Des-gamma-carboxyprothrombin is a favorable biomarker for the early diagnosis of alfa-fetoprotein-negative hepatitis B virus-related hepatocellular carcinoma.

Introduction

Hepatocellular carcinoma (HCC) is one of the most common malignant tumors in adults in Asia and the third leading cause of cancer-related death.[1,2] There are about 470,000 new cases of HCC in China every year, accounting for 55% of the worldwide incidence.[3,4] The occurrence of HCC in China is mainly related to hepatitis virus infection, hepatic fibrosis, and biliary cirrhosis,[5,6] and about 1% to 5% of cases of liver cirrhosis per year develop into HCC.[7] Treatments for HCC include surgical resection, interventional embolization, radiofrequency ablation, liver transplantation, and chemotherapy.[8] However, the postoperative recurrence and metastasis rates remain high and the long-term survival rate for more than 5 years is low.[9]

Chronic hepatitis B virus (HBV) infection (CHB) is the major risk factor for HCC in China.[1] Although technologies such as ultrasound imaging, contrast-enhanced computed tomography, magnetic resonance imaging, and percutaneous transhepatic biopsy have greatly improved the rate of HCC diagnosis, these methods have some disadvantages, such as high cost, invasiveness, and a low efficiency for small lesions. Testing for serum tumor markers is thus an important complement for HCC diagnosis. Alpha-fetoprotein (AFP) is the most frequent biomarker used in the clinic; however, 30% to 40% of HCCs may present as AFP-negative.[10,11] Furthermore, increased AFP levels may also be detected in patients with chronic active hepatitis.[12] It thus remains difficult to improve the efficiency of the early diagnosis of AFP-negative HCC, especially HBV-related HCC.

Des-gamma-carboxyprothrombin (DCP; also known as protein induced by vitamin K deficiency or antagonist-II (PIVKA II)) is a prothrombin precursor synthesized by the liver. Liebman et al.[13] first reported serum DCP in patients with liver cancer in 1984, and >90% of patients with HCC have since been shown to have strong positive expression of DCP.[14] Sterling et al.[15] found that more than half of patients with AFP-negative liver cancers had elevated DCP, while liver cancer could be ruled out in patients with normal DCP levels, despite positive nodules on ultrasound results. DCP is therefore considered to be a diagnostic marker for AFP-negative HCC and an independent prognostic indicator.[16] This retrospective study aimed to evaluate the early diagnostic value of serum DCP for distinguishing AFP-negative HBV-related HCC.

Materials and methods

Study population and DCP measurement

This study retrospectively enrolled patients treated at Beijing YouAn Hospital between March 2010 and July 2018. The participants were divided into four subgroups: CHB group, liver cirrhosis (LC) group, early-stage HCC group, and advanced-stage HCC group. The inclusion criteria for the HCC groups in this study were: (1) diagnosis confirmed by histopathological examination, (2) chronic HBV infection, (3) age 20 to 80 years, (4) serum collected preoperatively, and (5) serum AFP level ≤20 ng/mL. The exclusion were: (1) intrahepatic cholangiocarcinoma or mixed HCC, (2) hepatitis C virus, alcoholic liver disease, or biliary cirrhosis, and (3) other cancers. Early-stage HCC was defined as a single lesion or up to three lesions ≤3 cm. Advanced-stage HCC was defined as lesions >3 cm or accompanied by metastasis. CHB and LC were diagnosed according to the Guideline of Prevention and Treatment for Chronic Hepatitis B (2015 update).[17] This study was approved by the research ethics committee of Beijing YouAn Hospital. Written informed consent was obtained from all participants. Serum AFP and DCP levels were measured using an automated chemiluminescent enzyme immunoassay system (LUMIPULSE® G1200; Fujirebio Inc., Tokyo, Japan), according to the manufacturer’s instructions.

