figshare
Browse
ORT487267_sm_suppl table.pdf (463.35 kB)

Supplementary Material for: Cadherin Expression Shift Could Well Distinguish Esophageal Squamous Cell Carcinoma from Non- Cancerous Esophageal Tissues

Download (463.35 kB)
dataset
posted on 2018-05-15, 12:15 authored by Shengtao Zhu, Juan Liu, Li Min, Xiujing Sun, Qingdong Guo, Hengcun Li, Zheng Zhang, Yu Zhao, Junchao Gu, Shutian Zhang
Summary Background: Biomarkers for esophageal squamous cell carcinoma (ESCC) identification with high sensitivity are not well established. Since abnormal expression of cadherins has been widely reported in cancer, we explored its feasibility as an ESCC biomarker. Methods: Expression of E-cadherin and N-cadherin were detected in 150 esophageal tissues by immunohistochemistry. Staining strength and percentage in different subcellular structures of each specimen were evaluated by 2 independent pathologists. A logistic regression-based classifier derived from E-cadherin and N-cadherin staining was generated. Results: E-cadherin exhibited decreased membrane expression in ESCC, while N-cadherin exhibited decreased expression in the nucleus but elevated expression in the cytoplasm. Both E-cadherin and N-cadherin could distinguish ESCC tissues from non-cancerous tissues (area under the curve (AUC) = 0.748, 0.801, respectively). E-cadherin and N-cadherin staining scores could be merged into a cadherin (CDH) logistic index, which showed better discrimination (AUC = 0.909) than E-cadherin or N-cadherin alone. Further investigation indicated that the CDH logistic index was significantly correlated with tumor size and differentiation in ESCC. Conclusion: Both E-cadherin and N-cadherin had a strong expression shift in ESCC compared with non-cancerous tissues. The CDH logistic index, a parameter integrating the expression data of both cadherins, could be used as a marker with high sensitivity and specificity in the identification of ESCC.

History

Usage metrics

    Oncology Research and Treatment

    Categories

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC