CADTH Reimbursement Review

Givosiran (Givlaari)

Sponsor: Alnylam Netherlands B.V.

Therapeutic area: Acute hepatic porphyria in adults

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Clinical Review

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Introduction

The porphyrias are a group of metabolic disorders caused by altered activities of enzymes within the heme biosynthetic pathway. The acute hepatic porphyrias (AHPs) include acute intermittent porphyria (AIP), aminolevulinic acid dehydratase-deficient porphyria (ADP), hereditary coproporphyria (HCP), and variegate porphyria (VP). Each type of AHP is characterized by a specific genetic mutation involved in the synthesis of heme in the liver; however, these types are clinically indistinguishable.¹ Information about the prevalence and incidence of AHP that is specific to Canadians is not available. The estimated prevalence of AHP in the Canadian provinces (excluding Quebec) is |||| per million population²; however, it was noted by both the sponsor and the clinical experts consulted for this review that patients with AHP are underdiagnosed due to a lack of available treatment options and poor awareness of the disease.

In patients with AHP, the altered enzymatic activity within this pathway results in an excessive accumulation of the intermediate porphyrin precursors, aminolevulinic acid (ALA) and porphobilinogen (PBG), which are neurotoxins that can precipitate an acute attack.^1,3 During an acute attack, patients report a significant increase in pain that can gradually build over hours and last for a number of days.³ Some patients report pain in the chest, back, or extremities; however, severe abdominal pain is more typical. Long-term complications with recurrent acute attacks may include chronic pain, chronic kidney failure, and liver damage.

Treatment for AHP is mainly targeted at preventing acute attacks. The British and Irish Porphyria Network (BIPNET) recommends that recurrent acute attacks be managed through general measures and the avoidance of precipitating factors, gonadotropin-releasing hormone (GnRH) analogues, prophylactic heme arginate, and/or liver transplantation.⁴ The clinical experts consulted for this review indicated that the avoidance of triggers is sufficient for many patients. One of the biggest contributors to an acute attack is a change in progesterone levels during the menstrual cycle. In some cases, GnRH analogues may be used (with expert guidance) to suppress the menstrual cycle; however, the clinical experts consulted for this review reported that long-term use of hormone therapy is typically not an option due to the severity of the side effects. Despite the narrow indication for the treatment of acute attacks, hemin is often considered for prophylactic use outside of the indication in patients who exhibit recurrent attacks with the preventive measures described previously; however, the use of hemin is also associated with complications resulting from the requirement for venous access, iron overload, and difficulty withdrawing from treatment.⁴ Liver transplantation may be considered in patients with recurrent attacks who are frequently hospitalized and exhibit diminishing health-related quality of life (HRQoL).

The drug under review, givosiran, is a double-stranded, small, interfering RNA that results in a reduction of liver 5'-aminolevulinate synthase 1 (ALAS1) mRNA. In Canada, givosiran is indicated for the treatment of AHP in adults.⁵ It is available as a solution for subcutaneous injection (189 mg givosiran/mL) and does not require additional reconstitution or dilution before administration. The recommended dose is 2.5 mg/kg once monthly, based on body weight. The objective of this review is to perform a systematic review of the beneficial and harmful effects of givosiran 2.5 mg/kg once monthly for the treatment of AHP in adults.

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient groups who responded to CADTH’s call for patient input and from clinical experts consulted by CADTH for the purpose of this review.

Patient Input

CADTH received 2 patient group submissions for this review from the Canadian Association for Porphyria/Association Canadienne de Porphyrie (CAP) and the American Porphyria Foundation (APF). CAP is a national voluntary charity whose mission is to deliver evidence-based information and support to patients with porphyria, their families, health care providers, and the general public. APF provides programs to raise awareness and educate health care professionals and the general public in 76 countries around the world. Of its international members, more than 300 are located in Canada. To obtain input for this review, CAP distributed a survey to its members in February 2021. The survey was restricted to Canadian patients and caregivers with experience with AHP. In total, 22 patients and 4 caregivers responded to the survey. CAP also requested support from the British Porphyria Association, which shared 3 interviews from individuals who had received givosiran. APF used its social media platforms and online newsletters to connect with Canadian patients about their experiences with porphyria. It also collected responses by telephone and email. Some of the responses in the APF submission were collected during an Alnylam Patient Advisory Board meeting. Twelve individual patient submissions were collected from Canadians.

In both submissions, respondents noted that they had experienced the following symptoms, among others: pain, fatigue, nausea, weakness, paralysis, neuropathy, seizures, anxiety, and depression. More than 80% of patients from the CAP survey had experienced symptoms at least once a month, with many reporting that these symptoms occurred more than 20 days per month. The group also reported that 86% of respondents had at least 1 attack in the past year and 36% had at least 10. Furthermore, 55% of patients had gone to the emergency room at least once in the past year due to an attack, while 18% had gone at least 10 times. Porphyria attacks can prevent patients and caregivers from being able to work, lead to poorer quality of life, and negatively affect relationships. The patient input submissions described how symptoms and efforts to avoid triggers could strain social relationships and make it difficult to care for their families. Both groups emphasized the negative effects that porphyria had on their daily lives and mental health.

