Detection of drug resistance mutations of hepatitis C virus in patients with failure of the treatment with direct acting antivirals

Full Text

Introduction

The hepatitis C virus (HCV) is a leading cause of chronic liver disease¹. Acute hepatitis develops approximately in 20% of cases after initial infection with HCV; around 15–45% of patients spontaneously clear the virus within 6 months of infection without treatment [1].

Absence of treatment and progression of the chronic process lead to replacement of the healthy liver tissue with fibrous tissue, to distortion of the parenchyma architecture, and, eventually, to liver fibrosis and cirrhosis in 25–30 years. Once the disease has progressed to the state of cirrhosis, the risk of developing hepatocellular carcinoma over a year is 1–5% [2].

Historically, combination therapy with pegylated interferon (weekly subcutaneous injections) and ribavirin (daily orally) during 24-48 weeks was the standard treatment of chronic hepatitis C (CHC). This treatment produced a sustained virological response (SVR) — a PCR undetectable viral load during the treatment, 12 weeks (SVR12) and 24 weeks (SVR24) after the treatment — in more than 50% of patients with HCV genotype 1 after the 12-month treatment and in more than 80% of patients with HCV genotypes 2 and 3 after the 6-month treatment [3].

The successful treatment strategy resulting in complete eradication of the virus can be limited both due to the presence of the therapeutically challenging HCV genotype and due to other factors. Such common side effects as flu-like, asthenic vegetative, and neurological syndromes, weight loss, leukopenia, and thrombocytopenia can lead to dosage reduction, specific medication therapy, and even to premature treatment termination. Higher treatment rates have been demonstrated by young people, female patients, Caucasians, and patients with a low HCV viral load in blood plasma and low levels of preexisting fibrosis [4]. Patients with clinically apparent liver cirrhosis demonstrated a successful response to interferon and ribavirin treatment in 21.5% of all cases and in 11.3% of cases with HCV genotype 1, thus proving low efficiency of treatment for the above cohort of patients [5]. Non-responding patients were recommended to switch to long-term interferon monotherapy, which decreased serum aminotransferase levels, though it did not eradicate the virus [6].

The advent of direct acting antivirals (DAAs) has revolutionized treatment of CHC patients. Interferonfree therapy has significantly increased SVR rates (over 90%), reduced the treatment duration (to 12 weeks), reduced the number of side effects (headaches and fatigue were most common among them), and permitted more convenient dosage regimens (once daily orally)². Novel antivirals have demonstrated their efficiency in treatment of patients with liver cirrhosis and other complications, resulting in high SVR rates, which were difficult to achieve with the previous regimens [7].

The HCV genome encodes 4 structural proteins (core, E1, E2, P7) and 6 non-structural proteins (NS2, NS3, NS4A, NS4B, NS5A, NS5B) [8, 9]. Unlike interferon agents indirectly affecting replication and development of viral particles, DAAs directly target the virus and disrupt its life cycle.

Currently, there are 4 groups of antivirals targeting 3 viral non-structural proteins: NS3 protease inhibitors, NS5A replication complex inhibitors, NS5B RNA-polymerase nucleotide and non-nucleotide inhibitors [10]. In 2011, telaprevir and boceprevir, NS3 protease inhibitors, became the first approved DAAs [11]. Triple therapy with pegylated interferon, ribavirin and NS3 protease inhibitor improved treatment efficiency by 25% when administered to treatment-naïve patients with HCV genotype 1. At the same time, patients with liver cirrhosis and non-responders showed rather low SVR rates [12].

In Russia, therapeutic agents from all four groups of inhibitors are used in clinical practice. A combination of antivirals is the main strategy in treatment of CHC patients [13].

Pangenotypic sofosbuvir-based regimens are currently representing the main treatment options [14]. These regimens are most efficient for patients with HCV genotypes 1, 2 and 3, which are most typical of Russia³. In the meantime, some of the patients (around 5%) do not respond antiviral therapy. Failures can be associated with the phenomenon of virus resistance to antivirals [15]. According to some sources, 83% of treatment failure cases were characterized by a single or multiple resistance mutations found in genomic regions encoding DAA targeted proteins [16, 17].

