Endocrine Abstracts

Introduction: Prolactinomas are mostly benign tumours usually managed with pharmacological treatment. Some, however, seem to be resistant to dopamine agonists (DA) for unclear reasons. The relationship between T2W signal intensity (T2WSI) and response to treatment with somatostatin analogs is well described in acromegaly patients. Some evidence suggests that prolactinoma’s T2W hypointensity might be related to higher baseline prolactin levels and more resistance to DA.

Aim: Evaluate the relationship between MRI T2WSI and response to treatment with DA in adult patients with prolactinoma.

Methods: Retrospective study including data from patients with prolactinoma diagnosis followed at our outpatient clinic from 2005 to 2018. Signal intensity on MRI was observed by the same physician and compared to the normal pituitary gland. Response to DA was evaluated by mean dose necessary to normalize prolactin levels (bromocriptine and cabergoline considered separately) and time to normalization. When normal levels were not reached (resistant cases), the highest dose used was considered.

Results: 70 patients were identified, 37 with feasible data, with a mean age at diagnosis of 45 years, 26 (70%) females. Most had macroprolactinoma (n=26; 70%). Mean initial prolactin levels were 1490 ng/mL; they normalized after a mean time of 9 months (mean prolactin 13.8 ng/mL). Four patients did not reach normal levels. Men had larger tumours and higher baseline prolactin levels, but only the last with statistical significance (P 0.04). There was no relationship between T2WSI and baseline prolactin levels (P 0.96). In 23 patients, the association between T2WSI and mean time for prolactin level normalization was evaluated: hyperintense lesions had a slightly longer mean time, but not statistically significant (P 0.89). In 10 patients treated with cabergoline and 22 treated with bromocriptine, there was no significant association between T2WSI and mean dose reached (P 0.55; P 0.63, respectively).

Conclusion: MRI signal intensity is related to tumour characteristics. Prolactinomas are usually not surgically removed, so it would be very useful if data from imaging studies could help predict clinical and biochemical behaviour, especially when it comes to response to treatment. We found that hyperintense lesions in T2W take slightly longer to normalize prolactin levels, but this was not statistically significant. Our results were not in accordance with other data in literature. This might be possibly related to differences in study design and small sample size in this study. Relationship between T2WSI and response to DA is still not clear. Larger, well designed studies will be needed to assess this.