EASL Recommendations on Treatment of Hepatitis C 2018

Other causes of chronic liver disease, or factors which are likely to affect the natural history or progression of liver disease and therapeutic choices, should be systematically investigated. All patients should be tested for other blood-borne viruses, particularly hepatitis B virus (HBV), and for human immunodeficiency virus (HIV). HBV and hepatitis A virus (HAV) vaccination should be proposed for patients who are not protected. Alcohol consumption should be assessed and quantified, and specific counselling to stop harmful alcohol consumption should be given. In addition, HCV may cause a variety of extra-hepatic manifestations which need to be considered in the work-up of HCV-infected patients. Thus, assessments should be carried out for possible comorbidities, including alcoholism, cardiac disease, renal impairment, autoimmunity, genetic or metabolic liver diseases (for instance genetic hemochromatosis, diabetes mellitus or obesity) and the possibility of drug-induced hepatotoxicity.

Assessment of liver disease severity is necessary prior to therapy. Identifying patients with cirrhosis (METAVIR score F4) or advanced (bridging) fibrosis (METAVIR score F3) is of particular importance, as the choice of treatment regimen and the post-treatment prognosis depend on the stage of fibrosis. Assessment of the stage of fibrosis is not required in patients with clinical evidence of cirrhosis. Patients with cirrhosis need to be assessed for portal hypertension, including oesophageal varices. Patients with advanced fibrosis and those with cirrhosis need continued post-treatment surveillance for HCC every 6 months. Since significant fibrosis may be present in patients with repeatedly normal ALT, evaluation of disease severity should be performed regardless of ALT levels.

In chronic hepatitis C, non-invasive methods should be used instead of liver biopsy to assess liver disease severity prior to therapy. Liver stiffness measurement can be used to assess liver fibrosis and the presence of portal hypertension in patients with chronic hepatitis C. Consideration must be given to factors that may adversely affect its performance, such as obesity, high ALT levels, or post-prandial testing. Well-established panels of fibrosis biomarkers can also be applied. Both liver stiffness measurement and biomarkers perform well in the identification of cirrhosis or no fibrosis, but they perform less well in resolving intermediate degrees of fibrosis. Cut-offs used with common non-invasive markers to establish the different stages of fibrosis in patients with chronic hepatitis C prior to therapy are shown in Table 2.^{,  ,  ,  ,  ,}  In low- and middle-income countries, as well as in settings where treatment expands outside of specialty clinics, aspartate aminotransferase to platelet ratio index (APRI) and fibrosis-4 (FIB-4) are generally available, simple and cheap, and the information they provide is reliable. Notably, non-invasive tools should not be used to assess the fibrosis stage after therapy, as they are unreliable in this setting.

The combination of blood biomarkers or the combination of liver stiffness measurement and a blood test improve accuracy.^,  Liver biopsy may be required in cases of known or suspected mixed aetiologies (e.g. metabolic syndrome, alcoholism or autoimmunity).

HCV RNA detection or detection/quantification in serum or plasma is indicated for patients who undergo antiviral treatment. HCV RNA assessment should be made by a reliable sensitive assay, and HCV RNA levels should be expressed in IU/ml.

HCV core antigen detection and quantification by means of EIA can be performed when HCV RNA tests are not available and/or not affordable. HCV core antigen quantification should be made with a reliable assay and core antigen levels should be expressed in fmol/L.

Together with prior treatment experience and the presence of cirrhosis, the HCV genotype, including genotype 1 subtype (1a or 1b), is still useful to tailor the treatment regimen and its duration. Genotyping/subtyping should be performed with an assay that accurately discriminates subtype 1a from 1b, i.e. an assay using the sequence of the 5′ untranslated region plus a portion of another genomic region, generally the core-coding or the NS5B-coding regions. The most widely used method is based on reverse hybridization with the line probe assay. A kit based on deep sequencing will soon be available.

With pan-genotypic HCV drug regimens, it is possible to treat individuals without identifying their HCV genotype and subtype. This may be particularly useful in regions where virological tests are not available or their cost exceeds that of antiviral treatment, or to simplify therapy in other regions, in order to improve access to care.

No standardized tests for resistance of HCV to approved drugs are available as purchasable kits. Resistance testing mostly relies on in-house techniques based on population sequencing (Sanger sequencing) or deep sequencing. A limited number of laboratories have made such tests available in Europe and elsewhere. HCV resistance testing may be technically difficult, in particular for genotypes other than 1 and 4, and the performances of the available in-house assays vary widely. A kit based on deep sequencing is currently at the developmental stage.

Access to reliable HCV resistance testing is limited and there is no consensus on the techniques, interpretation and reporting of these tests. In addition, highly efficacious treatments are now available for patients with detectable pre-existing resistance-associated substitutions (RASs) at baseline. Thus, systematic testing for HCV resistance prior to treatment in direct-acting antiviral (DAA) drug-naïve individuals is not recommended.

The current EASL recommendations suggest treatment regimens that do not necessitate any resistance testing prior to first-line therapy. In areas where these regimens are not available or not reimbursed, physicians who have easy access to reliable resistance tests can use these results to guide their decisions, according to the EASL Recommendations for Treatment of Hepatitis C 2016.

Sofosbuvir should be administered at the dose of 400 mg (one tablet) once per day, with or without food. Approximately 80% of sofosbuvir is renally excreted, whereas 15% is excreted in faeces. The majority of the sofosbuvir dose recovered in urine is the dephosphorylation-derived nucleoside metabolite GS-331007 (78%), while 3.5% is recovered as sofosbuvir. Renal clearance is the major elimination pathway for GS-331007, with a large part actively secreted. Thus, currently, no sofosbuvir dose recommendation can be given for patients with severe renal impairment (eGFR <30 ml/min/1.73 m²) or with end-stage renal disease because of higher exposures (up to 20-fold) of GS-331007. However, there is accumulating evidence on safe use of sofosbuvir-based regimens in patients with an eGFR <30 ml/min/1.73 m², including patients on haemodialysis. Sofosbuvir exposure is not significantly changed in patients with mild liver impairment, but it is increased 2.3-fold in those with moderate liver impairment.

Sofosbuvir is well tolerated over 12 to 24 weeks of administration. The most common adverse events (≥20%) observed in combination with ribavirin were fatigue and headache. Slight elevations of creatine kinase, amylase and lipase without clinical impact were also observed.

Sofosbuvir is not metabolised by cytochrome P450, but is transported by P-gp. Drugs that are potent P-gp inducers significantly decrease sofosbuvir plasma concentrations and may lead to a reduced therapeutic effect. Thus, sofosbuvir should not be administered with known inducers of P-gp, such as rifampicin, carbamazepine, phenytoin or St. John’s wort. Other potential interactions may occur with rifabutin, rifpentine and modafinil. No significant drug-drug interactions have been reported in studies with the antiretroviral agents emtricitabine, tenofovir, rilpivirine, efavirenz, darunavir/ritonavir and raltegravir, and there are no potential drug-drug interactions with other antiretrovirals.

Sofosbuvir-based regimens are contraindicated in patients who are being treated with the anti-arrhythmic amiodarone because of the risk of life-threatening arrhythmias. Indeed, bradycardia has been observed within hours to days of starting the DAA, but cases have been observed up to 2 weeks after initiating HCV treatment. The mechanism of interaction and the role of other co-medications (e.g. β-blockers) is still unclear, although a number of potential mechanisms have been proposed involving P-gp inhibition, protein binding displacement and direct effects of sofosbuvir and/or other DAAs on cardiomyocytes or ion channels. Toxicity is likely the result of a combination of mechanisms. Because of the long half-life of amiodarone, an interaction is possible for several months after discontinuation of amiodarone. If the patient has no cardiac pacemaker in situ, waiting 3 months after discontinuing amiodarone before starting a sofosbuvir-based regimen is recommended. Sofosbuvir-containing regimens have also been implicated in cardiac toxicity in the absence of amiodarone, but this remains controversial. In the absence of specific drug-drug interaction data, caution should be exercised with antiarrhythmics other than amiodarone.

