Use of the Novel INNO-LiPA Line Probe Assay for Detection of Hepatitis B Virus Variants That Confer Resistance to Entecavir Therapy

Hepatitis B virus (HBV) infection is a common cause of chronic liver disease, and the risk of disease progression is associated with elevated HBV DNA levels. With the introduction of new HBV polymerase inhibitors, such as entecavir (ETV) and tenefovir (TDF), major advances have been made in the antiviral management of chronic hepatitis B (CHB) (13). However, the widespread use of these agents had led to the increasing emergence of drug-resistant mutations in HBV.

ETV is a potent deoxyguanosine analog, although it is less effective against lamivudine (LVD)-resistant HBV variants than against wild-type HBV (1). ETV resistance requires substitutions at residues T184, S202, and M250 in reverse transcriptase (rtT184, rtS202, and rtM250, respectively) in HBV isolates containing preexisting, LVD resistance-conferring substitutions (10-12).

This study assesses a prototype line probe assay (LiPA), the INNO-LiPA DR (version 3) assay (LiPA Innogenetics, Ghent, Belgium), for the detection and characterization of the HBV polymerase mutations associated with ETV resistance. A practical feature of this assay is that the single-stage PCR product can be used for combined hybridization of strips from the INNO-LiPA DR assay (version 3) and INNO-LiPA DR assay (version 2), the latter of which is used to detect the main variants resistant to LVD and adefovir-dipivoxil (ADV) (7), because both assays use the same set of primers. The INNO-LiPA DR assay was performed according to the manufacturer's instructions. Briefly, HBV DNA extracted from 200 μl of serum with a QIAamp DNA blood kit (Qiagen, Venlo, The Netherlands) was amplified by a single round of PCR with the kit-supplied primer mix: sense primer (nucleotides 255 to 278; 5′-CGTGGTGGACTTCTCTCAATTTTC-3′) and antisense primer (nucleotides 1122 to 1099; 5′-AGAAAGGCCTTGTAAGTTGGCGA-3′), both of which included domains A to F of the HBV polymerase. The amplified product was used for reverse hybridization with specific oligonucleotide probes immobilized on nitrocellulose strips. After hybridization, the strips were incubated with streptavidin conjugate. Colored lines were interpreted for the presence of mutations in the HBV polymerase. The probes fixed in the INNO-LiPA DR assay strips cover known HBV polymerase variations associated with ETV resistance (codons rt184, rt202, and rt250). The strips contain 2 control lines and 22 HBV probe lines representing wild-type or mutant variants: for codon rt184, 1 line reactive to the normal variant (amino acid T), 3 lines reactive (indistinctly) to mutated amino acid S, C, G, or A (marked as SCGA lines), and 2 lines to mutated amino acid I, L, F, or M (marked as ILFM lines); for codon rt202, 1 line reactive to the normal variant (amino acid S) and 1 line to each mutated G, C, or I variant; and for codon rt250, 2 lines reactive to the normal variant (amino acid M) and 2 lines to each mutated V, I, or L variant. Moreover, the strips allow the detection of variants rtA194T (two lines) and rtI233V (two lines) associated with resistance to TDF and ADV, respectively. The INNO-LiPA DR assay can detect a variant in a mixed population when the variant accounts for more than 5% of the total population with a sensitivity of 990 copies/ml at a 95% confidence interval (technical support staff of LiPA Innogenetics, personal communication).

The HBV polymerase gene in samples with ETV-resistant variants detected by the INNO-LiPA DR assay was sequenced by the use of dRhodamine Dye Terminator chemistry (ABI Prism 310 instrument; Applied Biosystems, Foster City, CA), as described previously (5). A fragment of the HBV polymerase region that is included in the same domains analyzed by the INNO-LiPA DR assay was amplified for sequencing. To test for false-positive reactivity, 10 HBV DNA clones (4) with no LVD-, ADV-, TDF-, or ETV-resistant variants were analyzed; all clones showed the wild-type patterns for all variants.

The utility of the INNO-LiPA DR assay was tested by retrospectively analyzing serum samples from 59 patients with CHB who had not responded to antiviral treatment with a nucleoside or nucleotide analog. In all cases, samples were obtained pretreatment (naïve) and when a treatment nonresponse was detected: 35 were treated with LVD, 20 were treated with ADV, and 4 were treated with ETV. One of the last four patients, included in a previous study (9), had been sequentially treated with LVD, ADV, and finally, ETV. A total of 119 serum samples were analyzed.

Before treatment, five (8.6%) cases (four who were further treated with LVD and one who was further treated with ADV) presented substitutions related to ETV resistance (two at position rt184, two at position rtS202, one at position rt250); none had LVD-resistant variants (Table 1).

After treatment, among the 35 LVD nonresponders (all of whom were infected with HBV isolates with LVD resistance-conferring mutations), variants with ETV resistance-related variations emerged in 4 (11%) (at position rt184 in two patients and at position rt202 in two patients) (Table 1; Fig. 1). The variants with ETV resistance-related variations detected before treatment remained detectable at the end of therapy. Among the 20 nonresponders to ADV therapy, no emergence of variants with ETV resistance-related variations was observed. In one of these patients, a variant with a variation at rt184 was detected before treatment and remained during follow-up. Of the four nonresponders to ETV, the sequentially treated patient developed a virological breakthrough, with the detection of the variant with the ETV resistance-related variation at codon rtS202S/G (Fig. 1).

