Summary Basis of Decision for Vosevi

Clinical Pharmacology

The clinical pharmacology included reports on the human pharmacodynamic and pharmacokinetic (PK) studies. The clinical pharmacological data support the use of Vosevi for the specified indication.

A total of 15 Phase I studies characterized the PK of Vosevi (sofosbuvir/velpatasvir/voxilaprevir [SOF/VEL/VOX]). The PK of SOF and VEL were previously characterized as part of the new drug submission for the drug product Epclusa; therefore, the current review was focused on the PK of VOX and Vosevi.

The systemic exposure of VOX values were 299% and 500% higher in subjects with moderate and severe hepatic impairment, respectively, as compared to subjects with normal hepatic function; therefore, Vosevi is not recommended in patients with moderate and severe hepatic impairment. Population PK analysis in HCV-infected subjects indicated that subjects with compensated cirrhosis had a 73% higher VOX exposure than those without cirrhosis. This increase is not considered clinically relevant and no dose adjustment of VOX is required in these patients.

The systemic exposure and maximum plasma concentration of VOX were approximately 71% and 45% higher, respectively, in subjects with severe renal impairment as compared to subjects with normal renal function. Based on these results, dose adjustments in patients with severe renal impairment are not warranted.

The three medicinal ingredients (SOF, VEL, and VOX) are substrates of drug transporters P-glycoprotein and breast cancer resistance protein (BCRP); VOX, and to a lesser extent VEL, are also substrates of organic anion transporting polypeptide 1B1 (OATP1B1) and OATP1B3. In vitro, slow metabolic turnover was observed for VEL, primarily by cytochrome P450 (CYP) enzymes CYP2B6, CYP2C8, and CYP3A4, and for VOX, primarily by CYP3A4. Based on these properties, Vosevi is contraindicated in the presence of potent inducers of P-glycoprotein and/or moderate to potent inducers of CYP2B6, CYP2C8, or CYP3A4, such as rifampin, St. John's Wort, phenytoin and phenobarbital, all of which would decrease the exposure of Vosevi, resulting in loss of therapeutic effect. Co-administration of Vosevi with rosuvastatin is contraindicated due to the inhibition of BCRP and OATP1B by Vosevi, resulting in an increased risk of statin-related myopathy, including rhabdomyolysis. Co-administration of Vosevi with dabigatran etexilate is contraindicated due to P-gp inhibition by Vosevi, resulting in an increased risk of dabigatran etexilate-related risk of bleeding. A Grade 3 alanine aminotransferase (ALT) elevation was reported in one, healthy subject receiving Vosevi and ethinyl estradiol, suggesting the coadministration of these drugs may increase the risk of ALT elevation and monitoring for ALT levels may be warranted. Supratherapeutic doses of VOX (900 mg) did not have an effect on the QTc interval.

For further details, please refer to the Vosevi Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

The clinical efficacy of Vosevi (sofosbuvir/velpatasvir/voxilaprevir) was evaluated in two Phase III studies (POLARIS-1 and POLARIS-4) with patients who had genotype 1 to 6 hepatitis C virus (HCV) infection without cirrhosis or with compensated cirrhosis and who had previously been treated with direct-acting antivirals (DAAs).

Sustained virologic response defined as HCV ribonucleic acid (RNA) less than the lower limit of quantification (LLOQ) at 12 weeks after the cessation of treatment (SVR12), was the primary endpoint to determine the HCV cure rate. Serum HCV RNA values were measured during the clinical studies using the COBAS AmpliPrep/COBAS Taqman HCV test (version 2.0) with a LLOQ of 15 IU per mL.

The POLARIS-1 study was a Phase III, randomized, double-blind, placebo-controlled, multicentre study that compared 12 weeks of Vosevi treatment to 12 weeks of placebo treatment in 415 DAA-experienced patients with chronic HCV infection who had previously been treated with an NS5A inhibitor. The efficacy results of the study showed that the Vosevi group met the primary efficacy endpoint of an SVR12 rate that was statistically superior relative to the pre-specified performance goal of 85% (p<0.001). The SVR12 rate in the Vosevi group was 96.2% (253 of 263 patients; 95% confidence interval [CI]: 93.1% to 98.2%). A total of 10 of 263 patients (3.8%) in the Vosevi group did not achieve SVR12. One patient, with genotype 1a HCV infection, had on-treatment virologic failure (breakthrough) that was consistent with non-adherence. Three patients had relapse determined at post-treatment Week 4; all had genotype 3 HCV infection with cirrhosis. Three other patients (HCV genotypes 1a, 3, or 4, all with cirrhosis) achieved SVR at 4 weeks after the cessation of treatment, but had relapse at the post-treatment Week 12 visit. Three additional patients did not achieve SVR12 (2 withdrew consent and 1 was lost to follow-up).