The virus resistance is associated with nucleotide substitutions occurring randomly in HCV NS3, NS5A, NS5B regions during each replication cycle. Some substitutions prevent antiviral molecules from binding to their protein targets, thus making the virus resistant to the antiviral agent. Such substitutions are known as resistance-associated substitutions. Most of such mutations in the genome interfere with the ability of the wild-type virus to survive and reproduce under normal conditions; however, in DAA therapy, mutant strains have an advantage over the wild-type virus [18]. Detection of such mutations in patients who failed the treatment and in treatment-naïve patients is of critical importance.

At the moment, Russia and European countries do not have any routine test to detect DAA resistance mutations in HCV. Nevertheless, it is being developed in many countries, including Russia [19]. According to the EASL Recommendations for Treatment of Hepatitis C (2018), physicians who have easy access to reliable resistance tests can use these results to guide their decisions and select the further treatment strategies both for initial treatment and after relapse [15].

The purpose of the study is to detect drug resistance mutations in HCV patients who failed DAA treatment.

Materials and methods

The study was performed on plasma samples from 3 CHC patients who failed to respond the combination treatment with sofosbuvir + daclatasvir. The samples were provided by the Clinic of Infectious Diseases at the Kirov Military Medical Academy (St. Petersburg, Russia) and the hospital of the Krasnoyarsk Scientific Center of the Siberian Branch of the Russian Academy of Sciences in 2018. All the patients demonstrated virological treatment failure.

This study was performed at the HIV Infection Immunology and Virology Laboratory of the Pasteur Research Institute of Epidemiology and Microbiology in 2018–2019. The study was approved by the local ethics committee. All the patients gave written informed consent to participate in the study.

Biochemical values, METAVIR-scored fibrosis stage, viral loads and virus genotyping were assessed for all the patients. The data are shown in the Table.

 

Patients data

Data	Patient 1	Patient 2	Patient 3
Age, years	43	55	55
Sex	Male	Male	Male
ALT, U/liter	190	66	131,1
Total bilirubin, pmol/liter	29,5	5,9	7,7
Fibrosis stage (METAVIR)	F1	F1	F2
Viral load before treatment, copies/ml	2,1 x 106	1,6 x 106	2,6 x 106
Genotype	1b	1b	1b
Treatment regimen	SOF+DAC	SOF+DAC	SOF+DAC

 

Following the EASL Recommendations, all the patients were administered combination treatment with sofosbuvir, 400 mg + daclatasvir, 60 mg, 1 tablet taken once a day after a meal [15].

The treatment efficiency as well as SVR12 and SVR24 rates were assessed by measuring the viral load 12 and 24 weeks after the beginning of treatment. Highly sensitive tests did not detect any viral load in patient 1 during the screening performed 12 weeks after the start of antiviral therapy. The screening after 24 weeks showed a viral load of 1.1 × 10⁶ copies/mL. During the screening performed 12 weeks after the beginning of treatment of patients 2 and 3, the viral load was 3.2 × 10⁵ and 4.2 × 10⁵ copies/mL, respectively.

The first analytical stage included assessment of the viral load and identification of the virus genotype by using AmpliSens HCV-monitor-FL and AmpliSens HCV-genotype-FL reagent kits, respectively. The RNA was isolated from blood plasma by using the RIBO-prep kit for extraction of RNA/DNA from clinical materials (Central Research Institute of Epidemiology, Moscow).

The next stage included a reverse transcription reaction with the REVERTA-L reagent kit for cDNA synthesis from an RNA template (Central Research Institute of Epidemiology, Moscow).

Genotype-specific primers were used to obtain sequences of three viral regions with resistance-associated mutations: NS3, NS5A, and NS5B. The amplification products were purified and analyzed for the fragment size and concentration. The ABI Prism 3500 genetic analyzer (Applied Biosystems) was used to identify nucleotide sequences. The primary analysis of the obtained consensus nucleotide sequences was completed by using the NCBI BLAST program against the nucleotide sequences available in the international GenBank database. The fragments of NS3, NS5A, NS5B regions were assessed for presence of nucleotide substitutions causing drug resistance mutations; the assessment was performed with the Geno2pheno [resistance] 3.4 interpretation system⁴.

Results

By using direct sequencing, we obtained satisfactory 800, 750 and 600 bp nucleotide sequences of target NS3, NS5A, NS5B genes for all 3 samples.

