Salivary Glycopatterns as Biomarkers for Differential Diagnosis of HBV-Induced Cirrhosis and Hepatocellular Carcinoma

Abstract

Background: Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related death worldwide, however, there has been little success in developing effective blood-based methods for early detection of HCC. The present study systematically investigated the alterations in salivary glycopatterns related to HBV-infected chronic liver diseases, and evaluated the possibility of saliva-based protein glycopatterns as potential biomarkers for diagnosis of HBV-induced cirrhosis (HC) and HCC. Methods: The correlation of alterations in salivary glycopatterns related to HBV-infected chronic liver diseases were investigated in the discovery cohort. The diagnostic models of HC and HCC were constructed based on the alterations of salivary glycopatterns in training and validation cohort. This study enrolled 491 patients with HBV-infected chronic liver diseases and 156 age- and sex-matched healthy volunteers between January 2010 and May 2017. The availability of the diagnostic models was confirmed by an independent test in a single-blind cohort with 104 subjects from January 2018 to January 2019. A small-scale lectin microarray based on the candidate lectins that exhibited significant alterations of protein glycopatterns in saliva with the development of HCC, and the constructed models of HC and HCC were used for diagnosis of HC and HCC. Findings: The diagnostic models of HC indicated that Model HC referred to three lectins (MAL-I, AAL and DSA) had higher diagnostic accuracy (AUC: 0·76, sensitivity: 0·82 and specificity: 0·73) in the validation cohort and a blind-test accuracy of 0·856 for distinguishing HC from HB and HCC patients. The diagnostic models of HCC indicated that Model HCC referred to four lectins (MAL-I, RCA120, LTL, and MAL-II) had higher diagnostic accuracy (AUC: 0·85, sensitivity: 0·71 and specificity: 0·82) in the validation cohort and a blind-test accuracy of 0·875 for distinguishing HCC from HB and HC patients. Notably, 20 cases of 21 HCC with AFP(-) were identified as HCC-positive by Model HCC. Interpretation: The salivary glycopatterns could be used as biomarkers for differential diagnosis of HBV-infected chronic liver diseases and should be further examined in larger studies. Funding Statement: This research is supported by National Natural Science Foundation of China (No. 81372365, to Li and 81670577, to Guo). Declaration of Interests: The authors declared no competing interests. Ethics Approval Statement: The collection and use of saliva for the research presented here were approved by the Ethical Committee of Northwest University, Shaanxi Provincial People’s Hospital, Xi’an Jiaotong University and Fourth Military Medical University (Xi’an, China). Written informed consent was received from participants.