Introduction: Gonadotroph tumours are the most common type of pituitary neuroendocrine tumour (PitNETs). Before routine immunohistochemical staining for transcription factors, hormone-negative gonadotroph tumours were often misclassified as null cell tumours, a distinct and more clinically aggressive entity1,2. Steroidogenic factor 1 (SF-1) is a lineage-specific marker for gonadotroph pitNETs, with weaker staining potentially reflecting less differentiated tumours3. Gata3 is a surrogate marker for Gata2, required for development of gonadotrophs and a useful adjunct in classification of PitNETs4. The correlation of SF-1, GATA34,5 and CAM5.2, an epithelial cytokeratin biomarker for NETs6, with characteristics of gonadotroph pitNETs remains to be elucidated.
Objective: To evaluate whether reduced staining for SF-1, GATA3 and/or CAM5 correlates with a higher clinicopathological score (CPC)7 and/or the presence of radiological or histological evidence of tumour invasiveness and proliferation in a cohort of gonadotroph pitNETs.
Methods: Retrospective analysis of 59 gonadotroph tumours resected between 2021 – 2024. Sections were reported for percentage of positive cells and staining intensity (strong vs weak) for SF-1, GATA3 and CAM5 using tissue immunohistochemistry analysis software (Indica Labs, Halo). Clinical notes and imaging were reviewed for CPC and knosp score.
Results: Lower GATA3 nuclear staining was observed in radiologically invasive (41.6%) compared to non-invasive tumours (59.3%). Within radiologically invasive tumours, SF-1 strong positive staining was lower in proliferative compared to non-proliferative tumours (28.9% vs 56.1%) and tumours with higher knosp scores showed a reduced GATA3 strong positive nuclear staining pattern. The combined percentage of positive staining for SF1, CAM5 and GATA3 was lower in radiologically invasive gonadotrophs compared to non-invasive. Histological invasiveness and CPC alone had no correlation with staining patterns.
Conclusion: Our results indicate that immunohistochemical expression of SF1, CAM5 and GATA3 correlates more closely with radiological invasiveness, as assessed by Knosp grade, than with other traditional parameters used to define 'high-risk' gonadotroph PitNETs.