by Xijun Min, Dr Yan Li, Liuyan Chen, Chengwei Yao, Dr Xi Qin
As a traditional tumour marker for the diagnosis of early primary hepatic carcinoma (PHC), alpha fetoprotein (AFP) has a low positive rate. Serum glypican 3 (GPC3) is a novel biomarker, and the combined examination of GPC3 and AFP will provide a new strategy for the diagnosis of PHC. We aimed to compare and evaluate the clinical value of GPC3 and AFP alone or in combination for the diagnosis of PHC. The concentrations of AFP and GPC3 in serum were measured in 181 healthy people, 121 patients with chronic hepatitis B and 143 patients with confirmed PHC by chemiluminescence and non-competitive ELISA solid-phase immunoassay. The Spearman rank correlation test was utilized to analyse the correlation of AFP and GPC3. The ROC curve was drawn and the diagnostic value of GPC3 and AFP in PHC were analysed and compared. There was a significant correlation between GPC3 and AFP (Rs=0.757, P<0.05). The positive rate of GPC3 was 59.38% in patients with confirmed PHC whose AFP was negative. The sensitivity and accuracy of combination detection (88.73%, 90.56%) were significantly higher than those of GPC3 alone (74.83%, 88.54%) and AFP (85.24%, 81.80%) (P<0.05). The Youden index (0.7841) of the combined detection was significantly higher than that of the individual application of GPC3 (0.699) and AFP (0.496) (P<0.05). Our results suggest that the single detection of GPC3 has higher clinical value than that of AFP. Combined GPC3 and AFP detection has higher sensitivity and accuracy than single detection, and has higher application value in clinical diagnosis of PHC.
Primary hepatic carcinoma (PHC) is among the most common malignant tumours. In recent years, its incidence ranks fifth among all malignant tumour types, with about 900 000 new cases occurring annually in the world [1,2]. In China, PHC ranks fourth in the incidence and second in the mortality of common malignant tumours, respectively [3,4]. The early symptoms of PHC are not obvious, and about 60% of cases are found at an advanced stage, often when tumour metastasis has occurred . Therefore, early detection, early diagnosis and early treatment are vital for the improvement of the survival rate of PHC patients. Currently, the most widely applied means of clinical diagnosis of early PHC is the detection of alpha fetoprotein (AFP) in serum combined with imaging examination. However, liver inflammation and cirrhosis sometimes interfere with the early diagnosis of PHC, and therefore the optimum time for treatment can be missed . AFP has been used in the clinical diagnosis and prognosis evaluation of PHC: research demonstrated that 400μg/L is considered to be the best clinical cut-off value for PHC in patients with suspected chronic liver disease, the sensitivity and specificity of AFP are 49–71% and 49–86% respectively, and the AFP level of 32–59% patients with hepatic carcinoma is within the standard scope . Therefore, the early diagnosis of hepatic carcinoma is prone to be either missed or misdiagnosed.
At present, AFP, AFP L3 (Lens culinaris-agglutinin-reactive fraction of AFP) and DCP [des-gamma carboxyprothrombin, also known as protein induced by vitamin K absence/antagonist-II (PIVKA-II)] have been identified as serum molecular markers for the diagnosis of hepatic carcinoma. The “Guidelines for the diagnosis and treatment of hepatocellular carcinoma (2019 edition)” propose that in addition to AFP, AFP L3, DCP and cell-free plasma microRNA can also be identified as molecular markers for early diagnosis of hepatic carcinoma, especially for people with negative serum AFP . The American Association for the Study of Liver Diseases (AASLD), the European Association for the Study of the Liver (EASL) and the Japanese Society of Hepatology (JGH) suggest that although AFP, AFP L3 and DCP tests cannot serve as indispensable conditions for the confirmation of the diagnosis of hepatic carcinoma, they can play a major role in diagnosis and recurrence monitoring [9–11]. Nevertheless, these tumour markers are not sensitive to small hepatic carcinoma with lumps ≤2cm in diameter, and patients with chronic liver disease are also likely to have positive test results. A study by Edoo and colleagues shows that the joint detection of AFP, carcinoembryonic antigen (CEA) and cancer antigen (CA)19-9 for the screening and diagnosis of hepatic carcinoma is not better than AFP alone .
Therefore, there is an urgent need to find a novel diagnostic indicator of significant clinical value that is more sensitive and specific than AFP for the diagnosis of early hepatic carcinoma, especially for hepatic carcinoma with negative AFP. Serum glypican 3 (GPC3) is a novel biomarker. Studies have detected high expression of GPC3 in more than 70% of hepatic carcinomas, even in the early stage of hepatic carcinoma progression, but not in normal adult tissues. GPC3 also has a bearing on the metastasis and prognosis of hepatic carcinoma, and is expected to be applied as a new target for early diagnosis, prognosis evaluation and immunotherapy of hepatic carcinoma [13,14]. El-Saadany et al. conducted a study on 80 patients with primary hepatic carcinoma in Egypt, by adopting flow cytometry (FCM) and ELISA for the detection of GPC3 and AFP, respectively; they concluded that GPC3 was more sensitive for the diagnosis of hepatic carcinoma than AFP and the combined detection of GPC3 and AFP showed high sensitivity (98.5%) and specificity (97.8%) . In recent years, it has been found that a single test analyte cannot meet the diagnostic needs of early hepatic carcinoma, which implies that the combination of multiple test indicators is of great utility for improving the positive detection rate. Our study took hepatic carcinoma patients, chronic hepatitis B patients and healthy people as the research subjects. The purpose of this study was to explore the clinical diagnostic value of GPC3 and AFP and their combination in primary hepatic carcinoma, as well as the clinical diagnostic significance of GPC3 in AFP-negative hepatic carcinoma.
