Benchmarking robust spatial transcriptomics approaches to capture the molecular landscape and pathological architecture of archived cancer tissues

Abstract

Applying spatial transcriptomics (ST) to explore a vast amount of formalin-fixed paraffin-embedded (FFPE) archival cancer tissues has been highly challenging due to several critical technical issues. In this work, we optimised ST protocols to generate unprecedented spatial gene expression data for FFPE skin cancer. Skin is among the most challenging tissue types for ST due to its fibrous structure and a high risk of RNAse contamination. We evaluated tissues collected from ten years to two years ago, spanning a range of tissue qualities and complexity. Technical replicates and multiple patient samples were assessed. Further, we integrated gene expression profiles with pathological information, revealing a new layer of molecular information. Such integration is powerful in cancer research and clinical applications. The data allowed us to detect the spatial expression of non-coding RNAs. Together, this work provides important technical perspectives to enable the applications of ST on archived cancer tissues.

Competing Interest Statement

The authors have declared no competing interest.

Footnotes

* https://stlearn.readthedocs.io/en/latest/