**2.1 High grade astrocytoma**

High grade astrocytoma (HGA) includes anaplastic astrocytoma (AA), glioblastome multiforme (GBM), giant cell GBM and gliosarcoma according to WHO (World Health Organization) classification system (Louis et al., 2007). Whilst AA is grade III, rests are grade IV tumors. AA and GBM account for 60-65% of all gliomas (Sloan et al., 2005). The overall survival for untreated GBM is only 2-3 months which increases to mean 9-12 months with addition of gross total resection and RT. Addition of chemotherapy to this modality brings approximately 5 more months. Currently, overall survival for GBM following surgical resection and RT increased to 14-19 months by addition of a latterly popularized chemotherapeutic agent temazolamide (Combs et al, 2005). Median survival for AA is about 2- 3 years with surgical resection, RT and chemotherapy. 5 years survival rate for AA is reported 18%. Most of the AA cases transform into GBM during the course of disease. The treatment approaches for HGA remains palliative, not curative. There is a general consensus for a classification system for evaluating the response of the tumor to SRS treatment (Table 2).


*(CR+PR+NC = Tumor Control Rate (TCR), CR+PR = Effectiveness)* 

Table 2. Classification of responsiveness of the tumor to SRS treatment.

Stereotactic Radiosurgery for Gliomas 281

Furthermore, they pointed that the boost radiosurgery is associated with increased long term toxicity. They also reported that there is not sufficient evidence yet to show the effectiveness of SRS on recurrent or progressive malignant glioma (Anker et al., 2010; Tsao

A multicentric study including 46 patients on CyberKnife comparing the use as a boost with salvage reported median overall survival of 11.5 and 21 months for GBM respectively. This study also suggested no significant difference of survival between boost SRS and not to perform SRS (Villavicencio et al., 2009). In another study including 48 GBM patients, the use of SRS as boost or salvage was related with median survival of 15.1 and 17.1 months respectively. Difference in survivals was also statistically significant in this study (Pouratian et al., 2009). Contrarily, median survivals for GBM was found 10 and 16.7 months with boost and salvage SRS respectively in another study including 51 GBM patients in which the difference was statistically not meaningful (Hsieh et al., 2005). A study including 32 recurrent GBM patients treated with LINAC radiosurgery following conventional approach (surgery + RT) reported median 10 months of PFS following initial conventional treatment. SRS has contributed an additional 5 months of PFS to patients and a median 22 months of overall survival has been achieved. Survival rates of the study for 1st, 2nd and 3rd years are 88%, 41% and 19%, respectively (Combs et al., 2005). Besides, current studies on efficacy of repetitive SRS for multiple recurrences suggest no benefit on overall survival (Yoshikawa et

Increased tumor volume is associated with increased complication rates in SRS. Treatment dose should be decreased while tumor volume increases to avoid the complications such as radionecrosis and edema; which weakens the effectiveness of the treatment (Combs et al., 2007; Niyazi et al., 2011). Despite the lack of a definite threshold, SRS is not recommended for lesions larger than 3 cm diameter. Kong et al have reported the <10 ml tumor volume as the most important prognostic factor for SRS for malignant glioma in a series of 114 patients (Tsao et al., 2005). While adverse radiation effects occur rarely for tumors under 10 ml volume, Cho et al reported a high late complication rate of 30% for treatment of mean 30 ml

HGAs are classified as grade III and IV tumors. Various studies suggested the significant effect of histological grade on SRS treatment outcome. Yoshikawa et al reported an effectiveness rate and TCR of 27.2% and 63.3% for GBM respectively at least four weeks after SRS. Nevertheless, they found 18.2% and 45.5% for AA. Another study reported by Kong et al suggested a significant increase in overall median survival rate with SRS for GBM group and no difference in AA group as compared with control group (Kong et al., 2008).

Extent of surgical resection and effective post-operative RT are important prognostic factors for HGA. However, extensive surgical resection is not always possible particularly for tumors located in eloquent areas as optic nerves, brainstem and midbrain. Surgical

These results suggest that SRS may have a potential benefit on grade IV HGA.

**2.1.4 Tumor location and extent of surgical resection** 

et al., 2005).

al., 2006).

**2.1.2 Tumor volume** 

**2.1.3 Histological grade** 

tumors with mean 17 Gy (Cho et al., 1999).

A review by Yoshikawa et al on malignant glioma included seven clinical studies of RT plus SRS comparing with four clinical studies of RT only showed 20.2 and 11.1 months median overall survivals, respectively. Also the progression free survival (PFS) is found a median 281 days for SRS and 130 days for RT group (Yoshikawa et al., 2006). (Table 3).


Table 3. Review of studies comparing SRS with conventional RT by means of survival rates. *(NA\*: Not available)* 

Current multimodal treatment regimen for HGA includes a diagnostic or cytoreductive surgery followed by boost RT. For that reason, it's not so possible to meet with cases only treated with SRS without RT. Preliminary results of cases treated only with SRS for HGA suggested poor outcomes (Crowley et al., 2006). Certain indications and guidelines for patient selection criteria is not established yet on SRS for HGA. However long term outcome results of randomized controlled trials for SRS in HGA is not well reported yet, some helpful criteria standing out are described below.

