Markers of tumor invasion are major predictive factors for the long-term outcome of corticotroph microadenomas treated by transsphenoidal adenomectomy

Abstract

OBJECTIVE: To assess the postsurgical outcome of patients with corticotroph microadenomas and to define predictors of the long-term outcome, with special emphasis on markers of tumor extension. DESIGN: Prospective study of 53 corticotroph microadenomas treated by enlarged adenomectomy. Patients followed for at least 2 years were classified into two groups: those in long-term remission and uncured patients (immediate failures and recurrences). Pre-, per- and postoperative parameters were analyzed as predictors of the long-term outcome. METHODS: Baseline hormone assessments were performed preoperatively, 8 days after surgery and every 6-12 months thereafter. Pituitary magnetic resonance imaging (MRI) allowed analysis of possible tumor extension to adjacent structures. Apparent completeness of the surgical removal was determined, and fragments labeled either 'tumor' or 'surrounding pituitary tissue' were submitted to serial sectioning. RESULTS: Immediate control of hypercortisolism was achieved in 43/53 patients (81%). However, later recurrences were observed in five patients (9%). Preoperative MRI showed tumor extension into adjacent structures with good specificity (91%) for prediction of surgical failure. Evidence of local invasion at surgery was also significantly predictive of the long-term outcome. A corticotroph adenoma was found at histological examination in 96% of the patients, and 26% had irregular limits, a feature significantly correlated with a poor outcome. Immediate postoperative plasma cortisol did not allow discrimination between long-term remissions and recurrences. CONCLUSION: Surgical failure was best predicted by signs of tumor 'invasiveness' at MRI, confirmed by peroperative examination and histology.