Fig. 1

Study design overview and pathway hallmarks in PitNET PBMCs. a Schematic representation of the overall study framework, integrating large-scale transcriptomic and single-cell datasets from PitNET tumor tissues, PitNET PBMC and normal PBMC samples. ^*The gene expression data from normal pituitary tissues (n = 107) were obtained from the genotype-tissue expression database (GTEx)⁸³ of the Broad Institute. ^#For normal PBMCs (n = 175), nine of them were in-house generated, and others were obtained from public database. b Principal Component Analysis (PCA) plots using protein-coding genes. In the left panel, samples are colored by tissue type: yellow indicates tumor tissues, red denotes PitNET PBMCs, and blue corresponds to normal PBMCs. In the right panel, tumor tissues are further classified by hormone groups (PRL, GH, ACTH, and NFPA). The % value indicates the explained variance. c PCA plot focusing exclusively on PBMCs, illustrating the separation between PitNET PBMCs (red) and normal PBMCs (blue). d Volcano plot showing DEGs between PitNET PBMCs and normal PBMCs. Significantly upregulated genes (red) were identified by adjusted P-value < 0.05 and log₂(fold change) > 0.58 (fold change > 1.5), whereas significantly downregulated genes (blue) satisfy adjusted P-value < 0.05 and log₂(fold change) < -0.58. Other genes (gray) are considered not significant. Statistical analyses were performed using the limma³⁶ R package, and P-values were adjusted with Benjamini-Hochberg correction. Two-sided P-values were used. Dot plots highlighting significantly enriched pathways using upregulated (e) and downregulated (f) DEGs in PitNET PBMCs compared with normal PBMCs. Each dot represents a pathway calculated via Enrichr,⁸¹ colored by enrichment P-values (calculated by Enrichr) and sized according to gene count. Red dots mark pathways significantly enriched for upregulated genes, while blue dots mark those enriched for downregulated genes. Two-sided P-values were used