Sofosbuvir and ledipasvir are available in a two-drug fixed-dose combination containing 400 mg of sofosbuvir and 90 mg of ledipasvir in a single tablet. The recommended dose of the combination is one tablet taken orally once daily with or without food.

Biliary excretion of unchanged ledipasvir is the major route of elimination with renal excretion being a minor pathway (accounting for approximately 1%), whereas sofosbuvir is principally excreted renally, as noted above. Following administration of sofosbuvir/ledipasvir, the median terminal half-lives of sofosbuvir and its predominant metabolite GS-331007 were 0.5 and 27 h, respectively. Neither sofosbuvir nor ledipasvir are substrates for hepatic uptake transporters; GS-331007 is not a substrate for renal transporters.

Ledipasvir plasma exposure (area under the curve [AUC]) was similar in patients with severe hepatic impairment and control patients with normal hepatic function. Population pharmacokinetics analysis in HCV-infected patients indicated that cirrhosis (including decompensated cirrhosis) had no clinically relevant effect on the exposure to ledipasvir.

While no dose adjustment of sofosbuvir and ledipasvir is required for patients with mild or moderate renal impairment, the safety of the sofosbuvir-ledipasvir combination has not been assessed in patients with severe renal impairment (eGFR <30 ml/min/1.73 m²) or end-stage renal disease requiring haemodialysis, but there is growing evidence of acceptable risk-benefit. Relative to patients with normal renal function (eGFR >80 ml/min/1.73 m²), the sofosbuvir AUC was 61%, 107% and 171% higher in patients with mild, moderate and severe renal impairment, while the GS-331007 AUC was 55%, 88% and 451% higher, respectively. Thus, no dose adjustment is required for patients with mild or moderate renal impairment, but no dose recommendation can currently be given for patients with severe renal impairment (eGFR <30 ml/min/1.73 m²) or with end-stage renal disease. Pangenotypic drug combinations that are not cleared by the kidney are available, thus obviating the need for sofosbuvir-based regimens where appropriate drugs are available.

The most common adverse reactions reported with this combination were fatigue and headache.

Since the combination contains ledipasvir and sofosbuvir, any interactions identified with the individual drugs will apply to the combination. The potential (limited) interactions with sofosbuvir have been previously outlined. Since both ledipasvir and sofosbuvir are transported by intestinal P-gp and breast cancer resistance protein (BCRP), any co-administered drugs that are potent P-gp inducers will not only decrease sofosbuvir but also ledipasvir plasma concentrations, leading to reduced therapeutic effect. Although co-administration with drugs that inhibit P-gp and/or BCRP may increase the exposure of sofosbuvir and ledipasvir, clinical consequences are unlikely.

Ledipasvir may also be the perpetrator of drug interactions by inhibiting P-gp and/or BCRP, potentially increasing the intestinal absorption of co-administered drugs. Thus, caution is warranted with well-studied P-gp substrates such as digoxin and dabigatran, but also potentially with other drugs which are, in part, transported by these proteins (e.g. aliskerin, amlodipine, buprenorphine, carvedilol, cyclosporine). Co-administration of amiodarone with sofosbuvir/ledipasvir is contraindicated because of a serious risk of symptomatic or even fatal bradycardia or asystole (see above, mechanism of interaction is unknown). The use of rosuvastatin is also not recommended (because of potential inhibition of hepatic OATP by ledipasvir) and interactions with other statins cannot be excluded. It is important to monitor carefully for statin-related adverse reactions. Since ledipasvir solubility decreases as pH increases, drugs that increase gastric pH (antacids, H2-receptor antagonists, proton pump inhibitors) are likely to decrease concentrations of ledipasvir. H2-receptor antagonists can be given simultaneously or 12 h apart at a dose not exceeding that equivalent to famotidine 40 mg and proton pump inhibitors can be given simultaneously, at a dose comparable to omeprazole 20 mg (Table 5). Real-world data have suggested slightly reduced SVR rates in patients receiving high-dose proton pump inhibitors, reinforcing the need for caution when treating patients on such drugs with sofosbuvir and ledipasvir.

Sofosbuvir/ledipasvir may be given with all antiretrovirals. However, because of an increase in tenofovir disoproxil fumarate (TDF) concentrations when a pharmacokinetic enhancer (ritonavir or cobicistat) is present in an antiretroviral regimen, these combinations (i.e. atazanavir/ritonavir, darunavir/ritonavir, lopinavir/ritonavir, elvitegravir/cobicistat, atazanavir/cobicistat, darunavir/cobicistat, all in combination with tenofovir disoproxil fumarate/emtricitabine) should be used with caution, with frequent renal monitoring if other alternatives are not available. The interaction is not mitigated by staggering administration by 12 h. Tenofovir levels are also increased in efavirenz-containing regimens and caution is required. The recent approval of tenofovir alafenamide (TAF), which results in considerably lower plasma tenofovir levels, means that there is less concern about an interaction leading to increased tenofovir exposure.

Sofosbuvir and velpatasvir are available in a two-drug fixed-dose combination containing 400 mg of sofosbuvir and 100 mg of velpatasvir in a single tablet. The recommended dose of the combination is one tablet taken orally once daily with or without food.

Velpatasvir is metabolised in vitro by CYP2B6, CYP2C8 and CYP3A4. However, because of the slow turnover, the vast majority of drug in plasma is the parent drug. Importantly, velpatasvir is transported by P-gp and BCRP and, to a limited extent, by organic anion transporting polypeptide (OATP) 1B1. Biliary excretion of the parent drug is the major route of elimination. The median terminal half-life of velpatasvir following administration of sofosbuvir and velpatasvir is approximately 15 h.

Velpatasvir plasma exposure (AUC) is similar in subjects with moderate and severe hepatic impairment compared to subjects with normal hepatic function. Cirrhosis (including decompensated cirrhosis) has no clinically relevant effect on velpatasvir exposure in a population pharmacokinetic analysis in HCV-infected individuals.

The pharmacokinetics of velpatasvir were studied in HCV-negative patients with severe renal impairment (eGFR <30 ml/min/1.73 m²). Relative to individuals with normal renal function, velpatasvir AUC was 50% higher, which was not considered to be clinically relevant.

The safety assessment of sofosbuvir and velpatasvir was based on pooled phase III data. Headache, fatigue and nausea were the most commonly reported adverse events, at as similar frequency to placebo-treated patients.

Because of the disposition profile of velpatasvir, there are some contraindications in relation to co-medications. Drugs that are potent P-gp or potent CYP inducers (e.g., rifampicin, rifabutin, carbamazepine, phenobarbital, phenytoin, St John’s wort) are contraindicated, because of the decrease in sofosbuvir and/or velpatasvir exposure with the potential loss in efficacy. However, there are also drugs that are moderate P-gp or CYP inducers (such as modafinil) which can reduce velpatasvir exposure. Currently, this combination would not be recommended with sofosbuvir and velpatasvir.

Similar to ledipasvir, there is some concern about the inhibition of P-gp and/or BCRP by velpatasvir, such that there is an increase in exposure of a co-medication that is a substrate for these transporters. Current thinking is that the sofosbuvir and velpatasvir combination may be co-administered with P-gp, BCRP, OATP and CYP substrates, but there clearly needs to be some caution with co-medications that have a narrow therapeutic window and in which an increase in drug exposure could potentially have clinical consequences. The colour coding for sofosbuvir/velpatasvir in Table 4A, Table 4B, Table 4C, Table 4D, Table 4E, Table 4F, Table 4G reflects this (e.g. for digoxin, dabigatran, ticagrelor, carvedilol, amlodipine, diltiazem, aliskiren).

