The Diagnostic Value of Serum PIVKA-II Alone or in Combination with AFP in Chinese Hepatocellular Carcinoma Patients

Background At present, the diagnostic accuracy of alpha-fetoprotein (AFP) for hepatocellular carcinoma (HCC) surveillance is insufficient. It remains controversial whether prothrombin induced by vitamin K absence II (PIVKA-II) has a better diagnostic value than AFP for HCC patients. Objective To investigate the diagnostic role of PIVKA-II alone or in combination with AFP in Chinese HCC patients. Methods Serum AFP and PIVKA-II levels were detected and analyzed in 308 HCC afflicted patients and 120 unafflicted controls. The receiver operator curve (ROC) and area under the curve (AUC) were conducted to evaluate the clinical value of AFP and PIVKA-II for diagnosing HCC and early HCC. Results In the whole HCC cohort, the diagnostic values of PIVKA-II were better than that of AFP. The AUC of PIVKA-II and AFP was 0.90 (95% CI 0.87-0.94) and 0.79 (95% CI 0.74-0.84), respectively. “AFP + PIVKA-II” yielded a high sensitivity of 95.1% and a specificity of 83.3%, with the AUC 0.89 (95% CI 0.85-0.93). In the early stage HCC group, the diagnostic accuracy of PIVKA-II was also better than that of AFP. The AUC of PIVKA-II and AFP was 0.83 (95% CI 0.77-0.89) and 0.75 (95% CI 0.68-0.81), respectively. “AFP + PIVKA-II” achieved the sensitivity of 83.3% and specificity of 89.1%, with an AUC of 0.86 (95% CI 0.81-0.91). Moreover, for AFP-negative HCC patients, serum PIVKA-II showed good diagnostic performance, with an AUC of 0.804 (95% CI 0.720-0.887). Besides, elevated PIVKA-II level was a strong independent risk factor for HCC patients with portal vein tumor thrombus (PVTT) (OR = 4.890, P = 0.020). Conclusion PIVKA-II is superior to AFP in HCC screening, and AFP in combination with PIVKA-II significantly improves the diagnostic value for Chinese HCC patients. PIVKA-II could effectively indicate HCC accompanied by PVTT and help to optimize the therapeutic strategy.


Introduction
Hepatocellular carcinoma (HCC) is the sixth frequently diagnosed malignancy and the third leading cause of cancerrelated death worldwide. Although the management of HCC has improved in recent years, the long-term overall survival remains poor [1][2][3]. Thus, sufficient methods are urgently needed for screening HCC at an earlier stage. Among traditional diagnostic and prognostic biomarkers, serum alpha-fetoprotein (AFP) level is the most frequently used for HCC detection and surveillance. However, the diagnostic accuracy of AFP for HCC patients is unsatisfactory.
Previous results showed that serum AFP yielded 39-65% sensitivity and 76-94% specificity in detecting HCC, which is still far from ideal for clinical application [4,5].
Prothrombin induced by vitamin K absence II (PIVKA-II) is firstly described in 1984 as a biomarker specific for HCC [6]. Increasing evidence has indicated that PIVKA-II demonstrated good diagnostic and prognostic values for HCC surveillance [7][8][9]. However, there are also some studies which supported that AFP performs better than PIVKA-II for HCC screening [10,11]. Thus, results regarding the diagnostic efficiency of PIVKA-II in comparison or in combination with AFP still remain controversial. Currently, serum PIVKA-II level is considered a promising diagnostic biomarker in Japanese HCC patients. However, the clinical applications of serum PIVKA-II to assess Chinese HCC patients are limited. Thus, more studies are necessary to provide more evidence investigating the diagnostic accuracy of PIVKA-II in Chinese HCC patients.
Portal vein tumor thrombus (PVTT) is the most common macrovascular invasion and a frequent complication of HCC. Previous studies reported that the incidence of PVTT in HCC patients was 44%-62.2%. HCC complicated by PVTT is associated with limited treatment approaches, increased risk of recurrence, and poor prognosis [12][13][14]. Thus, predicting portal vein tumor thrombus is crucial for making cure strategies and prognosis assessment. Accumulating evidence suggested that PIVKA-II could stimulate overexpression of vascular endothelial growth factor (VEGF) and related to PVTT occurrence [15,16]. However, lately, few studies explored the predictive role of PIVKA-II for HCC accompanied by PVTT.
Therefore, this present study is aimed at investigating the diagnostic role of serum PIVKA-II alone or in combination with AFP for Chinese HCC patients, and to determine the relationship between serum PIVKA-II level and PVTT.

Study Settings and
Patients. The present retrospective study enrolled a total of 428 subjects, including 308 HCC patients, 60 patients with HBV-related liver cirrhosis (LC), and 60 patients with benign liver disease (BLD). HCC was defined according to the American Association for the study of Liver Disease (AASLD) Practice Guidelines for the management of hepatocellular carcinoma (updated version, 2010). BLD cases were patients with hepatic cysts or hepatic hemangioma confirmed by ultrasound or computed tomography. The BCLC (Barcelona Clinic Liver Cancer) staging system was used to assess the stage of HCC patients, and early HCC was defined as BCLC 0-A stage. Sun Yat-sen University Cancer Center approved this retrospective, anonymous analysis of data, and the requirement for written informed consent was waived. The flowchart depicting the selection process of the enrolled participants is shown in Figure 1.

