Glecaprevir/pibrentasvir+sofosbuvir: an optimal retreatment strategy in the setting of HCV NS5A resistance

Abstract

Some individuals do not achieve a cure of their hepatitis C virus (HCV) infection due to non-adherence or resistance associated substitutions. Salvage options that are optimised for resistance profiles are essential. We report a 56-year-old Caucasian man with fatigue, depression and confusion in the setting of untreated HCV genotype 3a infection. He received ruzasvir and uprifosbuvir for 12 weeks within a clinical trial. The patient relapsed 4 weeks after the end of treatment and at this time resistance testing showed multiple resistances including a NS5A Y93H mutation. Given that this mutation confers resistance to first line salvage options, sofosbuvir and glecaprevir/pibrentasvir was used for 12 weeks and the patient was cured of HCV infection 12 weeks after the end of treatment. This shows that sofosbuvir and glecaprevir/pibrentasvir is a viable, effective option for second line/salvage therapy of HCV infection in the setting of resistance to NS5A inhibitors with the Y93H mutation.

Background

About 71.1 million people worldwide are affected by chronic hepatitis C virus (HCV) infection, and 10%–20% of these individuals will develop liver complications such as cirrhosis and hepatocellular carcinoma.¹ However, with the advent of all oral direct-acting antiviral (DAA) therapy, up to 95% of HCV infections can be cured with minimal side-effects and complications.² Even in this era, real-world sustained virological response 12 weeks after the end of therapy (SVR12 or cure) rates are often below 90%. This is partially due to poor adherence, losses to follow-up and an emergence of resistance associated substitutions (RASs) leading to treatment failures with some DAA regimens.³ Particularly with specific genotypes, certain mutations can confer high-grade resistance. For example, in individuals infected with HCV genotype 3, the Y93H mutation in the non-structural 5A region (NS5A) of the viral genome induces resistance to daclatasvir and velpatasvir.⁴

This is clinically significant as the currently approved second line/salvage therapy for HCV infection in British Columbia is the coformulation of sofosbuvir (SOF)/velpatisvir (VEL)/voxilaprevir (VOX). This formulation consists of a nucleotide analogue (SOF), non-structural protein 5A inhibitor (NS5A, VEL) and a non-structural 3/4A protease inhibitor (NS3/4A, VOX). While SOF/VEL/VOX has been highly effective in many settings (cure rates >95%), there has been documentation of adverse side effects leading to its discontinuation in some patients.⁵ The presence of RASs may also reduce its efficacy, making the choice of the right salvage regimen critical, although SOF/VEL/VOX remains a potent and viable salvage option in most cases, justifying its positioning as the first line salvage therapy.

Glecaprevir (NS3/4A inhibitor, GLE)/pibrentasvir (NS5A inhibitor, PIB) is a newly approved regimen for the treatment of chronic HCV in Canada. However, its potential as a salvage therapy has not yet been well documented. In recent publications, GLE/PIB is being utilised as a salvage therapy, but data from this cohort did not include patients infected with genotype 3 infection, especially in the setting of the Y93H RAS.^{6 7} It may be that GLE/PIB, especially if it is combined with another potent third agent such as SOF, may be particularly useful in this setting.

We report a man with a history of injection drug use, infected with HCV genotype 3a with a Y93H RAS, successfully treated with SOF/GLE/PIB as second line/salvage therapy.

