Long-term tumor control following stereotactic radiosurgery for jugular paraganglioma using 3D volumetric segmentation

OBJECTIVE The morbidity of gross-total resection of jugular paraganglioma (JP) is often unacceptable due to the potential for irreversible lower cranial neuropathy. Stereotactic radiosurgery (SRS) has been used at the authors' institution since 1990 for the treatment of JP and other benign intracranial tumors. Conventional means of assessing tumor progression using linear measurements or elliptical approximations are imprecise due to the irregular shape and insinuating growth pattern of JP. The objective of this study was to assess long-term tumor control in these patients by using slice-by-slice 3D volumetric segmentation of serial MRI data. METHODS Radiographic data and clinical records were reviewed retrospectively at a single, tertiary-care academic referral center for patients treated from 1990 to 2017. Volumetric analyses by integration of consecutive tumor cross-sectional areas (tumor segmentation) of serial MRI data were performed. Tumor progression was defined as volumetric growth of 15% or greater over the imaging interval. Primary outcomes analyzed included survival free of radiographic and clinical progression. Secondary outcomes included new or worsened cranial neuropathy. RESULTS A total of 85 patients were treated with Gamma Knife radiosurgery (GKRS) for JP at the authors' institution over the last 27 years. Sixty patients had pretreatment and serial posttreatment contrast-enhanced MRI follow-up suitable for volumetric analysis. A total of 214 MR images were analyzed to segment tumor images in a slice-by-slice fashion to calculate integral tumor volume. The median follow-up duration was 66 months (range 7-202 months). At 5 years the tumor progression-free survival rate was 98%. Three tumors exhibited progression more than 10 years after GKRS. Estimated survival free of radiographic progression rates (95% confidence interval [CI]; n = number still at risk) at 5, 10, and 15 years following radiosurgery were 98% (95% CI 94%-100%; n = 34), 94% (95% CI 85%-100%; n = 16), and 74% (95% CI 56%-98%; n = 6), respectively. One patient with tumor progression required treatment intervention using external beam radiation therapy, constituting the only case of clinical progression. Two patients (3%) without preexisting lower cranial nerve dysfunction developed new ipsilateral vocal fold paralysis following radiosurgery. CONCLUSIONS SRS achieves excellent long-term tumor control for JP without a high risk for new or worsened cranial neuropathy when used in primary, combined modality, or recurrent settings. Long-term follow-up is critical due to the potential for late radiographic progression (i.e., more than 10 years after SRS). As none of the patients with late progression have required salvage therapy, the clinical implications of this degree of tumor growth have yet to be determined.