The Prevalence and Serological Association of Hepatitis D Virus Genotypes in Taiwan

Hepatitis Delta Virus (HDV) is an RNA virus that requires the presence of hepatitis B surface antigen (HBsAg) to propagate into hepatocytes, with Genotype I being more prevalent globally. However, the prevalence of HDV genotypes in Taiwan is unknown. Accordingly, a cohort including 24 chronic HBV patients who received nucleos(t)ides (NUCs) between January 2002 and July 2018 was used to determine HDV genotypes and genotype specific serological association in chronic HBV carriers. HDV-positive genotypes in 18/24 (75%) males and 6/24 (25%) females were identified among chronic HBV patients. Viremia was lower in HDV-IV patients than in patients affected with other HDV genotypes (1.34 log10 copies/mL vs. 3.30 log10 copies/mL; p = 0.009). A logistics regression analysis revealed that HDV-IV was inversely proportional to HDV RNA (odds ratio [OR]/95% confidence intervals [CI]: 0.370/0.164–0.830; p = 0.017). The serologic association study indicated lower levels of creatinine (p = 0.047) and HDV-RNA (p = 0.009) in the HDV-IV group than the non-HDV-IV group but did not indicate any significant differences in the AST, ALT, bilirubin levels or other laboratory test factors. The three genotypes evident in Taiwan were HDV-I (4/24, 16.7%), HDV-II (6/24, 25.0%), and HDV-IV (14/24, 58.3%), and HDV-IV is the predominant HDV genotype in Taiwan. These results anticipate a clear understanding of HDV genotype serological association in chronic HBV carriers.


Introduction
The hepatitis delta virus (HDV) was discovered and detected in the nuclei of hepatocytes of patients who tested seropositive for hepatitis B virus (HBV) surface antigen (HBsAg) [1]. This virus is unique because it utilizes the hosts' RNA polymerase and HBsAg of the HBV to propagate and then egress into liver cells and has been theorized to originate from plant viroids or/and host cell pre-mRNA through a splicing pathway [2]. The virus is approximately 35-37 nm in diameter with a small, single circular RNA genome approximating 1672-1700 nucleotides, enveloped by HBsAg-a key feature that perpetuates the ability of the virus to hijack the machinery of the hosts' cellular functions [2,3]. It is, therefore, referred to as a defective RNA virus [4]. This notwithstanding, key modified pathways related to fibrosis, epigenetic changes, immune response, specific dysregulation of long non-coding RNA, and proteomic changes have all been suggested to promote hepatocellular carcinoma (HCC) development [5,6]. HDV has two clinical presentations: co-infection and superinfection with HBV, where co-infection is the ability of HDV and HBV to infect a host simultaneously in an acute phase, and superinfection refers to chronic HBV infection and acquiring HDV later in the clinical presentation of the disease [4].
Globally, the high prevalence of HBV has led to higher incidences of infection by HDV [7]. With vaccine intervention for HBV, it was anticipated that HDV would decline; however, due to the unpredictability and minimal knowledge on the nature of the HDV species, it has increased over the last two decades; out of 400 million HBV carriers, 5% (15-20 million) have probably been exposed to HDV [3]. Dual infection with HBV and HDV leads to aggressive progression to end-stage liver diseases. In Taiwan, chronic hepatitis B prevalence in the younger generation has decreased drastically to below 1% after the countrywide execution of the HBV vaccination program from 1986 [8,9]. HDV global prevalence ranges from 0.16-0.8% in the general population and ranges from 4.5-13% in HBV carriers [10,11], although the prevalence of HDV is decreasing and estimated to be 1.15% in the general population in Taiwan [12]. Currently, eight genotypes of HDV have been isolated based on the homology of viral nucleotide sequencing, where Genotype I is distributed globally, and others are distributed regionally. Genotypes I, II, and IV are found in Taiwan, Japan, Russia, and other Eastern Asian countries; Genotype III is more localized to the Amazon region (Peru, Colombia, Ecuador and Brazil), while Genotypes V, VI, VII and VIII are found in Africa [3].
Liver disease is complex and can manifest through a variety of pathways, from acute to severe status. Liver disease can then progress to fibrosis, liver cirrhosis, and finally total morphological changes leading to cancer, HCC. HBV/HDV coinfection increases the risk liver cirrhosis and rapid progression to HCC [13], although co-infection or superinfection increases the risk of hepatic failure in contrast to HBV infection alone [14]; regardless, HDV is not currently placed on the list of oncogenic agents along with the already enlisted HBV and hepatitis C virus (HCV). The mechanism by which HDV-related HCC occurs is not clear, and there is ongoing research on this phenomena; for example, previous research has pointed out that high levels of HDV DNA, old age, male gender, family history of HCC, serostatus of hepatitis B e-antigen (HBeAg), high serum alanine aminotransferase (AST) level, high quantitative HBsAg levels, basal core promoter (BCP) mutations, and HBV genotype C are all major risk factors in the development of liver cirrhosis [14]. Since our previous study elucidated the serial serological changes of HDV in chronic HBV patients receiving NUCs analogue therapy in Taiwan [15]. Through this study, we aim to determine the serological HDV genotypes in chronic HBV carriers and the associated factors in a cohort of the general population.

