Spatial transcriptomics reveals gene expression characteristics in invasive micropapillary carcinoma of the breast.

Invasive micropapillary carcinoma (IMPC) is a special histological subtype of breast cancer, featured with extremely high rates of lymphovascular invasion and lymph node metastasis. Based on a previous series of studies, our team proposed the hypothesis of "clustered metastasis of IMPC tumor cells". However, the transcriptomics characteristics underlying its metastasis are unknown, especially in spatial transcriptomics (ST). In this paper, we perform ST sequencing on four freshly frozen IMPC samples. We draw the transcriptomic maps of IMPC for the first time and reveal its extensive heterogeneity, associated with metabolic reprogramming. We also find that IMPC subpopulations with abnormal metabolism are arranged in different spatial areas, and higher levels of lipid metabolism are observed in all IMPC hierarchical clusters. Moreover, we find that the stromal regions show varieties of gene expression programs, and this difference depends on their distance from IMPC regions. Furthermore, a total of seven IMPC hierarchical clusters of four samples share a common higher expression level of the SREBF1 gene. Immunohistochemistry results further show that high SREBF1 protein expression is associated with lymph node metastasis and poor survival in IMPC patients. Together, these findings provide a valuable resource for exploring the inter- and intra-tumoral heterogeneity of IMPC and identify a new marker, SREBF1, which may facilitate accurate diagnosis and treatment of this disease.