Embedded Files
Introduction
Introduction
Colorectal cancer (CRC) is the fourth most common cancer in the United States and the second leading cause of cancer-related deaths [1]. While the five-year survival rate for early-stage CRC (stages I–II) is approximately 91%, it drops dramatically to only about 13% for patients with stage IV metastatic disease [2]. The liver is the most common site of CRC metastasis because tumor cells frequently spread through the portal vein, which drains directly from the colon to the liver [3]. This metastatic process involves several distinct steps (Figure 1), including local invasion from the primary tumor, intravasation into the bloodstream, circulation, extravasation out of the bloodstream, and finally, colonizing distant tissues [4].
Figure 1. Illustrated Overview of the Metastatic Cascade. Adapted from Hapach et al. (2019) [4]
Mouse models of cancer are essential for studying tumor biology and testing potential therapies in a whole-organism context. Although both implantation-based models and genetically engineered mouse models are used to study CRC, robust genetically engineered models that reliably produce metastases from CRC remain limited [5]. Implantation models, like injected cell-line derived xenografts (CDX), are commonly used to study metastatic disease, but these approaches may not fully capture the natural, clinically relevant metastatic process [5]. Introducing tumor cells directly into foreign microenvironments could potentially bypass the selective pressures that shape natural metastasis, making the model fail to replicate the metastatic evolutionary trajectory in vivo.
In a 2017 study by Bocuk et al., CMT-93 cells (an invasive mouse colon tumor cell line) were injected into the liver of a syngeneic mouse via the portal vein, and RNA-seq analysis was used to identify differentially expressed genes (DEGs) between the original cell line, liver metastases, and tumor-free liver tissue [6]. They limited their analysis to 119 metastasis-associated genes based on the commercially available Qiagen product Tumor Metastasis RT2 Profiler PCR Array and reported 32 of these genes were differentially expressed between the cultured CMT-93 cell line and tumor cells collected from liver four weeks after injection. They did not compare their results to any clinical data from human patient samples of colorectal tumors or liver metastases to confirm the translational validity of their mouse model.
To better assess how well this cell-line derived implantation model reflects the transcriptional profiles of human colorectal cancer metastasis, I reanalysed human clinical RNA-seq data from Lee et al. (2016) to check human expression of these 32 reported genes [7]. This data set included matched primary colorectal tumors, liver metastases, and adjacent normal colon and liver tissues from five patients. After identifying differentially expressed genes between primary colon tumors and liver metastatic tumors in the human clinical samples, I compared the direction and significance of the mouse-model-identified genes in the human data to assess whether the transcriptional changes identified in these mice reflect clinically relevant CRC liver metastasis biology. Finally, I used colon adenocarcinoma expression data from The Cancer Genome Atlas (TCGA) to explore the clinical impacts of candidate genes found in my analysis [8].
Figure 2. One million CMT-93 cells were injected into the surgically exposed portal vein (brought to front in diagram for visibility). Tumors and adjacent normal tissue were collected from liver 4 weeks later. (Created in BioRender.com)
Page updated
Google Sites
Report abuse