Case presentation

A 56-year-old man was referred to our infectious diseases clinic in November 2016 with symptoms of fatigue, depression and confusion in the setting of untreated HCV genotype 3a infection which was initially diagnosed in 1993. On social history, he reported using injection drugs in the past but has been free of recreational drug use for over 20 years. In 2016, he smoked ¾ of a pack per day. All relevant blood work and investigations are summarised in table 1 and figure 1. On the initial visit, appropriate bloodwork and investigations were performed and his HCV viral load was 6.6 log₁₀ IU/mL with stage 0 fibrosis, as determined by transient elastography (FibroScan). Additionally, he was found to be immune to hepatitis A but not to hepatitis B for which immunisation was initiated. Physical examination of the cardiovascular and respiratory systems was unremarkable. Additionally, there were no signs of Dupuytren’s contractures, Terry’s nails, spider angiomata, hepatomegaly, splenomegaly, ascites or lower extremity oedema. In February 2017, he was enrolled in a clinical trial using the pangenotypic combination of ruzasvir and uprifosbuvir for the treatment of his HCV infection. In March 2017, he initiated treatment. At week 4 of therapy, the patient’s viral load was undetectable, and the rest of his laboratory evaluations were unremarkable. He had no specific symptoms and was 100% adherent to medication. Viral load remained undetectable until the end of therapy at 12 weeks, and all other laboratory evaluations were normal. The only side effect of HCV treatment was transient slight dizziness. At week 4 after the end of therapy (PT4, July 2017) an early virological relapse was detected with a viral load of 5.8 log₁₀ IU/mL. Resistance testing via genome deep sequencing was performed and the isolate was found to be resistant to grazoprevir, velpatasvir, elbasvir and daclatasvir (table 2). In September 2018, he initiated second line/salvage therapy with SOF/GLE/PIB for 12 weeks at a baseline viral load of 4.13 log. At week 1, 2 and 3 of follow-up viral load was steadily decreasing, while liver enzymes were normalising. The only side effects reported were mild fatigue and headache. By week 4, the patient’s viral load was undetectable. During the 12 weeks of therapy, this patient remained 100% adherent to the medications and did not experience any adverse events leading to treatment discontinuation. Viral load remained negative 12 weeks post-therapy indicating a complete cure of infection. At follow-up on 29 May 2019, he remains cured of HCV.

Table 1.

Blood work and investigation

Date of Blood work	Visit	Albumin (g/L)	Total bilirubin (umol/L)	ALP (U/L)	GGT (U/L)	ALT (U/L)	AST (U/L)	HCV RNA (Logs)	FibroScan score (kPa)
November 2016	First visit	41	6	81	20	67	43	6.68	3.7
March 2017	Baseline	48	7	97	23	64	46	7.16	4.9
April 2017	Week 4	40	5	96	15	10	15	0	TND
June 2017	Week 12	45	8	95	11	14	20	0	TND
July 2017	PT4	43	7	95	16	68	47	4.76	TND
August 2017	PT8	42	8	97	21	91	53	5.27	TND
October 2017	Awaiting salvage	41	6	105	18	53	38	5.8	3.2
July 2018	Awaiting salvage	41	5	127	17	59	43	4.93	TND
September 2018	Baseline	43	5	117	20	66	46	4.13	4.8
October 2018	Week 4	TND	TND	TND	TND	21	24	0	TND
December 2018	Week 12	42	7	104	10	20	20	0	TND
December 2018	PT2	TND	TND	TND	TND	21	19	0	TND
February 2019	SVR12	42	6	107	8	12	17	0	4.5
May 2019	Follow-up	TND	TND	TND	TND	14	18	0	TND

ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase; HCV, hepatitis C virus; TND, test not done.

Figure 1.

ALT, AST and HCV RNA values over time. ALT, alanine aminotransferase; AST, aspartate aminotransferase; HCV, hepatitis C virus.

Table 2.

Resistance analysis

Resistance marker	Mutation	Drug resistance
HCV NS3	Q168Q	Grazoprevir
HCV NS5A	A30K, L31V, Y93H	Daclatasvir, elbasvir, velpatasvir
HCV NS5B	None	None

HCV, hepatitis C virus.

Treatment

Initially, this patient was treated with trial combination of ruzasvir and uprifosbuvir for 12 weeks which was not successful in curing their HCV infection. He was later treated with SOF, GLE and PIB for 12 weeks which cured his HCV infection.

Outcome and follow-up

12 weeks after the end of salvage therapy, this patient became cured of his HCV infection. At follow-up on 29 September 2019, this patient remained cured of HCV.

