**2.2.2 Tumor location and pattern**

Even though the primary treatment for LGA is the surgical resection, SRS following pathological diagnosis serves as an option for tumors located in eloquent areas and for unresectable tumors. However, the treatment dose should be diminished to avoid damage to surrounding tissues for tumors in close proximity to eloquent tissues, which results in reduction of effectiveness of the treatment. SRS is preferable instead of whole brain irradiation for LGA because of the locally invasive nature of these tumors.

Brainstem gliomas account for less than 2% of adult and 10-20% of pediatric age glial tumors. Although 52-69% of brainstem gliomas are low grade, they carry greater potential for malignant transformation with respect to other locations (Bricolo, 2009). 80% TCR during 78 month follow-up was reported in a series of 20 unresectable focal brainstem gliomas with gamma knife with mean 12.8 Gy doses (Yen et al., 2007). Another study comparing TCR for LGA between brainstem and other locations reported 59% and 67% TCR, respectively (Hadjipanayis et al., 2003). Progression rate following SRS was also found 45% for brainstem versus 10% for other locations. Major reasons for lower success rate of SRS for brainstem gliomas are the more aggressive nature of tumor at this location and the requirement of dose reduction. Unresectable low grade optic glioma may also benefit from fractionated SRS. Effective TCR and prevention of progressive visual symptoms were reported for optic gliomas (Kurt et al., 2010).

LGAs may include solid or cystic components. Better response to the SRS for solid LGA was reported in various series. Furthermore, half of the progressive patients have only cyst enlargement without solid enlargement. TCR for pure solid tumors was 84% in a study including both solid and cystic tumors with overall 68% TCR (Hadjipanayis et al., 2002a). 1,

SRS may either be performed alone or as a boost in combination with RT for residual tumor in early post-operative period; or as salvage treatment at the time of recurrence. Whether or not to perform and when to perform is the moot point. Boost SRS concurrent with RT was found to cause more adverse radiation effect in comparison with salvage (adjuvant) SRS (Wang et al., 2006). 10 year median survival rate was found 88.9% for PA patients underwent partially resection or biopsy followed by SRS alone as the principal treatment. The ratio was also found 44.5% for PA patients received delayed SRS for recurrent disease. Delayed SRS for recurrent tumor seems to be associated with poor PFS (Kano et al., 2009b). On the other hand, it doesn't seem so reasonable to make a generalization for timing of SRS because the tumors highly tended to recur already have poor prognosis. Another study reported TCR of 56.3 months for boost SRS versus 44.4% for late SRS. However, this difference was not statistically significant (Park et al., 2010). The beginning of shrinkage following SRS occurs between a median 13-16 months (range; 3-92.4) for LGA (Yen et al., 2007; Kano et al. 2009a & 2009b). In case of progression, the mean time from SRS to the beginning of progression has been found about 23 months (Hadjipanayis et al., 2002a). That's why the patients should be periodically followed-up in a long time period. Despite the lack of large series on effectiveness of repetitive SRS for recurrent LGA; achievement of effective tumor control has been reported for sporadic cases. More studies are needed intended to timing of SRS for LGAs. Available literature suggests better tumor control for

Even though the primary treatment for LGA is the surgical resection, SRS following pathological diagnosis serves as an option for tumors located in eloquent areas and for unresectable tumors. However, the treatment dose should be diminished to avoid damage to surrounding tissues for tumors in close proximity to eloquent tissues, which results in reduction of effectiveness of the treatment. SRS is preferable instead of whole brain

Brainstem gliomas account for less than 2% of adult and 10-20% of pediatric age glial tumors. Although 52-69% of brainstem gliomas are low grade, they carry greater potential for malignant transformation with respect to other locations (Bricolo, 2009). 80% TCR during 78 month follow-up was reported in a series of 20 unresectable focal brainstem gliomas with gamma knife with mean 12.8 Gy doses (Yen et al., 2007). Another study comparing TCR for LGA between brainstem and other locations reported 59% and 67% TCR, respectively (Hadjipanayis et al., 2003). Progression rate following SRS was also found 45% for brainstem versus 10% for other locations. Major reasons for lower success rate of SRS for brainstem gliomas are the more aggressive nature of tumor at this location and the requirement of dose reduction. Unresectable low grade optic glioma may also benefit from fractionated SRS. Effective TCR and prevention of progressive visual symptoms were reported for optic

LGAs may include solid or cystic components. Better response to the SRS for solid LGA was reported in various series. Furthermore, half of the progressive patients have only cyst enlargement without solid enlargement. TCR for pure solid tumors was 84% in a study including both solid and cystic tumors with overall 68% TCR (Hadjipanayis et al., 2002a). 1,

irradiation for LGA because of the locally invasive nature of these tumors.

**2.2.1 Timing of SRS** 

residual PA with early SRS.

gliomas (Kurt et al., 2010).

**2.2.2 Tumor location and pattern** 

3 and 5 year PFS rates were found 75%, 50% and 50% for solid, and 88.9%, 17.8% and 0% for mixed solid-cystic tumors respectively in a study (Kano et al, 2009a). Another study reported 3, 5 and 10 year PFS rates of 100%, 94.4% and 85% for solid, and 53.1%, 21.3% and 0% for mixed solid-cystic tumors respectively (Kano et al., 2009b). Peripheral contrast enhancement and cystic changes on MRI are related with poor prognosis (Park et al., 2010). SRS may also be performed for multicentric LGA, but the prognosis of multicentric tumors is poorer than the solitary tumors (Hadjipanayis et al., 2002a).
