Summary Basis of Decision for Bylvay

As outlined in the What steps led to the approval of Bylvay? section, the clinical review of the New Drug Submission for Bylvay was conducted as per Method 3 described in the Draft Guidance Document: The Use of Foreign Reviews by Health Canada .

Clinical Pharmacology

Odevixibat is a reversible, selective inhibitor of the ileal bile acid transporter (IBAT). It acts locally in the distal ileum to decrease the reuptake of bile acids from the terminal ileum and to increase the clearance of bile acids through the colon, thereby reducing the concentration of bile acids in the serum.

The pathophysiology of pruritus, a common symptom in patients with progressive familial intrahepatic cholestasis (PFIC), is not completely understood. Although the complete mechanism by which odevixibat improves pruritus in PFIC patients is unknown, it may involve inhibition of the IBAT.

The clinical pharmacology profile of odevixibat was mainly characterized in healthy adult volunteers and pediatric patients (6 months to 17 years of age) with cholestatic pruritus.

Odevixibat is minimally absorbed following oral administration in both healthy adults and pediatric patients. In healthy adults treated with a single 7.2 mg dose, the peak odevixibat plasma concentration was reached within 1 to 5 hours and the mean half-life was 2.36 hours following administration. In pediatric patients administered oral doses of 10 to 200 mcg/kg odevixibat, the resulting plasma concentrations, when quantifiable, were low and variable. Dose proportionality could not be reliably assessed. No accumulation of odevixibat was observed following once-daily dosing.

Odevixibat is minimally metabolized and is primarily eliminated as unchanged parent drug via fecal clearance, with minor renal clearance. Hepatic or renal impairment is not expected to have a significant effect on the pharmacokinetics of odevixibat.

Given that odevixibat acts locally in the distal ileum, its pharmacodynamic effect is unrelated to its systemic exposure. Following 42 days of once-daily dosing of 10 to 200 mcg/kg odevixibat in pediatric patients, decreases in mean serum bile acids and changes in pharmacodynamic biomarkers (i.e., decreases in fibroblast growth factor 19 [FGF19] and increases in 7α-hydroxy-4-cholensten-3-one [C4]) from baseline were observed across all doses. No dose-response relationship was observed for any of these effects. The extent of decrease in serum bile acids was consistently similar between 40 and 120 mcg/kg in pediatric patients with PFIC.

Bile acid binding resins may bind odevixibat in the gut, which may reduce the efficacy of odevixibat. The Bylvay Product Monograph recommends administration of bile acid binding resins at least 4 hours before or 4 hours after taking Bylvay. Treatment with Bylvay may affect the absorption of fat-soluble vitamins and lipophilic medicinal products as a consequence of increasing bile acid excretion in the feces.

A dedicated QT study was not conducted. Findings from non-clinical safety pharmacology and Phase I clinical studies in healthy volunteers indicate a low risk for QT prolongation.

The administration of Bylvay following a high-fat, high-calorie meal decreased the rate and extent of absorption and prolonged the time to reach maximum concentrations, relative to fasted conditions. Similarly, the administration of Bylvay, opened and sprinkled on applesauce, resulted in decreases in the rate and extent of absorption and prolonged the time to reach maximum concentrations compared to whole capsules administered under fasted conditions. However, the effect of food on the systemic exposure of odevixibat did not have a negative impact on its pharmacodynamic effects. A greater pharmacodynamic effect, as measured via changes in the plasma concentrations of the biomarker C4, was observed following a high-fat meal versus fasted conditions. Based on these clinical pharmacology data, administration of odevixibat with food may be associated with a greater pharmacodynamic effect and less systemic toxicity compared to fasted conditions. Based on the above finding, Bylvay should be administered in the morning with a meal.

Overall, the clinical pharmacology data support the use of Bylvay for the recommended indication. Key clinical pharmacology findings, relevant risks, and uncertainties are properly addressed in the Bylvay Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

The clinical efficacy of Bylvay for the treatment of pruritus in patients with progressive familial intrahepatic cholestasis (PFIC) was primarily evaluated in two Phase III studies: the pivotal randomized, double-blind, placebo-controlled PEDFIC1 study (A4250-005), and the supportive long-term, open-label extension PEDFIC2 study (A4250-008).

