ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Endocrine Abstracts (2019) 63 GP243 | DOI: 10.1530/endoabs.63.GP243

Histological classification of pituitary neuroendocrine tumors: genomic insights on cell lineage

Chiara Villa1,2,3, Mario Neou2, Roberta Armignacco2, Anne Jouinot2,4, Amandine Septier2, Karine Perlemoine2, Michèle Bernier1, Marie Laure Raffin-Sanson5,6, Bertrand Baussart7, Jérôme Bertherat2,8, Stéphan Gaillard7 & Guillaume Assié2,8


1Foch Hospital, Department of Pathological Cytology and Anatomy, Suresnes, France; 2Cochin institute – Inserm U1016 - Cnrs UMR8104 - University Paris Descartes, Paris, France; 3University of Liège, CHU de Liège, Department of Endocrinology, Liége, Belgium; 4Department of Medical Oncology, Cochin Hospital, Paris Descartes University, CARPEM, AP-HP, Paris, France; 5Department of Endocrinology and Nutrition, Centre Hospitalier Universitaire Ambroise Paré, AP-HP, Boulogne-Billancourt, France; 6INSERM U1173, Université de Versailles Saint-Quentin-en-Yvelines, Montigny-le-Bretonneux, France; 7Foch Hospital, Department of Neurosurgery, Suresnes, France; 8Department of Endocrinology, Center for Rare Adrenal Diseases, Hôpital Cochin, AP-HP, Paris, France.


The 2017 World Health Organization (WHO) classification of pituitary adenomas is based on cell lineage and transcription factors (TFs). Pituitary progenitors expressing Pit-1 are driven towards the somato-lacto-thyrotroph differentiation, T-Pit towards corticotroph, and SF-1 towards gonadotroph. We recently generated a multi-genomic classification of pituitary neuroendocrine tumors (PitNETs) (abstract submitted to ECE2019 by Neou M.). Transcriptome classification identified six groups: corticotroph overt Cushing; lactotroph; silent corticotroph; gonadotroph and null-cell; thyrotroph, Pit-1 plurihormonal and sparsely-granulated somatotroph; somatotroph and mixed GH-PRL.

Aim: To characterize the lineage differentiation in PitNETs.

Methods: Transcriptome of 134 PitNETs (RNA sequencing) was used to determine TFs expression at mRNA level, and to provide a canonical transcriptome signature for each cell-type (cortico-, lacto-, somato-, thyro- and gonadotroph). Pathological study of the present serie of PitNETs included the histological examination and the immunohistochemical tests for all pituitary hormones, proliferation markers and TFs including GATA3.

Results: Pit1 lineage. As expected, Pit1 mRNA showed expression exclusively in somatotroph (23/23), mixed GH-PRL(8/8), thyrotroph (6/6), lactotroph (16/16) and Pit-1 plurihormonal (9/9) PitNETs. Immunohistochemistry confirmed this observation. Based on transcriptome classification, accurate thresholds of immunoexpression for GH, PRL and TSH were established in order to define the different Pit-1 subtypes:

PRL >10%, with GH ≤5% for lactotroh

GH >10%, with PRL ≤5% for somatotroph

TSH >10%, with GH and PRL <5% for thyrotroph

PRL and GH positivity >5% for mixed GH-PRL

Combined PRL, GH, TSH, FSH, and LH positivity for Pit-1 plurihormonal.

T-Pit lineage

T-Pit mRNA showed the expected expression in corticotroph PitNETs (35/35), lower in silent ones (Wilcoxon P<10–5). Immunohistochemistry confirmed this profile.

SF1 lineage

SF1 mRNA expression showed the expected high expression in gonadotroph PitNETs (29/29), but also in a subset of somatotroph PitNETs (9/21). SF1 immunopositivity was confirmed in this somatotroph subgroup.

The cell-type specific transcriptome signatures revealed an unexpected gonadotroph signature in silent corticotroph PitNETs, accounting for 40% (median) of silent corticotroph signature. GATA3 TF was one of the gonadotroph signature genes. Immunohistochemistry for GATA3 confirmed positivity in gonadotroph (29/29) and in silent corticotroph PitNETs (6/8).

Conclusion: This study showed for the first time on a large series of PitNETs the correlation between pathological and molecular classifications. The Pit1 and T-Pit TFs are strong determinants of PitNETs molecular classification. In contrast, gonadotroph lineage is more complex, with possible transdifferentiation in silent corticotroph and in a subset of somatotroph PitNETs.