Background Invasive ductal carcinoma (IDC) is the most common type of breast cancer, accounting for 80% of all breast cancer diagnoses. Formalin-fixed paraffin-embedded (FFPE) samples are a valuable resource for diagnosing breast cancers. FFPE samples are often available with patient metadata and clinical outcomes, thus providing a critical sample resource for disease studies. Here we demonstrate the potential of single-cell transcriptomics on dissociated FFPE tissues. We also assessed the same FFPE breast IDC tissue block using the Chromium Single Cell Fixed RNA Profiling and Visium Spatial Gene Expression assays for more accurate estimation of cell-type proportions for each Visium spot.

Methods FFPE scrolls (2 x 25-µm) of human breast IDC were dissociated into a single cell suspension using a combination of enzymatic digestion and mechanical dissociation. Cells were then processed using the Chromium Fixed RNA Profiling workflow. A section from the same FFPE IDC block was processed using the Visium Spatial Gene Expression Solution for FFPE.

Results Clustering analysis of the single cell FFPE data identified 17 distinct cell populations, including luminal epithelial progenitors, mature luminal epithelia, basal cells, fibroblasts, pericytes, and immune cells. Applying the Prediction Analysis of Microarray 50 (PAM50) subtype predictor to these clusters, we identified one cluster showing a strong correlation with a basal-like cancer subtype whereas all other clusters correlated with a normal signature. To compare single cell and spatial data, we utilized Robust Cell Type Decomposition to estimate cell-type proportions for each Visium gene expression spot. Distinct regions were observed for mature and progenitor luminal cells. Interestingly, the mature luminal cells mapped specifically to the region annotated by a pathologist as invasive carcinoma while T and B cells colocalized outside of the carcinogenic region, which may have implications on the patient‘s prognosis.

Conclusions Our studies show that transcriptome-wide gene expression at a single-cell resolution on FFPE-preserved tissue adds a whole new axis of information to clinical studies, providing new opportunities for the analysis of large biobanked cohorts and longitudinal samples of clinical relevance. Additionally, assessing the same FFPE tissue blocks by combining Chromium Single Cell Fixed RNA Profiling with Visium Spatial Gene Expression assays enables a comprehensive dataset with more precise cell type profiling in the spatial context.