Integrative Clinical, Hormonal and Molecular Data associate with Invasiveness in Acromegaly: REMAH Study

INTRODUCTION

Growth-hormone (GH)-secreting pituitary-tumors (GHomas) are the most common acromegaly cause. At diagnosis, most of them are macroadenomas, and up to 56% display cavernous sinus invasion. Biomarkers assessment associated with tumor-growth and invasion are important to optimize their management.

OBJECTIVES

To identify clinical/hormonal/molecular-biomarkers associated with tumor-size and invasiveness in GHomas, and to analyze the influence of pre-treatment with somatostatin-analogs or dopamine-agonists in key molecular biomarkers expression.

METHODS

Clinical/analytical/radiological-variables were evaluated in 192 patients from the REMAH-study (ambispective multicenter post-surgery study of the Spanish Society of Endocrinology and Nutrition). Expression of somatostatin/ghrelin/dopamine-systems components, and key pituitary/proliferation-markers were evaluated in GHomas after the first surgery. Univariate/multivariate regression studies were performed to identify association between variables.

RESULTS

80% of patients harbor macroadenomas (63.8% with extrasellar-growth). Associations between larger and more invasive GHomas with younger age, visual-abnormalities, higher IGF1-levels, extrasellar/suprasellar-growth and/or cavernous-sinus invasion were found. Higher GH1 and lower PRL/POMC/CGA/AVPR1B/DRD2T/DRD2L expression levels (p<0.05) were associated to tumor invasiveness. LASSO´s penalized regression identified combinations of clinical and molecular features with AUCs between 0.67-0.82. Preoperative therapy with dopamine-agonist or somatostatin-analogs did not alter the expression of any of the markers analyzed except for DRD1/AVPR1B (up-regulated with dopamine-agonist), and FSHB/CRHR1 (down-regulated with somatostatin-analogs).

CONCLUSIONS

A specific combination of clinical/analytical/molecular variables was found to be associated with tumor invasiveness and growth capacity in GHomas. Pre-treatment with first-line drugs for acromegaly did not significantly modify the expression of the most relevant biomarkers in our association model. These findings provide valuable insights for risk stratification and personalized management of GHomas.