Regulatory Decision Summary for Livmarli

The purpose of this Supplement to a New Drug Submission (SNDS) was to obtain market authorization, pursuant to section C.08.004 of the Food and Drugs Regulations, for Livmarli filed by Mirum Pharmaceuticals Inc.

This Supplement was filed for Livmarli (Maralixibat) for a new indication in the treatment of progressive familial intrahepatic cholestasis (PFIC), a new strength oral solution (19 mg/mL) to be used for the proposed new indication, and a new tablet dosage form (in four strengths: 10 mg, 15 mg, 20 mg, and 30 mg) as an alternate dosage form to be used for both the currently approved indication in Alagille Syndrome (ALGS) and the newly proposed indication.

Upon review of the submitted data package, Health Canada authorized Livmarli with the following modification to the indication: the treatment of cholestatic pruritus in patients aged 12 months or older with progressive familial intrahepatic cholestasis (PFIC), the new strength oral solution, and the new tablet dosage form.

The proposed indication was supported by the pivotal efficacy study MRX-502, which was a 26-week randomized, double-blind, placebo-controlled study in patients with PFIC aged ≥ 12 months. The study enrolled 93 patients. Efficacy was evaluated in 64 patients (PFIC cohort). An additional 29 patients that did not meet the inclusion criteria for the PFIC cohort were not assessed for efficacy (e.g., patients with truncated PFIC2 mutation [t‑PFIC2], or unknown variants). The PFIC cohort enrolled 64 patients with genetically confirmed diagnosis of biallelic variants of PFIC: 31 with non-truncated PFIC2 (nt-PFIC2) and 33 patients with PFIC1 (n = 13), PFIC3 (n = 9), PFIC4 (n = 7), or PFIC6 (n = 4). Patients had to have persistent pruritus (> 6 months), abnormal liver tests or/and progressive liver disease, and an average pruritus severity score ≥ 1.5 (Itch Observer-Reported Outcome : ItchRO[Obs]). Patients with decompensated cirrhosis, alanine transaminase (ALT) > 15× the normal value (ULN) or total bilirubin (TB) > 15× ULN at screening were excluded. Patients were randomized to receive Livmarli orally twice daily (BID) at escalating doses starting at 142.5 µg/kg BID to up to 570 µg/kg BID (or maximum tolerated dose) by week 4 (n = 33) or placebo twice daily (n = 31). Dose reductions were allowed for safety/tolerability.

The primary efficacy endpoint was the change in pruritus as measured by the change from baseline to week 15-26 in morning ItchRO(Obs) score, which relies on a scale from 0 (none) to 4 (very severe pruritus) rated by the child’s caregiver. The ItchRO(Obs) score and a minimum clinically relevant decrease of 1 were adequately validated. The key secondary efficacy endpoint was the change in total serum bile acid (sBA) levels from baseline.

Approximately 53% of patients were female with a median age of 3 years (64% < 6 years of age). They had significant growth delay, were mostly on stable ursodeoxycholic acid or rifampicin, and had markedly elevated liver parameters. Baseline morning ItchRO(Obs) averaged 2.8 and sBA 263 μmol/L. Overall, no major issues were noted in the statistical analyses or handling of missing values, and 94% of patients completed treatment.

Livmarli was statistically significantly superior to placebo in improving pruritus with a decrease from baseline in itch score of ‑1.8 with Livmarli and -0.6 with placebo, for a clinically relevant difference of -1.2 (p<0.0001). The decrease in sBA levels was also significantly larger with Livmarli (-158 µmol/L) compared to placebo (+3 µmol/L), for a difference of ‑160 µmol/L (p<0.0001). The data also suggested that the proportions of responders for the reduction in pruritus and the reduction of sBA was larger with Livmarli (differences in responders rates of between 30% to 39% in favor of Livmarli). Other secondary/exploratory endpoints were generally supportive of the primary endpoint results. Sensitivity analyses confirmed the robustness of the results, and subgroup analyses did not reveal any significant divergences precluding the approval in any subgroup by region, age, sex, baseline median sBA levels, baseline UDCA, or rifampicin usage, with the exception of t‑PFIC2 patients (truncating PFIC2 mutation with total loss of bile salt export pump (BSEP) function) in which Livmarli failed to improve pruritus. Therefore, a limitations of use for this population was recommended accordingly. The analyses in the other PFIC types tended to be in favor of Livmarli.

No clinically significant benefit was observed in major liver-associated outcomes or in children’s growth. Laboratory results showed small increases in ALT and AST (+14 to +17 U/L) and a decrease in bilirubin (‑2 mg/dL) with Livmarli compared to placebo. Levels of 7αC4, a biomarkers of bile acid synthesis, also increased (+46 ng/mL). While this may suggest increase bile acid synthesis, the clinical relevance is unclear. As a result, the indication was revised from “treatment of PFIC” to “treatment of cholestatic pruritus in PFIC”.

Study MRX-801 was a 13-week open-label study conducted in 10 infants < 12 months of age with PFIC, followed by a long-term extension phase. The dosing regimen was similar to that of the pivotal trial. Nine patients completed the 13-week period, and 1 patient was discontinued due to the serious adverse event (AE) of ALT increased. Only 3 patients were treated with 570 mcg/kg BID for 12 months or more. The efficacy data failed to show any improvement in pruritus. Therefore, the indication was restricted only to patients aged 12 months or older.

