *J Neurosurg*, Vol.107, pp 7–12. **Part 2**  neurocytoma. *J Neurosurg*, Vol.107, pp 7–12. **Part 2** 

**Functional and Vascular Disorders** 

100 Gamma Knife Radiosurgery

Yen, C.P., Sheehan, J., Patterson, G., Steiner, L. (2007). Gamma Knife surgery for

**6** 

*Chile* 

**Clinical, Anatomo-Radiological and Dosimetric** 

**Features Influencing Pain Outcome After** 

*1Department of Neurosurgery, Pontificia Universidad Católica de Chile, Santiago* 

Gamma Knife radiosurgery has become nowadays a well validated and accepted option for the treatment of trigeminal neuralgia and numerous publications support its role in the management of this disease. The results achieved with this technique in terms of pain outcome are quite similar to other ablative treatments such as radiofrequency thermocoagulation, balloon micro-compression or glycerol gangliolysis, nevertheless,

In contrast to the mentioned treatments, radiosurgery does not mitigate pain immediately, existing a "latency period" for pain relief of about 2 to 6 weeks. Initially, favorable results (Barrow Neurological institute I -IIIb) are obtained in more than 80% of the cases, then, because of recurrences over time, at 3 to 5 years after the treatment, the percentage of patient which maintains this outcome is near 50%. (Kondziolka, 2010; Longhi, 2007; lópez, 2004a;

From the evidence based medicine point of view, either for radiosurgery as well as for all other medical and surgical treatment options for trigeminal neuralgia, there is lacking of comparative randomized prospective trials and the majority of publications correspond to observational data, categorized in Class III studies, then, in general the level of evidence is

The aim of the present chapter consists in a systematic review of the literature searching and categorizing information about the prognostic factors involved in pain improvement after

A search in Pub med was done crossing the key words: Gamma Knife, and Trigeminal Neuralgia. Secondarily, other key words were introduced: Radiosurgery, Multiple

**1. Introduction** 

complications seem to be less frequent (López 2004a).

poor (Cruccu, 2008; López, 2004b; Zakrzewska, 2007).

Gamma knife radiosurgery for trigeminal neuralgia.

2004b; Pollock, 2002; Regís, 2009; Sheehan, 2005; Verheul, 2010).

*3Department of Radiation Oncology, Fundación Arturo López Pérez, Santiago* 

**Gamma Knife Treatment of** 

José Lorenzoni1,2, Adrián Zárate1, Raúl de Ramón2,

Leonardo Badínez2,3, Francisco Bova2 and Claudio Lühr2

**Trigeminal Neuralgia** 

*2Centro Gamma Knife de Santiago, Santiago* 
