ABSTRACT

Objectives:

The hepatitis delta virus (HDV) is an enveloped, negative-sense, single-stranded RNA virus in the genus Deltavirus. In this study, it was aimed to evaluate the ratio of anti-HDV signal/ cut-off (S/CO) in the diagnosis of HDV infection in patients who were positive for anti-HDV.

Materials and Methods:

Between August 2014 and December 2018, 156 patients, who were detected hepatitis B surface antigen (HBsAg) and anti-HDV positivity and were analyzed HDV-RNA, were included in the study. Anti-HDV antibody and HBsAg tests were analyzed by micro-ELISA method in serum samples and HDV-RNA was studied by real-time polymerase chain reaction method in plasma samples.

Results:

HDV-RNA was detected in 42.9% (67/156) of the anti-HDV positive patients. The mean S/CO value of anti-HDV in HDV-RNA positive group (8.99±3.53) was significantly higher than HDV-RNA negative group (5.99±3.73) (p<0.001). When the S/CO value was determined as 6.13 by receiver operating characteristic (ROC) analysis; sensitivity, specificity, positive, and negative predictive values were 79.8%, 59.7%, 72.4%, and 68.9%, respectively.

Conclusion:

In this study, the optimal cut-off value which provides the maximum sum of sensitivity and specificity for the test was found 6.13 by ROC analysis. The reporting with the S/CO ratio of anti-HDV ELISA results and determining each laboratory’s optimal cut-off value may be helpful for the diagnosis of HDV infection.

Introduction

The hepatitis delta virus (HDV), which is the only member of the only species in the genus Deltavirus, is an enveloped, negative-sense, single-stranded RNA virus (1,2). HDV requires the simultaneous presence of hepatitis B virus (HBV) to complete its life cycle (3). Although HDV suppresses HBV replication, it causes variable clinical presentation ranging from mild disease to fulminant liver failure (4). It is estimated that 18 million hepatitis B surface antigen (HBsAg) carriers in the world also have anti-HDV antibodies, which is 5% of HBV-infected individuals (5). The Amazon region of South America, the Middle East, the Mediterranean region, West and Central Africa are highly endemic areas for HDV (6). Turkey is considered a moderate endemic area for HDV infection with regional differences (7,8). HDV can cause co-infection with HBV or superinfection in people infected with HBV (9). Diagnosis of coinfection or superinfection with HDV is based on serological and molecular methods. The diagnosis of acute HDV co-infection is based on the detection of HDV-Ag, HDV-RNA and anti-HDV antibodies, together with markers of HBV infection. HBV-DNA, an indicator of HBV replication, could be suppressed during acute HDV infection and could not be detected (10,11). Therefore, all HBsAg positive patients should be analyzed anti-HDV antibody testing and especially should be recommended in case of acute hepatic exacerbation (10). The diagnosis of chronic HDV infection is determined by the absence of immunoglobulin M (IgM) antibody against hepatitis B core antigen and the presence of HBV and HDV infection markers. HDV-RNA is the gold standard method for the diagnosis of HDV infection due to sensitivity and specificity problems in HDV-Ag tests (11).

The first step in the diagnosis of HDV is anti-HDV antibody screening against HDV-Ag in individuals with positive HBsAg (9). Although the sensitivity of ELISA tests is high, populations with low prevalence have lower positive predictive values and higher false positivity rates. The aim of this study was to evaluate the anti-HDV signal/cut-off (S/CO) ratio in the diagnosis of HDV infection in patients with anti-HDV positivity.

Materials and Methods

Study Group

This retrospective study included 156 patients, who were detected HBsAg and anti-HDV positivity and were analyzed HDV-RNA, between August 2014 and December 2018. Ethical approval for this study was obtained by the Ethics Committee of İstanbul University Faculty of Medicine (approval number: 2018/1766/84).

Serological Analysis

Anti-HDV and HBsAg tests in serum samples were performed on Triturus Enzyme Immunoassay Analyzer (Grifols, Spain) using micro-enzyme immunoassay (ELISA) kits (Dia. Pro, Diagnostic Bioprobes, Milano, Italy). The cut-off value was calculated according to the manufacturer’s recommendations. Test results were calculated by proportioning the sample absorbance to the cut-off value. If S/CO ratios are ≥1.00, the result of the test is evaluated as positive.

Molecular Analysis

Between August 2014 and October 2016, HDV-RNA extraction was performed using the High Pure Viral Nucleic Acid kit (Roche Applied Science, Basel, Switzerland) in plasma samples. The RNA molecule was transformed into complementary DNA (cDNA) using the Transcriptor First Strand cDNA Synthesis V6 kit (Roche Diagnostics, Mannheim, Germany). The cDNA was amplified by the LightCycler 2.0 instrument (Roche Diagnostics GmbH, Switzerland) using a polymerase chain reaction (PCR) mixture prepared with the TIB Molbiol HDV GmbH kit (Berlin, Germany) and the Light Cycler FastStart Master HybProbe kit (Roche Diagnostics, Germany). Between October 2016 and December 2018, HDV-RNA was obtained using the Qiagen EZ1 virus mini kit V2 nucleic acid extraction kit (Qiagen, Germany). Reverse transcription and amplification of RNA molecules was performed using a PCR mixture prepared with Fluorion HDV QNP 1.0 Real-Time PCR Kit (Iontek, İstanbul, Turkey) by Rotor-Gene Q instrument (Qiagen, Germany).