Pivotal Study

The PEDFIC1 study was a 24‑week, randomized, double-blind, placebo-controlled study conducted in 62 pediatric patients. Eligible participants were 6 months to 17 years of age with genetically confirmed PFIC type 1 or 2 (PFIC1 or PFIC2), significant pruritus (a mean scratch score greater than or equal to 2), and elevated serum bile acids (≥100 µmol/L). Of the enrolled patients, 73% had PFIC2 and 27% had PFIC1. At study entry, 89% were taking either ursodeoxycholic acid or rifampicin. Mean patient age was 3.2 years (range 0.5 to 15.9 years) and there were equal amounts of male and female participants. Three patients were older than 12 years of age with the oldest being 15.9 years old.

Reasons for patient exclusion included pathologic variations of the ABCB11 gene that predict the complete absence of the bile salt export pump (BSEP) protein, liver transplant, history of other liver disease, hepatic decompensation, abnormal intestinal motility or malabsorption, active infection, international normalized ratio (INR) greater than 1.4, alanine aminotransferase (ALT) greater than ten times the upper limit of normal (ULN), or bilirubin greater than ten times the ULN. Of the enrolled patients, 13% had prior biliary diversion surgery. Patients completing the PEDFIC1 study were eligible to enroll in the 72-week, open-label extension PEDFIC2 study.

The 62 patients were randomized 1:1:1 to receive Bylvay 40 mcg/kg/day (total number [n] = 23), Bylvay 120 mcg/kg/day (n = 19), or placebo (n = 20). Patients were stratified by PFIC type (1 or 2) and age (6 months to 5 years, 6 to 12 years, and 13 to 17 years).

The primary endpoint of the study was the proportion of patients with at least a 70% reduction in fasting serum bile acid levels or who achieved a level of 70 μmol/L or lower at Week 24, the end of treatment. This primary endpoint was chosen for all regions and was accepted by the European Medicines Agency. Canada was considered part of Europe and the “rest of the world” for whom the primary endpoint was a significant decrease in serum bile acids. The United States Food and Drug Administration (FDA) recommended that the primary endpoint be modified to a symptom-based endpoint, particularly because pruritus drives patients to biliary diversion surgery or liver transplantation. In response to this feedback, the primary endpoint for the United States was changed to the proportion of positive pruritus assessments at the patient level over the 24-week treatment period based on an observer-reported outcome (ObsRO) instrument. Given the patients’ young age, ObsRO was used to measure patients’ scratching as observed by their caregiver twice daily (once in the morning and once in the evening). Scratching scored from 0 (no scratching) to 4 (worst possible scratching). A positive pruritus assessment was a score of 1 or lower or at least a 1-point improvement from baseline. This was considered a secondary endpoint for Canada.

The primary endpoints for both the “rest of the world” and the United States FDA were statistically significant. The end-of-study proportion of serum bile acid responders were as follows: 43.5% in the 40 mcg/kg/day Bylvay group (p = 0.0015 versus [vs.] placebo), 21.1% in the 120 mcg/kg/day Bylvay group (p = 0.017 vs. placebo), and 33% overall among patients treated with Bylvay (p = 0.0015 vs. placebo) as compared to 0% of patients in the placebo group. With respect to the secondary endpoint (the primary endpoint for the United States FDA), the end-of-study proportion of positive pruritus assessments were as follows: 58.3% in the 40 mcg/kg/day Bylvay group (p = 0.0019 vs. placebo), 47.7% in the 120 mcg/kg/day group (p = 0.016 vs. placebo), and 53.5% overall amongst all patients treated with Bylvay (p = 0.0019 vs. placebo) as compared to 28.7% of patients in the placebo group.