In the Vosevi group, 78.8% of patients had baseline NS3 and/or NS5A resistance-associated variants (RAVs). The NS5A RAVs were the most common RAVs in patients across genotypes, observed in 75.4% of patients. The presence of baseline RAVs did not impact the SVR12 rate in the Vosevi group, with an SVR12 rate of 97.1% for patients with RAVs, compared with an SVR12 rate of 97.7% for patients without RAVs. Of the 7 patients with virologic failure, 1 patient developed treatment-emergent RAVs L31M and Y93H; this patient, with genotype 1a, experienced virologic breakthrough at the end of treatment and had PK data consistent with nonadherence. No NS3, NS5A and NS5B RAVs emerged in any of the other 5 patients who relapsed with data available.

The second study (POLARIS-4) was a Phase III, randomized, open-label, multicentre study that evaluated 12 weeks of Vosevi (sofosbuvir/velpatasvir/voxilaprevir) treatment and 12 weeks of treatment with sofosbuvir/velpatasvir (SOF/VEL, brand name Epclusa) in 333 DAA-experienced patients with chronic HCV infection who had previously been treated with an HCV regimen without an NS5A inhibitor. Patients whose only DAA exposure was a NS3/4A protease inhibitor were excluded. The SVR12 rate for the Vosevi group was statistically superior relative to the SVR12 performance goal of 85% at the significance level of 0.025 (p<0.001), while the SVR12 rate for the SOF/VEL group was not statistically superior relative to the SVR12 performance goal of 85% at the significance level of 0.025. The SVR12 rates were as follows:

• Vosevi group: 97.8% (95% CI: 94.5% to 99.4%) of patients (178 of 182) achieved SVR12.
• SOF/VEL group: 90.1% (95% CI: 84.1% to 94.3%) of patients (136 of 151) achieved SVR12.

In the Vosevi group, 4 of 182 patients (2.2%) did not achieve SVR12. Of these, 1 patient relapsed and 3 patients were categorized as "Other". Patients were categorized as "Other" because they did not have post-treatment Week 12 assessments due to death (1 patient) or missed the post-treatment Week 12 visit (2 patients). Only 1 patient in the Vosevi group experienced virologic failure, precluding any meaningful subgroup analysis. In the SOF/VEL group, 15 of 151 patients (9.9%) did not achieve SVR12. Of these, 1 patient had on-treatment virologic failure (breakthrough) at Week 8 and 14 patients relapsed. Of the 14 patients who relapsed following SOF/VEL treatment for 12 weeks, 8 patients had genotype 3 HCV infection, and 7 of these subjects also had cirrhosis. The remaining 6 patients who relapsed had genotype 1 HCV infection (3 patients with genotype 1a with cirrhosis, 2 patients with genotype 1a without cirrhosis, and 1 patient with genotype 1b without cirrhosis who completed only 56 days of study treatment).

The SVR12 rates in the Vosevi group were >93.0% across all key subgroups, including those associated with traditional negative predictive factors. Within this treatment group, 46.2% of the patients had cirrhosis, and 97.6% (82 of 84 patients) achieved SVR12. The SVR12 rates were also high regardless of prior DAA experience. The majority of patients had prior DAA exposure to sofosbuvir (SOF) or to SOF and simeprevir (SMV), and the SVR12 rates for these patients were 97.0% (129 of 133 patients) and 95.8% (23 of 24 patients), respectively. Overall, the SVR12 rates in the SOF/VEL group (90.1%) were lower than the Vosevi group (97.8%).

Seven of the 15 patients who did not achieve SVR12 in the SOF/VEL group had genotype 3 HCV infection and cirrhosis; the SVR12 rate in this subgroup was 76.7% (23 of 30 patients). For patients in the Vosevi group, 96.8% of patients (30 of 31) with genotype 3 HCV infection and cirrhosis achieved SVR12. The SVR12 rates, overall and for most subgroups, were higher following 12 weeks of Vosevi treatment compared with 12 weeks of SOF/VEL treatment. In patients with genotype 1a HCV infection, the SVR12 rates were 98.1% and 88.6% in the Vosevi and SOF/VEL groups, respectively. In patients with genotype 1b HCV infection, the SVR12 rates were 95.8% and 95.5% in the Vosevi and SOF/VEL groups, respectively. In patients with genotype 2 HCV infection, the SVR12 rates were 100.0% and 97.0% in the Vosevi and SOF/VEL groups, respectively. In patients with genotype 3 HCV infection, the SVR12 rates were 96.3% and 84.6% in the Vosevi and SOF/VEL groups, respectively. Among cirrhotic patients, the SVR12 in the Vosevi group was also higher compared with the SOF/VEL group (97.6% vs. 85.5%). For patients with genotype 4 HCV infection, all 19 patients in the Vosevi group achieved SVR12.

