Identifying prognostic predictors for postoperative pituitary neuroendocrine tumour recurrence: an integrated clinical, radiological, and immunohistochemistry assessment.

OBJECTIVE:Pituitary neuroendocrine tumours (PitNETs) are the second most common type of intracranial tumour. Several studies have explored the prognostic factors for PitNETs. However, prognostic factors for postoperative PitNET recurrence remain not fully understood. This study aimed to explore potential prognostic factors for PitNET recurrence, such as surrounding tissue invasion and the extent of surgical resection in patients with postoperative PitNETs. METHODS:We included 106 patients who underwent PitNET surgery between 2013 and 2018, dividing them into two groups: those with recurrence and those without recurrence. Tumours were classified based on demographics, neuroradiological, and immunohistological characteristics. Univariate and multivariate analyses were used to determine factors predicting recurrence. Kaplan-Meier plots and log-rank tests were used to analyse each independent factor based on the cumulative 5-year recurrence rate. RESULTS:During the 5-year follow-up period, 29.2% of the patients (n = 31) had disease recurrence. Univariate analysis showed that predictors of recurrence included cavernous and sphenoid sinus invasions, optic chiasm compression, larger tumour volume, giant adenoma >4 cm, and gross total resection (GTR). Multivariate analysis showed that lactotroph tumour type, sphenoid sinus invasion, and GTR were independent predictors. Kaplan-Meier analysis revealed significant differences in the 5-year recurrence rate among the three independent predictors, with significantly lower recurrence rate in patients with lactotroph tumours and GTR, and a significantly higher recurrence risk in patients with sphenoid sinus invasion. CONCLUSIONS:Lactotroph tumour type, sphenoid sinus invasion, and GTR are independent predictors of postoperative PitNET recurrence. This study provides insights into the factors affecting postoperative PitNET recurrence.