Respondents would like a cure for porphyria, but many believe a more realistic goal is to have a treatment that prevents attacks and reduces symptoms, particularly pain, nerve damage, and paralysis. Patients and caregivers would like to see additional options that are more effective, have fewer side effects, offer an easier mode of administration, can be administered outside of a hospital, and will lead to improvements in quality of life. Other limitations to accessing treatments that were identified include the need for travel, the requirement for venous access, and the lack of access to specialists and proper diagnostic testing.

Clinician Input

Input From Clinical Experts Consulted by CADTH

One of the major goals in the management of AHP is to reduce the frequency of AHP attacks. According to the clinical experts consulted for this review, most patients with recurrent attacks will continue to have recurrent attacks with currently available treatment strategies. The experts noted that while prophylactic hemin can be used to reduce the rate of AHP attacks, with case reports of improvement, the use of prophylactic hemin is outside of the Health Canada–approved indication and has not been studied well. GnRH may also be used to prevent AHP attacks, but it is not approved for prolonged use and is associated with side effects and loss of bone mineral density.

As per feedback from the clinical experts consulted for this review, givosiran would be used in patients that have recurrent attacks because there is no evidence to support its use in asymptomatic individuals or acute attacks. The clinical experts felt that givosiran would not be used as a first-line treatment or to treat the first AHP attack. They recommended that patients with AHP try other approaches to treatment, such as avoidance of triggers, before givosiran. The experts expected givosiran to provide an alternative therapy for a small subset of patients with frequent or recurrent attacks who would otherwise require hospitalization and hemin administration. The experts recommended that givosiran be reserved for patients with recurrent symptoms or flares that are consistently affecting their HRQoL. Givosiran was also described by the experts as an appropriate treatment for patients who qualify for hemin prophylaxis but cannot adhere to treatment due to toxicity or lack of convenience.

The following outcomes were noted by the clinical experts as those that are used to determine response to treatment in clinical practice: reduced attack rate, reduced hospitalization, reduced need for hemin, frequency of neurovisceral flares, and improved patient-reported outcomes, such as daily symptoms, HRQoL, and work-life productivity. The clinical experts suggested that patients be assessed for response to treatment every 6 months or annually. All of the experts agreed that 1 year would be a sufficient amount of time to assess a patient’s response to treatment; however, the variable presentation of the disease—as evidenced by yearly fluctuations in attack frequency—was noted as a limitation in this assessment.

In general, the clinical experts felt that patients treated with givosiran would continue with treatment until there was a reason for discontinuation, such as safety concerns or an increase (or lack of improvement) in rate of attacks with treatment, which may indicate that treatment is not working. The clinical experts also indicated that menopause would be a potential reason to trial treatment discontinuation in patients with stable disease. However, it was challenging for the clinicians to specifically define response to treatment due to the heterogeneous nature of AHP among patients. The clinical experts also noted that if attacks recurred following discontinuation, restarting treatment with givosiran would be a possibility.

Clinician Group Input

CADTH did not receive any input from clinician groups for this review.

Drug Program Input

The drug programs inquired about the requirements for diagnosis of types of AHP, the use of givosiran outside of the criteria used in Study 003 (see Description of Studies in the next section), discontinuation of therapy, the use of givosiran in combination with hemin for an acute attack, and generalizability issues for non-AIP types of AHP. The clinical experts noted that the biochemical tests for urinary ALA and PBG are specific to AHP and, along with clinical evidence consistent with porphyria attacks, are sufficient to make a diagnosis; genetic tests are not required. The clinical experts indicated that treatment decisions would be made on a case-by-case basis using clinical judgment, but would generally be guided by the criteria outlined in the pivotal trial. The clinical experts did not express concern about the use of givosiran in combination with hemin. The results of the trial in patients with AIP were considered generalizable to all patients with AHP.

