Reframing thymic epithelial tumors through single-cell transcriptomics: a narrative review

Thymic epithelial tumors (TETs) are rare neoplasms originating from thymic epithelial cells (TECs). They exhibit histological, molecular, and immunological diversity shaped by aberrant interactions between neoplastic TECs and immune cells. World Health Organization (WHO) histological classification effectively captures their clinicopathological features, whereas inter- and intra-subtype heterogeneity persists. In this point, single-cell transcriptomics is well-suited to resolve cellular heterogeneity in TETs. This narrative review synthesizes findings from recent single-cell studies that illuminate epithelial subpopulations, intra-tumoral immune landscape, and disease-specific aberrations such as myasthenia gravis (MG)-associated phenotypes in TETs.

We conducted a focused narrative review on three key studies that applied single-cell RNA sequencing (scRNA-seq) to primary human TETs.

Single-cell analyses identified a distinct TEC subpopulation ectopically expressing neuromuscular antigens, termed neuromuscular medullary TECs (nmTECs), in MG-associated thymomas, implicating aberrant antigen presentation and immune cell recruitment in autoantibody production. Other studies delineated tumor epithelial programs that shape the immune microenvironment, linking TEC transcriptional identities to the persistence of immature T-cells or dominance of mature T-cells. Bioinformatic approaches of ligand-receptor networks supported pathogenic crosstalk between neoplastic TECs and immune cells. Notably, independent groups converged on functionally oriented classifications that align across studies, stratifying TETs by epithelial lineage programs and immune composition with correspondence to biological behavior and prognosis.

Single-cell approaches have revealed TET heterogeneity encompassing abnormal epithelial cell states, immune development trajectories, and autoimmune-associated molecular programs with high resolution. The ongoing accumulation of single-cell transcriptomics datasets would refine a cell-centric framework for TET classification and pathogenesis, while also deepening insight into normal TEC biology.