Like ledipasvir, the solubility of velpatasvir decreases as pH increases. Therefore, it is important to be aware of the recommendations concerning the co-administration of antacids, H2-receptor antagonists and proton pump inhibitors. For most patients, proton pump inhibitors should be avoided during sofosbuvir/velpatasvir treatment. If considered necessary, sofosbuvir/velpatasvir should be given with food and taken 4 hours before the proton pump inhibitor, at a maximum dose comparable to omeprazole 20 mg (Table 5).

In HIV-HCV coinfected patients, sofosbuvir/velpatasvir may be given with most antiretrovirals, the exceptions being the inducing drugs efavirenz, etravirine and nevirapine. Efavirenz causes a 50% decrease in velpatasvir exposure. Sofosbuvir/velpatasvir also increases tenofovir exposure by inhibiting P-gp. This means that patients on a regimen containing TDF will need to be monitored for renal adverse events.

Sofosbuvir, velpatasvir and voxilaprevir are available in a three-drug fixed-dose combination containing 400 mg of sofosbuvir, 100 mg of velpatasvir and 100 mg of voxilaprevir in a single tablet. The recommended dose of the combination is one tablet taken orally once daily with food, as voxilaprevir plasma exposure (AUC) and maximum concentration (C_max) were 112% to 435%, and 147% to 680% higher, respectively, in the presence of food.

The specific pharmacokinetic information related to sofosbuvir and velpatasvir individually is discussed in previous sections. Voxilaprevir is metabolised in vitro by CYP3A4, with the vast majority of drug in plasma being the parent drug. Velpatasvir and voxilaprevir are both inhibitors of drug transporters P-gp, BCRP, OATP1B1 and OATP1B3. Biliary excretion of the parent drug is the major route of elimination for voxilaprevir. The median terminal half-life of voxilaprevir following administration of sofosbuvir, velpatasvir and voxilaprevir is approximately 33 h.

Population pharmacokinetic analysis of voxilaprevir in HCV-infected patients indicated that patients with compensated (Child-Pugh A) cirrhosis had 73% higher exposure of voxilaprevir than those without cirrhosis. Thus, no dose adjustment of sofosbuvir, velpatasvir and voxilaprevir is required for patients with compensated cirrhosis. The pharmacokinetics of single-dose voxilaprevir were also studied in patients with moderate and severe hepatic impairment (Child-Pugh B and C, respectively). Relative to patients with normal hepatic function, the voxilaprevir AUC was 3-fold and 5-fold higher in patients with moderate and severe hepatic impairment, respectively. Thus, the combination of sofosbuvir, velpatasvir and voxilaprevir is not recommended in patients with moderate hepatic impairment (Child-Pugh B) and contraindicated in those with severe hepatic impairment (Child-Pugh C).

The pharmacokinetics of voxilaprevir were studied in HCV-negative patients with severe renal impairment (eGFR <30 ml/min/1.73 m²). Relative to subjects with normal renal function, voxilaprevir AUC was 71% higher in subjects with severe renal impairment, which was not considered to be clinically relevant.

The safety data of sofosbuvir, velpatasvir and voxilaprevir was based on data from phase II and III clinical trials. Headache, diarrhoea and nausea were the most commonly reported adverse events. The risk of gastrointestinal side effects is greater than with the combination of sofosbuvir and velpatasvir alone.

Because velpatasvir and voxilaprevir are both inhibitors of P-gp, BCRP, OATP1B1 and OATP1B3, co-administration of sofosbuvir, velpatasvir and voxilaprevir with medicinal products that are substrates of these transporters may increase the exposure of the co-medications. This means that those for which elevated plasma levels are associated with serious events are contraindicated and others may require dose adjustment or additional monitoring. Rosuvastatin is contraindicated because of a 19-fold increase in plasma exposure of the statin. As this effect is likely to be attributed more to the BCRP transporter, other drugs that are a BCRP substrate, including methotrexate, mitoxantrone, imatinib, irinotecan, lapatinib, sulfasalazine and topotecan, are also not recommended. Dabigatran is contraindicated because of a near 3-fold increase in AUC. This is caused by P-gp inhibition by both velpatasvir and voxilaprevir. Other substrates of P-gp may need to be dose-adjusted or monitored for increased exposure, including digoxin, ticagrelor, carvedilol, diltiazem and aliskiren. Similar caution is required with OATP1B inhibitors, such as cyclosporin, as voxilaprevir plasma exposure increases 19-fold, or with OATP1B substrates, such as edoxaban, as voxilaprevir inhibition is expected to increase the exposure of the factor Xa inhibitor. Neither of these combinations are recommended.

Concomitant use with medicinal products that are strong P-gp and/or strong CYP inducers such as rifampicin, rifabutin, St. John's wort, carbamazepine, phenobarbital or phenytoin are contraindicated due to the decrease in sofosbuvir, velpatasvir and/or voxilaprevir exposure with the potential loss in efficacy. However, there are also drugs that are moderate P-gp or CYP inducers (such as modafinil, efavirenz, oxcarbazepine and others) which can also reduce exposure of this DAA and are not currently recommended.

For women of childbearing age, concomitant use with ethinylestradiol-containing contraception is contraindicated because of the risk of ALT elevations. Progestogen-containing contraception is allowed.

The solubility of velpatasvir decreases as pH increases. Therefore it is important to be aware of the recommendations concerning the co-administration of antacids, H2-receptor antagonists and proton pump inhibitors. Proton pump inhibitors can be given with sofosbuvir/velpatasvir/voxilaprevir at a dose that does not exceed doses comparable to omeprazole 20 mg (Table 5). Sofosbuvir/velpatasvir/voxilaprevir should be given with food and taken 4 hours before the proton pump inhibitor if possible.

In HIV-HCV coinfected patients, sofosbuvir/velpatasvir/voxilaprevir is not recommended with the inducing drugs efavirenz, etravirine and nevirapine, and the protease inhibitors atazanavir/ritonavir and lopinavir/ritonavir. Caution is required with twice daily darunavir/ritonavir, darunavir/cobicistat and atazanavir/cobicistat as there are no data. Efavirenz causes a 50% decrease in velpatasvir exposure and atazanavir causes a 4-fold increase in voxilaprevir exposure. Sofosbuvir/velpatasvir/voxilaprevir also increases tenofovir exposure by inhibiting P-gp. This means that patients on a regimen containing TDF need to be monitored for renal adverse events.

Paritaprevir is a protease inhibitor which is metabolised primarily by CYP3A4 and is given with a low dose of the CYP3A inhibitor ritonavir as a pharmacokinetic enhancer. This enables once daily administration and a lower dose than would be required without ritonavir. Ombitasvir is an NS5A inhibitor given in a fixed-dose combination with paritaprevir/ritonavir. The recommended dose of this combination is two tablets of ritonavir/paritaprevir/ombitasvir (50 mg/75 mg/12.5 mg per tablet) taken orally once daily with food. Dasabuvir is a non-nucleoside inhibitor of HCV RNA-dependent RNA polymerase administered in 250 mg tablets twice daily, in combination with ritonavir/paritaprevir/ombitasvir in genotype 1 patients.

Paritaprevir is excreted predominantly into the faeces. Ombitasvir shows linear kinetics, and is predominantly eliminated in the faeces. Dasabuvir is metabolised in the liver, and its predominant metabolite is mainly cleared via biliary excretion and faecal elimination with minimal renal clearance.

