Delineating histomorphological signatures of iCMS subtypes in colorectal cancer using multiplexed fluorescence-in-situ hybridization.

Ignasius Joanito,Li Bo,Grace Hui Ting Yeo,Vairavan Lakshmanan,Yurike Laurensia,Iain BeeHaut Tan,Shyam Prabhakar
DOI: https://doi.org/10.1200/jco.2024.42.16_suppl.e15603
IF: 45.3
2024-06-01
Journal of Clinical Oncology
Abstract:e15603 Background: Our recent single-cell transcriptomic data indicate that colorectal cancer (CRC) can be classified into two major epithelial subtypes: iCMS2 and iCMS31. iCMS3 CRC includes MSI-H and one-third of MSS tumors. The latter are enriched in KRAS, PIK3CA, and BRAF mutations. The iCMS2 subtype often exhibits chromosome instability, including copy number changes in specific chromosome regions. These genomic alterations may be associated with distinct histomorphological features, which can be assessed using hematoxylin and eosin (H&E) staining. Methods: To test this hypothesis, we acquired and analyzed multiplexed Fluorescence in situ Hybridization (mFISH) and immunofluorescence (IF) staining data from 84 samples from 32 patients. For the mFISH data, we targeted 464 genes, encompassing major cell lineage markers and genes differentially expressed between iCMS2 and iCMS3 patient subtypes. Images were acquired over 31 hybridization rounds. Following the acquisition of the mFISH data, the same tissue section was stained with anti-EPCAM, anti-CD45 and anti-Vimentin antibodies to delineate cell boundaries and distinguish between the major cell lineages. Subsequently, all crypts were segmented based on their EPCAM and VIM stains, and parameters such as crypt size (number of pixels), crypt depth (distance to the nearest VIM signal), and fractal dimension were calculated for each crypt. The output of crypt segmentation was then integrated with mFISH RNA data to elucidate cell type proportions and gene expression profiles across and within individual segmented crypts. Results: Here, we found that iCMS2 crypts, in contrast to iCMS3 crypts, are significantly larger, deeper, and have higher fractal dimension. Additionally, our mFISH data demonstrated that iCMS2 tumors contain a higher proportion of undifferentiated (stem-like) cells, while iCMS3 tumors have a higher proportion of goblet (terminally differentiated) cells. The presence of larger and deeper crypts in iCMS2 may suggest a distinct cell state. Indeed, we identified specific genes whose expression was correlated with crypt depth. Conclusions: Our preliminary results suggest that these two iCMS subtypes exhibit distinct histomorphological features, which can be employed as practical and cost-effective companion diagnostics to assist in the pathological discrimination of these subtypes. References: 1 Joanito, I., Wirapati, P., Zhao, N. et al. Single-cell and bulk transcriptome sequencing identifies two epithelial tumor cell states and refines the consensus molecular classification of colorectal cancer. Nat Genet 54, 963–975 (2022).
oncology
What problem does this paper attempt to address?