J Neurol Surg B Skull Base 2018; 79(S 01): S1-S188
DOI: 10.1055/s-0038-1633594
Oral Presentations
Georg Thieme Verlag KG Stuttgart · New York

Visual Changes following Cyberknife Radiosurgery for Skull Base Chordomas and Chondrosarcomas

Swapnil S. Mehta
1   Stanford University, Stanford, California, United States
,
Yagmur Muftuoglu
1   Stanford University, Stanford, California, United States
,
Geoffrey Appelboom
1   Stanford University, Stanford, California, United States
,
Kevin Kumar
1   Stanford University, Stanford, California, United States
,
Justin Moore
1   Stanford University, Stanford, California, United States
,
Scott Soltys
1   Stanford University, Stanford, California, United States
,
Steven D. Chang
1   Stanford University, Stanford, California, United States
› Author Affiliations
Further Information

Publication History

Publication Date:
02 February 2018 (online)

 

Background Chondrosarcomas and chordomas of the skull base are locally aggressive tumors associated with visual symptoms given proximity to the optic apparatus. Although these tumors are frequently considered together, chondrosarcomas are generally more responsive to treatment with better progression-free and overall survival. We aimed to compare their response to robotic stereotactic radiosurgery (rSRS) and impact on visual symptoms.

Methods We retrospectively evaluated 21 patients with chordoma (13) or chondrosarcoma (9) who underwent primary adjuvant rSRS at our institution. Average clinical follow-up from initial surgery was 37.2 ± 25.3 months or 59.3 ± 31.4 months; average follow-up from rSRS was 33.8 ± 26.2 or 47.6 ± 35.12 months; rSRS was delivered in one to five fractions at a mean prescription dose of 36.5 ± 6.9 Gy for chordomas and two to five fractions at 29.9 ± 3.7 Gy for chondrosarcomas. At an alpha/beta ratio of 2.45, this translated to a median biologically effective dose of 34.1 Gy (27.0–259.1 Gy) and 36.3 Gy (20.7–70.8), respectively. Visual symptoms of diplopia, blurry vision, or visual field defects were evaluated prior to surgery, prior to rSRS, and at final follow-up. Changes in tumor volume after rSRS were assessed by radiologist report. Statistical analyses were performed by Pearson's chi-squared test, with p < 0.05 considered significant.

Results Both chordomas (76.9%) and chondrosarcomas (62.5%) resulted in a high rate of visual defects. Surgical resection resulted in an 80% decrease in visual symptoms for patients with chordomas as compared with a 20% increase for those with chondrosarcomas prior to CKRS (p = 0.006). We compared extent of resection as a proxy for proximity of tumor to critical structures and found higher rates of gross total resection (GTR) for chordomas (61.5%) as compared with chondrosarcomas (12.5%; p = 0.03). Chondrosarcomas were more likely to lie adjacent to or involve the cavernous sinus (p = 0.004), rendering GTR difficult to achieve given proximity to cranial nerve VI. Three out of seven patients (42.9%) with subtotal resection (STR, N = 6) or near-total resection (NTR, N = 1) for chondrosarcoma experienced decreased tumor volume, with an average time to diminution of 8.59 ± 6.7 months. All of these patients experienced resolution of visual symptoms. In contrast, no patients with STR (N = 4) or NTR (N = 1) for chordoma experienced a reduction in tumor volume. In fact, patients with chordoma experienced new visual symptoms after rSRS attributable to treatment in three cases and tumor progression in one case. Further, one patient with chondrosarcoma experienced a new visual symptom attributable to rSRS. At last follow-up, 46.2% patients with chordoma and 50% with chondrosarcoma displayed visual symptoms (p = 0.86).

Conclusion Consistent with previous results, chondrosarcomas have a greater sensitivity to rSRS, but both result in roughly equivalent visual symptoms at last follow-up. Conservative resection of chondrosarcomas owing to tumor location and tumor progression with poor response to CKRS for chordomas likely explains these results. Ultimately, an assessment of tumor location, type, and extent of resection must be optimized to improve visual outcomes.