Pharmacokinetic results from hepatic impairment studies have shown that, in patients with severe hepatic impairment (Child-Pugh C), the AUC of paritaprevir was increased 9.5-fold, whereas ombitasvir was reduced 54% and dasabuvir was increased 3.3-fold. In Child-Pugh B, there is an increase in paritaprevir exposure of 62% with a decrease in ombitasvir of 30%. Thus, no dose adjustment is required for patients with mild hepatic impairment (Child-Pugh A), but the combination of ritonavir-boosted paritaprevir and ombitasvir with or without dasabuvir should not be used in patients with moderate hepatic impairment (Child-Pugh B) or in those with severe hepatic impairment (Child-Pugh C).

The AUC of paritaprevir was increased 45% in patients with severe renal impairment (creatinine clearance 15–29 ml/min), that of ritonavir 114%, and dasabuvir 50%. Currently, no dose adjustment is required for patients with mild, moderate or severe renal impairment. Paritaprevir, ombitasvir and dasabuvir can also be used in dialysis settings.

The most common side effects reported with the combination of ritonavir-boosted paritaprevir, ombitasvir and dasabuvir were fatigue and nausea.

Paritaprevir is primarily metabolised by CYP3A4, whereas dasabuvir is primarily metabolised by CYP2C8 and ombitasvir undergoes hydrolysis. However, both ombitasvir and dasabuvir can be metabolised by CYP3A4. Transporters seem to play an important role in the disposition of these drugs, with paritaprevir inhibiting OATP1B1/B3, P-gp and BCRP. Dasabuvir and ritonavir may also inhibit P-gp and BCRP. Given the metabolic profile of the drugs and the presence of ritonavir, there is a potential for many drug-drug interactions. A comprehensive drug-drug interaction programme has been undertaken based on regulatory guidance from both the European Medicines Agency and the US Food and Drug Administration. It is important to consider the drug interaction profile of the compounds as a combination (either with or without dasabuvir), because the drugs have mutual effects on each other.

Ritonavir is a strong inhibitor of CYP3A4; thus, co-administration with drugs metabolised by this enzyme may result in markedly increased plasma concentrations. A number of drugs are contraindicated because elevated plasma exposure would lead to serious adverse events, including: alfuzosin, amiodarone, astemizole, terfenadine, cisapride, ergot derivatives, lovastatin, simvastatin, atorvastatin, oral midazolam, triazolam, quetiapine, quinidine, salmeterol, sildenafil when used for pulmonary arterial hypertension. Also contraindicated are enzyme inducers that might compromise virological efficacy, e.g. carbamazepine, phenytoin, phenobarbital, rifampicin, St John’s wort, enzalutamide, and enzyme inhibitors that might increase paritaprevir exposure, e.g. azole antifungals, some macrolide antibiotics.

In addition to the contraindications, there are other drugs where caution needs to be exercised and there may be a requirement for a dosage adjustment, altered timing of administration or additional monitoring. Drug interactions need to be carefully considered in the setting of coinfection with HIV. Atazanavir and darunavir should be taken without ritonavir and other protease inhibitors are contraindicated. Efavirenz, etravirine and nevirapine are contraindicated, and rilpivirine should be used cautiously with repeat ECG monitoring. The exposure of raltegravir and dolutegravir may be increased, but this is not linked to safety issues. Cobicistat-containing regimens should not be used because of the additional boosting effect.

Grazoprevir and elbasvir are available in a two-drug fixed-dose combination containing 100 mg of grazoprevir and 50 mg of elbasvir in a single tablet. The recommended dose of the combination is one tablet taken orally once daily with or without food.

Grazoprevir and elbasvir are partially metabolised by CYP3A4, but no circulating metabolites are detected in plasma. The principal route of elimination is biliary and faecal with <1% recovered in urine. Grazoprevir is transported by P-gp and OATP1B1, while elbasvir is a substrate for P-gp. Both elbasvir (>99.9%) and grazoprevir (98.8%) are extensively bound to plasma proteins. The terminal half-life values are approximately 24 and 31 h, respectively.

Pharmacokinetic data from hepatic impairment studies in non-HCV-infected individuals have demonstrated a decrease in elbasvir AUC in Child-Pugh A (40%), Child-Pugh B (28%) and Child-Pugh C (12%) cirrhosis. In contrast, grazoprevir exposure is increased in Child-Pugh A (70%), Child-Pugh B (5-fold) and Child-Pugh C (12-fold) cirrhosis. Based on these data, there is a contraindication for elbasvir/grazoprevir in patients with moderate (Child-Pugh B) or severe (Child-Pugh C) hepatic impairment.

No dose adjustment is required in patients with mild, moderate or severe renal impairment (including patients on haemodialysis or peritoneal dialysis). There is an increase in elbasvir (65%) and grazoprevir (86%) exposure in non-HCV infected individuals with an eGFR <30 ml/min/1.73 m², but this is not considered to be clinically significant.

The safety of elbasvir/grazoprevir is based on phase II and III clinical studies with the most commonly reported adverse reactions being fatigue and headache. Rare cases (0.8%) of substantial ALT level elevations were reported, slightly more frequently in female, Asian and elderly patients. Less than 1% of subjects treated with elbasvir/grazoprevir with or without ribavirin discontinued treatment because of adverse events.

Since elbasvir and grazoprevir are substrates of CYP3A and P-gp, inducers of these proteins such as efavirenz, etravirine, phenytoin, carbamazepine, bosentan, modafinil and St John’s wort may cause a marked decrease in plasma exposure of both DAAs and are therefore contraindicated. Strong inhibitors of CYP3A (e.g. boosted protease inhibitors, some azole antifungals), which may markedly increase plasma concentrations, are either contraindicated or not recommended. In addition to inhibition of CYP3A, grazoprevir plasma concentrations may also be markedly increased by inhibitors of OATP1B1 (including boosted protease inhibitors, cobicistat, cyclosporin, single-dose rifampicin). However, there is no effect of acid reducing agents on the absorption of either DAA.

The potential for grazoprevir/elbasvir to affect other medications is relatively low, although grazoprevir is a weak CYP3A inhibitor (approximately 30% increase in midazolam exposure) and elbasvir a weak inhibitor of P-gp. There needs to be some caution when co-administering drugs that use CYP3A and P-gp in their disposition, especially in the presence of a narrow therapeutic index (e.g. tacrolimus, some statins, dabigatran, ticagrelor), or drugs with large ranges such a quetiapine, where those on higher doses may need additional monitoring, dose reduction and/or ECG.

Based on the findings above, there are limitations on which antiretrovirals can be co-administered with elbasvir/grazoprevir. Currently the antiretrovirals that can be used are the nucleotide reverse transcriptase inhibitors abacavir, lamivudine, tenofovir (either as TDF or as TAF), emtricitabine, rilpivirine, raltegravir, dolutegravir and maraviroc (Table 4A).

Glecaprevir and pibrentasvir are available in a two-drug fixed-dose combination containing 100 mg of glecaprevir and 40 mg of pibrentasvir. The recommended dose is three tablets taken orally once daily with food, as glecaprevir plasma exposure increases 83%-163% in the presence of food compared to the fasted state.

Biliary excretion is the major route of elimination for glecaprevir and pibrentasvir. The half-lives of glecaprevir and pibrentasvir are approximately 6 and 23 h, respectively.

