Risk Factors for Biochemical Recurrence in Prolactinomas After Withdrawal of Dopamine Agonists: A Retrospective Cohort Study

Abstract Context: Dopamine agonists (DA), mainly cabergoline, are the primary therapy for prolactinomas. Risks factors related with biochemical recurrence after DA withdrawal are not completely understood. Objective: To assess hyperprolactinemia recurrence risk factors after dopamine agonist (DA) withdrawal in prolactinomas. Design: Retrospective, comparative, observational cohort study. Survival, and multivariate regression analyses were performed to estimate biochemical remission (prolactin (PRL) 3-25 ng/mL after at least 6 months of DA withdrawal), and recurrence (>25 ng/ml and/or tumor regrowth) at follow-up. Results: 1140 patients with hyperprolactinemia were evaluated. Prolactinoma was confirmed in 182 cases. Micro- (n=95), and macroprolactinoma (n=75) were more frequent in women (99% and 73%, respectively), while giant prolactinomas (n=12) were more common in men (75%). Cabergoline was the most frequently (92-95%) DA used with a median dose of 1.2 (0.25-4.0), 2.2 (1.5-3.5), and 2.8 (1.5-8.0) mg/week (p<0.001), respectively. Higher tumor size at diagnosis was significantly related to biochemical recurrence (p<0.02). However, independently of tumor size at diagnosis, and the DA cumulative dose received, once patient have achieved an 80% volume reduction, tumors were unlikely to decrease further (p=0.02). Among cases under remission at last follow-up, almost half with microprolactinomas (n=51, 54%), and macroprolactinomas (n=35, 46%) persisted with tumor remnants. Multivariate regression analysis confirmed that tumor remnant at last follow-up was not significantly associated with recurrence risk (p=0.81). Duration of therapy was the only variable significantly associated with remission (p<0.01). Conclusion: Probability of biochemical recurrence after withdrawal of DA therapy in micro and macroprolactinomas was not significantly associated with tumor remnant. Duration of DA therapy was the only variable significantly associated with remission.