Seedlabs
Source research

Systematic benchmarking of imaging spatial transcriptomics platforms in FFPE tissues

Huan Wang, Ruixu Huang, Jack Nelson, Ce Gao, Miles Tran, Anna H. Yeaton et al. · 2025 · 112 citationsRead the paper

Abstract Emerging imaging spatial transcriptomics (iST) platforms and coupled analytical methods can recover cell-to-cell interactions, groups of spatially covarying genes, and gene signatures associated with pathological features, and are thus particularly well-suited for applications in formalin fixed paraffin embedded (FFPE) tissues. Here, we benchmarked the performance of three commercial iST platforms on serial sections from tissue microarrays (TMAs) containing 23 tumor and normal tissue types for both relative technical and biological performance. On matched genes, we found that 10x Xenium shows higher transcript counts per gene without sacrificing specificity, but that all three platforms concord to orthogonal RNA-seq datasets and can perform spatially resolved cell typing, albeit with different false discovery rates, cell segmentation error frequencies, and with varying degrees of sub-clustering for downstream biological analyses. Taken together, our analyses provide a comprehensive benchmark to guide the choice of iST method as researchers design studies with precious samples in this rapidly evolving field.

What the full text shows

Working prototype
  • Xenium demonstrated the highest sensitivity and specificity for shared genes across all tissue types and panels. (Matched genes across 33 FFPE tissue types)
  • CosMx 2024 chemistry showed a significant increase in sensitivity compared to 2023 chemistry. — ~2-fold decrease in transcript count difference vs Xenium (compared to 12-15 fold in 2023) (tTMA1 and tTMA2 samples)
  • CosMx and Xenium achieved higher cell segmentation precision in dense cell distributions than MERSCOPE. — 0.90 precision vs 0.83 (p < 0.01) (Dense cell distributions in tTMA1 (24))
  • CosMx showed the highest correlation to reference bulk RNA-seq (TCGA) and single-cell RNA-seq data. (Breast cancer tTMA1 (24))
  • Xenium provided the highest transcript counts per cell for shared genes in the majority of cancer types. — 21 out of 22 TMA-cancer type combinations (Shared genes across platforms)
  • Xenium has the shortest and least hands-on workflow. — 2-3 days hands-on time vs 5-7 days (MERSCOPE) and 2 days (CosMx) (Standard slide processing)
Setup
33 FFPE tissue types (17 tumor, 16 normal) across three TMAs (tTMA1, tTMA2, nTMA) with >31,400 manually annotated cells for segmentation validation.
How
The study benchmarked three iST platforms (Xenium, MERSCOPE, CosMx) using serial sections of three FFPE tissue microarrays (TMAs). Performance was evaluated via pseudo-bulk gene expression correlation, manual cell segmentation ground-truth comparison, and concordance with reference RNA-seq datasets.
Limitations
MERSCOPE imaging area limited to 1 cm2, leaving some cores unimaged. · CosMx workflow required demanding manual selection of ROIs for each core. · Xenium batch processing limited by the need for separate thermocyclers per slide pair. · Segmentation accuracy decreased for elongated and sparse cell distributions across all platforms.

Summarized by Gemma 4 from the open-access full text · 12 Oct 2026. Read it