Statistical analysis

The data were analyzed using SPSS version 22.0 (IBM, Armonk, NY, USA) and GraphPad Software version 7 (Graph Pad Software, San Diego, CA, USA). Differences within subgroups were compared using the Mann–Whitney test (for non-normally distributed data) and Student’s t-test (for normally distributed data). The clinicopathological characteristics were compared with Pearson’s χ2 test. The diagnostic value of DCP was evaluated by analyzing the receiver-operating characteristic curves (ROC) and areas under the curves (AUC). A value of P < 0.05 was considered significant.

Results

Patient characteristics

The baseline characteristics of the study population are presented in Table 1. A total of 210 patients were enrolled, including 87 patients with AFP-negative HBV-related HCC and 123 with CHB or LC. There was a male predominance in both groups (P = 0.008) (Table 1). The Child–Pugh grade was well-matched between the groups. There were significant differences among the HCC, LC, and CHB groups in terms of alanine aminotransferase, aspartate aminotransferase, total bilirubin, and total protein (all P < 0.05) (Table 1). Serum DCP levels were significantly higher in patients with early- or advanced-stage HCC compared with the CHB and LC groups (P = 0.032) (Figure 1).

Table 1.

Clinical features of the study population.

Feature	CHB	LC	Early-stage HCC	Advanced-stage HCC	P value
Number	51	72	62	25
Sex
Male	30	47	52	21	0.008
Female	21	25	10	4
Age (year)
≤55	45	44	39	10	<0.001
>55	6	28	23	15
Child–Pugh
A	51	51	52	21	0.183
B	0	17	10	3
C	0	4	0	1
ALT, U/L
≤50	48	67	47	18	0.02
>50	3	5	15	7
AST, U/L
≤40	46	54	47	13	0.03
>40	5	18	15	12
TBil, µmol/L
≤21	41	37	44	19	0.04
>21	10	35	18	6
TP, g/L
≤65	3	27	18	9	0.01
>65	48	45	44	16
Cirrhosis
Yes	0	72	43	15	<0.001
No	51	0	19	10
DCP(mAU/mL, median, range)	24 (13, 49)	24.5 (7, 12,391)	34 (11, 26,858)	406 (12, 75,000)	<0.001

HCC, hepatocellular carcinoma; LC, liver cirrhosis; CHB, chronic hepatitis B virus infection; TBil, total bilirubin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TP, total protein; DCP, des-γ-carboxy prothrombin.

Figure 1.

Serum DCP levels in patients with AFP-negative HBV-related HCC, LC, and CHB. **P<0.001; *P<0.01. ns, not significant; HCC, hepatocellular carcinoma; HBV, hepatitis B virus; DCP, des-γ-carboxy prothrombin; AFP; alpha-fetoprotein. Note: Extreme values were removed.

Diagnostic value of DCP

The diagnostic value of DCP for AFP-negative HBV-related HCC was evaluated by ROC curves. Based on the AUCs, DCP was able to distinguish between all HCCs and CHB or LC, all HCCs and LC, early-stage HCC and CHB or LC, early-stage HCC and LC (Table 2, Figure 2).

Table 2.

Diagnostic values of DCP for AFP-negative HBV-related HCC.

Groups compared	DCP
	Cutoff (mAU/mL)	AUC (95%CI)	Sen (%)	Spe (%)	PPV (%)	NPV (%)
All HCC vs LC or CHB	45	0.731 (0.657–0.805)	50.6	94.3	86.3	73.0
All HCC vs LC	41	0.728 (0.649–0.806)	50.6	91.7	88.0	60.6
Early-stage HCC vs LC, CHB	45	0.685 (0.596–0.774)	43.5	94.3	79.4	76.8
Early-stage HCC vs LC	41	0.683 (0.591–0.775)	43.5	91.7	81.8	65.3

HCC, hepatocellular carcinoma; LC, liver cirrhosis; CHB, chronic hepatitis B virus infection; DCP, des-γ-carboxy prothrombin; AUC, area under curve; Sen, sensitivity; Spe, specificity; PPV, positive predictive value; NPV, negative predictive value.