Supportive Study

The PEDFIC2 study was an open-label, 72-week follow-up of PEDFIC1 to provide long-term efficacy and safety data. The study included 56 patients who completed the PEDFIC1 study (Cohort 1), and 56 other patients with any type of PFIC (Cohort 2), including PFIC1 (n = 35), PFIC2 (n = 66), PFIC3 (n = 7), PFIC4 (n = 2), and PFIC6 (n = 2), but not PFIC5 (n = 0). All patients were treated with 120 mcg/kg/day Bylvay. The median patient age was 3.6 years and 44% of patients were female. The interpretation of data on efficacy in the long-term follow-up study was limited by the lack of a concurrent control and the allowance for adjustment of potentially effective medications that are generally felt to be unsatisfactory on their own (ursodiol, rifampicin, and antihistamines).

In patients with PFIC1 or PFIC2 who started in the randomized, placebo-controlled PEDFIC1 study, the mean proportion of positive pruritus assessments from the PEDFIC2 study baseline increased following treatment with Bylvay from 31.4% during weeks 0 to 24 to 38.6% during weeks 0 to 72.

In Cohort 2, the mean proportion of positive pruritus assessments from weeks 0 to 72 was 92.4% in patients with PFIC3. In the two patients in Cohort 2 with PFIC4, scratching scores decreased from a baseline score of 4 to a score of 2.8 at weeks 13 to 16 for one patient, and from a baseline score of 2.8 to a score of 1.4, at weeks 1 to 4 for the other patient (the last assessments reported as of the cut-off date). In the two patients in Cohort 2 with PFIC6, scratching scores decreased from 3 and 2.1 at baseline, to 0 and less than 1 at weeks 71 to 72. The acceptance of Bylvay’s indication in all PFIC subtypes was supported by this data and the shared aspects of PFIC pathophysiology among all PFIC subtypes, in the context of PFIC’s rarity (especially subtypes other than PFIC1, PFIC2, and PFIC3) and an unmet medical need.

Because of the unexplained tendency towards superiority of the 40 mcg/kg/day dose, the recommended initial dose is 40 mcg/kg/day dose. However, an increase to 120 mcg/kg/day may be considered after 3 months following an inadequate response, as 13 of the 16 responders to 40 mcg/kg/day responded within 12 weeks of treatment onset, and 6 of the 7 patients who did not respond to 40 mcg/kg/day responded to a dose of 120 mcg/kg/day.

Indication

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

Bylvay (odevixibat) is indicated for the treatment of Progressive Familial Intrahepatic Cholestasis (PFIC) in patients ≥6 months of age.

To support safe and effective use of the product, Health Canada approved the following indication:

Bylvay (odevixibat) is indicated for the treatment of pruritus in Progressive Familial Intrahepatic Cholestasis (PFIC) in patients ≥6 months of age.

Limitations of Use:

Bylvay may not be effective in a subgroup of PFIC type 2 patients with specific ABCB11 variants resulting in non-functional or complete absence of the bile salt export pump protein.

The indication was changed to the treatment of pruritus in PFIC, to reflect the nature of the study participants who were required to have a history of significant pruritus, and to reflect that the outcome that defined clinically relevant benefit with Bylvay was pruritus, a cardinal, often-debilitating feature of PFIC that is the primary reason for the invasive surgical treatment of children with PFIC.

Due to the rarity of PFIC, especially of pre-surgical PFIC in older ages, and the common pathophysiology of PFIC in children, in adolescents, and in adults, extrapolation of data from mostly pre-adolescent children to adolescents and adults was considered acceptable.

Patients with PFIC2 and specific ABCB11 variants resulting in non-functional or complete absence of the bile salt export pump protein were excluded from the Phase III studies as Bylvay was not expected to be effective in their treatment. For the same reason, a limitation on the use of Bylvay was included as part of the indication for Bylvay.

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

Clinical Safety

The clinical safety of Bylvay was evaluated in a 117 PFIC patients who were enrolled in the two Phase III clinical studies (PEDFIC1 and PEDFIC2) described in the Clinical Efficacy section.

