Prevention and management of drug resistant hepatitis B virus infections

Abstract

In the past decade, broadened therapeutic options of oral direct antiviral agents for the treatment of chronic hepatitis B infection include: Lamivudine, Adefovir Dipivoxil, Telbivudine, Entecavir and Tenofovir Disoproxil Fumarate. These direct oral antiviral agents effectively suppress the replication of the virus and reduce the risk of potential liver-related complications. However, prolonged use of these nucleos(t)ide analogues has been associated with drug resistance that compromises the initial clinical benefits. Moreover, the oncogenic risk of mutations due to prolonged nucleos(t)ide analogue therapy needs to be further investigated by in vitro and in vivo studies. In the current era of potent nucleotide analogues, new data are emerging, we are still facing the pool of patients who have developed resistance to the prior generation of nucleos(t)ide analogues. This paper aims to focus on incidence of antiviral drug resistance and virological breakthrough, prudent selection of initial therapy, on-treatment monitoring for drug resistance and revise treatment strategies for patients with resistant virus.

Background

Hepatitis B virus (HBV) infects 400 million people worldwide; more than 75% of those reside in the Asia-Pacific area.¹ Chronic hepatitis B (CHB) infection is the major risk factor for cirrhosis, end-stage liver disease and hepatocellular carcinoma (HCC).^2,3 Increasing serum HBV DNA level is a strong predictor associated with progression to cirrhosis and HCC.^4,5 Substantial advances have been made during the last decade with the worldwide clinical use of nucleos(t)ide analogues (NA). Since 1998, five NA have been licensed: Lamivudine (LAM), Adefovir dipivoxil (ADV), Telbivudine (LdT), Entecavir (ETV) and Tenofovir Disoproxil fumarate (TDF). These direct oral antiviral agents effectively suppress the replication of HBV and reduce the risk of potential liver-related complications. Sustained and profound suppression of viral replication together with reversal of fibrosis and cirrhosis are achievable.⁶ Based on specific viral kinetics and low sustaining off-treatment response, long-term NAs treatment is necessary.^7,8 However, the major concern of prolonged NA treatment of CHB can result in development of antiviral resistance with reduced sensitivity to the antiviral agents, thus compromising the initial clinical benefits.⁹

Close insight of anti-HBV resistance

Definitions, patterns and pathways of drug resistance.  The “gold standard” of confirmed genotypic resistance is based on in vitro phenotypic analysis.¹⁰ The definition of terminology for anti-HBV drug resistance is shown in Table 1. With over decade of NAs administration, cross and multi-drug resistant viruses have been observed.¹⁶ It is noteworthy that rtA181T/V was gradually selected to be the marker of cross resistance across NA groups, i.e. L-nucleosides (LAM and LdT) and alkylphosphonates (ADV and TDF).^17,18 The major ADV resistance substitution, rtN236T, also confers three- to four-fold reduced susceptibility to TDF in vitro.¹⁹ However, sub-analyses of prospective randomized trials showed no impact of ADV resistant mutations on virological response to TDF.^20,21 The role of the rtA194T mutation potentially related to TDF resistance remains controversial as 5 years of clinical trial data have not shown any resistance.^22–25

Table 1. Definitions of antiviral resistance

Terms	Definitions
Definition above is based on a treatment compliant patient.
Primary treatment failure (nonresponse)	The failure to achieve a 1 log10 copies/mL (or 1.0 log10 IU/mL) decline in viral load after 12 weeks of therapy.^11
Secondary treatment failure (virological breakthrough, VBT)	Confirmed increase in HBV DNA level of more than 1 log10 IU/mL (1 log10 copies/mL) from the lowest detectable level or redetection of serum HBV DNA at levels > 10-fold of the lower limit of detection of the HBV DNA assay after having an undetectable result on therapy.^12,13
Partial virological response	Decrease in HBV DNA of more than 1 log10 IU/mL but detectable HBV DNA by real-time PCR assay at week 48.^13
Suboptimal virological response	Residual HBV DNA levels ≥ 3.3 log10 IU/mL at week 24.^14
Genotypic resistance	Detection of mutations in the HBV genome that has been shown to decrease susceptibility to antiviral drugs identified by phenotypic assay.^10,15
Phenotypic resistance	In vitro confirmation that the mutation decreases susceptibility to treatment^10
Cross resistance	Decreased susceptibility to more than one antiviral drug conferred by the same amino acid substitution or combination of amino acid substitutions.^10,13,15
Multi-resistance	Complex patterns and clusters of specific mutations in HBV polymerase associated with multidrug failure.^13

