Integrating spatial omics with routine haematoxylin and eosin in formalin-fixed paraffin-embedded: a step-by-step...

Haematoxylin and eosin (H&E) remain the foundation of tissue diagnosis, yet many clinical questions, tumour-immune architecture, spatial heterogeneity, and predictors of therapy response, require molecular context that routine slides cannot provide. Spatial omics closes this gap by mapping RNA and proteins in situ while preserving morphology, and recent platforms are increasingly compatible with formalin-fixed paraffin-embedded (FFPE) tissue, paving the way for its future potential use in routine pathology and retrospective cohorts. However, significant challenges related to cost, complexity, reproducibility, and regulation currently remain before widespread routine deployment, underscoring that it is not yet ready for In Vitro Diagnostic or routine clinical application. This mini-review offers a pragmatic, step-by-step workflow for integrating spatial assays with H&E: define the clinical decision; select a fit-for-purpose modality (whole-transcriptome spot/grid vs targeted in situ RNA; multiplex proteomics); lock pre-analytics aligned to histology (sectioning, staining, de-crosslinking, storage); pre-specify regions of interest (ROIs), registration, and segmentation rules; analyse with quality-assurance gates (normalisation, deconvolution, batch handling, spatial statistics); and validate and report using orthogonal assays and multi-site replication. FFPE-ready platforms and typical use-cases are summarised, with emphasis on pre-analytical factors that materially affect signal and analysis "recipes" distilled from recent benchmarks. Brief clinical exemplars illustrate how H&E-anchored spatial maps change decisions by pinpointing actionable niches (e.g., immune neighbourhoods, vascular niches, layer-specific programmes). Common limitations are also outlined, including technology trade-offs, pre-analytics, sampling bias, segmentation and deconvolution error, batch effects, cost, turnaround, and regulatory considerations. Future directions include standards and metadata, cross-platform integration, prospective evidence, automation and quality assurance, and multi-omic detection. Overall, the goal is to support pathology and translational teams in adopting spatial omics in FFPE with both discipline and rigor, guiding the necessary steps to ensure reproducibility and credibility for eventual clinical impact.