Clinical Evidence

Pivotal Studies and Protocol-Selected Studies

Description of Studies

One multi-centre, placebo-controlled, double-blind (DB), phase III study (Study 003) was included in the CADTH systematic review. Study 003 was designed to evaluate the efficacy and safety of givosiran administered once monthly in patients with AHP. Included patients had to be at least 12 years old with a documented diagnosis of AIP, HCP, VP, or ADP, had experienced at least 2 composite porphyria attacks within 6 months before screening, and had to be willing to abstain from prophylactic use of hemin during the trial. The primary objective was to evaluate the effect of subcutaneous givosiran compared to placebo in terms of the rate of porphyria attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home over 6 months in patients with AIP. The annualized rate of porphyria attacks in patients with AHP and the following assessments in patients with AIP were included as secondary outcomes: urinary ALA and PBG levels, hemin use, daily worst scores for symptoms (including pain, fatigue, and nausea), and HRQoL as measured using the 12-item Short Form Health Survey (SF-12). Opioid use, the Porphyria Patient Experience Questionnaire (PPEQ), and the ability to work or attend school, as well as the secondary end points analyzed in patients with AHP, were included as exploratory outcomes. Study 003 implemented a statistical hierarchy to control for multiple testing, where the first outcome to be tested was the annualized attack rate (AAR) in patients with AIP over the 6-month DB period followed by the following outcomes (conducted in patients with AIP unless indicated otherwise): urinary ALA levels at 3 months; urinary ALA levels at 6 months; urinary PBG levels at 6 months; annualized rate of administered hemin doses over the 6-month DB period; AAR in patients with AHP over the 6-month DB period; daily worst pain score; fatigue score; nausea score over the 6-month DB period; and, change from baseline in the Physical Component Summary (PCS) of the SF-12 at 6 months.

A total of 94 patients were randomized in Study 003, 89 (95%) of whom had AIP. Patients with AIP were between the ages of 19 years and 65 years (mean = 37.3 years to 40.7 years); 89% to 91% were female; and 35% to 40% resided in North America. Between 40% and 44% of patients had prior experience with prophylactic hemin, and based on the composite definition of porphyria attacks, the median historical AAR was 8 attacks (range = 4 to 34) and 8 attacks (range = 0 to 46) in the givosiran and placebo treatment groups, respectively. While not having a porphyria attack, 48% to 56% of patients reported having chronic symptoms, and 28% to 30% reported chronic opioid use. Baseline characteristics in patients with AHP were similar to those reported for patients with AIP.

Efficacy Results

A summary of key results from the pivotal trial is provided in Table 2. The description of results provided here will focus on analyses conducted in the modified full analysis set (mFAS) for patients with AIP. Results based on the full analysis set (FAS) in patients with all types of AHP will only be described if there is a notable difference from the results based on the mFAS.

The primary end point of the pivotal trial was the annualized rate of porphyria attacks in patients with AIP over the 6-month DB period, where porphyria attacks were defined as events requiring hospitalization, urgent health care visits, or IV hemin administration at home. The mean AAR based on the composite end point was 3.22 (95% confidence interval [CI], 2.25 to 4.59) and 12.52 (95% CI, 9.35 to 16.76) for patients in the givosiran treatment group and placebo treatment group, respectively. This corresponded to a 74% reduction in the rate of porphyria attacks for patients in the givosiran group relative to patients receiving placebo (rate ratio = 0.26; 95% CI, 0.16 to 0.41; P < 0.001). The number of attacks for each component of the primary outcome was also reported. Treatment with givosiran corresponded to a 49% rate reduction in attacks that required hospitalization [rate ratio = 0.51 (95% CI, 0.25 to 1.04)], and an 84% rate reduction in attacks requiring an urgent health care visit (rate ratio = 0.16; 95% CI, 0.09 to 0.31). A total of 3 attacks required IV hemin administration at home for patients in the givosiran group, and 32 attacks required IV hemin administration at home for patients in the placebo group (rate ratio was not assessed due to n < 10 in the givosiran group).

HRQoL was evaluated using the SF-12, the EuroQol 5-Dimensions 5-Levels questionnaire (EQ-5D-5L), and the Patient Global Impression of Change (PGIC). Each of these HRQoL outcomes is widely used in clinical trials; however, evidence of validity, reliability, and responsiveness, or a minimally important difference (MID) in patients with AHP, were not identified. All of the HRQoL outcomes were reported as exploratory except for the PCS of the SF-12, which was a secondary outcome in Study 003. At month 6, the least squares (LS) mean change from baseline in the PCS score was 5.37 (standard error of the mean [SEM] = 1.17) for the givosiran treatment group and 1.43 (SEM = 1.22) for the placebo treatment group. The between-groups difference in the LS mean PCS score for givosiran compared to placebo was 3.94 (95% CI, 0.59 to 7.29; P = 0.0216). Due to a failure higher in the statistical testing hierarchy, the reported P value cannot be interpreted as statistically significant. The results of the change from baseline in the domains scores for the SF-12 suggest that the PCS score was driven by the bodily pain (BP) and role physical (RP) domains. The Mental Component Summary (MCS) score of the SF-12 was reported descriptively. At month 6, the mean changes from baseline in MCS score were 3.55 (standard deviation [SD] = 10.08) and 1.30 (SD = 8.54) for patients receiving givosiran and placebo, respectively. For the EQ-5D-5L index, the LS mean changes from baseline at month 6 were |||||||||||||||||||||||||||||||||| and |||||||||| |||||||||||||||||| for the givosiran and placebo treatment groups, respectively. For the EQ-5D-5L Visual Analogue Scale (VAS), the LS mean change from baseline at month 6 was |||||||||||||||||||||||||||| and |||||||||||||||||||||||||||||| for the givosiran and placebo treatment groups, respectively. At month 6, the percentages of patients who reported that their status had improved from the start of the study through the PGIC were 88.9% and 37.1% for those in the givosiran and placebo treatment groups, respectively.