Discussion

With the use of DAAs, HCV is now an easily curable and immediate treatment is recommended for all viraemic patients by The European Association for the Study of the Liver (EASL) and The American Association for the Study of Liver Diseases (AASLD) guidelines.^{8 9} In recent years, pangenotypic therapy has become available, simplifying access to treatment and increasing cure rates (>95%). In cases of virological failure with DAAs, there are salvage regimens that also achieve high (>95%) cure rates. Currently in British Columbia, Canada, one of the standards of care for initial DAA therapy is the pangenotypic combination of SOF/VEL which has been shown to be highly effective in clinical trial such as ASTRAL 1–5.⁹ For those who experience virological failure, the salvage therapy of SOF/VEL/VOX was shown to be effective in a large number of patients in the POLARIS 1 and 4 trials.¹⁰ This salvage therapy is currently the regimen that is recommended by EASL 2018 guidelines.⁸ However, as the prevalence of RASs continues to rise among patients failing initial therapy, current standards of care need to be re-evaluated in the setting of mutations conferring resistance. Clinically relevant mutations such as the substitution from tyrosine to histidine at the 93rd codon (Y93H) in the NS5A gene induces resistance to regimens including daclatasvir and velaptasvir.⁴ Furthermore, in genotype 3 infections, the prevalence of the Y93H RAS can be as high as 10% for individuals with no prior NS5A inhibitor exposure. In cases of virological failure following NS5A inhibitor exposure, the RASs can emerge in the majority of individuals.¹¹ Y93H RASs are associated with a lower SVR rates in patients with HCV genotype 3 who are being treated with SOF/VEL for 12 weeks. In a phase 3 clinical trial, the SVR rate was 84% for patients with Y93H at baseline, and 97% for patients without such baseline mutations. For all genotype 3a patients with virological failure, Y93H was observed at the time of failure. Even though the combination of ruzasvir and uprifosbuvir is not a currently available regimen, the mutation documented in this patient is a common cause of virological faliure in patients treated with current first line regimens such as SOF/VEL. Additionally, in vitro data suggests that a Y93H mutation confers >100× fold change to required VEL concentrations but does not significantly impact PIB potency.¹² Data are also limited in patients with the Y93H RAS treated with SOF/VEL/VOX but in vitro data suggests that GLE may be more potent with a median half-maximal inhibitory concentration of 0.3 nm in vitro compared with 6.6 nm for VOX. This potentially places SOF/GLE/PIB as a more potent salvage therapy for patients with previous treatment experience—particularly those with specific RASs such as Y93H. In this particular patient, RASs were documented after failure of ruzasvir and uprifosbuvir treatment in a clinical trial at our centre and despite the unavailability of this regimen, the mutation conferred is highly clinically relevant as its not exclusively seen only in patients with this regimen but also with current first line treatments. Although resistance testing was not performed before enrolment in the clinical trial, it is possible that these RASs was treatment-induced. Although this patient was not cirrhotic, choosing the regimen with the best chance of achieving cure in second line therapy was imperative as failure on salvage regimens reduces options for subsequent ‘third line’ therapy, and the efficacy of such ‘deep salvage’ approaches is likely to be further reduced. The present study suggests that SOF/GLE/PIB is a viable, potent second line/salvage regimen for patients with treatment-emergent Y93H RASs, as occurred here after initial treatment failure on ruzasvir and uprifosbuvir, alanine aminotransferase and aspartate aminotransferase rose along with HCV RNA but quickly normalised when treatment was initiated with SOF/GLE/PIB, in association with rapid and sustained virological suppression. Some investigators have now used the combination of SOF/GLE/PIB in clinical practice with encouraging results.¹³ In addition, recent unpublished data from our centre outlines a cohort of 24 individuals with a history of virological faliure on DAA therapy that all achieved SVR using GLE/PIB or SOF/GLE/PIB as a salvage therapy.

Learning points.

• Patients who initially fail initial direct-acting antiviral therapy have more than one option for second line/salvage therapy.

• Although current guidelines recommend using sofosbuvir (SOF)/velpatisvir/voxilaprevir as initial salvage therapy, there may be better options in certain cases, such as among patients with hepatitis C virus genotype 3a infection and Y93H RASs.

• SOF/glecaprevir (GLE)/pibrentasvir (PIB) may be more potent and associated with higher SVR12 rates in specific situations.

• Larger trials are needed to evaluate SOF/GLE/PIB as a salvage therapy in a broader range of clinical and virological settings.