Predictors of extent of resection and recurrence following endoscopic endonasal resection of craniopharyngioma.

OBJECTIVE: Craniopharyngioma is a benign but surgically challenging brain tumor. Controversies exist regarding its ideal treatment strategy, goals of surgery, efficacy of radiation, and the long-term outcomes of these decisions. The authors of this study performed a detailed analysis of factors predictive of the extent of resection and recurrence in large series of craniopharyngiomas removed via an endoscopic endonasal approach (EEA) with long-term follow-up. METHODS: From a prospective database of all EEAs done at Weill Cornell Medical College by the senior author from 2004 to 2022, a consecutive series of histologically proven craniopharyngiomas were identified. Gross-total resection (GTR) was generally the goal of surgery. Radiation was often given if GTR had not been achieved. The stalk was preserved if not infiltrated with tumor but was sacrificed to achieve GTR. Intentional subtotal resection (STR) was performed in select cases to avoid hypothalamic injury. RESULTS: Among the 111 identified cases were 88 adults and 23 children. Newly diagnosed cases comprised 58.6% of the series. GTR was attempted in 77.5% of the patients and among those cases was achieved in 89.5% of treatment-naive tumors and 72.4% of recurrent tumors. An inability to achieve GTR was predicted by prior surgical treatment (OR 0.13, 95% CI 0.03-0.6, p = 0.009), tumor diameter ≥ 3.5 cm (OR 0.11, 95% CI 0.02-0.53, p = 0.006), and encasement of the optic nerve or a major artery (OR 0.11, 95% CI 0.01-0.8, p = 0.03). GTR with stalk preservation maintained some anterior pituitary function in 64.5% of cases and prevented diabetes insipidus in 25.8%. After a median follow-up of 51 months (IQR 17-80 months), the recurrence rate after GTR was 12.5% compared with 38.5% after non-GTR. The median recurrence-free survival was 5.5 years after STR, 8.3 years after near-total resection (≥ 98%), and not reached after GTR (p = 0.004, log-rank test). GTR was the strongest predictor of recurrence-free survival (OR 0.09, 95% CI 0.02-0.42, p = 0.002), whereas radiation did not show a statistically significant impact (OR 1.17, 95% CI 0.45-3.08). In GTR cases, the recurrence rate was higher if the stalk had been preserved (22.6%) as opposed to a sacrificed stalk (4.9%; OR 5.69, 95% CI 1.09-29.67). CONCLUSIONS: The study data show that GTR should be the goal of surgery in craniopharyngiomas if it can be achieved safely. Although stalk preservation can maintain some endocrine function, the risk of recurrence is higher in such cases. Radiation may not be as effective as previously reported.