2.2.
Measurements of AFP and PIVKA-II. Serum AFP levels were detected by the electrochemiluminescence immunoassay according to the manufacturer's manual (Roche Diagnostics, GmbH, Mannheim, Germany). Serum PIVKA-II levels were detected by the chemiluminescence enzyme immunoassay according to manufacturer's manual (Fujirebio, Inc., Tokyo, Japan). The cut-off values for serum AFP and PIVKA-II levels were 25 ng/ml and 40 mAU/ml, respectively.
2.3. Statistical Analysis. All statistical analyses were performed using SPSS version 22 software (IBM, Armonk, USA). The Mann-Whitney test, Kruskal-Wallis test, chisquared test, and Fisher's exact test were used, as appropriate, to facilitate comparisons of differences in data between groups. Receiver operating characteristic (ROC) curves and area under the curve (AUC) with 95% CIs were conducted to elucidate the diagnostic values of AFP and PIVKA-II. Diagnostic efficiency-related parameters including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), Kappa value, and accuracy were calculated. Logistic regression analyses were used to facilitate investigations of the relationship between PIVKA-II levels and recorded clinical variables of HCC patients. P < 0:05 was regarded as statistically significant.

Patient Characteristics.
A total of 428 participants including HCC patients (n = 308), LC patients (n = 60), and BLD cases (n = 60) were recruited from June to September 2019. Among 308 HCC patients, 276 were HBV-related HCC. The median PIVKA-II and AFP levels were significantly higher in HCC patients compared with LC and BLD patients (P < 0:05). The clinicopathologic parameters of all participants are summarized in Table 1.

Discussion
Currently, AFP is the most clinically applicable biomarker for HCC detection. However, its unsatisfactory diagnostic efficiency is far from meeting the clinical requirement. Our research is aimed at identifying and validating effective non-invasive biomarkers for HCC management, and our previously published study has indicated that hypermethylation of SCAND3 and Myo1g gene was a potential diagnostic biomarker for hepatocellular carcinoma [17]. Recently, PIVKA-II is emerging as a promising marker for diagnosing HCC, but it is not as the routine analysis for HCC patients in China [18][19][20]. Besides, the results if PIVKA-II performs better than AFP for HCC diagnosis and prognosis have been controversial up to now. Therefore, we enrolled a sufficient sample size of HCC patients as well as integrated controls including LC and BLD, to identify the performance of PIVKA-II, AFP, and their combination for HCC diagnosis.
In the present study, our results revealed that the diagnostic performance of PIVKA-II was better than AFP for Chinese HCC patients. In the whole HCC cohort, the sensitivity and AUC were 89.0% and 0.90 (95% CI 0.87-0.94) for PIVKA-II and 68.8% and 0.79 (95% CI 0.74-0.84) for AFP. In the early stage HCC patients, the sensitivity and AUC were 74.5% and 0.83 (95% CI 0.77-0.89) for PIVKA-II and 60.9% and 0.75 (95% CI 0.68-0.81) for AFP. When AFP and PIVKA-II were used in combination, the diagnostic accuracy significantly improved not only in whole HCC cohort but also in the early stage group. Moreover, in the AFPnegative HCC patients, PIVKA-II still showed excellent diagnostic performance with the AUC 0.81 (95% CI 0.72-0.89). Our results are similar to the previously published studies [9,21,22]; serum PIVKA-II was obviously superior to AFP for HCC screening and provides further evidence that "AFP + PIVKA-II" was an effective blood-based biomarker facilitating diagnoses of HCC. An elevated PIVKA-II level was reported to function as a predictive factor for microvascular invasion [18,23]. However, our data indicated that there was no relationship between PIVKA-II levels and microvascular invasion. More large clinical studies are needed to verify these results in the future. Interestingly, our current study found that serum PIVKA-II level was significantly correlated with an increased risk of PVTT (OR = 4:890, 95% CI 1.361-38.169; P = 0:020), which may help to predict the prognosis of patients with HCC and guide therapeutic strategy.
Compared with previous research, this study offers several strengths. First, many previous studies focused on HCC patients in Japan as well as Western countries [15,16,19], and the diagnostic accuracy of PIVKA-II was debatable due to different demography and etiology of liver diseases. There, we enrolled a total of 428 Chinese participants including HCC patients, LC, and BLD cases, and the sample size is more sufficient to elucidate the diagnostic role of PIVKA-II and AFP for Chinese HCC patients. Second, we fully and comprehensively validated serum PIVKA-II and AFP levels as diagnostic biomarkers in three groups, including whole HCC cohort, the early stage group, and AFP-negative HCC patients. Third, recently published studies rarely reported the relationship between serum PIVKA-II expression level and portal vein tumor thrombus. Our data revealed that elevated PIVKA-II serum level could independently predict an increased risk of HCC accompanied by PVTT.
However, there are some limitations in our study. First, our study was designed retrospectively. More prospective

Disease Markers
clinical trials are required to validate the results. Second, the underlying biological mechanism of PIVKA-II in HCC and the reasons why increased PIVKA-II expression level is correlated with PVTT have not yet been elucidated. Third, we did not perform further follow-up to determine the association between dynamic changes of PIVKA-II and HCC prognostic value.

Conclusion
The current study confirmed that PIVKA-II is an excellent diagnostic biomarker for Chinese HCC patients and could be used as a predictor for HCC accompanied by PVTT. Notably, "AFP + PIVKA-II" achieved high specificity and sensitivity for detecting HCC, particularly in the detection of early stage HCC. Based on the evidences of the current study, we therefore strongly recommend that PIVKA-II should be routinely tested for managing Chinese HCC patients.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon reasonable request.

Disclosure
The sponsors or funding organizations had no role in the design or conduct of this research.

Conflicts of Interest
No conflicting relationship exists for any authors.

Authors' Contributions
Fei Xu and Lulu Zhang contributed equally to this work.