Colorectal cancer (CRC) metastases frequently recur due to minimal residual disease and persistent micrometastases after therapy. Here, we performed spatial multimodal profiling using spot-level and high-resolution spatial transcriptomics, multi-regional whole-genome sequencing following laser-capture microdissection, and high-plex protein imaging to map 49 tumors from 19 patients, encompassing paired primary CRC and matched liver (CLiM) and lung (CLuM) metastases. Phylogenetic reconstruction revealed that liver micrometastases (CLiMi) represent early clonal divergences that maintain a stem-like, quiescent phenotype consistent with metastatic dormancy. Spatially, we uncovered distinct stromal barriers: while macrometastases were encapsulated by myofibroblasts, micrometastases were surrounded by an immunosuppressive niche characterized by T-cell exhaustion and unique ligand-receptor signaling networks. Notably, we identified a CLiMi-specific six-gene signature that robustly predicts MRD status and disease-free survival across three independent cohorts. These findings elucidate the spatial evolution of CRC metastases and provide a tissue-based foundation for precision surveillance and therapeutic targeting.
Software Versions
The following software and tools were used in this study:
Seurat v5.2.1, BayesSpace v1.10.1, CytoTRACE v0.3.3, ggraph v2.2.1, survminer v0.5.0, survival v3.7.0, survcomp v1.54.0, GSVA v1.52.3, NMF v0.27, SpatialInferCNV v1.0.1, CellChat v2.1.2, CNVkit v0.9.12
cell2location v0.1.4, Stardist v0.8.3, Cellpose v4.0.4, SMURF v1.0.2
Meta_data
The Meta_data directory contains pathological annotations for: Visium and Visium HD spatial transcriptomics samples
These annotations include region labels, pathological classifications, and associated metadata used for downstream spatial analyses.
All additional data used for figure generation, including processed matrices and intermediate outputs, have been deposited on Zenodo.
Zenodo repository: 👉 [https://zenodo.org/records/17796737]
For any questions, please leave your comment in GitHub or contact Yang Liu (yliu47@mdanderson.org). We will help address the issues as soon as possible.