Contrast-Enhanced CT Parameters of Gastric Adenocarcinoma: Can Radiomic Features Be Surrogate Biomarkers for HER2 Over-Expression Status?

Autor: Wang, Na, Wang, Xinxin, Li, Wenya, Ye, Huajun, Bai, Hongzhao, Wu, Jiansheng, Chen, Mengjun
Předmět:
Zdroj: Cancer Management & Research; Feb2020, Vol. 12, p1211-1219, 9p
Abstrakt: aim of this study was to determine the role of contrast-enhanced computed tomography (CE-CT) parameters in predicting the expression status of HER2 in gastric adenocarcinoma (GAC) patients before radical gastrectomy. Materials and Methods: A total of 460 GAC patients who underwent non-contrast CT (NC-CT) and CE-CT examinations before radical resection were enrolled in this retrospective study. The radiologists reviewed their CT scans and recorded parameters, including CT attenuate value (CAV) and corrected CAV (cCAV). The pathologist identified the postoperative HER2 expression status, and HER2 expression status was evaluated by immunohistochemical staining (IHC). The association between CE-CT parameters and HER2 expression status was analyzed. Results: Among the 460 patients, 84 patients had HER2 over-expression status, at a prevalence of 18.3%. The CAVs were significantly different between the 2 different HER2 expression groups in the non-contrast and arterial phases (non-contrast phase: p = 0.005; arterial phase: p < 0.001). Besides, there was a significant difference in the cCAVs between the 2 groups in the arterial phase (arterial phase: p = 0.003). Univariate and multivariate logistic regression analyses identified that the maximum diameter of tumor, differentiation degree, CAV in non-contrast, arterial, and portal phases, and cCAV in the arterial phase were predictive factors of HER2 expression status. Conclusion: Our analyses showed that the CE-CT parameters were significantly different between different HER2 expression groups. CE-CT parameters could serve as simple, objective predictive factors of HER2 expression status of GAC patients. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index