Population pharmacokinetic analysis in HCV-infected subjects showed that following administration of glecaprevir/pibrentasvir in HCV-infected individuals with compensated (Child-Pugh A) cirrhosis, exposure of glecaprevir was approximately 2-fold higher whilst pibrentasvir exposure was similar to patients without cirrhosis. When compared to patients with normal hepatic function, glecaprevir AUC was 33% higher in patients with compensated cirrhosis (Child-Pugh A), 100% higher in those with moderate hepatic impairment (Child-Pugh B), and increased to 11-fold in those with severe hepatic impairment (Child-Pugh C). Thus, glecaprevir/pibrentasvir is contraindicated in patients with Child-Pugh B or C.

Glecaprevir/pibrentasvir was studied in HCV-negative individuals with mild, moderate, severe, or end-stage renal impairment not on dialysis and compared to subjects with normal renal function. The AUCs were increased by less than 56% in all patients, which was not clinically significant. Glecaprevir/pibrentasvir AUC was also similar with and without dialysis.

The safety of pibrentasvir and glecaprevir was evaluated in phase II and III clinical trials. Headache and fatigue were the most commonly reported adverse events.

Glecaprevir and pibrentasvir are inhibitors of P-gp, BCRP and OATP1B1 and OATP1B3. Co-administration with glecaprevir/pibrentasvir may increase the concentration of co-medications that are substrates of P-gp (e.g. dabigatran etexilate which is contraindicated because of a 2.4-fold increase in dabigatran exposure), BCRP (e.g. rosuvastatin which requires a dose reduction), or OATP1B1/3 (e.g. atorvastatin or simvastatin which are contraindicated). For other P-gp, BCRP, or OATP1B1/3 substrates, dose adjustment should be considered.

Glecaprevir/pibrentasvir concentrations may be decreased by strong P-gp and CYP3A inducing drugs such as rifampicin, carbamazepine, St. John's wort or phenytoin, leading to reduced therapeutic effect or loss of virologic response. Co-administration with these, or other potent inducers, is contraindicated. A similar effect cannot be ruled out with moderate inducers, such as oxcarbazepine and eslicarbazepine, and co-administration of these drugs is not recommended. Co-medications that inhibit P-gp and BCRP may increase plasma exposure of glecaprevir/pibrentasvir. Similarly OATP1B1/3 inhibitors, such as cyclosporin, darunavir and lopinavir, may also increase glecaprevir concentrations.

The potential for glecaprevir/pibrentasvir to affect other medications is relatively low, although glecaprevir is a weak CYP3A inhibitor (approximately 27% increase in midazolam exposure). There needs to be some caution when co-administering drugs that use CYP3A in their disposition in the presence of a narrow therapeutic index (e.g. tacrolimus) or drugs with large ranges such a quetiapine, whereas patients on higher doses may need additional monitoring, dose reduction and/or ECG.

For women of childbearing age, concomitant use with ethinylestradiol-containing contraception is contraindicated because of the risk of ALT elevations. Progestogen-containing contraception is allowed.

Similar to other DAAs, the solubility of glecaprevir decreases as pH increases. C_max of glecaprevir decreases on average by 64% when co-administered with omeprazole 40 mg. The license states that no dose changes are recommended. However, prescribing doses of omeprazole greater than 40 mg or equivalent (Table 5) with glecaprevir and pibrentasvir has not been studied and may lead to a greater decrease in glecaprevir concentrations.

In HIV-HCV coinfected patients, because of the mechanisms described above, glecaprevir/pibrentasvir is contraindicated with atazanavir-containing regimens and is not recommended with other HIV protease inhibitors. Similarly, the inducing non-nucleoside reverse transcriptase inhibitors efavirenz, etravirine and nevirapine are not recommended because of an expected reduction in plasma exposure of glecaprevir/pibrentasvir. All other antiretroviral drugs can be co-administered, including cobicistat when used with integrase inhibitor elvitegravir.

Four treatment options are available in 2018 for patients infected with HCV genotype 1a (Table 6, Table 7, Table 8). These options are considered equivalent, and their order of presentation does not indicate any superiority or preference, unless specified:

• Sofosbuvir/velpatasvir.
• Glecaprevir/pibrentasvir.
• Sofosbuvir/ledipasvir.
• Grazoprevir/elbasvir.

Comments: This recommendation is based on the results of the phase III ASTRAL-1 trial in patients with HCV genotype 1 infection (22% with cirrhosis, 66% treatment-naïve, 34% treatment-experienced, 44% of whom exposed to previous DAA) treated with the fixed-dose combination of sofosbuvir and velpatasvir for 12 weeks. An SVR12 was observed in 98% (206/210; one relapse) of patients infected with genotype 1a. These results were confirmed in real-world studies.^, 

In the ASTRAL-5 trial in treatment-naïve or treatment-experienced patients with or without cirrhosis infected with genotype 1a and coinfected with HIV, the SVR12 rate with the same regimen was 95% (63/66; 2 relapses).

The triple combination of sofosbuvir, velpatasvir and voxilaprevir administered for 8 weeks failed to achieve non-inferiority compared to sofosbuvir/velpatasvir for 12 weeks in the POLARIS-2 phase III trial, which included approximately 20% of patients with cirrhosis and 25% of treatment-experienced patients. The SVR12 rates in patients infected with genotype 1a were 92% (155/169; 14 relapses) after 8 weeks of sofosbuvir/velpatasvir/voxilaprevir and 99% (170/172; one relapse) after 12 weeks of sofosbuvir/velpatasvir. Thus, the triple combination of sofosbuvir, velpatasvir and voxilaprevir for 8 weeks is not recommended in patients infected with HCV genotype 1a.

Comments: This recommendation is based on the results of two phase III trials in patients with HCV genotype 1a infection. In ENDURANCE-1, the SVR12 rate was 98% (150/152; one virological breakthrough, one non-virological failure) in treatment-naïve or treatment-experienced patients without cirrhosis receiving 8 weeks of glecaprevir/pibrentasvir, including 13 patients who were HIV-coinfected. Treatment-naïve and treatment-experienced genotype 1a-infected patients with compensated cirrhosis were studied in the EXPEDITION-1 trial. The SVR12 rate was 98% (47/48; one relapse) after 12 weeks of glecaprevir/pibrentasvir.

Comments: This recommendation is based on the results of the three phase III trials ION-1, ION-3 and ION-4,^{,  ,  ,}  on post hoc analyses of pooled data from phase II and III clinical trials and on real-world data reported at international medical conferences or published.

In ION-1, treatment-naïve genotype 1a patients, including approximately 15% with compensated cirrhosis, achieved SVR12 in 98% (141/144; one relapse) of cases after 12 weeks of the fixed-dose combination of sofosbuvir and ledipasvir. An integrated analysis of treatment-naïve genotype 1a patients with compensated cirrhosis treated with sofosbuvir/ledipasvir for 12 weeks in different phase II and III studies showed an overall SVR12 rate of 98% (84/86).

In ION-4, an open-label study in treatment-naïve or treatment-experienced genotype 1a patients with or without cirrhosis who were coinfected with HIV and received an antiretroviral regimen of tenofovir and emtricitabine with efavirenz, rilpivirine or raltegravir, the SVR12 rate was 96% (240/250; 8 relapses).

In ION-3 in treatment-naïve genotype 1a patients without cirrhosis, the SVR12 rates were 93% (159/171; 10 relapses) with sofosbuvir/ledipasvir for 8 weeks and 95% (163/172; 2 relapses) with sofosbuvir/ledipasvir for 12 weeks. These results were confirmed by real-world studies from Europe and the United States in the same subgroup of patients, showing similarly high SVR12 rates. One study showed that shortening sofosbuvir and ledipasvir treatment duration can be applied to patients with an HCV RNA <6,000,000 IU/ml (6.8 Log₁₀ IU/ml) at baseline. A pooled analysis of patients from different real-world studies included 566 treatment-naïve genotype 1a-infected patients without cirrhosis; 527 of them were eligible to receive 8 weeks of sofosbuvir-ledipasvir, as per FDA labelling. The SVR12 rate was 98% (518/527; 9 relapses). Logistic regression analysis identified male sex, African-American origin and a fibrosis stage F3 as independent predictors of post-treatment relapse. The effect of F3 fibrosis was not confirmed in later studies.^, 

SVR12 rates of the same order as in the clinical trials were observed in patients with or without compensated cirrhosis in real-world studies from various continents.

