RESUMO
BACKGROUND: There are no guidelines regarding post-treatment surveillance specific to skull base chordomas. OBJECTIVE: To determine an optimal imaging surveillance schedule to detect both local and distant metastatic skull base chordoma recurrences. METHODS: A retrospective review of 91 patients who underwent treatment for skull base chordoma between 1993 and 2017 was conducted. Time to and location of local and distant recurrence(s) were cataloged. Existing chordoma surveillance recommendations (National Comprehensive Cancer Network [NCCN], London and South East Sarcoma Network [LSESN], European Society for Medical Oncology [ESMO], Chordoma Global Consensus Group [CGCG]) were applied to our cohort to compare the number of recurrent patients and months of undiagnosed tumor growth between surveillances. These findings were used to inform the creation of a revised imaging surveillance protocol (MD Anderson Cancer Center Chordoma Imaging Protocol [MDACC-CIP]), presented here. RESULTS: Thirty-four patients with 79 local/systemic recurrences met inclusion criteria. Mean age at diagnosis and follow-up time were 45 yr and 79 mo, respectively. The MDACC-CIP imaging protocol significantly reduced the time to diagnosis of recurrence compared with the LSESN and CGCG/ESMO imaging protocols for surveillance of local disease with a cumulative/average of 576/16.9 (LSESN), 336/9.8 (CGCG), and 170/5.0 (MDACC-CIP) months of undetected growth, respectively. The NCCN and MDACC-CIP guidelines for distant metastatic surveillance identified a cumulative/average of 65/6.5 and 51/5.1 mo of undetected growth, respectively, and were not significantly different. CONCLUSION: The MDACC-CIP for skull base chordoma accounts for recurrence trends unique to this disease, including a higher rate of leptomeningeal spread than sacrococcygeal primaries, resulting in improved sensitivity and prompt diagnosis.