Transcriptome analysis of archived tumors by Visium, GeoMx DSP, and Chromium reveals patient heterogeneity

Yixing Dong ¹, Chiara Saglietti ², Quentin Bayard ³

¹Biomedical Data Science Center, Lausanne University Hospital; University of Lausanne, Lausanne, Switzerland.
²Institute of Pathology, Department of Laboratory Medicine and Pathology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland.
³Owkin, Paris, France.
⁴Department of Oncology, Lausanne University Hospital; Swiss Cancer Center Leman, Lausanne, Switzerland.
⁵Ludwig Insitute for Cancer Research, Lausanne, Switzerland.
⁶Agora Translational Research Center, Lausanne, Switzerland.
⁷Institute of Pathology, Department of Laboratory Medicine and Pathology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland. Laurence.DeLeval@chuv.ch.
⁸Biomedical Data Science Center, Lausanne University Hospital; University of Lausanne, Lausanne, Switzerland. Raphael.Gottardo@chuv.ch.
⁹Agora Translational Research Center, Lausanne, Switzerland. Raphael.Gottardo@chuv.ch.
¹⁰Swiss Institute of Bioinformatics, Lausanne, Switzerland. Raphael.Gottardo@chuv.ch.
¹¹School of Life Sciences, Ecole Polytechnique Fédérale de, Lausanne, Switzerland. Raphael.Gottardo@chuv.ch.
¹²Department of Oncology, Lausanne University Hospital; Swiss Cancer Center Leman, Lausanne, Switzerland. Krisztian.Homicsko@chuv.ch.
¹³Ludwig Insitute for Cancer Research, Lausanne, Switzerland. Krisztian.Homicsko@chuv.ch.
¹⁴Agora Translational Research Center, Lausanne, Switzerland. Krisztian.Homicsko@chuv.ch.
¹⁵Owkin, Paris, France. elo.madissoon@owkin.com.

⁰Biomedical Data Science Center, Lausanne University Hospital; University of Lausanne, Lausanne, Switzerland.

Nature Communications +

|

May 12, 2025

View abstract on PubMed

Summary

New spatial transcriptomics technologies for FFPE tissues offer high-quality data. Visium and Chromium excel at discovering tumor heterogeneity and drug targets, making them ideal for large-scale projects.

Area Of Science

Genomics
Molecular Biology
Cancer Research

Background

Advancements in probe-based, full-transcriptome technologies like Visium CytAssist, Chromium Flex, and GeoMx DSP allow analysis of archival Formalin-Fixed Paraffin-Embedded (FFPE) tissues.
These technologies facilitate extensive cohort studies but can be labor-intensive and costly, necessitating careful selection based on research goals.

Purpose Of The Study

To compare the performance of Visium CytAssist, Chromium Flex, and GeoMx DSP for spatial transcriptomics on FFPE tumor samples.
To evaluate data quality, reproducibility, cell mixture detection, and the ability to discover tumor heterogeneity and potential drug targets.

Main Methods

Comparative analysis of Visium CytAssist, Chromium Flex, and GeoMx DSP platforms.
Application of these methods to FFPE tumor samples from Breast cancer, Non-Small Cell Lung Cancer (NSCLC), and Diffuse Large B-cell Lymphoma (DLBCL).
Assessment of data quality, reproducibility, cellular composition, tumor heterogeneity, and drug target identification.

Main Results

All tested methods generated good-quality, highly reproducible data from FFPE tumor samples.
GeoMx DSP data exhibited cell mixtures, even after marker-based preselection.
Visium CytAssist and Chromium Flex demonstrated superior performance in identifying tumor heterogeneity and potential drug targets compared to GeoMx DSP.

Conclusions

Visium CytAssist and Chromium Flex are recommended for high-throughput and discovery-oriented spatial transcriptomics projects.
GeoMx DSP, despite being more labor-intensive, remains valuable for targeted analyses addressing specialized research questions.