There was no impact of baseline RAVs on SVR12 for patients in the Vosevi or SOF/VEL group. The single patient who relapsed in the Vosevi group did not have any treatment-emergent RAVs, consistent with the regimen having a high barrier to resistance. However, 10 of the 14 patients in the SOF/VEL group did have treatment-emergent RAVs, all of whom had NS5A variants at the Y93 position. These results demonstrate the contribution of VOX to the potent regimen of SOF/VEL for 12 weeks for retreatment of the DAA-experienced patients who have not previously received an NS5A inhibitor-containing regimen.

For more information, refer to the Vosevi Product Monograph, approved by Health Canada and available through the Drug Product Database.

Indication

The New Drug Submission for Vosevi was filed by the sponsor with the following indication:

• Vosevi (sofosbuvir/velpatasvir/voxilaprevir) is indicated for the treatment of chronic hepatitis C virus (HCV) infection in adult patients with any HCV genotype, without cirrhosis or with compensated cirrhosis, who have failed prior treatment with an HCV direct-acting antiviral (DAA).

Based on the review of submitted data the following indication was recommended:

• Vosevi (sofosbuvir/velpatasvir/voxilaprevir) is indicated for the treatment of chronic hepatitis C virus (HCV) infection in adult patients, without cirrhosis or with compensated cirrhosis, who have:

• genotype 1, 2, 3, 4, 5, or 6 infection and have previously been treated with an HCV regimen containing an NS5A inhibitor
• genotype 1, 2, 3 or 4 infection and have previously been treated with an HCV regimen containing sofosbuvir without an NS5A inhibitor

Clinical Safety

The clinical safety of Vosevi was primarily assessed in the two pivotal Phase III studies POLARIS-1 and POLARIS-4, described in the Clinical Efficacy section. The safety profile of Vosevi was established in patients infected with HCV, without cirrhosis or with compensated cirrhosis.

In the POLARIS-1 study, treatment with Vosevi for 12 weeks was generally well-tolerated. A higher percentage of patients in the Vosevi group experienced any adverse event (AE) (78.3%, 206 of 263) compared with the placebo group (70.4%, 107 of 152). The most common AEs (>10% of patients) were headache, fatigue, diarrhea, and nausea. Most AEs were Grade 1 or 2 in severity. Grade 3 or 4 AEs were reported for 1.9% of patients (5 of 263) in the Vosevi group and 2.6% of patients (4 of 152) in the placebo group. One patient (0.4%) in the Vosevi group and 1 patient (0.7%) in the placebo group had a Grade 4 AE. The patient in the Vosevi group had a Grade 4 AE of seizure and the patient in the placebo group had a Grade 4 AE of ventricular fibrillation. Both Grade 4 AEs were assessed as serious and not related to the study drug.

Serious adverse events (SAEs) were reported for 1.9% (5 of 263) of patients in the Vosevi group and 4.6% (7 of 152) of patients in the placebo group. No trends in SAEs were observed; no SAE was reported in >1 patient, and all SAEs were assessed as unrelated to study drug. In the Vosevi group, no patients had SAEs that led to discontinuation or interruption of study drug. In the placebo group, 1 patient had an SAE (schizophrenia) that led to discontinuation of the study drug, and 1 patient had an SAE (atrial fibrillation) that led to interruption of dosing. No patients died during the study. Four patients (Vosevi, 1 patient; placebo, 3 patients) permanently discontinued the study drug due to AEs. The patient in the Vosevi group had a Grade 3 AE of angioedema on Day 12 and the study drug was discontinued the same day. The event was attributed to a newly initiated concomitant medication ramipril, and not related to the study drug, and was considered resolved on post-treatment Day 4.

