Comparison of imaging based single-cell resolution spatial transcriptomics profiling platforms using formalin-fixed paraffin-embedded tumor samples

Our plain-English summary of Comparison of imaging based single-cell resolution spatial transcriptomics profiling platforms using formalin-fixed paraffin-embedded tumor samples, by N. Ozirmak Lermi, M. Molina Ayala, S. Hernandez, et al (Nature Communications, 2025).

The study

• Authors: N. Ozirmak Lermi, M. Molina Ayala, S. Hernandez, W. Lu, K. Khan, A. Serrano, I. Lubo, L. Hamana, K. Tomczak, S. Barnes, J. Dou, Q. Liang, T.E.A.M. Rti, M.G. Raso, X. Tang, M. Jiang, B. Sanchez-Espiridion, A. Weissferdt, J. Heymach, J. Zhang, B. Sepesi, T. Cascone, A. Tsao, M. Altan, R. Mehran, D. Gibbons, I. Wistuba, C. Haymaker, K. Chen, L.M. Solis Soto
• Journal: Nature Communications
• Published: 2025;16(1):8499
• Cited by: 34 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Imaging-based spatial transcriptomics (ST) is evolving as a pivotal technology in studying tumor biology and associated microenvironments. However, the strengths of the commercially available ST platforms in studying spatial biology have not been systematically evaluated using rigorously controlled experiments. We use serial 5 μm sections of formalin-fixed, paraffin-embedded surgically resected lung adenocarcinoma and pleural mesothelioma samples in tissue microarrays to compare the performance of the ST platforms (CosMx, MERFISH, and Xenium (uni/multi-modal)) in reference to bulk RNA sequencing, multiplex immunofluorescence, GeoMx, and hematoxylin and eosin staining data. In addition to an objective assessment of automatic cell segmentation and phenotyping, we perform a manual phenotyping evaluation to assess pathologically meaningful comparisons between ST platforms. Here, we show the intricate differences between the ST platforms, reveal the importance of parameters such as probe design in determining the data quality, and suggest reliable workflows for accurate spatial profiling and molecular discovery.

© 2025. The Author(s).

