We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Development and Validation of a Novel Prognostic Model for Predicting Lymph Node Metastasis in Early Colorectal Cancer.
- Authors
Ji Hyun Ahn; Min Seob Kwak; Hun Hee Lee; Jae Myung Cha; Hyun Phil Shin; Jung Won Jeon; Jin Young Yoon
- Abstract
Background/Aims Population screening for colorectal cancer (CRC) is expected to increase the number of early stage tumor. The identification of simplified prediction model for lymph node metastasis (LNM) for early CRC is urgently needed to determine the treatment and follow-up strategies. This study aims to develop an accurate prediction model for the LNM in early CRC. Methods We used data from the 2004 to 2016 Surveillance, Epidemiology, and End Results database to develop and validate the prediction models for LNM. Six models, namely, neural network, naïve-Bayesian, Support vector machines, logistic regression, decision tree, and random forest (RF) models were assessed. Results A total of 32,725 patients with a diagnosis of early CRC (Tis and T1) were analyzed in the study. The model included eight independent prognostic variables, including age at diagnosis, sex, race, primary site, histologic type, tumor grade, tumor size, and tumor depth. LNM was significantly more frequently in patients with larger tumor, in elderly patients, and in patients with more poorly differentiated tumor. The RF algorithm had the best predictive performance in comparison to the other methods, achieving an accuracy of 94.7%, a sensitivity of 71.4%, a specificity of 92.2%, and an area under the curve of 0.90. Conclusions Our data show that the age at diagnosis is the most important feature in predicting LNM of early CRC in the RF model. We established a simplified reproducible predictive model for LNM in early CRC that could be used to guide treatment decisions.
- Subjects
COLORECTAL cancer; LYMPH nodes; SUPPORT vector machines; TUMOR classification; METASTASIS
- Publication
Gut & Liver, 2019, Vol 13, Issue 6(suppl. 1), p7
- ISSN
1976-2283
- Publication type
Article