In terms of management of symptoms related to porphyria, the change in self-reported assessments of pain, fatigue, and nausea based on a numeric rating scale (NRS) were reported in Study 003. Post hoc non-parametric tests were used to evaluate daily worst pain following demonstration of a deviation from normality and failed statistical test using the analysis of covariance (ANCOVA) model. The medians of the area under the curve (AUC) for the change from baseline in the weekly mean score for daily worst pain over the 6-month treatment period were –11.5 (interquartile range [IQR] = –29.2 to 3.0) and 5.3 (IQR = –23.1 to 11.2) for the givosiran and placebo treatment groups, respectively. This indicated a decrease in the rating of daily worst pain for patients receiving givosiran and an increase for those receiving placebo. The treatment-group difference for rating of daily worst pain was –10.1 (95% CI, –22.8 to 0.9; P = 0.0455) for givosiran compared to placebo. At month 6, the changes from baseline in daily worst fatigue and daily worst nausea were also evaluated; a difference between treatment groups was not observed.

In Study 003, hemin was permitted only as a rescue medication for the treatment of acute porphyria attacks and was reported as days of hemin use. In patients with AIP, 54% of those in the givosiran treatment group and 23% of those in the placebo treatment group reported 0 days of hemin use over the 6-month treatment period. When compared to placebo, treatment with givosiran corresponded to a 77% rate reduction in days of hemin use based on a rate ratio of 0.23 (95% CI, 0.11 to 0.45; P < 0.001). Reported hemin use is consistent with the reduction in AAR reported for the primary outcome. The results for urinary levels of ALA and PBG were also consistent with the primary outcome. At month 6, urinary levels of ALA and PBG were lower among patients receiving givosiran than among those receiving placebo. This corresponded to a between-group difference of –19.14 mmol/mol creatinine (Cr) (95% CI, –26.04 to –12.24; P < 0.001) for ALA levels and –36.20 mmol/mol Cr (95% CI, – 49.71 to – 22.70; P < 0.001) for urinary PBG levels, both in favour of givosiran.

Opioid use, the PPEQ, and days of missed work or school were also reported as exploratory efficacy outcomes in Study 003. Reduced complications of AHP, hospitalization and health care use, and mortality were included in the systematic review protocol, but were not reported in the pivotal trial. However, attacks requiring hospitalization and health care use were incorporated in the composite definition of acute porphyria attacks, and mortality was reported as a safety outcome.

The primary and key secondary outcomes — AAR and change in urinary ALA levels — were analyzed by subgroups. The only subgroup analysis of interest to this review was by high or low historical AAR. The subgroup analyses were consistent with the results in the overall population. Additionally, a number of sensitivity analyses were conducted to account for variation in the primary end point based on reporting of porphyria attacks, which were all consistent with the primary analysis.

Harms Results

A summary of key safety results is provided in Table 2. In Study 003, 85% of patients with AIP experienced at least 1 adverse event (AE), with nausea, injection-site reaction, chronic kidney disease, fatigue, increase in alanine transaminase (ALT), and decrease in glomerular filtration rate more commonly reported among patients who received givosiran. Serious adverse events (SAEs) were reported more frequently among patients in the givosiran group (17%) than in patients in the placebo group (9%). Specific SAEs were infrequent, with the only SAEs reported by more than 1 person being chronic kidney disease (2 patients in the givosiran treatment group, 0 of those receiving placebo) and device-related infection (2 patients in the placebo treatment group, 1 of those receiving givosiran). A single patient randomized to receive givosiran withdrew from treatment due to an AE. The patient |||||||||||| discontinued treatment due to ALT elevation. No deaths were reported during the 6-month DB period of Study 003.

Motor neuropathy, hepatocellular carcinoma, injection-site reactions, transaminase elevation, and progression of renal impairment were included in the CADTH systematic review protocol as notable harms. As previously described, injection-site reactions and transaminase elevation were more common among patients receiving givosiran. Nerve compression and peripheral neuropathy were reported for motor neuropathy and were more common in the placebo group. There were no cases of hepatocellular carcinoma reported during the 6-month treatment period; however, 6 months may have not been a sufficient amount of time to observe this safety outcome.