Baseline Characteristics of HBV/HDV Superinfected Patients
The total cohort as shown in Table 1, the prevalence of HBV/HDV is higher in men than in women. The mean age of the patients was 52 years old; the mean levels of ALT, AST, and bilirubin were 211 IU/L, 226 IU/L, and 3.28 mg/dL, respectively. The mean HDV RNA level was 2.15 log 10 copies/mL. Two patients were positive for HBeAg, and five patients were positive for anti-HCV, while mean HBsAg and HBV DNA levels were 1376.7 IU/mL and 3.58 Log 10 IU/L, respectively. Furthermore, cirrhotic patients had significantly lower platelet (p = 0.003) and white blood cell (WBC) count (p = 0.024) than non-cirrhotic patients ( Table 1). The baseline factors associated with patients with or without liver cirrhosis in HBV/HDV co-infected patients showed no significant change in HBsAg and HBV DNA levels.

Factors Associated with HDV Genotypes in HBV/HDV Superinfection Patients
Specific-HDV primers were used to target highly conserved regions of the L-HDAg [16]. The first round PCR product band was 323 bp, and the second round PCR product was 234 bp by primer-specific targeting for L-HDAg (Figure S1A-D). Sample sequences retrieved from direct sequencing were aligned. The conservative sites showed unique diversity among the sample study cohort (Table S1 and Figure S2). HDV-genotyping and phylogenetic analyses revealed that the 24 patients exhibited HDV-Genotypes I, II and IV, which are associated with the general Taiwanese population ( Figure 1). Of the three genotypes observed, HDV-IV (14/24, 58.3%) was predominant followed by HDV-II (6/24, 25.0%) and HDV-I (4/24, 16.7%) (Table S2). Our findings demonstrate a higher prevalence of HDV Genotype IV in the general Taiwanese cohort.

Comparison of the Distribution of Different Biomarkers in HDV/HBV Patients with HDV-IV and Non-HDV-IV Genotype
A simple linear regression analysis between non-HDV-IV and HDV-IV genotypes was performed to observe the clinical association of biomarkers, including AST, ALT, bilirubin, and HDV RNA in the Figure 2. AST level was insignificantly associated with HDV-IV (109 IU/L) than non-HDV (390 IU/L), p = 0.062 ( Figure 2A); ALT level was insignificantly associated with HDV-IV (123 IU/L) than non-HDV (335 IU/L), p = 0.114 ( Figure 2B); bilirubin level was insignificantly associated with HDV-IV (1.934 mg/dL) than non-HDV (4.897 mg/dL) p = 0.181 ( Figure 2C), while HDV RNA was significantly decreased in HDV-IV patients (1.338 log 10 copies/mL) than non-HDV patients (3.280 log 10 copies/mL) p = 0.009 ( Figure 2D), demonstrating an inverse relationship between HDV-IV and HDV RNA levels. Our findings show that HDV-RNA levels may be HDV-genotype-specific in HBV/HDV superinfected patients.  Figure 2D), demonstrating an inverse relationship between HDV-IV and HDV RNA levels. Our findings show that HDV-RNA levels may be HDV-genotype-specific in HBV/HDV superinfected patients.

