Iatrogenic tumor seeding after open surgery for chordoma has been well described in the literature. The incidence and particularities related to endoscopic endonasal surgery (EES) have not been defined.
The authors retrospectively reviewed their experience with EES for clival chordoma, focusing on cases with iatrogenic seeding. The clinical, radiographic, pathological, and molecular characterization data were reviewed.
Among 173 EESs performed for clival chordomas at the authors’ institution between April 2003 and May 2016, 2 cases complicated by iatrogenic seeding (incidence 1.15%) were identified. The first case was a 10-year-old boy, who presented 21 months after an EES for a multiply recurrent clival chordoma with a recurrence along the left inferior turbinate, distinct from a right petrous apex recurrence. Both appeared as a T2-hypertintense, T1-isointense, and heterogeneously enhancing lesion on MRI. Resection of the inferior turbinate recurrence and debulking of the petrous recurrence were both performed via a purely endoscopic endonasal approach. Unfortunately, the child died 2 years later due progression of disease at the primary site, but with no sign of progression at the seeded site. The second patient was a 79-year-old man with an MRI-incompatible pacemaker who presented 19 months after EES for his clival chordoma with a mass involving the floor of the left nasal cavity that was causing an oro-antral fistula. On CT imaging, this appeared as a homogeneously contrast-enhancing mass eroding the hard palate inferiorly, the nasal septum superiorly, and the nasal process of the maxilla, with extension into the subcutaneous tissue. This was also treated endoscopically (combined transnasal-transoral approach) with resection of the mass, and repair of the fistula by using a palatal and left lateral wall rotational flap. Adjuvant hypofractionated stereotactic CyberKnife radiotherapy was administered using 35 Gy in 5 fractions. No recurrence was appreciated endoscopically or on imaging at the patient’s last follow-up, 12 months after this last procedure. In both cases, pathological investigation of the original tumors revealed a fairly aggressive biology with 1p36 deletions, and high Ki-67 levels (10%–15%, and > 20%, respectively). The procedures were performed by a team of right-handed surgeons (otolaryngology and neurosurgery), using a 4-handed technique (in which the endoscope and suction are typically passed through the right nostril, and other instruments are passed through the left nostril without visualization).
Although uncommon, iatrogenic seeding occurs during EES for clival chordomas, probably because of decreased visualization during tumor removal combined with mucosal trauma and exposure of subepithelial elements (either inadvertently or because of mucosal flaps). In addition, tumors with more aggressive biology (1p36 deletions, elevated Ki-67, or both) are probably at a higher risk and require increased vigilance on surveillance imaging and endoscopy. Further prospective studies are warranted to evaluate the authors’ proposed strategies for decreasing the incidence of iatrogenic seeding after EES for chordomas.