Critical Appraisal

One of the key limitations of the internal validity of the study was the use of a composite outcome for the primary end point, which was the AAR based on attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home. A specific MID was not identified for the AAR, although the clinical experts indicated that in general, a reduction in attacks is clinically meaningful. The frequency of attacks was reported descriptively for each of the individual components, which highlighted some variability in the treatment benefit associated with givosiran compared to the composite outcome. Variation in clinical practice and potential for unblinding or deduction of treatment allocation may have also biased treatment, which would affect the results of the individual components. As a result, there is notable uncertainty regarding the ability to interpret the individual components of the composite end point; however, the estimates of effect for each of the components were in the same direction and were not expected to have affected the overall composite outcome. A number of secondary outcomes were included and controlled for multiplicity using a statistical testing hierarchy; however, a failed statistical test for the change from baseline in daily worst pain rendered all subsequent secondary outcomes unadjusted for multiple testing. This included the evaluation of nausea, fatigue, and HRQoL through the PCS of the SF-12, which were all outcomes that were clinically relevant and important to patients. Further, all other HRQoL outcomes were exploratory and without an identified disease-specific MID, which hindered the interpretability of the results. Regarding the generalizability of the pivotal trial results, 95% of the study population were patients with AIP (1 of the 4 types of AHP); however, according to the clinical experts, there is no biological a priori to expect that the observed results would not be generalizable to different AHP types. According to the sponsor, “the study was enriched for attack frequency to ensure the ability to measure a difference in treatment effect on the primary composite porphyria attack end point.”⁶ The higher historical frequency of attacks at baseline for patients included Study 003; the inclusion criterion of at least 2 attacks in the past 6 months at baseline may limit the generalizability of the results to patients with less frequent attacks. These patients represent most of the patients in clinical practice, according to the clinical experts consulted for this review.

Indirect Comparisons

Indirect treatment evidence for givosiran was not identified in this review.

Other Relevant Evidence

Study 001 and Study 002

Description of Studies

Study 001 was a 3-part, multi-centre, placebo-controlled, phase I study of the safety and tolerability of subcutaneous givosiran for the treatment of adults with AIP. Parts A, B, and C were single-ascending dose, multiple-ascending dose, and multidose in design, respectively. The adaptive design allowed for different dosing regimens and dose levels to be assessed based on new safety, tolerability, and pharmacodynamic (PD) data. In total, 40 patients with AIP who are chronic high excreters (CHEs) were randomized to parts A and B (n = 23), while those with AIP who had recurrent attacks were randomized to part C (n = 17). Data were summarized for patients who received givosiran 2.5 mg/kg (part A, n = 3; part C, n = 3). Patients in the 2.5 mg/kg cohort of part C had a mean of 14.7 attacks (SD = 18.9 attacks) in the 12 months before the study, and one-third of patients were on prophylactic hemin.

Study 002 (N = 16) is a multi-centre, open-label, phase I/II study of the long-term safety and tolerability of subcutaneous givosiran for treatment of adults with AIP who completed part C of Study 001. The results of an interim analysis of Study 002 were summarized for this review. Patients received givosiran 2.5 mg/kg monthly or 5.0 mg/kg monthly or every 3 months until the safety review committee assessed safety, tolerability, and efficacy data, and agreed that all patients would be transitioned to receive a 2.5 mg/kg dose. Treatment duration is estimated to be up to 36 months; the estimated total time in study with screening and baseline will be up to 44 months. Nearly all patients (93.8%) in Study 002 had at least 1 porphyria attack in the 12 months preceding the study, with a mean of 13.0 porphyria attacks (SD = 13.1 attacks) attacks during that time period. All patients had used hemin during an acute attack, and half had used it prophylactically.

Efficacy Results

In Study 001, patients had fewer attacks during the treatment and follow-up phase compared to the run-in of part C for all attacks, attacks requiring hospitalization, and attacks requiring urgent health care visits. The cohort receiving givosiran 2.5 mg/kg monthly had a mean AAR of 2.9 (SEM = 1.91) for composite attacks and a mean annualized rate of hemin use of 2.9 days (SEM = 1.44 days) during the treatment and follow-up period. The placebo group had a mean AAR of 16.7 (SEM = 4.97) for composite attacks and mean annualized rate of hemin use of 23.4 days (SEM = 9.9 days) during the treatment and follow-up period.

In Study 002, patients had fewer composite attacks during the treatment period compared to the run-in (n = 9 and n = 72, respectively) and fewer attacks requiring hospitalization, urgent health care visits, and treatment with hemin at home. The mean composite AARs were 17.0 (SEM = 3.5) and 1.2 (SEM = 0.4) for the run-in period of Study 001, part C and the treatment period, respectively. The mean rate for annualized hemin use was 33.1 days (SEM = 7.0 days) during the run-in period compared to 1.1 days (SEM = 0.6 days) during the treatment period of Study 002.