Phylogenetic relationships between the studied samples collected from CHC patients in Russia and the reference sequences from the international GenBank database are shown in the Figure. For the purpose of comparison, we used HCV nucleotide sequences of genotypes 1a, 1b, and 3a.

The dendogram analysis suggests that sequences of HCV NS5B gene of three genotypes cluster separately from each other. All three samples with virological failure are distributed on the neighboring tree branches belonging to HCV nucleotide sequences of genotype 1b, thus correlating with the genotypes previously identified with the help of the commercial test system.

The analysis of drug resistance mutations in all the patients showed the following drug resistance mutations:

• patient 1: the Y93H mutation in the NS5A region; the mutation caused virus resistance to daclatasvir included in the regimen and to elbasvir, ledipasvir, ombitasvir, velpatasvir not included in the regimen;
• patient 2: the L31V mutation in the NS5A region resulted in decreased sensitivity to daclatasvir and ombitasvir;
• patient 3: similar to patient 2, the L31V nucleotide substitution was detected in the NS5A region. In addition, L159F and Y56F mutations were detected in the NS5B and NS3 regions, respectively. The L159F nucleotide substitution causes a decrease in sensitivity to sofosbuvir. The Y56F nucleotide substitution in the NS3 region causes a decrease in sensitivity to grazoprevir.

All the patients were administered combination treatment with glecaprevir (the NS3 inhibitor) + pibrentasvir (the NS5A inhibitor), as no resistance mutations to them were detected. Patients 2 and 3 reached SVR12 and SVR24 by following the regimen. Patient 1 decided not to follow the recommendations and discontinue the treatment.

Discussion

Treatment of CHC with pegylated interferon and ribavirin is not a totally efficient method for eradication of the virus and it may cause a number of problems. The solution to some of them was found with the advent of direct-acting antivirals. Sofosbuvir is an inhibitor of NS5B protein RNA-dependent RNA polymerase. In sofosbuvir-containing regimens, therapeutic efficacy is achieved in more than 95% of cases [20]. Sofosbuvir is characterized by pangenotypic, high antiviral activity, thus being the best-choice medication for most patients. Daclatasvir is a direct-acting inhibitor of the HCV NS5A protein. The combination with sofosbuvir suppresses the viral replication by targeting two regions at a time, thus providing therapeutic efficacy in more than 90% of the patients [16].

DAA combination regimens including NS5A inhibitors are most relevant. According to the EASL Recommendations (2016), HCV resistance testing prior to treatment makes sense if the regimen includes NS5A inhibitors [21].

Despite the high barrier to resistance due to the combination of two antiviral agents, in all three cases, the treatment failure evidenced as detected viral RNA in the plasma during the treatment can indicate the developing dominance of the resistant strain over the wild-type virus during antiviral treatment and, consequently, the occurring resistance to the administered regimen. It is quite possible that resistance mutations of the circulating HCV strain existed in these patients
before the beginning of treatment, thus being a cause of their non-response to the antiviral treatment. For patients 2 and 3, it is reasonable to assume that viral strains with DAA resistance mutations circulated before the treatment, thus explaining the non-response to the antiviral therapy and failure to achieve SVR12 rates. On the contrary, as SVR12 was achieved in patient 1, it can be assumed that the resistant HCV strain dominated over the wild type during the treatment.
However, there is a risk of reinfection of this patient with another viral strain. Thus, when performed before the treatment, the sequencing of HCV genes, in which nucleotide substitutions are associated with DAA resistance, makes it possible to detect mutations to therapeutic agents planned for administration. The analysis of nucleotide sequences helps avoid administration of inadequate regimens for patients, thus being of importance for a health care facility or individual patients with no need to change the regimen. Furthermore, the HCV sequencing prior to treatment helps assess the phylogenetic relatedness of viral strains in case of a viral breakthrough, thus making it possible to differentiate between on-treatment reinfection and disease relapse.

HCV strains harboring Y93H polymorphism detected in patient 1 are associated with a high resistance level and are found in NS5A-inhibitor treatment-naïve patients [22]. Strains with this mutation can persist for a long time even after the DAA treatment is discontinued. Moreover, the above nucleotide substitution causes increased production of DAA resistant viruses, though it is associated with higher sensitivity to protease inhibitors. Thus, the detailed profile of HCV variants with the Y93H polymorphism is required for making an optimum choice of treatment strategy for CHC patients.