Different HBV genotypes have a distinct geographic distribution. Genotypes B and C are mainly found in Asia; genotypes A and D are highly prevalent in Europe.^26,27 In a cohort of 2136 CHB patients were studied from SeqHepB, primary antiviral drug-resistance was detected in 456 patients. The dominant mutation patterns were distinct in different genotypes,^21,28 e.g. LAM resistant patterns is rtM204V (81%, P < 0.05) and rtM204I (54%, P < 0.05), respectively, in patients with genotype A and B HBV infection, ADV resistant patterns in genotype C HBV infected patients (rtA181T, 64%, P < 0.01) were also different from those in genotype D HBV infected patients (rtA181V + rtN236T, 53%, P < 0.05). Because of the distinct geographic distribution of HBV genotypes, rtM204I and rtA181T (which were respectively associated with sW196 stop and sW172 stop in overlapped hepatitis B surface antigen [HBsAg]) were detected exclusively in Asian patients.

Incidences of drug resistance and virological breakthrough.  The rate of genotypic drug resistance was rarely identical with the rate of virological breakthrough (VBT) (Table 2), partly due to different definitions of viral resistance and VBT.^34,35 Poor medication adherence is another major factor leading to VBT. Regarding resistance rate, it is impossible to conduct a direct comparison of resistance incidence associated with different NA due to differences in the populations studied. Comparison of long-term resistance rates is even more complicated due to different study designs.^36,37 The resistance profile of the subgroup of patients who started ADV at the first 48 weeks till the end of 144 weeks remains unreported.²⁴ Patients with genotypic resistance from a 2-year Telbivudine GLOBE trial³⁸ were excluded from the prolonged 3-year trial.³⁹ For long-term therapy with ADV (up to 5 years), patients with detectable serum HBV DNA at 48 weeks were excluded.⁴⁰ Global clinical trials of TDF allowed for the addition of Emtricitabine at week 72 in patients with detectable HBV DNA (≥ 400 copies/mL or 69 IU/mL). ETV registration in global trial only provided the intent-to-treat (ITT) analysis at 1 year, and patients without virological response after 1 year treatment were switched to another ETV trial (NCT01438424)⁴¹ with a higher dose (1 mg daily), which might alter the risk of drug resistance.⁴² Most of the data were from global clinical trials in which serum HBV DNA levels were quantified by assays with a lower detection limit of 300–400 copies/mL (approximately 60–80 IU/mL). Potential genotypic resistance was detected by sequencing. For some clinical trials, mutations were detected both at baseline and by annual surveillance of all patients with detectable viremia.^34,36,40,43 However, in some studies, genotypic resistance was only detected among the patients with VBT.^34,35,44,45 Additionally, the threshold of detectable serum HBV DNA levels varied from 15 IU/mL to 385 IU/mL.^46,47 Moreover, successful identification of drug resistance detection requires that quasi-species populations with resistant substitutions occupies 5%⁴⁸ (restriction fragment length polymorphism, RFLP) to 25%³⁵ (direct sequencing) of the viral load.