HRQoL was also assessed using the EQ-5D-5L in Study 001 and Study 002.

Harms Results

Most patients (66.7%) in part A of Study 001 and 100% of patients in both part C of Study 001 and in Study 002 experienced at least 1 AE. In part C, the most frequently reported AEs were abdominal pain, abdominal distension, nausea, and injection-site reaction. In Study 002, the most commonly reported AEs were abdominal pain, fatigue, nausea, and injection-site reaction. SAEs were reported in 100% of patients who received givosiran 2.5 mg/kg in part C of Study 001 and in 25% of those in Study 002. SAEs included functional gastrointestinal disorder, pyrexia, anaphylactic reaction, Clostridium difficile colitis, sinusitis bacterial infection, mental status changes, dyspnea, and deep vein thrombosis. There was 1 withdrawal due to adverse event (WDAE) in Study 002, and no deaths reported in the cohorts of interest.

Critical Appraisal

A key limitation of Study 001 was the single-blind, adaptive study design. Study 002 was limited by an open-label study design that selected for patients who were able to tolerate and adhere to treatment, which may bias the results in favour of givosiran. Both studies had small sample sizes; only a couple of patients were randomized to receive givosiran 2.5 mg/kg, the intended commercial dose, for a short duration of time.

Study 003 Open-Label Extension

Description of Studies

The 6-month, DB, placebo-controlled Study 003 was followed by an ongoing, open-label extension (OLE) period (Study 003 OLE). The OLE is expected to continue for 29 months⁷ and was designed to evaluate the long-term efficacy and safety of givosiran for treatment of adults with AIP. Patients who completed the DB portion of Study 003 (N = 94) were eligible to participate in the OLE phase. The baseline characteristics of the patients included in the OLE were similar to those reported for the DB treatment period, with a slightly higher mean historical AAR of 11.6 (SD = 9.0); prior prophylactic hemin use was reported by 44.2% of patients. Initially, patients received either 1.25 mg/kg or 2.5 mg/kg givosiran, but with protocol amendment 5 (after the cut-off date for the interim report), all patients received the latter dose.

Efficacy Results

After 18 months, the median follow-up, the mean number of attacks during givosiran treatment was 3.4 (SEM = 0.7), and appeared to be stable over time following treatment with givosiran. Mean (SEM) AARs for attacks requiring hospitalization, urgent health care visit, treatment with IV hemin at home, and treatment without IV hemin at home were |||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||| respectively. The mean (SEM) number of days of hemin use was |||||||||||||||| Urinary levels of ALA decreased from baseline by a mean (SD) of |||||||||||||||| mmol/mol and |||||||||||||||||| mmol/mol at month 12 and month 18, respectively. Urinary levels of PBG also decreased by an average (SD) |||||||||||||||||| mmol/mol and |||||||||||||||||| mmol/mol for the same time points from baseline. Patient-reported outcomes — including the SF-12, EQ-5D-5L, PGIC, and PPEQ, as well as daily worst symptom scores — were also reported during the OLE and were consistent with the results described in the DB treatment period.

Harms Results

Nearly all patients (94.8%) experienced at least 1 AE, with 32.5% reporting nausea, 27.3% injection-site reaction, 22.1% fatigue, 22.1% nasopharyngitis, and 19.5% headache. SAEs occurred in 24.7% of patients, with chronic kidney disease, device breakage, and urinary tract infection being reported by 2.6% of patients for each SAE. There was 1 WDAE. No deaths were reported.

Critical Appraisal

The OLE was subject to most of the limitations associated with the DB treatment period. Additional limitations of the extension period of Study 003 include the lack of a randomized comparison group and the open-label design, which may have influenced patients’ and clinicians’ perception of improvement (which, in turn, may be reflected in the patient-reported and safety outcomes). It is also worth noting that there was a dose change for those who initially enrolled under protocol amendment 3 and received givosiran 1.25 mg/kg. At month 13, patients who had inadequate disease control were able to increase their dose to 2.5 mg/kg from 1.25 mg/kg, and with protocol amendment 5, all patients were to receive givosiran 2.5 mg/kg (the intended commercial dose).

Study 005

Study 005 is an international program that will provide expanded access to givosiran to patients 12 years and older with AHP.⁸ It is ongoing. No additional information was available for this review.