Sequencing analysis of the 15 samples with variants with ETV resistance-related variations identified the same variants as the INNO-LiPA DR assay did in 11 samples. However, in the remaining four samples (two with rtS202S/G variants [one obtained pretreatment and one in the LVD resistance group], one rtS202S/I variant from the LVD resistance group, and one rtT184T/I/F/L/M variant obtained before treatment), only the wild-type variant was observed by sequencing (Table 1).

The INNO-LiPA DR assay has demonstrated optimal sensitivity and specificity, according to the results of the studies with serum and clonal HBV DNA. In relation to sensitivity, when the 15 samples positive for ETV resistance-associated mutations by the INNO-LiPA DR assay were tested by direct sequencing, sequences with ETV resistance-related mutations in 4 (26.7%) samples were identified as being of the wild type by sequencing. The absence of hybridization bands corresponding to mutated variants when clonal HBV DNA with wild-type sequences was processed excludes the possibility that these four cases were due to the false reactivity of the INNO-LiPA DR assay, thus indicating that this assay has optimal specificity. The fact that only 11 (73.3%) cases with ETV-resistant variants were detected by sequencing clearly indicates that the INNO-LiPA DR assay detects minority quasispecies in a more sensitive manner, prior to their expansion toward a dominant species. These findings are consistent with those of other studies that have reported that the INNO-LiPA DR assay is more sensitive than sequencing for the detection of nucleotide variations, allowing the detection of mixtures of wild-type isolates and ETV-resistant mutants (3, 7). Along these lines, a recent study by Pas et al. demonstrated the practicability and usefulness of HBV genotyping by the INNO-LiPA DR assay and reported that they achieved a higher sensitivity and a higher specificity with this technique than with sequencing and DNA chip technologies (8). However, one disadvantage of the INNO-LiPA DR assay is the limited scope of mutations represented in the assay, based on current knowledge of ETV resistance-associated mutations. The inclusion of newly recognized mutations will require the development of a new assay.

While investigating the applicability of the INNO-LiPA DR assay, drug-resistant variants were detected in treatment-naïve patients, confirming the results of our previous study in which we used a sequencing method (5). However, the proportion of ETV-resistant variants (8.6% by INNO-LiPA DR assay versus 2.7% by sequencing) and their characteristics (40% of variants with variations in each of codons rt184 and rt202 and 20% of variants with variations in codon rt250 by the INNO-LiPA DR assay versus 100% of variants with variations in codon rt202 by sequencing) were notably different by the two methods used. Similar results were observed for the emergence of these variants in LVD-refractory patients regarding both their proportion (14% by the INNO-LiPA DR assay versus 9% by sequencing) and their characteristics (50% of variants with variations in codon rt202 and 50% of variants with variations in rt184 by the INNO-LiPA DR assay versus 75% of variants with variations in codon rt202, 17% of variants with variations in codon rt184, and 8% of variants with variations in codon rt250 by sequencing). These differences are mainly due to the fact that the proportions of variants with variations in codons rt184 and rt250 seem to be underestimated by sequencing. This may be because these variants are present in percentages under the detection limit (<10% of the total viral population) of the sequencing method due to their low replication capacity relative to that of wild-type variants. Along these lines, it has been reported that the rtM250V substitution results in significant replication impairment (10). This is likely because codon rt250 is part of the highly conserved MGY motif in the human immunodeficiency virus (HIV) polymerase (codons rt230 to rt232) and the HBV polymerase (codons rt250 to rt252) involved in the polymerase and template primer interaction (6). With regard to position rt184, no in vitro experiments have been conducted to explain the possible low replication rate of variants with variations at this position. However, in support of this possibility, the amino acid T is conserved in both the HIV polymerase (codon rt165) and the HBV polymerase (codon rt184) (6), suggesting that its substitution in these polymerases could result in a lower replicative efficacy. The emergence of ETV-resistant variants during LVD treatment in the present study confirms previous observations suggesting that LVD can induce the selection of these variants. However, no emergence of ADV resistance has been observed in ADV-treated patients, suggesting that ADV does not stimulate the selection of such variants.

ETV resistance due to variations at position rt184, rt202, or rt250 alone seems to have a minor impact on ETV susceptibility (2, 12). However, it should be considered that addition of rtL180M and rtM204V induces a >70-fold decrease in ETV susceptibility; hence, the presence of variants resistant to ETV and LVD may be a factor predictive of ETV treatment failure.

Monitoring of drug-resistant variants is important for elucidation of the prevalence and mechanisms of resistance development and for the more effective management of treatment options. Our results provide additional evidence that the INNO-LiPA DR assay is useful for the detection of nucleotide variations and infections with virus with dual variations and is more sensitive than sequencing. Moreover, this study shows the applicability of the assay for monitoring for ETV-resistant variants. The ability to detect an ETV mutant virus population in the absence of ETV treatment (naïve patients and during LVD therapy) allows the selection of the appropriate antiviral agents.

• Open in new tab
• Download powerpoint

FIG. 1.

ETV resistance-related variations observed by the INNO-LiPA DR assay and HBV DNA polymerase sequencing for patients 9 and 10, in whom ETV-resistant variants emerged after LVD and ETV therapy, respectively. Strips A1 and B1 (basal sample), wild-type patterns; strip A2 (follow-up), ETV-resistant variant with a variation at position rt184 for patient 9, rtT184SCGA in INNO-LiPA DR assay strip and rtT184A by sequencing; strip B2 (follow-up), ETV-resistant variant with a variation at position rt202 for patient 10, rtS202S/G in the INNO-LiPA DR assay strip and by sequencing. The interpretation chart is included between strips A1-A2 and strips B1-B2. Conj., conjugation; Amp., amplification.

• Copyright © 2009 American Society for Microbiology