Editorial: Radiosurgery and cavernous malformations

Abstract

The successful obliteration of selected cerebral arteriovenous malformations (AVMs) using the Gamma Knife prompted the treatment of cavernous malformations (CMs) and aneurysms. For CMs, their small volume, lack of intervening normal brain tissue, and relatively low rate of clinically significant hemorrhage made them an appealing target for stereotactic radiosurgery (SRS). Radiosurgery for aneurysms was quickly abandoned as a failure, but radiosurgery for CMs persists despite controversy regarding its therapeutic value.13 Unlike for AVMs, radiosurgery for CMs often appears to have little to no effect evident on follow-up neuroimaging studies. In a previous report, a CM treated with Gamma Knife radiosurgery appeared unchanged on follow-up MRI 5 years afterwards.13 Extirpation of the lesion was undertaken, and histological examination revealed the lesion was partially obliterated with patency to a single persistent capillary channel (Fig. 1). Therefore, the correlation between radiographic and pathological changes may be unreliable following CM radiosurgery. From a series of more than 100 CMs treated at the University of Pittsburgh, Shin and colleagues report on the histopathological responses of 4 patients with CM who underwent resection 4 months to 7 years following radiosurgery.10 The CMs were treated with prescription doses of 14–15 Gy of radiation. In all 4 cases, radiosurgery failed and the patients suffered recurrent hemorrhages. Histopathological analysis revealed partially obliterated vessels and incomplete sclerosis as early as 4 months following radiosurgery. Partial obliteration could be a response to repeated microbleeds, recent hemorrhage, or past radiosurgery. Foci of neovascularization were also observed, as evidenced by staining for vascular endothelial growth factor (VEGF). In the case from our institution as well as the 4 from the University of Pittsburgh, we run the risk of drawing too much significance from a small and undoubtedly biased cohort. Nevertheless, the findings from Shin et al.’s study are some of the best available for elucidating the underlying radiobiological changes from radiosurgery on CMs. Coupling these histopathological findings with the neuroimaging and clinical outcomes published in the literature, it would seem safe to conclude that radiosurgery induces some of the same vessel sclerosis in CMs as it does in AVMs. However, the changes in AVM compared to CM vasculature following radiosurgery are substantially more robust and complete in the majority of patients.1,9,12 The incomplete response seen in CMs after radiosurgery may explain the potentially lower but still tangible risk of recurrent hemorrhage (8.8% annual risk for the first 2 years, then 1.1% annual risk after 2 years based on the group’s prior publication) in patients even years after radiosurgery.6 In contrast, in more than 1400 AVM patients treated using SRS, we have not observed a single hemorrhage in an AVM confirmed to be obliterated on angiography.11 The incomplete histopathological changes in CMs are consistent with the reduced but persistent risk of hemorrhage following radiosurgery. These findings raise 2 potential options for improving CM radiosurgery outcomes. First, the 4 patients in the current series were treated with a radiosurgical dose of 14–15 Gy to the margin of the CM. Prior preclinical and clinical studies have demonstrated more beneficial vascular changes with higher doses.1,5,7 However, such higher doses in patients with CM have also been associated with more complications. For example, in an early cohort of 22 CM patients treated with a mean margin dose of 18 Gy, 6 patients suffered neurological decline due to radiationinduced changes, and 5 (22.7%) had permanent deficits.4 Radiation-induced complications were defined as radionecrosis or transient radiation-induced changes demonstrated on CT or MRI and associated with a neurological decline. The radiation-induced complication rate in CMs was more than 2 times higher than expected for a simi-