Patients and methods
This study included 143 patients with hepatic carcinoma (who had not been treated surgically), who were hospitalized at Hainan Cancer Hospital (Haikou, China) from July to November 2018. Without tumour, node, metastasis (TNM) staging, the included patients received B-ultrasound and CT examination. The diagnosis of hepatic carcinoma was in accordance with the “Guidelines for diagnosis and treatment of primary liver cancer in China (2017 edition)” . The control group was made up of 121 patients with chronic hepatitis B at the same period, all cases were confirmed by Hepatitis B serologic test and clinical examination. Among them, 25 cases were among those with positive HBsAg, HBeAg, HBcAb markers, 96 cases were among those with positive HBsAg, HBeAb, HBcAb markers. The control group was composed of 181 healthy participants. The group of confirmed hepatic carcinoma patients included 108 males and 35 females, average age 56.4±12.7 years. The group of chronic hepatitis B included 106 males and 15 females, average age 51.85±14.1 years. The diagnosis of viral hepatitis B reached the diagnostic specification of the “Guideline of prevention and treatment for chronic hepatitis B (2015 update)” . The healthy population group included 147 males and 34 females, average age 54.11±13.4 years. There was no statistically significant difference between the groups in terms of age, sex and other general characteristics (P>0.05).
This study was approved by the Ethics Committee of Hainan Cancer Hospital, and informed consent forms were signed by all subjects enrolled.
Specimen collection and pretreatment
Fasting venous blood was taken from all subjects in the morning. After natural agglutination, all samples were centrifuged at 3700rpm for 10min, and the serum was separated and stored at −20°C for further experiments.
Determination of serum AFP and GPC3 concentration
AFP detection kit (cat. no. 37104503, Roche Applied Science) was used to detect the concentration of AFP by chemiluminescence assay (COBAS-e-602, Roche Applied Science), AFP ≥7ng/ml was positive. GPC3 detection kit (cat. no. 41651U, CanAg Diagnostics) was used to determine the content of GPC3 in serum by non-competitive ELISA solid-phase method, GPC3 ≥0.095ng/ml was positive. Absorbance was measured at 450nm using a microplate reader (MULTISKAN-FC, Thermo Fisher Scientific). The test operation steps and quality controls had been operated in strict accordance with the instructions of the kits.
SPSS 17.0 statistical software was used for the statistical analysis of data. The measurement data were processed by analysis of variance or t-test, and the counting data were analysed by nonparametric test. P<0.05 was used to express that the difference is significant. Spearman correlation analysis was used, P<0.05 was used to express that the difference is statistically significant. Receiver operating characteristic (ROC) curve was used to determine the best critical value of GPC3 in the diagnosis of hepatic carcinoma, and AUC was used to evaluate the diagnostic value of GPC3 and AFP in primary hepatic carcinoma.
Comparison of GPC3 and AFP levels among groups
The levels of GPC3 and AFP were 0.21±0.19ng/ml and 19.50±11.45ng/ml respectively in the group with confirmed hepatic carcinoma. The levels of GPC3 and AFP were 0.07±0.03ng/ml and 3.25±1.16ng/ml of chronic hepatitis B patients with positive HBsAg, HBeAg, HBcAb markers. The levels of GPC3 and AFP of chronic hepatitis B patients with positive HBsAg, HBeAb, HBcAb markers were 0.07±0.05ng/ml and 2.69±0.96ng/ml respectively. The levels of GPC3 and AFP of healthy subjects were 0.04±0.01 ng/ml and 1.96±1.52ng/ml respectively. The levels of serum GPC3 and AFP of patients with confirmed hepatic carcinoma were significantly higher than those in chronic hepatitis B patients with positive HBsAg, HBeAg, HBcAb markers, chronic hepatitis B patients with positive HBsAg, HBeAb, HBcAb markers and the healthy subjects (P<0.05). In the control group, among patients with chronic hepatitis B, the serum GPC3 level of those with positive HBsAg, HBeAb, HBcAb markers was not statistically significant compared with those with positive HBsAg, HBeAg, HBcAb markers (P>0.05), but showed an upward trend. Compared with healthy subjects, the GPC3 and AFP levels of chronic hepatitis B patients with positive HBsAg, HBeAb, and HBcAb markers and those with positive HBsAg, HBeAg, HBcAb markers showed no significant difference. The results are shown in Table 1.