Table 2. Patterns and pathways of antiviral drug resistance in chronic hepatitis B

Pathway	Amino acid substitutions in the rt domain	Proposed mechanism^29
L-Nucleoside (LAM/LdT)	M204I/V ± L180M^30	Catalytic site: allosteric change/steric hindrance
Acyclic phosphonate (ADV)	N236T^31	Triphosphate binding site: allosteric change
Shared (LAM, LdT, ADV, TDF)	A181T/V	Catalytic site: allosteric change
Double (ADV, TDF)	A181T/V + N236T^18,19
D-Cyclopentane (ETV)	L180M + M204V/I	Primer binding site: allosteric change,
	± I169 ± T184	Catalytic site: allosteric change
	± S202 ± M250
Multiple drug resistant pathway	V173L + L180M + A181V + N236T^16	Sequential NAs therapy
	L180M + M204V + I169T + T184A + S202G^32
	V173L + L180M + M204V + P237H^33

Clinical consequences and long term implications of antiviral drug resistance.  The silent emergence of drug-resistant viruses is typically followed by a VBT¹² (Table 1). Other clinical clues include: identification of known genotypic markers of drug resistance within viral polymerase, increasing serum alanine aminotransferase (ALT) levels, clinical deterioration and sometimes fatal exacerbation of liver diseases.^49,50 Generally, the generation and multiplication of drug resistant mutations ultimately amplified as VBT (> 1.0 log IU/mL increase from lower detection limit) or viral creep (< 1.0 log IU/mL increase from lower detection limit) in CHB patients treated by LAM and ADV, respectively.^9,51 LAM resistance is more profound in CHB patients with advanced liver diseases related to VBT in a prospective randomized study.⁵² Patients with Tyrosine, methionine, aspartate, aspartate (YMDD) mutation were more likely to have VBT compared with patients without YMDD mutations (62% vs 5%).⁵² Moreover, in spite of a better clinical prognosis compared with placebo group, patients with YMDD mutations are more likely to have an increased probability of disease progression than those without YMDD mutations (P < 0.001).⁵²

During long-term NA treatment, drug resistant substitutions in the HBV polymerase gene can concomitantly generate S gene mutations owing to overlapping reading frames.^53,54 For instance, rtA181T simultaneously generates a stop codon in the S gene (sW172stop), and it has a secretory defect and exerts a dominant negative effect when coexisting with the wild type and rtA181V.^18,55,56 A case report from Taiwan identified that the rtA181t/sW172 stop (surface truncation mutants) constituted major viral populations in a NA naïve patient with HCC who was HBsAg-negative, hepatitis B e antigen (HBeAg)-positive and in whom the serum HBV DNA was > 10⁵ copies/mL.⁵⁷ Overall, the study indicates the oncogenic potential related to NAs drug resistance and warrants careful re-evaluation of the current strategy of prolonged antiviral therapy. A systematic review of literature focusing on the incidence of HCC among CHB patients receiving NA showed that LAM resistance is related to higher cumulative HCC rates compared with NA naïve patients (42/594 or 7.1% vs 126/3287 or 3.8%, P = 0.001).⁵⁸ Even among cirrhotic patients, HCC occurred more frequently in patients with LAM resistance than NA naïve patients (18% vs 11%, P = 0.015). Furthermore, virological remission during rescue therapy did not seem to reduce the risk of HCC in patients with LAM resistance (P = 0.466).

Pre- and on-treatment strategies to prevent antiviral resistance

Timing and judicious selection of initial therapy.  Rapid and profound HBV DNA reduction to undetectable level is the first step to minimize long-term resistance. Always “hesitate” to active NA treatment until absolutely necessary and only after a thorough evaluation and counseling on an individual basis before starting therapy. Evaluations must including age, HBeAg status, serum HBV DNA level, ALT level, the severity of liver disease, the likelihood of response, possibility of drug resistance, the risk of disease progression, the likelihood of compliance and the patient's ability to pay for long-term therapy.^37,59

In considering initial therapy, aggressive or conservative strategies partly rely on physician's discretion. Updated major guidelines recommend that serum ALT values of 30 IU/mL for men and 19 IU/mL for women be used as the upper limit of normal when making decisions regarding initiation of therapy.^60,61 A retrospective review indicated over 37% of patients with persistent normal ALT (< 40 IU/mL) had significant fibrosis or inflammation. Also note that patients with CHB have spontaneously fluctuating HBV DNA and ALT levels. ALT may increase in different situations, such as spontaneous HBeAg seroconversion and abnormal lipid metabolism. Re-evaluation of serum HBV DNA and ALT levels several times before initiating therapy is highly recommended to avoid unnecessary antiviral treatment.

