Doo-Sik Kong, Cancer 2008;112:2046

Irradiation was attempted as a first-line or second-line treatment modality for brain tumors. Although there have been recent advances in novel chemotherapeutic regimens such as combined chemoradiotherapy using temozolomide or several small-molecule kinase inhibitors, reirradiation with FSRT or SRS is still an attractive modality because these modalities can deliver a high dose of radiation while sparing adjacent normal cerebral tissues.

However, it remains unclear whether radiosurgery provides any survival benefit for patients with malignant gliomas, including grade 3 gliomas and GBMs. Compared with the historic control group in this study, SRS did not significantly provide a survival benefit in patients with recurrent grade 3 gliomas (37.5 months vs 26 months), whereas it prolonged survival when used as a salvage treatment in patients with recurrent GBM (23 months vs 12 months). Although we acknowledge that a comparison study with a historic control group is statistically less powerful, this result still has merit for several reasons. First, lower-grade gliomas have a possibility of being less responsive to radiosurgery than higher-grade gliomas because grade 3 gliomas have slower progressive patterns. In addition, recurrence of grade 3 gliomas does not necessarily indicate malignant transformation into GBMs. Finally, the subtypes of grade 3 gliomas may have a very different response to SRS because this study used a heterogeneous group including AA, AO, AOA, and GC. Accordingly, to accurately determine the efficacy of SRS in grade 3 gliomas, survival should be analyzed according to each subdivision of grade 3 gliomas.

With regard to GBM, the results of the Radiation Therapy Oncology Group (RTOG) 93-05 study,which did not demonstrate a survival advantage, may not appear to be consistent with our results. However, that phase 3 trial examined the role of an upfront radiosurgery boost to conventional radiation and carmustine for newly diagnosed GBMs, and not for salvage radiosurgery, as was the case in the current study. Therefore, the results of the current study indicate that salvage radiosurgery may play a role in this patient population. In this study, however, the historic control group is problematic because radiation necrosis and pure tumor progression could not be distinguished radiographically. This might lead to interpretation errors in the comparison of PFS. This study also included some methodologic limitations. Because all of the patients in this cohort group had received a full course of external beam radiotherapy, it was possible that some of these patients had radiation necrosis in these regions, and not true tumor progression. However, a routine course of external beam radiotherapy below 6000 centigrays reportedly has a low incidence of radiation necrosis. Conversely, because this cohort group included tumors measuring 3 cm in maximal dimension, more biologically aggressive and infiltrative tumors might not be included in this study. This possibility can account for the lengthy survival of 8.5 months after progression from SRS. In contrast, the historic control group did not have any limitations with regard to the size of the lesion at the time of treatment. Therefore, control patients were more likely to have larger and more biologically aggressive/infiltrative tumors, with shorter survival times. Although this study cannot avoid the limitations of selection bias and not having an ideal control group, the result suggests evidence of a defined role of SRS in selected patients with recurrent GBMs.

It was interesting to note that the rate of macroscopic total resection was approximately 57.4% in the current series. This value was very high compared with previous studies (only 7%-21.8% of total patients or not described).In the current series, recurrent tumors might be smaller-sized lesions than those in other studies because approximately 50% of tumors were macroscopically/completely removed. This might result in a more improved survival outcome. Generally, radiosurgery has a limited role for large-sized tumors. Our univariate analysis demonstrated that recurrent tumors with smaller volume (<10 mL) had better survival after SRS.

In conclusion, SRS is a relatively safe treatment modality for patients with recurrent small-sized GBMs and can be efficiently used with acceptable morbidity in a highly selected patient population. The efficacy for recurrent grade 3 gliomas should be evaluated further in well-designed clinical studies.