Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Collectively Predicting Biomarkers for PD-1 Antibody in Gastric Cancer by Texture Analysis of Computed Tomography
ABSTRACT
Background:
The amplification of HER-2, expression of PD-L1, or microsatellite instability high (MSI-H) condition all indicate benefits of PD-1 antibody for advanced gastric cancer.
Objective:
The present study was performed to collectively evaluate this markers for PD-1 antibody.
Methods:
The imaging and pathological data of 461 patients underwent radical gastrectomy for advanced gastric cancer at our hospital between January 2020 and December 2022 were retrospectively reviewed. Patients were categorized into two groups: the PD-1 antibody panel positive group, comprising patients with HER-2 amplification, PD-L1 positive, or MSI-H condition, and the negative group. The texture features of the tumor region from arterial and portal vein phase CT images were extract. Prediction models were constructed by using the features from the arterial phase images, portal vein phase images, and their fusion, respectively.
Results:
Of the 461 patients, 147 patient (31.9%) were classified into the panel positive group. The clinical features were similar between the two groups. The results revealed predictive AUCs of 0.69 (95%CI 0.57 - 0.81) for the arterial phase model, 0.79 (95%CI 0.58 - 0.99) for the portal vein phase model, and a higher AUC of 0.82 (95%CI 0.68 - 0.95) for the combined fusion model.
Conclusions:
The status of HER-2, PD-L1, and MSI-H condition can be assessed collectively by CECT texture analysis. Of the features from varied phases, portal vein phase features are preferred over arterial phase features to construct the model, and a fusion of features from both phases may further improve the accuracy. Clinical Trial: NA.
Citation