Once selected during the initial therapy, the drug resistant HBV mutants that are retained in the virus population for up to 4 years after withdrawal of antiviral regimens³³ will limit future treatment options. Naturally occurring HBV mutants related to drug resistance may exist in up to 5% of NA naïve patients.⁶² Thus, early detection of these mutations may provide a timely adjustment of treatment to avoid virological and biochemical breakthroughs.⁶² Furthermore, patients harboring rtM204V + rtL180M are all infected with genotype C HBV, indicating that naturally occurring LAM-resistant HBV might be more frequent in genotype C HBV.⁶²

In the absence of financial concerns, when possible, the NA with the highest efficacy of viral suppression and genotypic barrier to resistance should be initiated. ETV and TDF are the preferred first-line NA.^{59,60,63–65} In NA naïve patients, ETV only shows 1.2% resistance rate after 6 years of treatment.⁹ Additionally, there is no report of drug resistance for up to 5 years in randomized clinical trials of TDF^25,46 (Table 3). Of note, a recent systematic review indicated that TDF is most effective for HBeAg-negative patients during the first year of treatment.⁶⁵

Table 3. Genotypic drug resistance and viral breakthrough among L-Nucleos(t)ide monotherapy in CHB patients

Year	LAM				LdT				ETV				ADV				TDF		LAM + ADV
	e(+)		e(−)		e(+)		e(−)		e(+)		e(−)		e(+)		e(−)		Naïve		ADVr rate in LAMrpts
	R	VBT	R	VBT	R	VBT	R	VBT	R	VBT	R	VBT	R	VBT	R	VBT	R	VBT	R	VBT
Sequencing:^2–5,8,35–40,42,67,71,73; RFLP:^42,47,48,68–70,76; INNO-LiPA HBV DR assay:^8,71. ^† Cumulative incidence.
1	27%^8				0%^43	0%^43			0%^34	2%^34	0%^45	2%^45	0%^43	NA	0%^36	NA	0%^23		1%^47^†	0%^47
	15%^44^†	17.5%^44	0%^44^†	0.7%^44			0%^44^†		Naive		LAMr		0%^66		0%^37^†		0%^67	2.3%^67
	11%^35	15.3%	10.7%^35	12.5%	5%^35	5.9%^35	2.2%^35	2.3%^35	0.2%^42^†	0.2%^42^†	6%^42^†	1%^42^†
	13%^34	18%^34
	23%^48		6.4%^45	8%^45
	32%^68
	14%^76
2	39.5%^38	47%^38	26%^38	31.7%^38	25.1%^38	28.8%^38	10.8%^38	12.2%^38	0.5%^42^†	0.2%^42^†	15%^42^†	11%^42^†	1.3%^66		3%^37^†		0%^23		2%^47^†	0%^47
	43%^69	44%^69
	46%^48																		1.6%^70
3	55%^48				11.3%^39		6.5%^39		1.2%^42^†	0.8%^42^†	36%^42^†	27%^42^†	5.4%^66		5.9%^37^†		0%^23	3%^23	4%^47^†	0%^47
									0%^71	2%^71	35.9%^72	21/31
4	71%^48								1.2%^42^†	0.8%^42^†	47%^42^†	41%^42^†	10.4%^66						4%^47^†	0%^47
5	65%^48								1.2%^42^†	0.8%^42^†	51%^42^†	43%^42^†	14.6%^66	18.5^73	29%^40^†	20%^40^†	0%^25
													20%^73
6									1.2%^74		57%^74	50%^74
8	76%^75