Figure 2.

Values of DCP for early diagnosis of AFP-negative HBV-related HCC. (a) All HCC vs CHB and LC; (b) all HCC vs LC; (c) early-stage HCC vs CHB and LC; (d) early-stage HCC vs LC. HCC, hepatocellular carcinoma; LC, liver cirrhosis; CHB, chronic hepatitis B virus infection.

Discussion

The present study evaluated the performance of serum DCP for diagnosing AFP-negative HBV-related HCC in 87 patients with AFP-negative HBV-related HCC and 123 with CHB or LC. Serum DCP levels were significantly higher in the patients with HCC compared with the CHB and LC groups, indicating that DCP might be an excellent diagnostic biomarker for AFP-negative HBV-related HCC. The results of ROC curve analysis showed that DCP could effectively distinguish between all AFP-negative HCCs and CHB and LC (AUC 0.731) and between early-stage AFP-negative HBV-related HCC and CHB or LC (AUC 0.685). These results confirmed the value of DCP as a promising marker for diagnosing AFP-negative HBV-related HCC.

Most previous research suggested that the diagnostic efficiency of DCP for HCC was inferior to that of AFP.[18-20] For instance, Choi et al.[21] conducted a longitudinal assessment of the diagnostic value of DCP for detecting HCC and reported an AUC for DCP of 0.71 for diagnosing HCC. However, regarding AFP-negative HCC, a large-scale multicenter study by Ji et al.[22] showed that DCP was superior to AFP for the surveillance of early HCC, and that DCP had an AUC of 0.856 for differentiating between patients with AFP-negative HCC and control subjects, including healthy individuals and patients with liver cirrhosis, liver hemangiomas, and chronic hepatitis B virus infection. The AUC for DCP in their study was higher than that in the current study, possibly because the former study included more patients with late-stage HCC and healthy controls. Another study reported by Wang et al.[23] showed an AUC for DCP of 0.731 among patients with AFP-negative HBV-related HCC, which was consistent with the current results.

The ratio of fucosylated to total serum paraoxonase 1 was previously reported to show superior potential compared with AFP-L3 and glypican-3 for distinguishing between AFP-negative HCC and LC.[24] In terms of other important markers, AFP-L3, as a specific type of AFP, was considered to be more specific than total AFP for diagnosing HCC, and might thus also improve the diagnosis of HCC in patients with low AFP levels.[25] Although we did not compare the diagnostic values of DCP and AFP-L3 for differentiating HCC, Zhang et al.[26] revealed that the maximum AUC of AFP-L3 for diagnosing AFP-negative HCC was 0.609 (95% confidence interval (CI) 0.599 to 0.799), which was much lower than that for DCP in the present study. Furthermore, Mao et al.[27] showed that the AUCs of peripheral plasma Dickkopf-1 and Tie2-expressing monocytes for differentiating between AFP-negative HCC and LC and CHB were 0.649 (95%CI 0.542 to 0.755) and 0.708 (95%CI 0.605 to 0.812), respectively. However, the AUC for DCP in the current study was higher than these previous results, indicating that DCP was more promising than other potential biomarkers for diagnosing AFP-negative HBV-related HCC.

Different from previous studies, we also evaluated the value of DCP for the early diagnosis of AFP-negative HBV-related HCC, and showed that serum DCP might be a suitable biomarker to aid the early diagnosis of AFP-negative HCC.

This study had some limitations. First, the total number of included participants was relatively small. Second, this was a retrospective study from single center and not a prospective study. Third, we only focused on patients with HBV-related HCC.

In conclusion, DCP shows relative high efficiency for differentiating between AFP-negative HBV-related HCC and CHB or LC. DCP may thus be a suitable biomarker to improve the rate of early diagnosis of AFP-negative HBV-related HCC.