Spatial Transcriptomic Analysis Reveals Associations between Genes and Cellular Topology in Breast and Prostate Cancers

Abstract

Background: Cancer is the leading cause of death worldwide with breast and prostate cancer the most common among women and men, respectively. Gene expression and image features are independently prognostic of patient survival; but until the advent of spatial transcriptomics (ST), it was not possible to determine how gene expression of cells was tied to their spatial relationships (i.e., topology).

Methods: We identify topology-associated genes (TAGs) that correlate with 700 image topological features (ITFs) in breast and prostate cancer ST samples. Genes and image topological features are independently clustered and correlated with each other. Themes among genes correlated with ITFs are investigated by functional enrichment analysis.

Results: Overall, topology-associated genes (TAG) corresponding to extracellular matrix (ECM) and Collagen Type I Trimer gene ontology terms are common to both prostate and breast cancer. In breast cancer specifically, we identify the ZAG-PIP Complex as a TAG. In prostate cancer, we identify distinct TAGs that are enriched for GI dysmotility and the IgA immunoglobulin complex. We identified TAGs in every ST slide regardless of cancer type.

Conclusions: These TAGs are enriched for ontology terms, illustrating the biological relevance to our image topology features and their potential utility in diagnostic and prognostic models.

Keywords: breast cancer; gene expression; histopathological images; image analysis; integrative analysis; prostate cancer; spatial transcriptomics; topological data analysis.

Figure 1

Workflow of the method from tissue preparation to identifying gene ontology terms associated with ITFs.

Figure 2

Heatmap of correlation matrix between 1-dimensional ITFs and TAG expression values for Parent Visium Human Breast Cancer (A), Parent Human Breast Cancer (B), FFPE Human Breast Cancer (C), Human Prostate Cancer (D), Prostate Acinar Carcinoma I (E), and Visium FFPE Human Normal Prostate (F).

Figure 3

Heatmap of correlation matrix between 0-dimensional image features and gene expression values for Parent Visium Human Breast Cancer (A), Parent Human Breast Cancer (B), FFPE Human Breast Cancer (C), Human Prostate Cancer (D), Prostate Acinar CarcinoI (E), and Visium FFPE Human Normal Prostate (F).

Figure 4

ITFs associated with cellular signaling. Here, 1-dimensional ITF121 is associated with CD8 T-cell interaction with tumor regions likely through MHC-I. Note that the periphery of the tumor region has greater ITF121. (A) H&E pathology image from an ST array. (B) Simplified diagram of T-cell and tumor signaling via MHC-I. (C) T-cell enrichment using DEGAS. Expression of marker genes (D) CD8, (E) MHC-I (HLA-C), and (F) TCR (TRAC). (G) ITF121 was calculated for an image patch around each ST spot. Red arrows indicate region enriched for CD8 T-cells interacting with the tumor.

Figure 5

Results for predicting expression levels of TAGs using machine learning models. (A) Correlation between TAG expression and ITFs. (B) Prediction of MHC-I (HLA-C) expression using an LightGBM model trained on the 12 ITFs in (A).