ECE2022 Poster Presentations Pituitary and Neuroendocrinology (127 abstracts)
1University of Turin, Division of Endocrinology Diabetology and Metabolism, Department of Medical Sciences, Turin, Italy; 2AOU Città della Salute e della Scienza di Torino University Hospital, Division of Endocrinology Diabetology and Metabolism, Department of Medical Sciences, Turin, Italy; 3University of Turin, Pathology Division 2U, Department of Medical Sciences, Turin, Italy; 4AOU Città della Salute e della Scienza di Torino University Hospital, Division of Neuroradiology, Department of Diagnostic and Interventional Radiology, Turin, Italy; 5Università Cattolica del Sacro Cuore, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Endocrinology and Diabetology Unit, Department of Translational Medicine and Surgery, Rome, Italy; 6Fondazione IRCCS Ca Granda Ospedale Maggiore Policlinico, University of Milan, Endocrinology Unit, Department of Clinical Sciences and Community Health, Milan, Italy; 7University Hospital G. Martino, University of Messina, Endocrine Unit, Department of Human Pathology of Adulthood and Childhood G. Barresi, Messina, Italy; 8General Hospital, Montebelluna, Department of Internal Medicine, Treviso, Italy; 9University of Ferrara, Section of Endocrinology and Internal Medicine, Department of Medical Sciences, Ferrara, Italy
Introduction: First-generation SRL (fg-SRL) represent the treatment of choice in acromegaly patients with post-neurosurgical adenomatous remnant and GH-hypersecretion. Anyway, approximately 60% of patients do not achieve adequate disease control. The main predictors of resistance include male sex, young age, invasiveness of the lesion and its hyperintensity on T2-weighted MRI scans; but also, SSTR2 expression, the cytokeratin pattern, Ki-67 and the presence of AIP gene mutation.
Aim: To identify the most relevant predictors of fg-SRL resistance among the most used clinical and histological parameters, specific to the Italian acromegaly population.
Methods: The MISS was an Italian multicenter, retrospective, case-control study, involving the Centers of Turin, Rome, Milan, Messina, Treviso and Ferrara (study duration 5/2018-12/2020). Non-response was defined after six months of full-dose fg-SRL treatment by the presence of both uncontrolled age-adjusted IGF-I, random GH levels and a tumor shrinkage <20%. Controls were collected in a ratio of 1 to 2 compared to resistant cases.
Results: Ninety-six patients were enrolled (63 resistant cases and 33 controls). Age at diagnosis was associated with the condition of fg-SRL resistance, even when corrected for IGF-I values (coefficient -0.04, OR 0.96, AUC 0.62, P=0.035). An iso/hyperintense signal in T2-weighted MRI scans resulted the strongest radiological predictor (coefficient 1.19, OR 3.3, AUC 0.64, P=0.027), even if corrected for the maximal tumor diameter at diagnosis. Both a low grade SSTR2 expression and a sparsely granulated (SG)/intermediate cytokeratin pattern proved to be predictors of the resistant condition (coefficient 1.52, OR 4.58, AUC 0.7, P=0.013; coefficient 0.97, OR 2.65, P=0.047, respectively); the latter resulting also to be superior to the T2-wheighted intensity on MRI (coefficient 1.71, OR 5.56, AUC 0.76, P=0.003). Among those patients undergone neurosurgery without any neoadjuvant treatment, the absence of an appreciable tumor remnant led to a negligible probability of non-response to medical treatment, even considering random GH at three months after surgery for the inclusion in the model (coefficient -3.09, OR 0.04, AUC 0.82, P=0.003).
Conclusions: A T2-iso/hyperintense MRI pattern was associated with a 3.3-fold greater probability of resistance to fg-SRL. Moreover, both a SG/intermediate granulation pattern and a low grade SSTR2 expression led to a 5 times greater probability of being resistant. Finally, the absence of an appreciable post-surgical remnant suggested a better response to fg-SRL. These factors deserve to be evaluated before setting up medical treatment with fg-SRL. Future guidelines should take this emerging evidence into account when making recommendations on therapeutic choice.