De novo combination of NA with different resistance profiles is another option for initial therapy. A global clinical trial (52 weeks) conducted in NA naïve HBeAg-positive CHB patients, found that combinations of LAM and LdT were inferior to LdT monotherapy in all clinical responses.⁷⁷ As such, a combination of NA with the same drug resistant pathways is not advisable. In a 48 week, prospective study in a single treatment center, involving NA naïve HBeAg-negative patients, no significant difference was observed in efficacy of ETV initial monotherapy compared with the combination of LAM and ADV.⁷⁸ Sung et al.⁶⁹ reported that de novo combination of LAM and ADV reduced the risk of LAM resistance compared with LAM monotherapy in NA naïve HBeAg-positive patients (43% vs 15%), but no adefovir monotherapy arm was included in this study. Actually, while de novo combination therapies of NA seem more logical than practical, currently no data support the superiority of de novo combination regimens.⁷⁹ European and Asian-Pacific guidelines recommend initial combination therapy to avoid resistance in subgroups of patients at high risk of developing drug resistance and potentially life-threatening associated diseases (cirrhosis).^59,63

On-treatment monitoring of HBV DNA levels and roadmap concept.  Periodic assessment of quantitative serum HBV DNA remains the most practical approach for the early detection of resistance. With the advances of sensitive real-time polymerase chain reaction (PCR), serum HBV DNA levels as low as 10 IU/mL can be detected.⁸⁰ Periodic monitoring during antiviral treatment of HBV DNA levels should be done every 2 to 3 months to assess primary treatment failure, suboptimal virological response¹⁴ and VBT.¹² The interval between HBV DNA testing can be extended to every 6 months after achieving complete response (HBV DNA undetectable by PCR) at the physician's discretion and the patient's convenience.^63,81,82 However, for safety concerns, it should be monitored more frequently in patients with advanced liver diseases.

Patients with an undetectable HBV DNA level at 24 weeks are highly likely to have a better clinical outcome.^63,81,82 Experts worldwide reach agreement that 24 weeks should be an ideal and appropriate time to assess both in-treatment and long-term antiviral efficacy. If the antiviral response at 24 weeks is not satisfactory, early switch to a more potent agent or adding an agent with a different drug resistance profile should be considered in a compliant patient.⁶³ However, such treatment paradigms based on data from the prior generation of NA should be re-evaluated. These are not suitable for the new generation of NA as ETV and TDF.⁷¹ A multicenter cohort 3-year study from Europe suggested that treatment adaptation is not necessary for the majority of NA naïve patients with partial virological response to ETV monotherapy.⁷¹

Detection of drug resistance.  In clinical trials, direct PCR-based virus population sequencing is the most common method to detect NA resistance, including unknown mutations. However, mutants can only be detected by direct sequencing when the mutants present up to 20% of the total viral population. To detect the known mutations, RFLP and reverse hybridization line-probe assays (Inno-LiPA DR, Innogenetics, Ghent, Belgium) are more sensitive. These assays have the ability to detect drug resistance strains at a proportion as low as 5% of the total population⁸³ (in contrast with 20% using direct sequencing). Ultra-deep pyrosequencing (UPDS, GS FLX platform, 454 life science) is probably the most sensitive method currently available to detect the minority of viral mutants as low as 0.1% of the total quasi-species pool.⁸⁴ Drug resistance testing results are crucial for clinical treatment decisions and have been shown to alter treatment decisions of 18 hepatologists in 52% of cases on NA medication.⁸⁵ In the area where the detection of genotypic resistance is not available or drug-resistant testing is not affordable, a confirmed VBT by repeating HBV DNA quantitation within a month is necessary.

