Molecular Phylogenetics of Hepatitis D Virus in New Zealand and the Implications for Pacific Island Countries

Hepatitis delta virus (HDV) is a small incomplete human hepatotropic virus that requires hepatitis B virus (HBV) surface antigen (HBsAg) for assembly, release from the infected cell, and infectivity [1]. HDV infects 2–8% HBsAg carriers [2] or 15–20 million people worldwide [3]. Recent publications, however, estimate that this may be closer to 48–72 million persons as it is largely underdiagnosed [4, 5]. HBV/HDV coinfection causes the most severe forms of viral hepatitis with increased risk of cirrhosis, hepatic decompensation, hepatocellular carcinoma, and fulminant hepatitis [2, 6-9].

HDV is transmitted by the parenteral route and may be acquired as a coinfection, simultaneous exposure/infection of HBV and HDV, or superinfection where an individual with an established HBV infection is exposed to HDV. Superinfections are more likely to develop into chronic HBV/HDV. The incubation period ranges from weeks to months, and the infection may present clinically as a worsening of hepatitis symptoms. Exposure to HDV is determined by antibody detection; however, HDV RNA detection in serum confirms current infection [10]. Until recently, HDV treatment has relied on administration of interferon-alpha but with limited success [11, 12]. Newer therapies are in development and should be available soon. Entry inhibitors such as bulevirtide, prenylation inhibitors such as lonafarnib, and capsid inhibitors are in clinical trials and should provide a more targeted sustained response [12].

HDV is found universally and with endemic pockets in Central and West Africa, northern Asia, Middle East, eastern Europe, Amazon Basin, and the Pacific Islands (PIs) [13-15]. Mongolia has an extremely high prevalence with a reported 6.7% blood donors and 56.5% of patients with chronic hepatitis with detectable HDV RNA [16]. There are 8 HDV genotypes, 1–8, which have up to 40% divergence in nucleotide sequence of the genome [17, 18].

The incidence of HDV in New Zealand (NZ) has not been documented recently, but a past report suggested that around 6.7% of HBsAg-positive subjects in Auckland had antibodies to HDV (anti-HDV) and that 61% of these were from the PI of Samoa [19]. NZ and Samoa are located in Polynesia, 1 of the 3 regions of the PIs. Polynesia also incorporates Hawaii, Tonga, Tuvalu, and Tokelau. Melanesia incorporates Papua New Guinea, Solomon Islands, Fiji, Vanuatu, and New Caledonia, while Micronesia includes Kiribati, Nauru, Guam, and Marshall Islands. There is historical serological evidence of HDV in the PIs implicating all 3 regions including Solomon Islands, Marshall Islands, and Niue [20, 21]. However, a paucity of recent publications and testing for current infection with HDV RNA assays precludes an assessment of the true prevalence in these PIs, with the exception of Kiribati. Studies from Kiribati have revealed a unique clonally derived subtype of HDV genotype 1, the Pacific clade, which infects up to 37.5–41% HBsAg-positive individuals in this small Micronesian country [14, 15]. A previous study documented the presence of this Pacific clade in Australia [22]. Thus, the aims of this study were to determine the HDV genotypes in NZ and investigate the link of strains to other PI countries and the world through phylogenetics.

Sixteen serum samples from 14 individuals collected between 2009 and 2014 at Auckland Hospital were coded and sent to the WHO Regional Reference for HBV and HDV, Melbourne, for sequencing, genotyping, and phylogenetic analysis. The method used has been described previously [14, 22] and is summarized here. RNA was extracted and a 420-bp product of the HDV delta antigen was amplified, purified, and sequenced. Sequences were aligned with 66 HDV sequences from GenBank [23], 7 sequences isolated from Kiribati samples and 8 representative sequences isolated from Australian samples [11] using MAFFT [24] and then manually corrected using BioEdit v7.0.5 [25]. Phylogenetic analyses were performed based on the maximum likelihood phylogenetic tree generated using the GTR + G + I substitution model, with 1,000 bootstrap replicates using MEGA5 [26].

Clinical and epidemiological data are summarized in Table 1. Eight of the 14 patients had cirrhosis, and there were 2 patients with HIV coinfection. Strains from 13 of 14 subjects were identified as genotype 1; the remaining 1 (S3) was genotype 5, an African genotype. Phylogenetic analysis shown in Figure 1 revealed that 11 of the genotype 1 strains clustered with the unique Pacific clade originally identified in Kiribati and Nauru [14, 15, 22]. These 11 NZ strains were from subjects born in Kiribati or another PI (Samoa, Tuvalu, or Niue). The S5 genotype 1 strain, isolated from an individual born in Ethiopia, clustered with African genotype 1 strains. S7, isolated from a Maori NZ health-care worker who acquired HDV after occupational exposure, clustered with other genotype 1 strains. The closest match, with 94% similarity was with an Italian strain. Figure 2 shows a map of the PIs where there is serological evidence of HDV and direct or indirect current HDV infection, as revealed by Australian [22] and NZ isolates.

