Colorectal cancer is driven by the accumulation of driver mutations, but the contributions of specific mutations to different steps in malignant progression are not fully understood. In this study, we generated mouse models harboring different combinations of key colorectal cancer driver mutations () in intestinal epithelial cells to comprehensively investigate their roles in the development of primary tumors and metastases. mutation caused intestinal adenomas and combination with mutation or deletion induced submucosal invasion. The addition of mutation yielded epithelial-mesenchymal transition (EMT)-like morphology and lymph vessel intravasation of the invasive tumors. In contrast, combinations of with and mutation were insufficient for submucosal invasion, but still induced EMT-like histology. Studies using tumor-derived organoids showed that was critical for liver metastasis following splenic transplantation, when this mutation was combined with either plus or deletion, with the highest incidence of metastasis displayed by tumors with a genotype. RNA sequencing analysis of tumor organoids defined distinct gene expression profiles characteristic for the respective combinations of driver mutations, with upregulated genes in tumors found to be similarly upregulated in specimens of human metastatic colorectal cancer. Our results show how activation of Wnt and Kras with suppression of TGFβ signaling in intestinal epithelial cells is sufficient for colorectal cancer metastasis, with possible implications for the development of metastasis prevention strategies. These findings illuminate how key driver mutations in colon cancer cooperate to drive the development of metastatic disease, with potential implications for the development of suitable prevention strategies. .

Background and study aims Decisions concerning additional surgery after endoscopic resection of T1 colorectal cancer (CRC) are difficult because preoperative prediction of lymph node metastasis (LNM) is problematic. We investigated whether artificial intelligence can predict LNM presence, thus minimizing the need for additional surgery. Patients and methods Data on 690 consecutive patients with T1 CRCs that were surgically resected in 2001 – 2016 were retrospectively analyzed. We divided patients into two groups according to date: data from 590 patients were used for machine learning for the artificial intelligence model, and the remaining 100 patients were included for model validation. The artificial intelligence model analyzed 45 clinicopathological factors and then predicted positivity or negativity for LNM. Operative specimens were used as the gold standard for the presence of LNM. The artificial intelligence model was validated by calculating the sensitivity, specificity, and accuracy for predicting LNM, and comparing these data with those of the American, European, and Japanese guidelines. Results Sensitivity was 100 % (95 % confidence interval [CI] 72 % to 100 %) in all models. Specificity of the artificial intelligence model and the American, European, and Japanese guidelines was 66 % (95 %CI 56 % to 76 %), 44 % (95 %CI 34 % to 55 %), 0 % (95 %CI 0 % to 3 %), and 0 % (95 %CI 0 % to 3 %), respectively; and accuracy was 69 % (95 %CI 59 % to 78 %), 49 % (95 %CI 39 % to 59 %), 9 % (95 %CI 4 % to 16 %), and 9 % (95 %CI 4 % – 16 %), respectively. The rates of unnecessary additional surgery attributable to misdiagnosing LNM-negative patients as having LNM were: 77 % (95 %CI 62 % to 89 %) for the artificial intelligence model, and 85 % (95 %CI 73 % to 93 %; P < 0.001), 91 % (95 %CI 84 % to 96 %; P < 0.001), and 91 % (95 %CI 84 % to 96 %; P < 0.001) for the American, European, and Japanese guidelines, respectively. Conclusions Compared with current guidelines, artificial intelligence significantly reduced unnecessary additional surgery after endoscopic resection of T1 CRC without missing LNM positivity.

Background and Aim: Recent advances in endoscopic technology have allowed many T1 colorectal carcinomas to be resected endoscopically with negative margins. However, the criteria for curative endoscopic resection remain unclear. We aimed to identify risk factors for nodal metastasis in T1 carcinoma patients and hence establish the indication for additional surgery with lymph node dissection. Methods: Initial or additional surgery with nodal dissection was performed in 653 T1 carcinoma cases. Clinicopathological factors were retrospectively analyzed with respect to nodal metastasis. The status of the muscularis mucosae (MM grade) was defined as grade 1 (maintenance) or grade 2 (fragmentation or disappearance). The lesions were then stratified based on the risk of nodal metastasis. Results: Muscularis mucosae grade was associated with nodal metastasis (P = 0.026), and no patients with MM grade 1 lesions had nodal metastasis. Significant risk factors for nodal metastasis in patients with MM grade 2 lesions were attribution of women (P = 0.006), lymphovascular infiltration (P < 0.001), tumor budding (P = 0.045), and poorly differentiated adenocarcinoma or mucinous carcinoma (P = 0.007). Nodal metastasis occurred in 1.06% of lesions without any of these pathological factors, but in 10.3% and 20.1% of lesions with at least one factor in male and female patients, respectively. There was good inter-observer agreement for MM grade evaluation, with a kappa value of 0.67. Conclusions: Stratification using MM grade, pathological factors, and patient sex provided more appropriate indication for additional surgery with lymph node dissection after endoscopic treatment for T1 colorectal carcinomas.

PurposeSubmucosal invasion depth (SID) in colorectal carcinoma (CRC) is an important factor in estimating risk of lymph node metastasis, but can be difficult to measure, leading to inadequate or over-extensive treatment. Here, we aimed to clarify the practical aspects of measuring SID in T1 CRC.MethodsWe investigated 568 T1 CRCs that were resected surgically at our hospital from April 2001 to December 2013, and relationships between SID and clinicopathological factors, including the means of measurement, lesion morphology, and lymph node metastasis.ResultsOf these 568 lesions, the SID was ≥1000 μm in 508 lesions. SIDs for lesions measured from the surface layer were all ≥1000 μm. Although lesions with SIDs ≥1000 μm were associated with significantly higher levels of unfavorable histologic types and lymphovascular infiltration than shallower lesions, a depth of ≥1000 μm was not a significant risk factor for lymph node metastasis (LNM) (6.7 vs. 9.8 %; P = 0.64), and no lesions for which the sole pathological factor was SID ≥1000 μm had lymph node metastasis. Protruded lesions showed deeper SIDs than other types.ConclusionsAlthough we found several problems of measuring SID in this study, we also found, surprisingly, that SID is not a risk factor for lymph node metastasis, and its measurement is not needed to estimate the risk of lymph node metastasis.

Alternative splicing, regulated by DEAD‐Box Helicase (DDX) families, plays an important role in cancer. However, the relationship between the DDX family and cancer has not been fully elucidated. In the present study, we identified a candidate oncogene DDX56 on Ch.7p by a bioinformatics approach using The Cancer Genome Atlas (TCGA) dataset of colorectal cancer (CRC). DDX56 expression was measured by RT‐qPCR and immunochemical staining in 108 CRC patients. Clinicopathological and survival analyses were carried out using three CRC datasets. Biological roles of DDX56 were explored by gene set enrichment analysis (GSEA), and cell proliferation in vitro and in vivo, cell cycle assays, and using DDX56‐knockdown or overexpressed CRC cells. RNA sequencing was carried out to elucidate the effect of DDX56 on mRNA splicing. We found that DDX56 expression was positively correlated with the amplification of DDX56 and was upregulated in CRC cells. High DDX56 expression was associated with lymphatic invasion and distant metastasis and was an independent poor prognostic factor. In vitro analysis, in vivo analysis and GSEA showed that DDX56 promoted proliferation ability through regulating the cell cycle. DDX56 knockdown reduced intron retention and tumor suppressor WEE1 expression, which functions as a G2‐M DNA damage checkpoint. We have identified DDX56 as a novel oncogene and prognostic biomarker of CRC that promotes alternative splicing of WEE1.