Assessment of medication adherence.  Poor medication adherence can lead to both primary treatment failure and VBT.^12,63 About 30% of VBT observed in clinical trials is associated with medication non-adherence.⁶⁴ A retrospective study of 148 CHB patients treated with NA in a single center found that 39 (26.4%) patients experienced VBT, among these patients, 24 (62%) patients had confirmed VBT and only 19 (49%) had confirmed genotypic resistance.¹² Concurrently with the single treatment center experience, a global clinical trial of TDF revealed that the vast majority of patients (85%) with VBT had documented non-adherence based on a direct assessment of adherence, i.e. plasma TDF levels.²³ Both poor adherence and emergence of drug-resistant HBV mutants are the major causes for VBT and subsequent reversal of clinical improvement. It is important to distinguish the true determinant of VBT in clinical practice to avoid unnecessary changes in antiviral medications.¹²

Poor understanding of the disease, uncertainty of the therapy's long-term clinical and financial outcomes and high treatment costs over a short period of time are the most common causes for non-adherence. The impetus for medication adherence rests largely on therapeutic educational programs. First, CHB can lead to serious complications. Such sequels can be reduced by antiviral treatment. Second, drug resistance may occur after long term NA therapy. Third, sticking to NA treatment every day is the easiest and most efficacious way to avoid antiviral resistance.⁸⁶

Furthermore, proper assessment of a patient's adherence to NA has to be addressed and reinforced at each appointment and by timely feedback on treatment progress.

Revise treatment strategies

Add on versus switching strategies.  Due to the relatively low cost and local reimbursement policy, LAM and ADV are still the most prescribed drugs in some countries of the Asian-Pacific area. The pool of the CHB patients with drug resistance to LAM and ADV continues to increase, thus constituting a considerable concern now and for the future. Rescue therapy in clinical practice largely points to LAM and/or ADV resistance. In a compliant patient with VBT, identification of HBV genotypic resistance can formulate a rescue strategy of either switching to or adding on a more potent drug without a cross-resistance profile (Table 4).

Table 4. Recue therapy for antiviral NAs¹³

Resistance	Rescue therapy
^† The long-term safety of these combinations is unknown. ^‡ Travada: a pill formation of emtricitabine 200 mg and Tenofovir 300 mg.
Lamivudine resistance	Add adefovir^47,87,88 or Tenofovir^89
	Switch to Tenofovir^90, or Travada^‡^79
	entecavir is not recommend, even with the dose of 1 mg/day^42
Adefovir resistance	Add a second drug
	Switch to Tenofovir^90,91
	Switch to Travada^‡^79
	If N236T, add lamivudine, entecavir^† or telbivudine^†^13
	If A181V/T, add entecavir^†^63
Telbivudine resistance	Add Tenofovir^34,64 (or adefovir)
Entecavir resistance	Add Tenofovir^63 (or adefovir)
Tenofovir resistance	Do genotyping and phenotyping in an expert lab to determine the cross-resistance profile
	Add entecavir^†, telbivudine^†, Lamivudine or switch to Truvada

Increasing evidence indicates that adding ADV is better than switching to ADV monotherapy for patients with LMV-resistant HBV.^47,87,88 Even for TDF, the newest and most potent antiviral agent, monotherapy may not be sufficient to suppress HBV when there are persisting ADV-resistant mutations.⁹¹ Current guidelines recommend that adding-on a second drug without cross-resistance should be the first-line rescue therapy strategy.⁶³ This has been argued in a recent retrospective, multicenter cohort study conducted in Germany and the Netherlands, which enrolled 131 NA experienced CHB patients (either incomplete response or genotypic resistance to LAM or ADV) during 5 years of treatment.⁹⁰ In this study, rescue therapy of TDF alone succeeded in suppressing HBV DNA to below 400 copies/mL in the majority of patients, and the response to TDF was less pronounced in patients with genotypic and viral resistance to ADV. Based on cost-effectiveness analysis, when facing previous suboptimal antiviral response, switch-to TDF monotherapy instead of add-on TDF will be more practical in the “real-world”.⁹² Regarding NA experience, a roadmap of rescue therapy should be individualized based on detection of genotypic resistance, HBeAg status, baseline HBV DNA and ALT level at the start of rescue therapy. For instance, in HBeAg-negative LAM refractory patients, the likelihood to achieve undetectable HBV DNA level is relatively low (in those with baseline HBV DNA > 6 log₁₀ copies/mL and abnormal ALT level⁹³).

