Single-Cell Atlas of Non-Functioning Gonadotroph Tumors and Anterior Pituitary Reveals Lineage Infidelity and Clinically Associated Transcriptional Programs

Non-functioning gonadotroph (NFG) tumors are the most common type of non-functioning pituitary adenomas and cause significant morbidity through mass effect. The molecular programs that distinguish NFG tumor cells from normal gonadotrophs and underlie tumor growth remain poorly understood. The aim of this study was to define transcriptional differences between NFG tumors and the normal anterior pituitary gland (APG) at single-cell resolution and to identify programs associated with tumor growth. We performed single-cell/nucleus RNA-sequencing of 21 NFG tumors and 8 APG samples, followed by integrative analyses of differential gene expression, regulatory networks, and cell-cell communication. We profiled 77,342 APG cells and 152,649 NFG tumor cells. Within the APG, we identified novel marker genes for each neuroendocrine cell type and distinct transcriptional signatures of anterior and posterior pituitary stem cell populations. Compared with normal gonadotrophs, tumor cells shared a core transcriptomic signature conserved across samples despite inter-tumor heterogeneity. Many conserved genes were markers of non-gonadotroph neuroendocrine and stem cells, suggesting that lineage infidelity is a recurrent feature of NFGs. Pseudobulk profiling identified genes positively correlated with tumor volume, including the known oncogenes CAD, BRF2, and SOX12. Finally, tumors showed increased proportions of myeloid, endothelial, and mural cells and crosstalk between tumor, immune, and stromal populations. This study defines transcriptomic signatures of cell identity in the normal APG and NFG tumors. Our findings reveal programs associated with NFG lineage infidelity and tumor growth, highlighting candidate predictors of patient outcomes and potential therapeutic targets.