The clinical significance of the ACRG classification in gastric adenocarcinoma and the establishment of a p53 expression prediction model based on biopsy histopathology
Objective To investigate the clinicopathological significance of the Asian Cancer Research Group (ACRG) classification in gastric adenocarcinoma and to develop a predictive model for p53 protein expression in gastric adenocarcinoma. Methods A total of 137 patients diagnosed with gastric adenocarcin...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Journal of Precision Medicine
2025-06-01
|
| Series: | 精准医学杂志 |
| Subjects: | |
| Online Access: | https://jpmed.qdu.edu.cn/fileup/2096-529X/PDF/1750385735658-887485893.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Objective To investigate the clinicopathological significance of the Asian Cancer Research Group (ACRG) classification in gastric adenocarcinoma and to develop a predictive model for p53 protein expression in gastric adenocarcinoma. Methods A total of 137 patients diagnosed with gastric adenocarcinoma who underwent surgery at the Affiliated Hospital of Qingdao University between 2015 and 2023 were included in this study. Postoperative pathological specimens were collected and immunohistochemically stained to determine the ACRG classification. Patients were grouped according to ACRG subtypes. These groups were compared in terms of age, sex, surgical approach, tumor location, differentiation status, pathological type, pre-chemotherapy Borrmann classification, clinical stage, tumor marker levels, tumor recurrence, and metastasis. Preoperative hematoxylin and eosin-stained endoscopic biopsy slides from all patients were digitized by scanning and randomly divided into a training set (109 cases) and a validation set (28 cases). Self-supervised learning and feature extraction were performed on the training set images using SimCLR with ResNet-18 as the backbone network. Subsequently, the images were input into a feature aggregation-based classification network to establish a predictive model for p53 protein expression in gastric adenocarcinoma. The model performance was eva-luated in the validation set. Receiver operating characteristic curves were plotted for both the training and validation sets, and the area under the curve, accuracy, sensitivity, specificity, positive predictive value, and negative predictive value were calculated. Results The ACRG classification of gastric adenocarcinoma was significantly associated with clinical stage (χ2=13.049,P<0.05). In the training set, the area under the curve, accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the p53 protein expression prediction model were 0.920, 0.719, 0.896, 0.724, 0.875, and 0.763, respectively; these values in the validation set were 0.890, 0.643, 0.938, 0.250, 0.625, and 0.750, respectively. Conclusion The ACRG classification of gastric adenocarcinoma is significantly correlated with clinical stage. The p53 protein expression prediction model can assist in preoperative ACRG classification, thereby improving the accuracy of diagnosis and treatment of gastric adenocarcinoma. |
|---|---|
| ISSN: | 2096-529X |