Clinicopathological evaluation of risk factors for recurrence in craniopharyngioma.

Abstract

Objective: Craniopharyngiomas are benign tumors that often exhibit aggressive clinical behavior. The aim of this study was to evaluate long-term surgical outcomes, identify risk factors for recurrence, determine predictors of subtotal resection (STR), and characterize multiple recurrence patterns to optimize treatment strategies.

Methods: The authors retrospectively analyzed 269 craniopharyngioma surgeries performed in 225 patients between 2010 and 2024, with a median follow-up of 73.0 months. Each surgery was classified as primary, first recurrence, second recurrence, or third and subsequent recurrences. Second recurrences onward were excluded from the main risk factor analysis for recurrence and analyzed only when characterizing features of multiple recurrences. Accordingly, the first recurrence group was labeled as the "recurrent group" for the main comparative analyses. Clinical characteristics, radiological features, surgical outcomes, and neuroendocrinological results were evaluated. Ki-67 indices were assessed in 223 cases using immunohistochemistry.

Results: Of the 269 craniopharyngioma surgeries, 181 were primary, 60 were for first recurrence, and 28 were for multiple recurrences. The 5-year recurrence rate was 27.9%, with gross-total resection achieved in 86.2% of cases. In primary tumors, Kaplan-Meier analysis identified STR as a significant predictor of recurrence, while pediatric onset and large tumor volume were potential risk factors. In the recurrent group, large tumor volume was the only factor associated with recurrence. Multivariable analysis confirmed STR (HR 3.60, 95% CI 1.56-8.33) and pediatric onset (HR 2.00, 95% CI 1.02-3.93) as independent predictors of recurrence in primary craniopharyngioma. Vascular encasement was a significant predictor of STR (OR 3.97, 95% CI 1.15-13.7). Endocrine function was maintained in 31.1% of cases with stalk preservation, without increased recurrence risk. Ki-67 indices showed weak correlation with recurrence. Cases with multiple recurrences demonstrated progressive patterns of declining gross-total resection rates and shorter recurrence intervals.

Conclusions: Extent of resection was an independent risk factor for craniopharyngioma recurrence, with pediatric patients undergoing STR representing a particularly high-risk group. The Ki-67 index was not a reliable predictor of recurrence. Multiple recurrences demonstrated a progressive trend of increasing recurrence rates and shortening recurrence intervals. These findings provide important insights for treatment planning and patient counseling in the management of craniopharyngioma.