Search

Background & Aims: Deep submucosal invasion (DSI) is considered a key risk factor for lymph node metastasis (LNM) and important criterion to recommend surgery in T1 colorectal cancer. However, metastatic risk for DSI is shown to be low in the absence of other histologic risk factors. This meta-analysis determines the independent risk of DSI for LNM., Methods: Suitable studies were included to establish LNM risk for DSI in univariable analysis. To assess DSI as independent risk factor, studies were eligible if risk factors (eg, DSI, poor differentiation, lymphovascular invasion, and high-grade tumor budding) were simultaneously included in multivariable analysis or LNM rate of DSI was described in absence of poor differentiation, lymphovascular invasion, and high-grade tumor budding. Odds ratios (OR) and 95% CIs were calculated., Results: Sixty-seven studies (21,238 patients) were included. Overall LNM rate was 11.2% and significantly higher for DSI-positive cancers (OR, 2.58; 95% CI, 2.10-3.18). Eight studies (3621 patients) were included in multivariable meta-analysis and did not weigh DSI as a significant predictor for LNM (OR, 1.73; 95% CI, 0.96-3.12). As opposed to a significant association between LNM and poor differentiation (OR, 2.14; 95% CI, 1.39-3.28), high-grade tumor budding (OR, 2.83; 95% CI, 2.06-3.88), and lymphovascular invasion (OR, 3.16; 95% CI, 1.88-5.33). Eight studies (1146 patients) analyzed DSI as solitary risk factor; absolute risk of LNM was 2.6% and pooled incidence rate was 2.83 (95% CI, 1.66-4.78)., Conclusions: DSI is not a strong independent predictor for LNM and should be reconsidered as a sole indicator for oncologic surgery. The expanding armamentarium for local excision as first-line treatment prompts serious consideration in amenable cases to tailor T1 colorectal cancer management., (Copyright © 2022 The Author(s). Published by Elsevier Inc. All rights reserved.)

Background & Aims: In accordance with guidelines, most patients with T1 colorectal cancers (CRC) undergo surgical resection with lymph node dissection, despite the low incidence (∼10%) of metastasis to lymph nodes. To reduce unnecessary surgical resections, we used artificial intelligence to build a model to identify T1 colorectal tumors at risk for metastasis to lymph node and validated the model in a separate set of patients., Methods: We collected data from 3134 patients with T1 CRC treated at 6 hospitals in Japan from April 1997 through September 2017 (training cohort). We developed a machine-learning artificial neural network (ANN) using data on patients' age and sex, as well as tumor size, location, morphology, lymphatic and vascular invasion, and histologic grade. We then conducted the external validation on the ANN model using independent 939 patients at another hospital during the same period (validation cohort). We calculated areas under the receiver operator characteristics curves (AUCs) for the ability of the model and US guidelines to identify patients with lymph node metastases., Results: Lymph node metastases were found in 319 (10.2%) of 3134 patients in the training cohort and 79 (8.4%) of /939 patients in the validation cohort. In the validation cohort, the ANN model identified patients with lymph node metastases with an AUC of 0.83, whereas the guidelines identified patients with lymph node metastases with an AUC of 0.73 (P < .001). When the analysis was limited to patients with initial endoscopic resection (n = 517), the ANN model identified patients with lymph node metastases with an AUC of 0.84 and the guidelines identified these patients with an AUC of 0.77 (P = .005)., Conclusions: The ANN model outperformed guidelines in identifying patients with T1 CRCs who had lymph node metastases. This model might be used to determine which patients require additional surgery after endoscopic resection of T1 CRCs. UMIN Clinical Trials Registry no: UMIN000038609., (Copyright © 2021 AGA Institute. Published by Elsevier Inc. All rights reserved.)