However, a recent case report described a treatment failure in a patient who sequentially received interferon, LAM, ADV, ETV, ETV-TDF combination during a 14-year follow up. Even the combination of the first-line oral antiviral medications recommended by guidelines resulted in treatment failure.⁹⁴ Furthermore, the long-term safety of the combination of NA, especially the combination of ETV and TDF, remains uncertain.⁶³ Severe lactic acidosis was observed in five of 16 CHB patients with liver cirrhosis in Germany, and these five patients had seriously impaired liver function (Model for End-Stage Liver Disease score ≥ 20).⁹⁵ The risk of lactic acidosis associated with ETV has been controversial based on the safety data of ETV in patients with decompensated cirrhosis reported by other studies.^96,97 TDF is associated with a small risk of nephrotoxicity and decreased bone mineral density.⁹⁸

Interferon based treatment for dealing with NA resistance.  A previous study showed peginterferon α-2a combined with LAM was not superior to peginterferon α-2a monotherapy.⁹⁹ In a small study Sarin et al.¹⁰⁰ suggested that 4 weeks of LAM given initially to decrease HBV DNA levels before switching to peginterferon lead to improved sustained virological response. Contrarily, will patients receiving NA regimens benefit from previous or simultaneous peginterferon introduction? What is the role of peginterferon in the rescue therapy of NA treatment failure? Lau et al.¹⁰¹ found that LAM combined with peginterferon was superior to monotherapy in reducing the incidence of YMDD mutations (4% vs 27%). Initial data suggested that interferon based regimens may provide a novel therapeutic option for the management of patients carrying LAM resistant virus.^59,64 However, due to the limitation of study designs and small sample sizes, it is difficult to make definitive conclusions. Sun et al.¹⁰² accomplished the first randomized, multi-center study to reveal the efficacy of peginterferon α-2a in patients harboring LAM resistant mutants. In spite of the sub-optimal virological response compared with treatment-naïve patients, HBsAg clearance was achieved in five patients at 24 weeks after cessation of peginterferon α-2a treatment. A case report from Italy described a particular therapeutic schedule formulated by add-on pegylated interferon α-2α to the pre-existing 4-year LAM administration for a subsequent period of 12 months, which succeeded in achieving HBsAg seroconversion together with sustained virological and biochemical response in up to 2 years of follow-up.¹⁰⁰ Moreover, sequential therapy from NA to pegIFNα-2a instead of combination therapy has also been shown to be safe and effective.¹⁰³ Interferon-based therapies are worthy of consideration in both prevention of de novo resistance and management of previous refractory NAs treatment.¹⁰⁴

Perspectives

While major advances in the treatment of CHB has been made in the past decade, the prevention and management of drug resistance is still very challenging. The role of periodical monitoring of quantitative HBsAg level and HBV genotypes remains controversial. Oncogenic potential of drug resistance needs further investigation. To ultimately overcome the cross-resistance and multiple resistances, the development of drugs with distinct antiviral mechanism of NA is essential. Chinese scientists recently indicated that combination of NA with Oxymatrine, a natural drug extracted from a Chinese herb (Kushen), could reduce the chance of developing drug resistance.¹⁰⁵ Such agents provide complementary treatment options. In this era of potent NA treatment, options have been expanded; various combination regimens including interferon-based therapies are available. However, the efficacy of the best combination, best timing of initiation, duration of therapy, potential add-on side effects due to drug interaction are uncertain. Although theoretically attractive, these combinations and applications from bench to bedsides still await justification from larger scale clinical studies and cost–benefit analyses.