BACKGROUND Intestinal and diffuse gastric carcinomas differ in morphology and growth behavior. Differentiated type gastric carcinoma (DGC), which corresponds roughly with the intestinal type of Lauren, can demonstrate phenotypic properties associated with mucin expression and brush border. However, their clinical significance is controversial. A classification based on mucin phenotype and brush border was performed to determine the clinicopathologic diversity of DGCs in their early stage. METHODS A total of 120 specimens from 116 DGC patients with definite submucosal invasion were evaluated both macroscopically and histologically. All sections were examined immunohistochemically with human gastric mucin, Muc‐2, and CD10 and with mucin histochemically with paradoxical concanavalin A staining and high iron Diamine‐Alcian Blue. They were classified into gastric type (G‐type), intestinal type (I‐type), mixed gastric and intestinal type (M‐type), or null type (N‐type) phenotypes. The immunoreactivity of E‐cadherin and β‐catenin also was investigated to determine the correlation between mucin phenotype and clinicopathologic factors. RESULTS The G‐type phenotype was found to be in contrast to I‐type: G‐type was an independent factor associated with lymph node metastasis. Significant correlations were observed between the G‐type phenotype and the complex type carcinoma found that was histologically: lymphatic invasion, lymph node metastasis, and the abnormal expression of E‐cadherin. A significant difference in the proportion of mucin phenotypes between papillary type and tubular type carcinoma was observed. G‐type was found to be the predominant phenotype in papillary carcinoma in contrast to tubular carcinoma. CONCLUSIONS The G‐type mucin phenotype and papillary adenocarcinoma should be distinguished from other types of DGCs because of their increased malignant potential in the incipient phase of invasion and metastasis. The significance of G‐type and papillary adenocarcinoma should be reflected in the treatment of patients with early stage DGCs, including endoscopic mucosal resection. Cancer 2000;89:724–32. © 2000 American Cancer Society.
Purpose: We aimed to develop an ovarian cancer-specific predictive framework for clinical stage, histotype, residual tumor burden, and prognosis using machine learning methods based on multiple biomarkers.Experimental Design: Overall, 334 patients with epithelial ovarian cancer (EOC) and 101 patients with benign ovarian tumors were randomly assigned to "training" and "test" cohorts. Seven supervised machine learning classifiers, including Gradient Boosting Machine (GBM), Support Vector Machine, Random Forest (RF), Conditional RF (CRF), Na€ ve Bayes, Neural Network, and Elastic Net, were used to derive diagnostic and prognostic information from 32 parameters commonly available from pretreatment peripheral blood tests and age.Results: Machine learning techniques were superior to conventional regression-based analyses in predicting multiple clinical parameters pertaining to EOC. Ensemble meth-ods combining weak decision trees, such as GBM, RF, and CRF, showed the best performance in EOC prediction. The values for the highest accuracy and area under the ROC curve (AUC) for segregating EOC from benign ovarian tumors with RF were 92.4% and 0.968, respectively. The highest accuracy and AUC for predicting clinical stages with RF were 69.0% and 0.760, respectively. High-grade serous and mucinous histotypes of EOC could be preoperatively predicted with RF. An ordinal RF classifier could distinguish complete resection from others. Unsupervised clustering analysis identified subgroups among early-stage EOC patients with significantly worse survival.Conclusions: Machine learning systems can provide critical diagnostic and prognostic prediction for patients with EOC before initial intervention, and the use of predictive algorithms may facilitate personalized treatment options through pretreatment stratification of patients.NOTE: There were too few early-stage EOC patients with residual tumor. A definition for the significance of bold is P value of < 0.05.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.