Real-life clinical impact of a five-tiered classification of pituitary tumors

International audienceIntroduction Usually benign, pituitary tumors (PT) can be invasive and aggressive with a propensity to progress and/or recur. Trouillas's clinicopathological classification attempts to predict the evolutionary risk of a PT. In this study, we assessed the prognostic value of this classification in an independent patient cohort and analyzed its impact on treatment strategies. Patients and methods In this study, 607 patients operated on between 2008 and 2018 for a PT were included. Grading was established based on invasion, proliferative activity (Ki-67, mitotic index) and p53 positivity. The therapeutic management following surgery was analyzed. Progression-free survival (PFS) of the graded tumors was estimated (Kaplan–Meier method and log-rank test) and a multivariate analysis was performed (Cox regression model). Results Grading identified non-invasive PT without (grade 1a: 303 cases) or with proliferative activity (grade 1b: 53 cases) and invasive PT without (grade 2a: 202 cases) or with proliferative activity (grade 2b: 49 cases). The mean follow-up was 47 ± 30 months (median: 38 months). Progression/recurrence occurred in 127 cases. Grades were significant and independent predictors of PFS (P < 0.001) with a 4.8-fold higher risk of progression/recurrence in grade 2b as compared to grade 1a. As second-line therapy, gamma knife or conventional radiotherapy controlled tumor growth in 91.6 and 100% of cases, respectively, irrespective of the grade. Proliferative tumors exposed the patient to a 9.5-fold higher risk of having ≥3 adjuvant therapeutic lines as compared to non-proliferative tumors. Discussion Grading of a PT according to Trouillas's classification predicts its risk of progression and should advocate for a personalized therapeutic approach in invasive and proliferative tumors. Significance statement This is the first study to assess, on a cohort of 607 well-characterized patients, the real-life therapeutic impact of the five-tiered clinicopathological classification of pituitary tumors. First, we validate that pituitary tumor grades predict the evolutionary risk of the tumor, with a significant higher risk of progression/recurrence in invasive and/or proliferative tumors (mean follow-up: 47 ± 30 months, median: 38 months). Moreover, our study provides evidence that patients with proliferative tumors have a higher risk to be retreated after primary surgery and point toward the fact that radiotherapy can successfully control tumor growth in case of progression or recurrence. Our findings advocate for a personalized therapeutic approach in clinically aggressive pituitary tumors