The combination of sofosbuvir and ledipasvir is not recommended in treatment-experienced patients infected with genotype 1a, because this regimen would require the addition of ribavirin, as explained in the EASL Recommendations for Treatment of Hepatitis C 2016.

Comments: This recommendation is based on the results of three phase III trials and subsequent post hoc analyses of pooled phase II and III clinical trial data.

In the C-EDGE-TN trial, in treatment-naïve patients infected with genotype 1a receiving grazoprevir and elbasvir for 12 weeks, the SVR12 rate was 92% (144/157; one breakthrough and 12 relapses), with compensated cirrhosis having no effect on SVR12 rate. In the open-label C-EDGE-COINFECTION trial, treatment-naïve patients coinfected with HIV with or without compensated cirrhosis were treated with grazoprevir and elbasvir for 12 weeks, with an SVR12 rate of 97% (139/144) in genotype 1a-infected patients. In a pooled efficacy analysis of treatment-naïve patients with genotype 1a infection from phase II and III trials treated with grazoprevir/elbasvir for 12 weeks, the SVR12 rate was 99% (121/122) in patients with an HCV RNA level ≤800,000 IU/ml, with no influence of pre-existing NS5A RASs at baseline on SVR (unpublished data provided to the panel by Merck).

In treatment-experienced patients included in the C-EDGE-TE phase III trial, including approximately 30% of patients with compensated cirrhosis, the SVR12 rate in genotype 1a patients was 92% (55/60) after 12 weeks of grazoprevir/elbasvir. In a pooled efficacy population of treatment-experienced patients with genotype 1a from phase II and III trials treated for 12 weeks, the SVR12 rate was 100% (14/14) in patients with an HCV RNA level ≤800,000 IU/ml (unpublished data provided to the panel by Merck).

With this regimen, the SVR12 rate was impacted by the presence of NS5A RASs at baseline in treatment-naïve and treatment-experienced patients with an HCV RNA level >800,000 IU/ml (unpublished data provided to the panel by Merck). Therefore, because resistance testing is not recommended prior to therapy, this regimen is not recommended in patients with an HCV RNA level >800,000 IU/ml.

Five treatment options are available in 2018 for patients infected with HCV genotype 1b (Table 6, Table 7, Table 8). These options are considered equivalent, and their order of presentation does not indicate any superiority or preference, unless specified:

• Sofosbuvir/velpatasvir.
• Glecaprevir/pibrentasvir.
• Sofosbuvir/ledipasvir.
• Grazoprevir/elbasvir.
• Ombitasvir/paritaprevir/ritonavir and dasabuvir.

Comments: This recommendation is based on the results of the phase III ASTRAL-1 trial in patients with HCV genotype 1 infection (22% with cirrhosis, 66% treatment-naïve, 34% treatment-experienced, 44% of whom exposed to previous DAA) treated with the fixed-dose combination of sofosbuvir and velpatasvir for 12 weeks. An SVR12 was observed in 99% (117/118; one relapse) of patients infected with genotype 1b. In the ASTRAL-5 trial in HIV-coinfected patients, the SVR12 rate with the same regimen was 92% (11/12; no virological failure) in treatment-naïve or treatment-experienced patients without cirrhosis or with compensated cirrhosis infected with genotype 1b. These results were confirmed in real-world studies.^, 

Comments: This recommendation is based on the results of two phase III trials in patients with HCV genotype 1b infection. In ENDURANCE-1, the SVR12 rate was 100% (198/198) in treatment-naïve and treatment-experienced patients without cirrhosis receiving 8 weeks of glecaprevir/pibrentasvir, including two patients who were HIV-coinfected. Treatment-naïve and treatment-experienced genotype 1b-infected patients with compensated cirrhosis were studied in the EXPEDITION-1 trial. The SVR12 rate was 100% (39/39) after 12 weeks of glecaprevir/pibrentasvir.

Comments: This recommendation is based on the results of the four phase III trials ION-1, ION-2, ION-3 and ION-4^{,  ,  ,}  and several post hoc analyses of pooled data from phase II and III clinical trials.

In ION-1, treatment-naïve patients infected with HCV genotype 1b, including approximately 15% with compensated cirrhosis, achieved SVR12 in 100% (66/66) of cases after 12 weeks of the fixed-dose combination of sofosbuvir and ledipasvir. An integrated analysis of genotype 1b patients with compensated cirrhosis treated with sofosbuvir/ledipasvir for 12 weeks in different phase II and III studies showed an overall SVR12 rate of 97% (72/74) in treatment-naïve and 96% (124/129) in treatment-experienced patients.

In ION-2, in treatment-experienced patients (previously treated with pegylated IFN-α and ribavirin, or with pegylated IFN-α, ribavirin and either telaprevir or boceprevir), including approximately 20% with cirrhosis, the SVR12 rate was 87% (20/23; 3 relapses) in patients infected with HCV genotype 1b.

In ION-4, an open-label study in treatment-naïve and treatment-experienced genotype 1b patients with or without cirrhosis who were coinfected with HIV and received an antiretroviral regimen of tenofovir and emtricitabine with efavirenz, rilpivirine or raltegravir, the SVR12 rate was 96% (74/77; 3 relapses).

In ION-3 in treatment-naïve patients without cirrhosis (F3 fibrosis was present in only 13% of patients with genotype 1 who underwent liver biopsy), the SVR12 rate was 98% (42/43; one relapse) after 8 weeks of sofosbuvir/ledipasvir in patients infected with genotype 1b. These results were confirmed by real-world studies from Europe and the United States in the same subgroup of patients, showing similarly high SVR12 rates. In a pooled analysis of patients from different real-world studies, the SVR12 rate after 8 weeks of sofosbuvir/ledipasvir as per FDA labelling was over 99% (235/237; 2 relapses) in genotype 1b patients.

Similar SVR12 rates as those achieved in the clinical trials were observed in treatment-naïve and treatment-experienced patients with or without compensated cirrhosis in real-world studies from various continents.

Comments: This recommendation is based on the results of four phase III trials, and subsequent post hoc analyses of pooled phase II and III clinical trial data, as well as of the STREAGER trial with a shorter treatment duration.

In the C-EDGE-TN trial, in treatment-naïve patients infected with genotype 1b receiving grazoprevir and elbasvir for 12 weeks, the SVR12 rate was 99% (129/131; one relapse). In the C-CORAL trial, performed in Russia and the Asia-Pacific region, the SVR12 rate was 98% (382/389; 5 relapses). In the open-label C-EDGE-COINFECTION trial, treatment-naïve patients coinfected with HIV with or without compensated cirrhosis were treated with grazoprevir and elbasvir for 12 weeks. The SVR12 rate was 95% (42/44) in genotype 1b-infected patients.

In treatment-experienced patients included in the C-EDGE-TE phase III trial, in which approximately a third of patients had compensated cirrhosis, the SVR12 rate in genotype 1b patients was 100% (34/34) after 12 weeks of grazoprevir/elbasvir.

A pooled analysis of all phase II and III trials showed an SVR rate of 97% (1040/1070; 15 relapses and 15 virological failures) in patients infected with genotype 1b treated for 12 weeks with this regimen.