The pivotal PEDFIC1 study was a randomized, double-blind, placebo-controlled, 24‑week study of two dose levels of Bylvay (40 mcg/kg and 120 mcg/kg) administered once daily. The PEDFIC2 study was an open-label 72-week extension study of the continued treatment 120 mcg/kg/day Bylvay for patients in the PEDFIC1 study as well as the enrollment of an additional cohort of patients with PFIC (n = 56). The median duration of exposure to Bylvay 120 mcg/kg/day in the PEDFIC2 study was 78.6 weeks.

Pivotal Study

The incidence of any treatment-emergent adverse events (TEAEs) and severe TEAEs was similar in the three treatment arms of the PEDFIC1 study. In the placebo group, 85% of patients experienced a TEAE as compared to 83% of patients treated with 40 mcg/kg/day Bylvay, and 84% of patients treated with 120 mcg/kg/day Bylvay. Severe TEAEs occurred in 10% of placebo patients, 4% of patients treated with 40 mcg/kg/day Bylvay, and 11% treated with 120 mcg/kg/day Bylvay.

In patients treated with Bylvay, it was more common to observe TEAEs that were drug-related, or associated with treatment interruption or discontinuation. Drug-related TEAEs occurred in 15% of placebo patients, 30% of patients treated with 40 mcg/kg/day Bylvay, and 37% of patients treated with 120 mcg/kg/day Bylvay. Treatment-emergent adverse events that led to treatment interruption occurred in 5% of placebo patients, 13% of patients treated with 40 mcg/kg/day Bylvay, and 32% of patients treated with 120 mcg/kg/day Bylvay. Treatment-emergent adverse events that led to treatment discontinuation were observed in 5% of the patients treated with 120 mcg/kg/day Bylvay and 0% in the other groups.

Very common TEAEs that occurred with an incidence of 10% or more were gastrointestinal symptoms, including diarrhea (occurring in 31% of Bylvay versus [vs.] 5% of placebo patients), vomiting (17% of Bylvay vs. 0% of placebo patients), and abdominal pain (14% of Bylvay vs. 0% of placebo patients).

One patient likely had gastroenteritis which caused severe dehydration and severely increased transaminases. As Bylvay may potentiate diarrhea and vomiting, the Bylvay Product Monograph recommends to consider interrupting Bylvay treatment during acute episodes of diarrhea and/or vomiting that risk dehydration.

One patient treated with Bylvay (2.4%) experienced a lower limb fracture, although causality was uncertain. There were no deaths in the pivotal study, and no serious or severe TEAEs clearly attributable to treatment with Bylvay.

Alanine aminotransferase (ALT) increased in 14% of Bylvay and 5% of placebo patients. Bilirubin increased in 12% of Bylvay and 10% of placebo patients. Aspartate aminotransferase (AST) increased in 7% of Bylvay and 5% of Bylvay patients. In one instance, the bilirubin increase was severe but likely not due to Bylvay.

In the PEDFIC1 study, trends were also noted towards decreased mean ALT, AST, gamma-glutamyl transpeptidase (GGT), bilirubin, and Model for End-stage Liver Disease (MELD)/Pediatric End-stage Liver Disease (PELD) scores from baseline to end-of-study. It was uncertain whether the variation seen in liver biochemistry and its causality were related to Bylvay. To address the potential risk of hepatoxicity, the Bylvay Product Monograph lists increased ALT, AST and bilirubin as possible abnormal laboratory findings that may occur with Bylvay treatment.

Other abnormal laboratory findings included vitamin A and vitamin E deficiency, which each occurred in one Bylvay patient (2%). Neither of these cases were refractory to supplementation. Furthermore, 7% of Bylvay-treated patients and 0% of placebo-treated patients experienced calcium shifts to low levels; however, these were not considered to be clinically significantly low.

No bile-diversion surgery or liver transplant took place during the pivotal PEDFIC1 study.