Conclusions

One DB, placebo-controlled, phase III randomized controlled trial (RCT), Study 003, evaluated the efficacy and safety of givosiran compared to placebo. Included patients had a diagnosis of AHP and had experienced at least 2 porphyria attacks in the 6 months before screening. Over the 6-month treatment period, a 74% reduction in the rate of acute porphyria attacks was demonstrated with givosiran compared to placebo in patients with AIP based on a rate ratio of 0.26 (95% CI, 0.16 to 0.41; P < 0.001). Similar results were reported in all patients with AHP. The primary outcome of porphyria attacks was a composite outcome that included attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home. This outcome did not have a defined minimum clinically important difference. However, the clinical experts identified that in general, a reduction in acute attacks was clinically relevant. A treatment difference in favour of givosiran was also reported for a reduction in annualized days of hemin use and change from baseline in urinary ALA and PBG levels. This difference supports the beneficial direction of the primary outcome. Management of pain was an outcome important to patients, but the results did not demonstrate a clinically meaningful difference in daily worst pain scores. Reported HRQoL outcomes were also important to patients but were subject to limitations that hindered the interpretability of the results. With regards to the safety assessment, the majority of patients in the trial experienced at least 1 AE, with nausea, injection-site reaction, chronic kidney disease, fatigue, increase in ALT, and decrease in glomerular filtration rate more commonly reported among patients who received givosiran. Reported SAEs and treatment discontinuations due to AEs were infrequent, and no deaths were reported. Although the DB treatment period was limited to 6 months on treatment, evidence from the OLE of Study 003 demonstrated maintenance of treatment effect for up to 18 months and did not detect any new safety concerns. Gaps in the evidence of the efficacy and safety of givosiran were identified in patients with concomitant prophylactic hemin use and in long-term safety and efficacy data beyond 18 months of treatment.

Introduction

Disease Background

The porphyrias are a group of metabolic disorders caused by altered activities of enzymes within the heme biosynthetic pathway. Altered enzyme activity is usually due to an inherited mutation in the gene for that enzyme. Porphyrias are classified as hepatic or erythropoietic based on whether pathway intermediates first accumulate in the liver or in the bone marrow, respectively.⁹ The types of AHP include AIP, ADP, HCP, and VP. Each is characterized by a specific genetic mutation involved in the synthesis of heme in the liver; however, the types are clinically indistinguishable.¹

In the liver, ALAS1 is the first, as well as a rate-limiting, enzyme in the heme biosynthetic pathway.^1,10 Upregulation of hepatic ALAS1 is an important feature during exacerbations of AHPs because it leads to an increase of intermediates in the heme biosynthetic pathway, such as ALA and PBG.^3,10 In patients with AHP, the altered enzymatic activity within this pathway results in an excessive accumulation of the intermediate porphyrin precursors (ALA and PBG), which are neurotoxins that can precipitate an acute attack.^1,3 AHP can be exacerbated through induction of the ALAS1 gene by drugs and other factors, such as stress, fasting, alcohol use, smoking, and hormones. Induction of hepatic heme oxygenase, which degrades heme, can also lead to induction of ALAS1.¹⁰ During an acute attack, patients report a significant increase in pain that can gradually build over hours and last for a number of days.³ Some patients report pain in the chest, back, or extremities; however, severe abdominal pain is more typical. Long-term complications with recurrent acute attacks may include chronic pain, chronic kidney failure, and liver damage.

Overall, these porphyrias cause acute and chronic symptoms due to effects on the nervous system, with the most common being neuropathic abdominal pain.⁹ The motor, sensory, and autonomic nervous systems are often affected, resulting in autonomic changes (e.g., tachycardia, hypertension), muscle weakness, sensory loss, and pain in the back, chest, and extremities. Linenberger and Fertrin (2020) refer to severe abdominal pain, peripheral neuropathy, and central or autonomic nervous system manifestations as the “classic triad,”¹¹ but even these severe symptoms may be discounted because they mimic other diseases, and physical findings are often minimal.⁹

AIP is the most common AHP prototype, although the symptoms are common to all AHPs. HCP and VP may also present with blistering skin lesions. AIP, HCP, and VP are autosomal dominant, inherited disorders with low penetrance and female predominance. One review notes that at least 90% or more of heterozygotes for disease-causing mutations remain asymptomatic for life.¹ ADP is autosomal recessive and extremely rare, usually with an onset of attacks in the early teenage years, and all cases exhibit elevated erythrocyte zinc protoporphyrin. Very rare cases of homozygous AIP, HCP, and VP have a completely different phenotype from the other AHPs.⁹ AIP occurs in all races, but may be most common in northern Europeans; males and females are equally likely to inherit a PBGD gene mutation, but AIP is more likely to manifest in women than men; AIP affects adults and typically presents in a patient’s thirties or forties¹⁰ (although the clinical expert said twenties or thirties).