In the STREAGER study, treatment-naïve genotype 1b-infected patients with a stage of fibrosis F0-F2 (excluding patients with advanced fibrosis or cirrhosis) treated with grazoprevir/elbasvir for 8 weeks achieved an SVR12 in 97% (66/68) of cases. Two patients relapsed post-treatment (updated data provided to the panel by Merck).

Comments: This recommendation is based on the results of several phase III trials. In the PEARL-3 trial, the SVR12 rate was 99% (207/209) in treatment-naïve patients without cirrhosis infected with subtype 1b receiving the triple combination of ritonavir-boosted paritaprevir, ombitasvir and dasabuvir for 12 weeks. In MALACHITE-1, the SVR12 rate in treatment-naïve patients without cirrhosis was 98% (81/83). In the TURQUOISE-1 study in treatment-naïve patients without cirrhosis coinfected with HIV-1 and stable on antiretroviral treatment containing atazanavir or raltegravir, SVR12 was achieved in 100% (7/7) of genotype 1b patients. Finally, in the GARNET study, the SVR12 rate was 97% (161/166) in treatment-naïve patients with genotype 1b infection and no cirrhosis (METAVIR score F0 to F3) after 8 weeks of treatment with ritonavir-boosted paritaprevir, ombitasvir and dasabuvir. Among the 15 patients with F3 fibrosis included in this study, two experienced a virological failure.

In treatment-experienced patients (pegylated IFN-α and ribavirin failures) without cirrhosis treated with this combination for 12 weeks in PEARL-2, SVR12 was achieved in 100% (95/95) of cases. In the TOPAZ-1 study, treatment-naïve and treatment-experienced patients without cirrhosis receiving the same regimen achieved SVR12 in 99% (738/745; 3 virological failures) of cases. A pooled analysis of several clinical trials showed a 99% SVR12 rate in 521 patients without cirrhosis (PEARL-2, PEARL-3, TOPAZ-2, MALACHITE-1). Similarly high SVR12 rates were achieved in Asian patients infected with genotype 1b with this combination.

In treatment-naïve and treatment-experienced patients with compensated cirrhosis included in the TURQUOISE-3 trial, SVR12 was achieved in 100% (60/60) of genotype 1b patients treated for 12 weeks with ritonavir-boosted paritaprevir, ombitasvir and dasabuvir.

Similar SVR12 rates as those achieved in the clinical trials were observed in a large number of real-world studies from various continents.

Two first-line treatment options are available for patients infected with HCV genotype 2 (Table 6, Table 7, Table 8). These options are considered equivalent, and their order of presentation does not indicate any superiority or preference, unless specified:

• Sofosbuvir/velpatasvir.
• Glecaprevir/pibrentasvir.

Comments: This recommendation is based on the results of the phase III ASTRAL-2 trial in patients with HCV genotype 2 infection (14% with compensated cirrhosis, 86% treatment-naïve, 14% treatment-experienced) treated with the fixed-dose combination of sofosbuvir and velpatasvir for 12 weeks without ribavirin, showing SVR12 in 99% (133/134) of patients. In ASTRAL-1, the SVR12 rate was 100% (104/104) in treatment-naïve (two-thirds) and treatment-experienced (one-third) patients, who included approximately 30% with cirrhosis. In the ASTRAL-5 trial in HIV-coinfected patients, the SVR12 rate with the same regimen was 100% (11/11) in genotype 2 patients.

Comments: This recommendation is in part based on the results of the phase II SURVEYOR-2 trial, showing an SVR12 rate of 98% (53/54; no virological failure) in treatment-naïve and treatment-experienced patients without cirrhosis infected with HCV genotype 2, receiving the fixed-dose combination of glecaprevir and pibrentasvir for 8 weeks. These results were confirmed in the CERTAIN-2 trial, showing an SVR rate of 98% (127/129, no virological failure) in Japanese patients infected with genotype 2, receiving the same treatment regimen for 8 weeks. In the EXPEDITION 2 trial, the SVR12 rate was 100% (12/12) after 8 weeks of glecaprevir/pibrentasvir in patients without cirrhosis with genotype 2 infection coinfected with HIV.

In the EXPEDITION-1 trial, 12 weeks of glecaprevir/pibrentasvir yielded SVR12 in 100% (31/31) of treatment-naïve or treatment-experienced genotype 2-infected patients with compensated cirrhosis. These results were confirmed in the CERTAIN-2 trial, showing an SVR rate of 100% (38/38) in Japanese patients with compensated cirrhosis infected with genotype 2 receiving the same treatment regimen for 12 weeks.

Three first-line treatment options are available for patients infected with HCV genotype 3 (Table 6, Table 7, Table 8). These options are considered equivalent, and their order of presentation does not indicate any superiority or preference, unless specified:

• Sofosbuvir/velpatasvir.
• Glecaprevir/pibrentasvir.
• Sofosbuvir/velpatasvir/voxilaprevir.

Comments: This recommendation is based on the results of the phase III ASTRAL-3 trial in patients with HCV genotype 3 infection (29% with compensated cirrhosis, 74% treatment-naïve, 26% treatment-experienced) treated with the fixed-dose combination of sofosbuvir and velpatasvir for 12 weeks. The SVR12 rates were 98% (160/163) in treatment-naïve patients without cirrhosis. Lower SVR12 rates were observed in patients who were treatment-experienced or had cirrhosis with this regimen: overall 90% (104/116; 12 virological failures); 93% (40/43) in treatment-naïve patients with compensated cirrhosis, 91% (31/34) in treatment-experienced patients without cirrhosis and 89% (33/37) in treatment-experienced patients with compensated cirrhosis. Thus, the addition of a third drug to this regimen is necessary, at least in patients infected with genotype 3 with compensated cirrhosis, justifying the use of the triple combination of sofosbuvir, velpatasvir and voxilaprevir in this group (see below).

In the ASTRAL-5 trial in HIV-coinfected patients, the SVR12 rate with the same regimen was 92% (11/12).

Comments: This recommendation is based on the results of the phase III ENDURANCE-3 trial showing an SVR12 rate of 95% (149/157; 5 relapses, one virological breakthrough) in treatment-naïve patients without cirrhosis infected with HCV genotype 3 receiving the fixed-dose combination of glecaprevir and pibrentasvir for 8 weeks. However, only 17% of patients in this study had advanced fibrosis (METAVIR score F3), the remaining 83% having mild to moderate fibrosis (F0-F2). Thus, more data must be generated to strengthen the recommendation of 8 weeks of glecaprevir and pibrentasvir as the ideal treatment duration in treatment-naïve patients with advanced (F3) fibrosis.In the EXPEDITION-2 trial, the SVR12 rate was 100% (22/22) after 8 weeks of glecaprevir/pibrentasvir in patients with genotype 3 infection coinfected with HIV without cirrhosis. An integrated analysis of phase II and III trials in patients infected with genotype 3 showed an SVR12 rate of 95% (198/208; 6 virological failures) after 8 weeks of glecaprevir/pibrentasvir in treatment-naïve patients infected with genotype 3 without cirrhosis.

In the same integrated analysis of phase II and III trials, the SVR12 rate after 12 weeks of glecaprevir/pibrentasvir in treatment-naïve patients with cirrhosis infected with genotype 3 was 97% (67/69; one virological breakthrough). In the SURVEYOR-2 study, the SVR12 rates were 91% (20/22; 2 relapses) and 95% (21/22; 1 relapse) in treatment-experienced patients without cirrhosis treated for 12 or 16 weeks, respectively; they were 98% (39/40; no virological failure) in treatment-naïve patients with cirrhosis treated for 12 weeks and 96% (45/47; 2 virological failures) in treatment-experienced patients with cirrhosis treated for 16 weeks. A pooled analysis of phase II and III clinical trials in patients infected with genotype 3 showed SVR12 rates of 96% (258/270) in treatment-naïve patients without cirrhosis treated for 12 weeks, 90% (44/49) in treatment-experienced patients without cirrhosis treated for 12 weeks, 96% (21/22) in treatment-experienced patients without cirrhosis treated for 16 weeks, 99% (64/65) in treatment-naïve patients with compensated cirrhosis treated for 12 weeks, and 94% (48/51) in treatment-experienced patients with compensated cirrhosis treated for 16 weeks. Data with 12 weeks of treatment with glecaprevir and pibrentasvir in treatment-experienced patients with cirrhosis are needed.

