Single-cell and spatial transcriptomic analyses reveal heterogeneity characteristics and specific cell subtype regulators in growth hormone-secreting pituitary adenomas

Background: Growth hormone-secreting pituitary adenomas (GHPA) display diverse biological behaviors and clinical outcomes, necessitating the identification of tumor heterogeneity and prognostically relevant markers. Methods: In this study, we performed single-cell RNA sequencing (scRNA-seq) on 10 GHPA samples, four of which also underwent spatial transcriptome sequencing, and used scRNA-seq data from four normal pituitary samples as controls. Cell subtype characterization in GHPA was analyzed using multiple algorithms to identify malignant bias regulators, which were then validated using a clinical cohort. Results: We constructed the first single-cell and spatial transcriptome profiles of GHPA, which contained 87 862 cells and revealed 16 tumor cell subtypes. Among the tumor cells, we identified distinct developmental trajectories and three malignant-biased subtypes (PIT1_C05, PIT1_C06, and PIT1_C10). The spatial distribution characteristics of these malignant-biased cells may influence the growth characteristics and prognosis of GHPA. We screened specific regulatory transcription factors, including FOXO1, GTF2IRD1, and MAX. Clinical cohort validation indicated that FOXO1 might be associated with tumor invasion and progression, while high expression of MAX could result in poor endocrine outcomes. Conclusion: GHPA exhibits rich heterogeneity and diverse cell subtypes, with specific transcription factors potentially regulating cell malignant bias, thereby influencing tumor characteristics and prognosis.