Supportive Study

Safety findings in the long-term follow-up PEDFIC2 study were consistent with those from the PEDFIC1 study. Two patients had severe TEAEs of elevated transaminases; one with increased GGT. Both TEAEs were assessed by investigators as being possibly related; however, were adjudicated as not being due to Bylvay. Residual uncertainty was reflected in the Bylvay Product Monograph for the potential risk of increased transaminases. Similarly, one long-term follow-up patient had a serious TEAE of cholestasis assessed as severe and unrelated to study drug, therefore the Bylvay Product Monograph includes the potential risks of jaundice and increased blood bilirubin. The Product Monograph also suggests to consider discontinuing Bylvay if a patient progresses to portal hypertension, cirrhosis, or demonstrated hepatic decompensation given the limited data available in these patients.

Appropriate warnings and precautions are in place in the approved Bylvay Product Monograph to address the identified safety concerns.

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

As outlined in the What steps led to the approval of Bylvay? section, the review of the non-clinical component of the New Drug Submission for Bylvay was conducted as per Method 3 described in the Draft Guidance Document: The Use of Foreign Reviews by Health Canada.

Pharmacodynamics

In vitro studies demonstrated that odevixibat is a highly potent and selective inhibitor of the human ileal bile acid transporter (IBAT) (concentration for 50% inhibition of effect [IC50] of 0.13 nmol/L).

In vivo, odevixibat showed a significant inhibitory effect on the absorption of intestinal bile acids. Following a single administration of odevixibat (0.463 mg/kg) to apolipoprotein E (ApoE) knockout mice, approximately 81% inhibition was observed for up to 3 hours. This fell to 28% inhibition at 10 hours post administration.

In ApoE/low-density lipoprotein (LDL) double knockout mice, odevixibat (0.463 mg/kg) reduced plasma cholesterol by 40%. This was correlated with a reduction in very low-density lipoproteins (VLDL) and LDL. High-density lipoproteins (HDL) remained unaffected.

In a mouse multidrug-resistant 2 protein (Mdr2-/-) model of cholestasis, odevixibat administered at 0.03% weight by weight in feed for 4 weeks reduced liver/body weight ratios, serum markers of liver damage (alanine aminotransferase [ALT], aspartate aminotransferase [AST], and alkaline phosphatase [ALP]), and cholestasis.

A cross-over study was conducted to evaluate the faeces of 4 dogs after oral administration of odevixibat (30 mg/kg on Days 1 and 4) in combination with the rectal administration of cholestyramine suspension or a placebo. The study demonstrated that rectal cholestyramine administration tended to normalize the odevixibat induced increases in the instances/number of defecations and faecal consistency relative to placebo controls. Therefore, the effects of odevixibat on the gastrointestinal system were largely reversible.

In male rats, a single administration of odevixibat (≤100 µmol/kg [≤74.09 mg/kg]) had no effect on total intestinal transit as assessed by intestinal length travelled and transit time (assessed via charcoal propulsion).

With respect to secondary pharmacodynamics, odevixibat (1 µM [740.9 mcg/L]) had no inhibitory effect in 17 enzyme/receptor models studied in vitro and no effect when evaluated in two isolated tissue models (L-type calcium channels or sodium channel site).

Safety Pharmacology

In the human IBAT transfected cell assay, there were no adverse effects on human Ether-à-go-go gene (hERG) channel conductance at a concentration of up to 1 µM (740.9 mcg/L), which was 7,700-fold higher than the IC50 (0.13 nM).

Following the oral administration of odevixibat (≤100 µmol/kg [≤74.09 mg/kg]) to rats, no adverse effects on the central nervous system system, respiration, renal function, or cardiovascular parameters including blood pressure and heart rate were observed.

No drug-related effects were observed on haemodynamics in dogs after the intravenous administration of odevixibat (≤0.1 µmol/kg [≤74.09 mcg/kg].

Pharmacokinetics

The pharmacokinetics and metabolism of odevixibat were studied in mouse, rat, dog, and human tissues in vitro and in mice, rats, dogs, marmosets, and humans in vivo. The studies demonstrated that odevixibat has minimal absorption following an oral dose, had low passive permeability in vitro, and low absolute oral bioavailability. Odevixibat was rapidly absorbed with a total time to maximum concentration (T_max) that ranged from 1 to 4 hours in all of the nonclinical species evaluated (mouse, rat, rabbit, dog, and marmoset). Odevixibat was highly protein bound at 98.0% in rabbit and more than 99.2% in all other species.