This study describes the HDV strains circulating in NZ which has a relatively low prevalence of HDV in NZ-born Maoris and Europeans and a high prevalence in those of Samoan origin [19, 27]. HDV viraemia was identified in 3 Samoans, and the HDV isolated was confirmed to belong to the Pacific strain (shown in Table 1; Fig. 1). Indeed, there were 11 individuals described here with the Pacific strain, which confirms the presence outside of the endemic Micronesian islands of Kiribati and Nauru [14, 28-31]. Although transmission within NZ is possible, transmission between the PIs before arriving in NZ is more likely with accumulating evidence of the Pacific strain in individuals from the Polynesian islands of Tuvalu (S6) and Niue (S15), as well as Samoa (S1, S13, and S14). This indirectly supports the historical serological evidence of HDV in these PIs depicted in Figure 2. Recent estimates suggest HDV prevalence in Samoa may be as high as 20% HBsAg-positive individuals (personal communication).

A study by Han et al. [14] reported an absence of HDV markers in Fiji, Vanuatu, and Tonga. A recent study in Australia confirmed the Pacific strain in an individual born in Fiji [22]. Fiji has historically low serological evidence of HDV [32] and is unlikely to be endemic. The sporadic presence of HDV in the Pacific (Fig. 2) reflects its irregular prevalence in the world. There are certainly endemic pockets with HDV viraemia confirmed in Kiribati at 41% HBsAg-positive individuals [15] and highly suspected in Nauru, where contract workers from Kiribati may have been implicated in the introduction of HDV [33]. There is also now circumstantial evidence of HDV infection in Samoa, Tuvalu, and potentially Niue, although the incidence of HDV in Niueans living in NZ appears to be very low. Serological evidence shown in Figure 1 suggests that HDV infection may be even more widespread in the PIs than originally thought. This is not surprising as there is considerable movement between the Pacific nations for trade, exploration, social, and economic reasons. Tattooing, piercings, and unhygienic dental practices may also play a role in the transmission of HDV in this region [15, 28].

The potential origin of the Pacific HDV strain has been discussed previously [14, 15]. The strain shares a common ancestor with African and Turkish HDV strains (Fig. 1) yet is clearly distinct from these and other genotype 1’s. The PIs were exposed to European colonization at different times in history, offering a hypothetical source of introduction. However, the delta strains from Europe also do not cluster with the PI strain. An alternative plausible theory is that there was a point introduction into Kiribati and the cultural practices at the time favoured transmission and establishment of endemicity. Almost half the population on Kiribati were refugees from Samoa, and this may explain the similarly high endemicity in Samoans. The authors are hopeful that future studies using molecular clock analyses may enable the age of the Pacific strain to be determined and provide a clue on the time and source into the PIs.

There is certainly a need to investigate the HDV prevalence further in other PI countries such as Niue, the Solomon Islands, Palau, Tokelau, and the Marshall Islands, where there is strong HDV serological evidence [34] and HBV infection is endemic. Investigations in Papua New Guinea are also needed as Brindle et al. [34] detected HDV antibody in 1 of 3 HBsAg-positive samples in 1987. The authors have detected 1 antibody-positive (HDV RNA-negative) sample in 50 HBsAg-positive samples from Popondetta (PNG) in 2019 (data not shown).

A previous study has shown that African genotype 1 strains tend to originate from East Africa such as Sudan and Egypt and genotype 5 strains from West Africa including Liberia and Sierra Leone [22]. This view was supported in this study where HDV genotype 5 was isolated from a Liberian-born patient (S3) and the African genotype 1 from an Ethiopian-born patient (S5). It is also interesting to note that the 2 African-born patients in NZ were also coinfected with HIV. A recent study in Nigeria reported that triple infection with HIV/HBV/HDV is rare but can occur at an incidence of around 0.7% [35]. This was confirmed by studies carried out in Uganda [36] and the USA where 2.2% of HBV/HIV coinfected subjects had HDV RNA detected [37].

The clinical history in Table 1 documents the severity of HBV/HDV coinfection with 8 of 14 subjects having cirrhosis and 3 developing hepatocellular carcinoma. The severity of disease reiterates the necessity of diagnosing HDV coinfection, particularly as antiviral treatment becomes available [38]. It is recommended that all HBsAg-positive individuals are screened for HDV, particularly those from the PIs and Africa.

This study reveals the presence of European and African HDV genotypes in NZ and confirms probable transmission of HDV Pacific clade throughout the PIs from Micronesia to Polynesia. Further studies are warranted to assess the prevalence in other PIs in order to provide appropriate clinical management and target the prevention of the ongoing transmission of this serious and often neglected infection.

The authors would like to acknowledge Dr. Lilly Yuen for assistance with Figure 2 and Dr. Fahimeh Rahnama for providing the samples.

Ethical approval was provided by the Auckland District Health Board and not subject to an external Ethics Committee. Patient consent was waived as the work was considered to be important for the public health of NZ.

The authors declare that there are no conflicts of interest.

The authors did not receive any funding.

Kathy Jackson: conceptualization, data curation, formal analysis, investigation, methodology, and roles/writing – original draft. Margaret Littlejohn: formal analysis, software, and writing – review and editing. Ed Gane: project administration and writing – review and editing. Stephen Locarnini: writing – review and editing.