In plain English

This laboratory study compared three commercial imaging-based spatial transcriptomics platforms: CosMx, MERFISH and Xenium, with Xenium run using two cell-outlining methods. These map gene activity cell by cell in tissue. The authors used stored tissue-microarray samples from MD Anderson: lung adenocarcinoma from 22 patients and pleural mesothelioma from 22 patients. The article does not study asbestos exposure. CosMx detected the most transcripts per cell. In the MESO2 mesothelioma array, 319 of its gene probes (31.9%) read at the same level as its negative controls. Xenium had false discovery rates below 0.09%, compared with 10% and 11% for CosMx in the lung arrays. Statistical comparisons of these rates between platforms were not significant. After quality filtering, MERFISH kept 22.6% of detected cells in ICON1, 40.0% in ICON2 and 64.4% in MESO2. In lung samples scored by three pathologists, Xenium multimodal scored best and MERFISH worst. For mesothelioma, labelling cell types using matched single-cell reference data worked better than manual labelling. The authors concluded that probe design affects data quality. They noted that tumor type and sample age could not be separated.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Laboratory comparison of three commercial imaging-based spatial transcriptomics platforms (CosMx, MERFISH, Xenium unimodal and multimodal) on serial tissue sections, benchmarked against bulk RNA sequencing, GeoMx DSP whole-transcriptome assay, multiplex immunofluorescence, H&E staining and pathologist review
• Who was studied: Formalin-fixed paraffin-embedded surgically resected tumor samples in four tissue microarrays from The University of Texas MD Anderson Cancer Center: lung adenocarcinoma (22 patients, 2 arrays, 53 cores; ICON1 and ICON2) and pleural mesothelioma (22 patients, 2 arrays, 49 cores; MESO1 and MESO2)
• Exposure: Not applicable: the study compared laboratory technologies and does not study or mention asbestos or any other exposure
• Main results:
  • CosMx detected the highest transcript counts and uniquely expressed gene counts per cell among all tissue arrays (p less than 2.2e-16)
  • On CosMx (1,000-gene panel), the number of target gene probes expressed at the same level as negative control probes was 8 (0.8%) in ICON1, 27 (2.7%) in ICON2, 196 (19.6%) in MESO1 and 319 (31.9%) in MESO2
  • False discovery rates for CosMx, calculated with negative control probes, were 10% and 11% in the lung (ICON) arrays and 5.8% and 6.2% in the mesothelioma (MESO) arrays
  • Xenium unimodal and multimodal had false discovery rates below 0.09% in both lung and mesothelioma arrays
  • Differences in false discovery rates between platforms were not statistically significant: CosMx versus Xenium p = 0.057, MERFISH versus Xenium p = 0.25, Xenium unimodal versus multimodal p = 1
  • The percentage of detected cells remaining after quality filtering on MERFISH was 22.6% in ICON1, 40.0% in ICON2 and 64.4% in MESO2
  • In the older lung array (ICON2), correlation with bulk RNA sequencing was lower for CosMx (R = 0.43) and MERFISH (R = 0.39) than for Xenium-MM
  • In the newest mesothelioma array (MESO2), correlation with bulk RNA sequencing was higher for CosMx (R = 0.58) and MERFISH (R = 0.64) than in ICON2
  • In the pathologists' evaluation of the ICON2 lung array, Xenium-MM had better F1-scores (median above 75%) than the other platforms for all cell types
  • MERFISH had the lowest F1-scores for all cell types in the pathologists' evaluation
  • For the pleural mesothelioma arrays, transferring cell-type labels from matched single-cell RNA sequencing reference data performed better than manual annotation for identifying sub-cell types
• The authors' conclusion: The authors concluded that the platforms differ in intricate ways and that parameters such as probe design are important in determining data quality
• Limitations the authors state:
  • Samples came from a single institution, covered few tumor types, and did not include whole tissue sections
  • Some assays lacked negative control or blank probes, which limited assessment of transcript count quality
  • Assays had to be run on adjacent serial sections rather than the same section
  • The older and newer arrays contained different tissue types, so findings about sample age are only informative
  • Differences in tumor location between the bulk RNA sequencing samples and the arrays could not be ruled out

Topics

The National Library of Medicine indexes this study under subjects our site covers: Lung cancer, Mesothelioma.

Medical terms in this study, explained in our Lexicon: Lung Adenocarcinoma, Pleural Mesothelioma, Discovery.

Related studies

• Comparison of imaging-based single-cell resolution spatial transcriptomics profiling platforms using formalin-fixed, paraffin-embedded tumor samples — N.O. Lermi, M.M. Ayala, et al. (2025)
• Numerical and structural genomic aberrations are reliably detectable in tissue microarrays of formalin-fixed paraffin-embedded tumor samples by fluorescence in-situ hybridization — H. Horn, J. Bausinger, et al. (2014)
• Validation of a Gene Expression Test for Mesothelioma Prognosis in Formalin-Fixed Paraffin-Embedded Tissues — A. De Rienzo, R.W. Cook, et al. (2017)
• Differential expression of N-cadherin in pleural mesotheliomas and E-cadherin in lung adenocarcinomas in formalin-fixed, paraffin-embedded tissues — A.C. Han, A. Peralta-Soler, et al. (1997)
• Discrepancies of asbestos body and fiber content between formalin-fixed and corresponding paraffin embedded lung tissue — B.K. Kuhn, F. Barresi, et al. (2021)
• Comparative analysis of asbestos body and fiber content in formalin-fixed vs. paraffin-embedded lung tissue — B.K. Kuhn, B. Vrugt, et al. (2025)

Browse more research in Nature Communications, or return to medical research.