An Institutional Review of Clinical Outcomes for Skull Base Chordoma: An Analysis of 263 Cases Over 19 Years

Background: Endoscopic endonasal approaches (EEA) have become the approach of choice for skull base chordoma (SBC). Both the approaches, as well the reconstructive techniques have evolved significantly over the years. In this report, we sought to review the clinical outcomes after EEA for SBC.

Methods: Medical records of patients who underwent surgical resection for chordoma within the electronic medical record from 2001 to 2020 were retrospectively reviewed. Statistical analyses were performed using SPSS. Cases were split into groups based on recurrence status. Chi square and t-tests were used to assess baseline differences between groups. Preoperative imaging variables and clinical characteristics were entered into a Cox proportional hazards model for progression-free survival following resection. The forward conditional method of independent variable entry was used. Kaplan–Meier survivorship curves were created to depict progression-free survival versus extent of tumor resection.

Results: This analysis included data from 194 patients who underwent a total of 263 operations. For 204 cases, follow-up length was >2 years; for 131 cases, follow-up length was >5 years; and for 62 cases, follow-up length was >10 years. Table 1 shows demographic characteristics of our sample. There were no significant baseline differences between patients with primary versus recurrent tumors. Patients with recurrent tumors were more likely than those with primary tumors to develop new postoperative cranial nerve palsies (p = 0.004) and be diagnosed with metastatic lesions from chordoma (p < 0.001). Primary tumors were more likely than recurrent tumors to involve the upper clivus (p = 0.012) and pan-clivus (0.008), but interestingly there were no preoperative imaging differences between groups (Table 2). Gross total resection was significantly more likely for primary tumors, while near total and subtotal resections were more likely for recurrent tumors (p < 0.001, Table 2). A Cox proportional hazards model indicates that factors associated with longer progression-free survival following chordoma resection include primary tumors (p < 0.001), gross total resections (p < 0.001), postoperative radiation (p = 0.003), and smaller tumor volume (p < 0.001). Kaplan–Meier analysis further supports that patients with gross total resection achieve longer progression-free survival time (p < 0.001, [Fig. 1A]). The survival benefit of gross total resection held true for both primary (p < 0.001, [Fig. 1B]) and recurrent tumors (p < 0.001, [Fig. 1C]). The learning curve had a significant effect on outcomes, with GTRs improving over time (p < 0.0001). Interestingly, the majority of this improvement was the ability to achieve complete resections in recurrent tumors. The GTR rate in the past 5 years was 69% for recurrent tumors and 90% for primary tumors. The improvement in outcomes was also related to a decrease in the rate of permanent new cranial neuropathies in both primary and recurrent tumors (p < 0.024).

Conclusion: The surgical outcomes for chordoma continue to improve, with improvement in GTRs and decreased complications.

No conflict of interest has been declared by the author(s).

Publication History

Article published online:
12 February 2021