The treatment failure with first-line antivirals should be followed by the second-line pangenotypic DAA regimen. Russia has registered elbasvir,⁵ ledipasvir,⁶ ombitasvir,⁷ and velpatasvir,⁸ which are included in combination pangenotypic therapy [22]. The Y93H nucleotide substitution leads to cross-resistance to the above antivirals. Thus, some mutations leading to drug cross-resistance can cause failure of treatment of CHC patients when both the first-line therapy and alternative DAA treatment options are administered. In this case, the pressing problems are those of pharmacoeconomic nature, as there can be need to change between multiple regimens and there can be complications resulting from absence of positive effect of the antivirals.

In patients 2 and 3, the L31V single nucleotide substitution in the NS5A region decreases sensitivity to some inhibitors. However, the combination of L31V and Q54L mutations can enhance sensitivity to other DAAs [23].

In clinical study P7977-2025, 16 (26%) out of 61 patients showed no response to the sofosbuvir + ribavirin treatment after the liver transplantation. Six (37.5%) patients out of 16 did not achieve the SVR12 rate and, importantly, had C316N amino acid substitution detected in the virus. The other 45 (74%) patients achieved SVR12 and had no C316N substitution [24]. In clinical study NEUTRINO, 4 (6%) patients out of 66 had HCV isolates with C316N mutation before the treatment. Among them, 50% of the patient did not respond the sofosbuvir + ribavirin treatment. In the above study, 85% of patients with HCV strain without C316N mutation achieved SVR12 [20, 25]. Therefore, post-transplantation patients who need DAA treatment should be tested for HCV resistance to DAAs to eliminate both the risk of a viral breakthrough during the treatment and the risk of non-response to the treatment.

In addition to drug resistance-associated mutations, one polymorphic variant was detected in the NS3 gene in patient 1, while patients 2 and 3 had 6 and 7, respectively. It should be noted that the F147S mutation was detected in all 3 cases, while S7A, A66G, K87A, and F147S mutations were detected in 2 out of 3 cases.

6 polymorphic variants were detected in the NS5B gene in patient 1, while patients 2 and 3 had 6 and 3 polymorphic variants, respectively. S231N and R254K mutations were detected in 2 out of 3 cases.

7 polymorphic variants were detected in the NS5A gene in patient 1, while patients 2 and 3 had 15 and 12 polymorphic variants, respectively. Notably, K6R, S17T, L34V, and K44R mutations were detected in all 3 cases, while T64S, V164A, V174T, and L183V mutations were detected in 2 out of 3 patients.

High genetic polymorphism is typical of all RNA-viruses, including HCV. Yet, the phenomenon of frequently occurring polymorphic variants needs further study, as such variants can cause both decreased sensitivity and decreased resistance to antivirals. The analysis of natural polymorphic patterns is highly important, as these patterns can potentially function as drug resistance mutations of the virus. With development of novel antivirals and novel regimens, attention should be given to frequently occurring nucleotide substitutions, which previously were not associated with resistance as drug resistance mutations. Statistical data on DAA treatment are being continuously updated and it is quite possible that a weaker response or non-response to existing antivirals can be caused by the mutations that are still not associated with drug resistance or by new mutations.

Although sofosbuvir-containing regimens are characterized by a high barrier to genetic resistance, treatment failures are still reported as caused by drug resistance mutations in one or several regions of the virus. The lack of routine HCV tests for resistance to antiviral agents not only aggravates the problem of optimum regimen selection, but also involves pharmacoeconomic aspects when other regimens have be administered after failed treatment with first-line antivirals. Such clinical cases emphasize the importance of the test for resistance to administered antivirals in case of a viral breakthrough and virological failure as well as before the administration of treatment.

The level of resistance to antivirals depends on several factors: The genetic barrier to drug resistance; the type of nucleotide substitution in the viral genome; the occurrence rate of the resistance-associated substitution in the virus population; existing combinations of nucleotide substitutions, which are present in the virus population. These factors can significantly increase the resistance to some treatment methods [26]. The high mutation rate and the speed of replication of RNA-containing viruses underlie the genetic diversity of the virus pool persisting in an individual. Therefore, mutations, including drug resistance-associated mutations, can be spontaneously generated in any patient. Thus, the pool of closely related virus variants known as quasispecies can contain a small population of drug-resistant variants before the beginning of antiviral treatment [27].

