Influence of parasellar extension of macroprolactinomas defined by magnetic resonance imaging on their responsiveness to dopamine agonist therapy

Summary Objective and Design The resistance of macroprolactinomas to dopamine agonist (DA) therapy, whether defined as an absence of PRL normalization or the lack of significant tumour shrinkage after prolonged treatment at high doses, is usually regarded as unpredictable. The aim of this retrospective study, conducted in a teaching hospital, was to determine whether cavernous sinus (CS) invasion assessed by magnetic resonance imaging (MRI) is associated with a higher rate of resistance to DA therapy. Methods Forty‐nine patients with a macroprolactinoma were included in this study and classified into four groups according to the percentage of encasement of the intracavernous internal carotid artery (ICA) by the tumour. All patients received DA as the primary treatment, mainly cabergoline (CAB). PRL normalization and tumour shrinkage during treatment were evaluated as a function of CS invasion. Results Tumours encasing more than three‐quarters of the intracavernous ICA (group 4) were less responsive to DA therapy, exhibiting a lower rate of early (≤ 3 months) PRL normalization (8% vs. 69% in the others groups; P < 0·01) under a higher dose of CAB (median: 3·5 mg vs. 1·0 mg per week; P < 0·01). CS invasion was a strongly significant and independent predictor of hormonal resistance to CAB ( P < 0·01). This hormonal resistance occurred in eight patients (16%), all but one belonging to group 4. Significant tumour shrinkage was observed in 31 out of 45 assessable cases (69%) and was more likely to occur in the case of PRL normalization ( P < 0·01). Conclusions Parasellar extension of macroprolactinomas, assessed on the basis of strict MRI criteria, may predict a negative response to DA. The responsiveness of noninvasive macroprolactinomas (over 90%) is similar to that reported in microprolactinomas, whereas invasive tumours are resistant to treatment in more than 50% of cases.