Charting the Heterogeneity of Colorectal Cancer Consensus Molecular Subtypes using Spatial Transcriptomics

Abstract

The heterogeneity of colorectal cancer (CRC) contributes to substantial differences in patient response to standard therapies. The consensus molecular subtypes (CMS) of CRC is the most widely-used gene expression-based classification and has contributed to a better understanding of disease heterogeneity and prognosis. Nevertheless, CMS intratumoral heterogeneity restricts its clinical application, stressing the necessity of further characterizing the composition and architecture of CRC. Here, we used Spatial Transcriptomics (ST) in combination with single-cell RNA sequencing (scRNA-seq) to decipher the spatially resolved cellular and molecular composition of CRC. In addition to mapping the intratumoral heterogeneity of CMS and their microenvironment, we identified cell communication events in the tumor-stroma interface of CMS2 carcinomas. This includes tumor growth-inhibiting as well as -activating signatures, such as RNF43-dependent modulation of the WNT pathway or the PLAU-PLAUR ligand-receptor interaction. Our data show the power of ST to bring the CMS-based classification of CRC to another level and thereby gain useful molecular insights for personalized therapy.

Tumor-agnostic transcriptome-based classifier identifies spatial infiltration patterns of CD8+ T cells in the tumor microenvironment and predicts clinical outcome in early- and late-phase clinical trials

“…Among desert-enriched genes, we retrieved the glucose transported GLUT10 (encoded by SLC2A10 ) and the Wnt pathway inhibitor NKD1, which showed preferential expression in tumor cells and fibroblasts ( figures 5B –D and supplemental figure S5B ). By using publicly available spatial transcriptomics data in CRC (Valdeolivas et al 2023 44 and Wu et al 2022 43 ), we confirmed an enrichment in NKD1 expression in CD8-desert as compared with CD8-high tumor areas. In addition, we observed the strongest expression of LAMP5 in the sample showing CD8 infiltration in the stromal area only, in line with LAMP5 being excluded-enriched ( supplemental figure S5C ).…”

Image-Based Consensus Molecular Subtyping in Rectal Cancer Biopsies and Response to Neoadjuvant Chemoradiotherapy

“…We thus undertook comprehensive simulation experiments on existing fully digitized clinical cohorts to capture the morphology related to transcriptional CMS calls and to address whether clinically established sampling protocols are sufficient to describe tumour heterogeneity at the gene-expression level. As expected and due to the spatially heterogeneous nature of CRC tumours [30,31], our results suggest that sampling less than five tumour biopsies is not sufficient to properly predict the CMS call that would be obtained from an equivalent resection specimen of the same tumour. Our results corroborate the 2021 ESGE recommendation as our experiments show on two independent external test datasets of 147 and 266 patients, that five or more standard tumorous biopsy fragments are sufficient to reliably capture the global tumour phenotype needed to achieve CMS classification performance with fidelity close to full resection specimens.…”

“…Color visualization is based on the highest voted tile-level predicted imCMS class among the five trained models that form the ensemble model of experiment [E3a] as indicated in the bottom-right legend. Classification maps illustrate the spatially resolved imCMS calls across biopsy samples including samples with some level of pervasive heterogeneity as previously described for both image-based as well as molecular analysis methods [13, 33, 34]. …”

Learning tissue representation by identification of persistent local patterns in spatial omics data

“…Kasumi representations can easily be adapted to explore tissue-specific common and differential patterns between conditions, even beyond association to clinical features. Furthermore, while we considered view compositions capturing a single spatial context, Kasumi can be deployed with more complex compositions addressing different spatial and functional contexts and different omics technologies [34][35][36][37][38] .…”