The Sanger sequencing method is efficient in detecting a major virus variant (a quasispecies), which prevails in the virus population of the studied clinical sample, though it does not detect minor genomes of drug-resistant quasispecies present in the sample and accounting for less than 15% of the total virus pool, while the next-generation sequencing provides highly reliable identification of minor mutations accounting for 1–2% of the population, helps assess the occurrence rate for each mutation in the virus population, and evaluates the possibility of several mutations combined in one viral genome. This sequencing method can be important for treatment efficiency and for treatment strategies [28].Nevertheless, direct sequencing is more affordable and economically feasible, especially when working with individual patients in the population where the level of drug resistance of the virus to antiviral agents is low.

Sequencing of viral genome regions will not only provide data on exiting resistance mutations, but also will be important for analyzing epidemiological situations in and outside the country through identification of virus genotypes and subgenotypes as well as through the phylogenetic analysis [29].

Conclusion

The development of a universal test for detecting DAA resistance in HCV is of high importance both for administration of treatment and for any changes in the treatment, if required. 

 

About the authors

D. E. Valutite

Saint Petersburg Pasteur Institute

Author for correspondence.
Email: dianavalutite008@gmail.com
ORCID iD: 0000-0002-0931-102X

Diana E. Valutite — doctor of clinical laboratory diagnostics, Department for diagnosing HIV infection and AIDS-related diseases

St. Petersburg

Russian Federation

A. V. Semenov

Saint Petersburg Pasteur Institute; Pavlov Saint Petersburg State Medical University; North-Western State Medical University named after I.I. Mechnikov

Email: fake@neicon.ru
ORCID iD: 0000-0003-3223-8219

Aleksandr V. Semenov — D. Sci. (Biol.), Head, Laboratory of virology and immunology of HIV, Deputy director for innovation, St. Petersburg Pasteur Institute; Assoc. Prof., Department of immunology, Pavlov First St. Petersburg State Medical University; Assoc. Prof., Department of clinical laboratory diagnostics, North-West State Medical University named after I.I. Mechnikov

St. Petersburg

Russian Federation

Yu. V. Ostankova

Saint Petersburg Pasteur Institute

Email: fake@neicon.ru
ORCID iD: 0000-0003-2270-8897

Yulia V. Ostankova — PhD (Biol.), senior researcher

St. Petersburg

Russian Federation

K. V. Kozlov

S.M. Kirov Military Medical Academy

Email: fake@neicon.ru
ORCID iD: 0000-0002-4398-7525

Konstantin V. Kozlov — D. Sci. (Med.), Assoc. Prof., Department of infectious diseases (with a course of medical parasitology and tropical diseases)

St. Petersburg

Russian Federation

A. G. Borisov

Krasnoyarsk Science Center of the Siberian Branch of the Russian Academy of Sciences

Email: fake@neicon.ru
ORCID iD: 0000-0002-9026-2615

Aleksandr G. Borisov — PhD (Med.), leading researcher, Laboratory of molecular cell physiology and pathology, Scientific Research Institute of Medical Problems of the North; Assoc. Prof., Department of infectious diseases, Krasnoyarsk State Medical University named after Prof. V.F. Voino-Yasenetsky

Krasnoyarsk

Russian Federation

V. D. Nazarov

Pavlov Saint Petersburg State Medical University

Email: fake@neicon.ru
ORCID iD: 0000-0002-9354-8790

Vladimir D. Nazarov — PhD (Med.), research fellow, Laboratory of autoimmune diagnostics, Center for molecular medicine

St. Petersburg

Russian Federation

A. A. Totolian

Saint Petersburg Pasteur Institute; Pavlov Saint Petersburg State Medical University

Email: fake@neicon.ru
ORCID iD: 0000-0003-4571-8799

Areg A. Totolian — D. Sci. (Med.), Prof., Academician of RAS, Head, Laboratory of molecular immunology, Director, St. Petersburg Pasteur Institute; Head, Department of immunology, Academician I.P. Pavlov First St. Petersburg State Medical University

St. Petersburg

Russian Federation

References

Supplementary files