1. As per the sponsor, maralixibat has been studied in approximately 200 children and 100 adults with cholestatic liver disease. Of these, 103 pediatric patients were treated at the proposed dose of 570 µg/kg BID (Studies MRX-502 and MRX-801). The pivotal trial safety data included 93 patients (47 treated with Livmarli and 46 with placebo for an average treatment duration of 177 days at an average daily dose of 955 µg/kg/day [84% of 570 µg/kg BID]). Almost all patients in the Livmarli group (46/47) escalated to the maximum dose (570 µg/kg BID), and 37 remained at that dose for 20 weeks. Out of 93 patients enrolled, 93% completed the study, and 7 discontinued (4 consent withdrawals, and 1 case each of AE, liver transplantation, and disease progression).

AEs occurred in almost all patients, with drug‑related AEs more common with Livmarli (38%) than placebo (4%). Most AEs were mild to moderate in severity. The most frequent drug‑related AEs with Livmarli were diarrhea (57% vs. 20% with placebo), abdominal pain (28% vs. 15%), transaminases increased (17% vs. 7%), rectal hemorrhage (9% vs. 2%), and bone fractures (7% vs. 0%). Serious AEs (SAE) occurred in 11% of Livmarli-treated patients and 7% with placebo. These included one case of Livmarli-related bilirubin elevation with cholangitis considered a possible drug induced liver injury (DILI), one case of worsening cholestasis (unrelated to Livmarli) with cholangitis resulting in liver transplant, and three cases of infections. One additional case of pruritus was deemed unrelated to Livmarli, and one patient (oldest age group) discontinued Livmarli permanently due to diarrhea. In the placebo group, serious events included one case of coagulopathy (with vitamin K deficiency and seizure) and one gastroenteritis. No deaths were reported in either group.

Rates of AEs and SAEs in the rollover open-label extension study (MRX-503) were somewhat higher compared to the pivotal study, but the difference may be due in part to the added 7 months of follow-up. Approximately 92% of 85 enrolled patients received the 570 mcg/kg BID dose and were treated on average for 13 months. One serious Livmarli-related AE of increased bilirubin, as well 3 discontinuations due to adverse reactions were reported: 1 case of fatal hepatic cirrhosis considered a possible DILI, 1 case of increase in ALT and bilirubin, and 1 case of diarrhea. A second fatality due to respiratory tract infection was probably not related to Livmarli. There were 3 liver-associated events: 1 listing for liver transplantation, biliary diversion surgery, and liver transplantation. In the overall Livmarli clinical development program, eight possible cases of DILI in patients with PFIC, and two possible and one probable case in patients with ALGS were reported by two adjudicating committees. In general, hepatotoxicity is difficult to differentiate from signs of disease progression or its complications. The Product Monograph (PM) was updated to add a contraindication in those with prior or active hepatic decompensation events and to add robust warnings and precautions.

With the new indication and higher dose, propylene glycol (PG) exposure from the oral solution increased 1.5-fold (16.6 to 24.2 mg/kg/day), approaching the acceptable daily intake of ≤ 25 mg/kg/day. Because PG toxicity is of greater concern in young children and in those with hepatic impairment, the dosing regimen was revised to decrease PG exposure mostly below 20 mg/kg/d (< 22 mg/kg/day in one age category). Furthermore, a detailed Warnings & Precautions on PG toxicity was also added to the PM.

The proposed dose escalation was considered too steep, therefore, an intermediate step (428 mcg/kg BID, as in the pivotal trial) was added to better reflect the dosing regimen of the pivotal clinical trial. Additionally, as no data supported the efficacy or safety of the proposed 57 mg maximum dose; the maximum dose in the PM was reduced to 38 mg for the solution, and 40 mg for the tablet.

As part of this submission, tablets were proposed as an alternate dosage form. Formal pharmacokinetic (PK) bioequivalence to the oral solution could not be demonstrated, with data suggesting ~20% lower systemic exposure. However, because Livmarli acts locally in the ileum and has minimal systemic absorption (~1%), these PK differences are not considered to be clinically meaningful. In addition, dose titration in the regimen allows adjustment should any unlikely differences arise. Moreover, the absence of PG in the tablets further supports their use, given that PFIC and ALGS patients may have (or develop) hepatic impairment, increasing susceptibility to PG toxicity.

Taken together, the clinical package provided sufficient efficacy and safety evidence of Livmarli in the treatment of cholestatic pruritus in patients aged 12 months or older with progressive familial intrahepatic cholestasis.

A Risk Management Plan (RMP) for Livmarli was submitted by Mirum Pharmaceuticals Inc. to Health Canada. Upon review, the RMP was considered to be acceptable with revisions. The RMP is designed to describe known and potential safety issues, to present the monitoring scheme and when needed, to describe measures that will be put in place to minimize risks associated with the product.

The chemistry and manufacturing information submitted for Livmarli (maralixibat oral solution 19 mg/mL) and Livmarli (maralixibat tablets 10 mg, 15 mg, 20 mg, and 30 mg) has demonstrated that the drug substance and drug product can be consistently manufactured to meet the approved specifications.

The final labelling and Product Monograph were considered acceptable.

Overall, the benefit-harm-uncertainty profile was favorable for Livmarli (maralixibat) for the approved indication of cholestatic pruritus in patients ≥ 12 years of age with PFIC, and the new 19 mg/mL oral solution and tablet dosage form, when used under the conditions of use recommended in the approved Product Monograph. Therefore, a Notice of Compliance (NOC) was recommended.

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