Distribution of HDV Genotypes among Baseline Factors by Group
The HDV-Ⅳ group of patients consisted predominantly of males (55.6%), even when compared with non-HDV-Ⅳ patients (53% vs. 48%, p = 0.269), as shown in Table 4. From the LC group, a total of 11 patients were diagnosed, with five (38.5%) being non HDV-Ⅳ and six (54.5%, p = 0.729) being HDV-Ⅳ. In the ALT group with ALT levels <400 IU/L or ≥400 IU/L, there was no significant difference (p = 0.139) between HDV-Ⅳ and non HDV-Ⅳ groups ; however, there were thirteen (65.0%) patients with ALT < 400 IU/L and HDV-IV genotype, while three (75.0%) patients with ≥400 were dominant in non-HDV IV. In the bilirubin group with <2 mL/dL or 2 mL/dL, there was no significant difference (p = 0.746) between HDV-Ⅳ and non HDV-Ⅳ groups, but number of patients with bilirubin level <2 mL/dL was higher in HDV-IV. Even in the HBV DNA group with <2000 IU/mL or ≥2000 IU/mL, no significant difference (p = 0.746) was observed between HDV-Ⅳ and non HDV-Ⅳ groups, but there was a higher number (8,57.1%) with HBV DNA ≥ 2000 IU/mL in the HDV-IV patient groups. Notably, seven patients were positive for HBsAg < 250 (58.3%) with HDV-Ⅳ and three patients in the range of HBsAg ≥ 250 were positive (50.0%) in both non HDV-Ⅳ and HDV-Ⅳ groups; (p-value = 0.737). Although there was a preponderance of the HDV-Ⅳ group in ALT ≥ 400, HBsAg < 250, and the LC group compared with non-HDV-Ⅳ, there were no significant differences.