The incidence of Grade 3 and 4 hematologic laboratory abnormalities was similar for both treatment groups; there were no clinically meaningful hematologic abnormalities. Patients in the Vosevi group had Grade 3 or 4 chemistry abnormalities of elevated aspartate aminotransferase (AST, Grade 3, 2 patients [0.8%]), elevated creatine kinase (Grade 3, 2 patients [0.8%]; Grade 4, 1 patient [0.4%]), elevated serum glucose (Grade 3, 4 patients [1.5%]), elevated lipase (Grade 3, 3 patients [1.1%]; Grade 4, 3 patients [1.1%]), and elevated total bilirubin (Grade 3, 1 patient [0.4%]). In the Vosevi group, all Grade 3 or 4 creatine kinase elevations were isolated events and were attributed to exercise. All Grade 3 or 4 lipase elevations were asymptomatic with no cases of clinical pancreatitis, and Grade 3 or 4 serum glucose elevations occurred in patients with a medical history of diabetes. The most commonly observed Grade 3 or 4 laboratory abnormalities observed in placebo-treated patients were elevated AST (Grade 3, 7 patients [4.6%]), consistent with untreated HCV infection, and elevated serum glucose (Grade 3, 7patientss [4.6%]) that mostly occurred in patients with a medical history of diabetes. No notable changes from baseline in vital signs were observed during the study. No patients had clinically significant electrocardiogram (ECG) abnormalities, and no pregnancies occurred during the study.

In the POLARIS-4 study, a similar percentage of patients in the Vosevi group experienced any AE (76.9%, 140 of 182 patients) compared with the SOF/VEL group (73.5%, 111 of 151 patients). A higher percentage of patients in the Vosevi group experienced a treatment-related AE (58.2%) compared with patients in the SOF/VEL group (51.0%). The most common AEs (Vosevi vs. SOF/VEL group) were headache (27.5% vs. 28.5%), fatigue (23.6% vs. 28.5%), diarrhea (19.8% vs. 4.6%), and nausea (12.1% vs. 7.9%). The type and incidence of common AEs were similar for the two treatment groups, with the exception of diarrhea which was reported for more patients in the Vosevi group compared with the SOF/VEL group.

Most AEs were Grade 1 or 2 in severity. A total of 4 patients experienced a Grade 3 or 4 AE; 2 patients (1.1%) in the Vosevi group and 2 patients (1.3%) in the SOF/VEL group. No Grade 3 or 4 AE was reported in >1 patient, and all Grade 3 or 4 AEs were considered not related to the study drug. Eight patients experienced SAEs; 4 patients (2.2%) in the Vosevi group and 4 patients (2.6%) in the SOF/VEL group. No trends in the SAE type or onset time were observed, and no SAE was reported in >1 patient. All of the SAEs were considered not related to the study drug and were resolved. One Grade 4 AE of illicit drug overdose (heroin and fentanyl) was reported in a patient in the Vosevi group that occurred 2 days post-dosing. This AE was considered serious and not related to the study drug and led to the only death in the study.

One patient (0.5%) in the SOF/VEL group experienced an AE of headache that led to premature discontinuation of study drug on Day 56; this event was considered Grade 2, related to study drug, and resolved following study drug discontinuation.

Most laboratory abnormalities were Grade 1 or 2 in severity. The most common Grade 3 hematology laboratory abnormality was decreased platelet count in the Vosevi and SOF/VEL groups (1.6% and 1.3%, respectively). No Grade 3 hematology laboratory abnormalities were associated with clinical symptoms or reported as AEs. No Grade 4 hematology laboratory abnormalities were observed. The most common Grade 3 chemistry laboratory abnormality was increased serum glucose in the Vosevi and SOF/VEL groups (2.2% and 2.0%, respectively). All of the patients with Grade 3 increased serum glucose had a history of diabetes. Four patients had a Grade 3 or 4 laboratory abnormality of increased lipase. These abnormalities were asymptomatic, with no cases of clinical pancreatitis. No patients reported pregnancies. Across treatment groups, there were no notable changes in vital sign measurements. No patients in either treatment group had a treatment-emergent clinically significant abnormal 12-lead ECG.

Overall, Vosevi has a tolerable and manageable safety profile. Appropriate warnings and precautions are in place in the approved Vosevi Product Monograph to address the identified safety concerns.

A warning for the potential of hepatitis B virus (HBV) reactivation has been included in a Serious Warnings and Precautions box in the Vosevi Product Monograph. Cases of HBV reactivation, including those resulting in fulminant hepatitis, hepatic failure, and death, have been reported during HCV treatment and/or post-treatment with regimens containing direct-acting antivirals in patients co-infected with HBV.

For more information, refer to the Vosevi Product Monograph, approved by Health Canada and available through the Drug Product Database.