Three oxidative metabolites; monohydroxylated (M2, M3 and M6) were characterized by liquid chromatography-mass spectrometry during the in vitro study, however, only parent drug was observed in human feces. No metabolites were formed by human hepatocytes that were not observed in hepatocytes from at least one other species.

Toxicology

The nonclinical safety profile of odevixibat was explored across the rodent (mouse and rat) and non-rodent (dog and marmoset) species. In all species, treatment with odevixibat was generally well tolerated, with in-life findings (periodic diarrhoea, emesis, salivation, reductions in body weight/weight gain and/or food consumption) and clinical pathology alterations (reductions in HDL, LDL, and total cholesterol).

In single-dose tests in mice and rats the maximum tolerated dose of odevixibat exceeded 2,000 mg/kg with no pathological findings. Mice experienced diarrhea and temporary weight loss, while rats remained asymptomatic.

In repeat-dose studies, there were no drug-related deaths in mice, rats, dogs and marmosets, with the exception of male mice given 300 mg/kg/day orally, which exhibited clinical deterioration before euthanasia. The cause for this deterioration was unclear, however it occurred at a systemic exposure level 490‑times higher than the human equivalent. Common clinical signs included post-dosing vomiting, soft feces, and increased salivation. Notable drug-related microscopic changes were primarily seen in rats. The no observed adverse effect level (NOAEL) was 300 mg/kg/day in rats and 150 mg/kg/day in dogs, significantly higher than recommended human doses.

Odevixibat showed no genotoxic potential, with negative mutagenic effects in bacterial strains and no direct deoxyribonucleic acid (DNA) damage. The rat micronucleus test confirmed non-clastogenic and non-aneugenic properties in vivo.

In two-year carcinogenicity studies, odevixibat was not tumorigenic in rats or mice at oral doses up to 100 mg/kg/day.

In pregnant rabbits, early delivery/abortion and fetal cardiovascular defects were observed at elevated exposure levels, indicating a potential risk to fetal development. Seven fetuses (1.3% of all fetuses from odevixibat exposed does) in all dose groups were found to have cardiovascular defects. These malformations were not observed in pregnant rats. Odevixibat was excreted in rat milk, but clinical significance remains uncertain. In juvenile rats, an NOAEL of 100 mg/kg/day was established, with only loose feces noted at the highest dose.

The results of the non-clinical studies as well as the potential risks to humans have been included in the Bylvay Product Monograph. In view of the intended use of Bylvay, there are no pharmacological or toxicological issues within this submission which preclude authorization of the product. Appropriate warnings and precautionary measures are in place in the Bylvay Product Monograph to address the identified safety concerns.

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

As outlined in the What steps led to the approval of Bylvay? section, the review of the quality component of the New Drug Submission for Bylvay was conducted as per Method 3 described in the Draft Guidance Document: The Use of Foreign Reviews by Health Canada.

The chemistry and manufacturing information submitted for Bylvay has demonstrated that the drug substance and drug product can be consistently manufactured to meet the approved specifications. Proper development and validation studies were conducted, and adequate controls are in place for the commercial processes. Changes to the manufacturing process and formulation made throughout the pharmaceutical development are considered acceptable upon review. Based on the stability data submitted, the proposed shelf life of 36 months is acceptable when the drug product is stored at room temperature (15 ºC to 30 ºC) in the original package to protect from light.

Proposed limits of drug-related impurities are considered adequately qualified (i.e., within International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use [ICH] limits and/or qualified from toxicological studies).

All sites involved in production are compliant with good manufacturing practices.

None of the non-medicinal ingredients (excipients, described earlier) found in the drug product are prohibited by the Food and Drug Regulations.

None of the excipients used in the formulation of Bylvay is of human or animal origin.