Serum Prolactin Levels are Unrelated to Pituitary Tumor Volume and Degree of Invasiveness In 114 subjects with Non-Functioning Pituitary Adenomas

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: MON 88-111-Cushing's Disease & Non-Functioning Hypothalamus-Pituitary Tumors
Monday, June 17, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board MON-105
John David Carmichael*1, Vivien Shelley Bonert2, Jaswinder Jutla3, Raymundo Garcia Lora3 and Adam N Mamelak4
1Cedars-Sinai Med Ctr, Los Angeles, CA, 2Cedars Sinai Med Ctr, Los Angeles, CA, 3Cedars-Sinai Medical Center, Los Angeles, CA, 4Cedars Sinai Med Cntr, Los Angeles, CA
Background: Hyperprolactinemia associated with non-prolactinoma pituitary adenomas, the “stalk effect,” reflects the hypothesis that impingement of the pituitary stalk interrupts dopaminergic tone causing increased prolactin (PRL) secretion. It has been proposed that hyperPRL is related to tumor mass effect, but no clear evidence has supported stalk deviation, tumor size, invasiveness, or compression of sellar structures as a definitive etiology. Prior studies evaluated effects of size and pressure, without direct assessment of tumor volume. The relationship between tumor volume, cavernous sinus extension, and serum PRL levels is unknown.

Methods: Subjects who harbored histologically proven non-functioning pituitary adenomas (NFPA) were selected from the Cedars-Sinai Pituitary Tumor Research Registry, a comprehensive database comprising clinical, pathological, and imaging data.  Preoperative hormonal assessments and imaging results were evaluated in the context of postoperative immunohistological findings in subjects with preoperative serum PRL, and imaging, and surgical pathology. Tumor volume was calculated using Osirix ® by identifying the tumor as a region of interest on contiguous cross sections of a T1 weighted coronal MRI.  Once the tumor was circumscribed, the imaging program calculated tumor volume.  Preoperative MRIs were scored on a Knosp scale for degree of invasiveness. Statistical analysis performed by SAS, nonparametric testing was used for all analyses.

Results: 114 subjects (45F/69M) harboring NFPA were included. Mean (±SD) age was 56 ±12.6 years. Median serum PRL level 19.4 ng/ml (Range 0.6 – 120.3) with one case >100 ng/ml. Median tumor volume was 3.9 cm3(Range 0.65-43.07). Histopathology included gonadotroph adenoma (45%) Null Cell (43%) Silent Corticotroph (7%) Plurihormonal adenoma (5%).  Knosp scores were: 0 (9%) 1 (30%) 2 (21%) 3 (33%) and 4 (5%).  Serum PRL levels did not correlate with either tumor volume (r=.02, p=.8), or Knosp score (r=.16, p=.1). Tumor volume correlated with Knosp score (r=.48, p<.001).

Conclusions: Serum prolactin levels were independent of either tumor volume or invasiveness in NFPA.  The results also confirm that a serum prolactin level of <100 ng/ml is highly suggestive of “stalk effect” in NFPA.  Lack of correlation between tumor volume and hyperprolactinemia suggests a potential humoral or neuroendocrine pathogenesis rather than a mechanical etiology for PRL hypersecretion.

Nothing to Disclose: JDC, VSB, JJ, RGL, ANM

*Please take note of The Endocrine Society's News Embargo Policy at