Comments: This recommendation is based on the results of the POLARIS-2 and -3 phase III trials. In POLARIS-2, which included approximately three-quarters of treatment-naïve and one-quarter of treatment-experienced patients and approximately 20% of individuals with cirrhosis, the SVR12 rate was 99% (91/92; no virological failure) after 8 weeks of the triple combination of sofosbuvir, velpatasvir and voxilaprevir. In POLARIS-3, 8 weeks of the triple combination yielded a 96% SVR12 rate (106/110; 2 relapses) in treatment-naïve and treatment-experienced patients with compensated cirrhosis. Because genotype 3 is more difficult-to-cure than other genotypes, and in the absence of data with 12 weeks of therapy, it appears to be safer to treat patients with genotype 3 infection who have cirrhosis for 12 weeks with this combination.

Four treatment options are available in 2018 for patients infected with HCV genotype 4 (Table 6, Table 7, Table 8). These options are considered equivalent, and their order of presentation does not indicate any superiority or preference, unless specified:

• Sofosbuvir/velpatasvir.
• Glecaprevir/pibrentasvir.
• Sofosbuvir/ledipasvir.
• Grazoprevir/elbasvir.

Comments: This recommendation is based on the results of the phase III ASTRAL-1 trial in patients with HCV genotype 4 infection (23% with cirrhosis, 55% treatment-naïve, 45% treatment-experienced) treated with the fixed-dose combination of sofosbuvir and velpatasvir for 12 weeks without ribavirin, showing SVR12 in 100% (116/116) of patients. In the ASTRAL-5 trial in HIV-coinfected patients receiving the same treatment regimen, the SVR12 rate was 100% (4/4).

Comments: This recommendation is partly based on the results of the phase II SURVEYOR-2 trial showing an SVR12 rate of 93% (43/46; no virological failure) in treatment-naïve and treatment-experienced patients without cirrhosis infected with HCV genotype 4 receiving the fixed-dose combination of glecaprevir and pibrentasvir for 8 weeks. In ENDURANCE-4, genotype 4 patients without cirrhosis treated for 12 weeks achieved SVR in 99% (75/76; no virological failures) of cases, whereas in EXPEDITION-1, 100% (16/16) of patients with cirrhosis infected with genotype 4 achieved SVR12.

In the EXPEDITION 2 trial, the SVR12 rate was 100% (16/16) after 8 weeks of glecaprevir/pibrentasvir in patients with genotype 4 infection coinfected with HIV without cirrhosis.

Comments: The SYNERGY trial assessed the efficacy and safety of the combination of sofosbuvir and ledipasvir in patients with genotype 4 infection. After 12 weeks of therapy, 95% (20/21; no virological failure) of them achieved an SVR. In another phase II trial, patients were treated with the combination of sofosbuvir and ledipasvir for 12 weeks. The SVR12 rates were 96% (21/22) in treatment-naïve and 91% (20/22) in treatment-experienced individuals; the split was 91% (31/34) in patients without cirrhosis and 100% (10/10) in those with cirrhosis.

Comments: This recommendation is based on the results of three phase III trials including a small number of patients infected with genotype 4 and on the analogy with data in patients infected with genotype 1. In the C-EDGE-TN trial, the SVR12 rate was 100% (18/18) in treatment-naïve patients infected with genotype 4 receiving grazoprevir and elbasvir for 12 weeks (including 12% with cirrhosis). In the open-label C-EDGE-COINFECTION trial, treatment-naïve patients with HCV genotype 4 coinfected with HIV, with or without compensated cirrhosis, were treated with grazoprevir and elbasvir for 12 weeks. The SVR12 rate was 96% (27/28; one relapse). In the C-CORAL trial, 3/3 treatment-naïve patients infected with genotype 4 achieved an SVR12 after 12 weeks of grazoprevir/elbasvir. The SVR12 rate was 100% (11/11) in the C-EDGE CO-STAR trial in PWIDs on opioid substitution therapy receiving the same treatment regimen.

Three treatment options are available in 2018 for patients infected with HCV genotype 5 (Table 6, Table 7, Table 8). However, the number of patients infected with genotype 5 treated in all of the trials was limited, making it difficult to make strong recommendations once the data are broken down by cirrhosis and prior treatment. These options are considered equivalent, and their order of presentation does not indicate any superiority or preference, unless specified:

• Sofosbuvir/velpatasvir.
• Glecaprevir/pibrentasvir.
• Sofosbuvir/ledipasvir.

Comments: This recommendation is based on the results of the phase III ASTRAL-1 trial in patients with HCV genotype 5 (14% with cirrhosis, 69% treatment-naïve, 31% treatment-experienced) treated with the fixed-dose combination of sofosbuvir and velpatasvir for 12 weeks, showing SVR12 in 97% (34/35) of them.

Comments: This recommendation is based on the results of the phase II SURVEYOR-2 trial in which 2/2 patients without cirrhosis infected with HCV genotype 5 receiving the fixed-dose combination of glecaprevir and pibrentasvir for 8 weeks achieved an SVR12. In ENDURANCE-4, genotype 5 patients without cirrhosis treated for 12 weeks achieved SVR in 100% (26/26) of cases, whereas in EXPEDITION-1, 2/2 patients infected with genotype 5 with cirrhosis achieved SVR12.

Comments: In a phase II trial, 41 treatment-naïve and treatment-experienced patients infected with HCV genotype 5, including 9 with compensated cirrhosis, were treated with sofosbuvir and ledipasvir without ribavirin for 12 weeks: 95% (39/41) achieved SVR12.

Three treatment options are available in 2018 for patients infected with HCV genotype 6 (Table 6, Table 7, Table 8). However, the number of patients infected with genotype 6 treated in all of the trials was limited, making it difficult to make strong recommendations once the data are broken down by cirrhosis and prior treatment. These options are considered equivalent, and their order of presentation does not indicate any superiority or preference, unless specified:

• Sofosbuvir/velpatasvir.
• Glecaprevir/pibrentasvir.
• Sofosbuvir/ledipasvir.

Comments: This recommendation is based on the results of the phase III ASTRAL-1 trial in patients with HCV genotype 6 (15% with cirrhosis, 93% treatment-naïve, 17% treatment-experienced) treated with the fixed-dose combination of sofosbuvir and velpatasvir for 12 weeks without ribavirin, of whom 100% (41/41) achieved SVR12. These results were confirmed by a 97% (35/36; one relapse) SVR rate in a phase III trial in patients infected with genotype 6 from Singapore, Malaysia, Thailand and Vietnam.

Comments: This recommendation is partly based on the results of the phase II SURVEYOR-2 trial, showing an SVR12 rate of 90% (9/10; no virological failure) in treatment-naïve and treatment-experienced patients infected with HCV genotype 6 without cirrhosis, who received the fixed-dose combination of glecaprevir and pibrentasvir for 8 weeks. In ENDURANCE-4, genotype 6 patients without cirrhosis treated for 12 weeks achieved SVR in 100% (19/19) of cases, whereas in EXPEDITION-1, 100% (7/7) of patients infected with genotype 6 with cirrhosis achieved SVR12.