Statistical Analysis

Data analysis was performed using SPSS 25 (SPSS Inc, Chicago, IL, USA) program. The suitability of the variables to normal distribution was examined by visual methods (histogram and probability graphs) and Kolmogorov-Smirnov test. Variables were compared using Student’s t-test or Mann-Whitney U test. Pearson chi-square or Fisher exact tests were used for qualitative variables. Sensitivity, specificity, negative predictive and positive predictive values were investigated by determining the significant cut-off values of the test by receiver operating characteristic (ROC) curve analysis. P value of less than 0.05 was considered statistically significant.

Results

Of the 156 anti-HDV positive patients with a mean age of 48.89±11.65 years (range: 19-90 years), 49.4% were male and 50.6% were female. The mean age of the male and female patients was 47.74±11.93 and 50.01±11.34, respectively (p=0.22). HDV-RNA positivity was detected in 42.9% (67/156) of anti-HDV positive patients.

The mean age of the viremic patients was 49.66±10.20 and the mean age of the non-viremic patients was 48.31±12.67 (p=0.48). The mean S/CO of anti-HDV positive patients was 7.70±3.90 (1.18-19.1). The mean S/CO value of anti-HDV in HDV-RNA positive group (8.99±3.53) was significantly higher than HDV-RNA negative group (5.99±3.73) (p<0.001) (Figure 1). A total of 67 (45.9%) samples with S/CO ≥ 1 by the Dia. Pro anti-HDV assay was false-positive.

When the S/CO value was 6.13 by ROC analysis, a total of 27 (27.6%) samples with S/CO  ≥6.13 was false-positive. When S/CO ≥ 6.13, the sensitivity, specificity, positive and negative predictive values were 79.8%, 59.7%, 72.4% and 68.9%, respectively. A total of 24 (28.2%) samples with S/CO ≥ 7.15 was false-positive. When S/CO ≥ 7.15, sensitivity, specificity, positive and negative predictive values were 68.5%, 64.2%, 71.8% and 60.6%, respectively. The area under the curve (AUC) was 0.72 (95% Cl: 0.64-0.80), p<0.001 (Figure 2). The performance results of the Dia. Pro anti-HDV test kit at different S/CO values are presented in Table 1.

Discussion

The first approach in the diagnosis of HDV infection is to investigate the antibodies against HDV-Ag in HBsAg-positive individuals. Anti-HDV antibody can be detected in immunocompetent patients with HDV infection (10). While anti-HDV immunoglobulin G (IgG) antibodies are detected in individuals with HDV infection even after clearance of the virus, anti-HDV IgM antibodies can be detected as a serological marker of HDV replication in the majority of patients with acute infection (12). Although anti-HDV IgM antibodies are indicative of acute HDV infection, anti-HDV IgM antibody tests may not be able to detect low antibody titers. Commercial anti-HDV antibody ELISA kits detected both IgM and IgG are preferred for HDV screening in HBsAg positive patients (13,14). However, HDV-RNA is the only reliable parameter of HDV replication.

In this study, HDV-specific total antibodies were investigated and reported semi-quantitatively by determining the S/CO ratio. The mean S/CO of anti-HDV in HDV-RNA positive group (8.99±3.53) was significantly higher than HDV-RNA negative group (5.99±3.73) (p<0.001). The results were consistent with rare studies on this subject (15). In a cross-sectional study performed using the anti-HDV radio-immunoassay kit (Abbott Laboratories, Chicago, IL, USA), it was demonstrated that high anti-HDV titers were correlated with HDV viremia (16).

Although studies on prediction of viremia of ELISA S/CO ratio have been performed mostly with the hepatitis C virus (HCV), such studies for HDV are quite limited. Anti-HCV S/CO ratios for prediction of HCV viremia in different studies, differences have shown depending on the size of the sample, HCV prevalence in the studied population, and kit differences (17).

It is of interest to define the best S/CO cutoff before using anti-HDV tests in the clinical routine. Thus, it could be decreased false-positive results. S/CO can vary across kits, and populations with different HDV infection rates. When S/CO ≥ 1 by the recommendation of the manufactory, 45.9% of samples had false-positive by the Dia. Pro anti-HDV assay. When S/CO ≥ 6.13, false-positivity decreased to 27.6%. In the present study, it was observed that the false-positive ratio decreased when the S/CO ratio increased.

In a study performed with the anti-HDV antibody ELISA kit (Hepanostika HDV, Organon Teknika, the Netherlands) in Turkey; when the ELISA index value was 100.10, sensitivity, specificity, negative and positive predictive values were 93%, 80%, 93%, 76% respectively. The AUC was found to be 0.934 (15). In this study, the optimal cut-off value which provides a total of the maximum sensitivity and specificity for the test was found 6.13. The AUC was found 0.72 (95% Cl: 0.64-0.80), p<0.001 by ROC curve analysis. Results below the optimal cut-off value should be retested with HDV-RNA and another anti-HDV kit. Determination of cut-off index value in test kits could be clinically important for predicting true HDV viremia.

Conclusion

This study showed that Dia. Pro anti-HDV antibody test kit had a good clinical performance for anti-HDV S/CO value of 6.13. The S/CO ratio in the HDV test could be used in the clinical decision-making process if it can correctly predict the diagnosis of HDV before HDV-RNA is analyzed. In addition to anti-HDV ELISA results, reporting of S/CO ratio and determining each laboratory’s optimal cut-off value may be helpful for the diagnosis of HDV infection.