### **2.1.1 Timing of SRS**

Timing of SRS for HGAs is controversial. While some of the authors have performed SRS for residual disease following surgical resection as a boost or in combination with RT, others have tended to perform as salvage for recurrence following RT. ASTRO (The American Society for Therapeutic Radiology and Oncology) has reported a comprehensive evidencebased review on SRS for HGA in 2005. They found level I-III evidence that the use of radiosurgery boost followed by RT and BCNU doesn't confer benefit in terms of overall survival, local tumor control or quality of life as compared with RT and BCNU.

A review by Yoshikawa et al on malignant glioma included seven clinical studies of RT plus SRS comparing with four clinical studies of RT only showed 20.2 and 11.1 months median overall survivals, respectively. Also the progression free survival (PFS) is found a median

> **Number of patients**

glioma 30 LINAC 13.9

glioma 1578 NA\* 11.3

glioma 17 NA 11.6

Masciopinto, 1995 GBM 31 LINAC 9.5

SRS Shrieve, 1999 GBM 78 LINAC 19.9 20.2

Yoshikawa, 2006 GBM 18 CyberKnife 20.7

RT Nwokedi, 2002 GBM 33 NA 13 11.1

Laws, 2003 GBM 413 NA 10.2 Table 3. Review of studies comparing SRS with conventional RT by means of survival rates.

Current multimodal treatment regimen for HGA includes a diagnostic or cytoreductive surgery followed by boost RT. For that reason, it's not so possible to meet with cases only treated with SRS without RT. Preliminary results of cases treated only with SRS for HGA suggested poor outcomes (Crowley et al., 2006). Certain indications and guidelines for patient selection criteria is not established yet on SRS for HGA. However long term outcome results of randomized controlled trials for SRS in HGA is not well reported yet, some

Timing of SRS for HGAs is controversial. While some of the authors have performed SRS for residual disease following surgical resection as a boost or in combination with RT, others have tended to perform as salvage for recurrence following RT. ASTRO (The American Society for Therapeutic Radiology and Oncology) has reported a comprehensive evidencebased review on SRS for HGA in 2005. They found level I-III evidence that the use of radiosurgery boost followed by RT and BCNU doesn't confer benefit in terms of overall survival, local tumor control or quality of life as compared with RT and BCNU.

glioma 15 Gamma

**SRS modality** 

knife <sup>26</sup>

knife <sup>25</sup>

knife 21.4

**Median survival after diagnosis (months)** 

**Mean survival of the group (months)** 

281 days for SRS and 130 days for RT group (Yoshikawa et al., 2006). (Table 3).

Kondziolka, 1997 GBM 64 Gamma

Nwokedi, 2002 GBM 31 Gamma

**Group First author, year Pathology** 

Gannett, 1995 Malignant

Prisco, 2002 Malignant

Curran, 1993 Malignant

helpful criteria standing out are described below.

*(NA\*: Not available)* 

**2.1.1 Timing of SRS** 

Prisco, 2002 Malignant

Furthermore, they pointed that the boost radiosurgery is associated with increased long term toxicity. They also reported that there is not sufficient evidence yet to show the effectiveness of SRS on recurrent or progressive malignant glioma (Anker et al., 2010; Tsao et al., 2005).

A multicentric study including 46 patients on CyberKnife comparing the use as a boost with salvage reported median overall survival of 11.5 and 21 months for GBM respectively. This study also suggested no significant difference of survival between boost SRS and not to perform SRS (Villavicencio et al., 2009). In another study including 48 GBM patients, the use of SRS as boost or salvage was related with median survival of 15.1 and 17.1 months respectively. Difference in survivals was also statistically significant in this study (Pouratian et al., 2009). Contrarily, median survivals for GBM was found 10 and 16.7 months with boost and salvage SRS respectively in another study including 51 GBM patients in which the difference was statistically not meaningful (Hsieh et al., 2005). A study including 32 recurrent GBM patients treated with LINAC radiosurgery following conventional approach (surgery + RT) reported median 10 months of PFS following initial conventional treatment. SRS has contributed an additional 5 months of PFS to patients and a median 22 months of overall survival has been achieved. Survival rates of the study for 1st, 2nd and 3rd years are 88%, 41% and 19%, respectively (Combs et al., 2005). Besides, current studies on efficacy of repetitive SRS for multiple recurrences suggest no benefit on overall survival (Yoshikawa et al., 2006).

### **2.1.2 Tumor volume**

Increased tumor volume is associated with increased complication rates in SRS. Treatment dose should be decreased while tumor volume increases to avoid the complications such as radionecrosis and edema; which weakens the effectiveness of the treatment (Combs et al., 2007; Niyazi et al., 2011). Despite the lack of a definite threshold, SRS is not recommended for lesions larger than 3 cm diameter. Kong et al have reported the <10 ml tumor volume as the most important prognostic factor for SRS for malignant glioma in a series of 114 patients (Tsao et al., 2005). While adverse radiation effects occur rarely for tumors under 10 ml volume, Cho et al reported a high late complication rate of 30% for treatment of mean 30 ml tumors with mean 17 Gy (Cho et al., 1999).