Distribution of HDV Genotypes among Baseline Factors by Group
The HDV-IV group of patients consisted predominantly of males (55.6%), even when compared with non-HDV-IV patients (53% vs. 48%, p = 0.269), as shown in Table 4. From the LC group, a total of 11 patients were diagnosed, with five (38.5%) being non HDV-IV and six (54.5%, p = 0.729) being HDV-IV. In the ALT group with ALT levels < 400 IU/L or ≥400 IU/L, there was no significant difference (p = 0.139) between HDV-IV and non HDV-IV groups; however, there were thirteen (65.0%) patients with ALT < 400 IU/L and HDV-IV genotype, while three (75.0%) patients with ≥400 were dominant in non-HDV IV. In the bilirubin group with <2 mL/dL or 2 mL/dL, there was no significant difference (p = 0.746) between HDV-IV and non HDV-IV groups, but number of patients with bilirubin level <2 mL/dL was higher in HDV-IV. Even in the HBV DNA group with <2000 IU/mL or ≥2000 IU/mL, no significant difference (p = 0.746) was observed between HDV-IV and non HDV-IV groups, but there was a higher number (8,57.1%) with HBV DNA ≥ 2000 IU/mL in the HDV-IV patient groups. Notably, seven patients were positive for HBsAg < 250 (58.3%) with HDV-IV and three patients in the range of HBsAg ≥ 250 were positive (50.0%) in both non HDV-IV and HDV-IV groups; (p-value = 0.737). Although there was a preponderance of the HDV-IV group in ALT ≥ 400, HBsAg < 250, and the LC group compared with non-HDV-IV, there were no significant differences. Table 4. Comparative grouped baseline factors of non HDV-IV and HDV-IV in HBV/HDV superinfected patients.
According to a previous study, since 1986, nationwide vaccination programs for HBV have been the standard clinical practice within the Taiwanese health care system [24]. Consistency in the reports on the prevalence of HDV superinfection in patients with active hepatitis has always been estimated to be about 5% to 8%. There was a decrease in HDV superinfection from 23% in 1983 to 4.2% in 1996 among high-risk patients due to the implementation of medical and social practices, such as active preventative measures directed at promiscuity and popular use of disposable needles for drug use [24,25]. Consistent with other studies, our findings also demonstrate a decreasing prevalence of HDV in patients; however, HDV cannot be eliminated as a causative agent for liver-associated complications [12]. Sequencing and phylogenetic analysis from the region (888-1122) of L-HDVAg revealed that the Taiwanese HDV isolates belonged predominantly to HDV-IV and frequently to HDV-II, followed by HDV-I. Our findings are consistent with other studies carried out by countries within Eastern Asia [16,20,26]. The nucleotide similarity of the isolate HDV-I showed 99% identity with isolate AB118848 from Japan; HDV-II isolates showed >95% identity with U19598 from Taiwan; HDV-IV isolates showed 95% identity to Pathogens 2021, 10, 1227 8 of 11 AF209859 from Taiwan; while four out of fourteen isolates in HDV-IV also showed 95% identity to the AB11847 isolate of Japan, which could be due mixing of genotypes.
We were interested in the clinical significance of HDV genotypes associated with clinical factors in HBV/HDV-superinfected patients. Patient samples were categorized into non-liver cirrhosis and liver cirrhosis groups. The liver cirrhosis group had an inverse relationship with platelet counts and WBC; however, all other clinical factors revealed no significant differences in association with liver cirrhosis. HDV is associated with increased risk of liver cirrhosis and advanced liver disease [27]. HBV/HDV co-infection demonstrates lower platelet counts compared with those in HBV mono-infection, regardless of liver disease stage; however, the functional mechanism is still unknown [27][28][29], whereby when altering immune-related genes and immune response, we observed lower integrity of immune response or white blood cells in our study. In addition, patient samples were grouped into non-HDV-IV and HDV-IV groups, where lower HDV loads were observed in the HDV-IV group than the non-HDV-IV group, but few studies have focused on HDV viral loads linked to genotypes. Lastly, patients were further grouped by HDV genotypes: HDV-I, HDV-II and HDV-IV, where low levels of creatinine and HDV-RNA were observed in HDV-IV. This suggests that HDV RNA levels may be genotype-specific.
There are several limitations in the study. Firstly, when associating the HDV genotypes with the clinical (grouped) factors, there was no significance observed among the groups. This might be due to the rise in Type I errors in Chi-square tests. Secondly, the clinical parameters were not accurately retrieved from patients during follow-up, possibly owing to small sample size, absence of patient records or inconsistencies in documentation; and thirdly, it could not be determined whether patients acquired the HDV simultaneously or throughout the course of chronic HBV and HDV infections. Owing to the fact that HDV is a coinfection of virus, it is problematically complex having an HDV-only clinical infection control group; nonetheless, different HDV genotypes have been demonstrated and possible risk factors that might be genotype-specific in Taiwan were explored. Possible future studies should include in-depth research to understand how the different HDV genotypes hijack the HBV mechanism for replication, which might further unveil such genotypic specific inhibitors.

Study Subjects
In total, 2580 HBV Taiwanese CHB patients receiving nucleos(t)ides (NUCs) therapy between January 2002 and July 2018 at the Kaohsiung Medical University Hospital were enrolled [15]. A total of 70 patients (2.7%) were identified as being seropositive for anti-HDV. Twenty-six of the 70 patients (37.1%) in this anti-HDV seropositive population were positive for HDV RNA, making up 1.0% (26/2580 patients) of the total HBV population. Among the 26 patients seropositive for HDV RNA, two patients with missing data during follow-up were excluded. The inclusion criteria were patients positive for HBsAg and HBV DNA > 6 months, and seropositive for antibodies to HDV (anti-HDV) and HDV RNA before anti-HBV therapy, with exclusion criteria being coinfection with human immunodeficiency virus (HIV). The experiment protocol conformed to regional guidelines and was approved by the Ethics Committee of Kaohsiung Medical University Hospital (KMUH). Written consent was obtained from each patient. We defined CHB as being HBsAg-positive for at least six months or more and with chronic hepatitis delta as viral infection for a period equal to or greater than six months. A small fraction of the patients presented with LC and HCC. Liver enzymes AST and ALT, platelets (PLT), bilirubin, and other biomarkers: HBsAg, HBeAg, anti-hepatitis Be (anti-HBe) and anti-HDV were collected as clinical parameters for each patient, as shown in Table 1. Liver cirrhosis was diagnosed by histology [30], transient elastography (FibroScan; Echosens, Paris, France) > 12 kPa) [31] or through the presence of clinical, radiological, endoscopic, or laboratory evidence of cirrhosis and/or portal hypertension [32].