In the absence of information about prevalence and incidence of AHP that is specific to Canadians, the sponsor reported an estimate of the prevalence of AHP in Canada based on European data from Elder et al. (2013).¹² The estimated prevalence of AHP in the Canadian provinces (excluding Quebec) was |||| per million population²; however, it was noted by both the sponsor and the clinical experts consulted for this review that patients with AHP are underdiagnosed due to a lack of available treatment options and poor awareness of the disease. Of note, endemic populations have been described in British Columbia, Manitoba, and Nova Scotia. The combined prevalence of these disorders is estimated to be approximately 5 per 100,000, with instances of higher prevalence due to founder effects.¹

Diagnosis

Patients with AHP exhibit nonspecific and variable symptoms, which may result in them being seen by a number of specialists before receiving a diagnosis. The challenge in diagnosing patients with AHP is arriving at the point where one suspects the disease. The clinical experts consulted for this review stated that ultimately, expert clinician judgment in the context of laboratory and clinical reviews is the best way to identify AHP in patients. The clinical experts also noted that once AHP is suspected, confirming a diagnosis is relatively straightforward. A diagnosis is made based on serum, urinary, and fecal biochemical tests, which includes testing for elevated levels of PBG, ALA, and other porphyrins.³ Urinary ALA, PBG, and porphyrin excretion is notably increased during porphyria attacks; therefore, it is recommended that a patient presenting with acute symptoms without a history of acute porphyria have urinary PBG tested as soon as possible.¹⁰ The caveat is that the biochemical tests are rarely performed in-house, and obtaining results may take time. False-negative biochemical tests are also common; thus, genetic studies with symptoms in keeping with AHP and documented responsiveness to first-line therapy are sufficient to support the diagnosis, as noted by the clinical experts. Genetic studies may also be conducted. These are not required, but are often done.

Further testing is required to distinguish AIP, HCP, VP, and ADP. Substantial elevation of plasma or urinary PBG and ALA during an attack is common to AIP, HCP, and VP, but elevated levels of PBG are not observed with ADP.¹⁰

Standards of Therapy

According to the clinical experts consulted for this review, the goals of treatment for Canadian patients living with AHP are to reduce the frequency of acute porphyria attacks and hospitalizations and improve HRQoL and work-life productivity. Currently, there are no specific Canadian clinical practice guidelines for the treatment AHP; therefore, expert opinion is primarily used to guide treatment, as per feedback from the clinical experts consulted for this review. Of note, the clinical experts reported that currently in Canada, there is a lack of sufficient specialists with extensive experience in managing patients with AHP. As such, some patients may lack access to optimal management.

Treatment for AHP is targeted mainly at preventing acute attacks. The BIPNET recommends that recurrent acute attacks be managed through general measures and avoidance of precipitating factors, through treatment with GnRH analogues and prophylactic heme arginate, and/or through liver transplantation.⁴ Patient education about avoidance of triggering factors is recommended for the prevention of acute attacks. Triggers can include certain medications, smoking, alcohol, illicit drugs, and periods of fasting or low carbohydrate intake.^4,13 The clinical experts consulted for this review indicated that avoiding triggers enables patients to mitigate or self-manage porphyria attacks, and is sufficient for many patients. Changes in the level of progesterone during the menstrual cycle are 1 of the biggest contributors to an acute attack; therefore, the use of GnRH analogues to suppress the menstrual cycle can be an option for treatment in some patients, along with expert guidance. Side effects are significant with hormone therapy and may include depression, hot flushes, reduced libido, osteoporosis, and other menopausal symptoms. The BIPNET recommends that patients using hormone therapy be reviewed often and receive regular gynecological and annual bone-density exams.⁴ The clinical experts consulted for this review reported that long-term use of hormone therapy is typically not an option due to the severity of the side effects.

In Canada, hemin for injection is indicated for the amelioration of recurrent episodes of AIP temporally related to the menstrual cycle in susceptible women after initial carbohydrate therapy is known or suspected to be inadequate. Limitations of use include consideration of an appropriate period of carbohydrate loading (i.e., 400 g of glucose per day for 1 to 2 days) before use.¹⁴ Despite the narrow indication for the treatment of acute attacks, hemin is often considered for prophylactic use outside of the indication in patients who exhibit recurrent attacks with the preventive measures described earlier. The clinical experts consulted for this review indicated that most patients do not require treatment between attacks, but that a small number who have recurrent attacks will require prophylactic hemin weekly. Hemin is also associated with complications resulting from the requirement for venous access, iron overload, and difficulty withdrawing from treatment.⁴ For patients who are unresponsive to treatments for recurrent attacks, are frequently hospitalized, and exhibit diminishing HRQoL, a liver transplant may be considered, and is highly effective in many cases without advanced motor neuropathy.¹⁵

Despite the preventive measures described, acute porphyria attacks can still occur. Typically, mild attacks may be managed at home with increased carbohydrate intake and analgesic medication, but moderate to severe attacks require hospitalization so that IV hemin can be administered. According to the clinical experts, IV glucose can be used if hemin is not available, but it is not as effective. Additionally, patients in hospital are treated medically for symptoms (pain, nausea, and vomiting) and are closely observed, with their salt and water balance monitored. The prognosis is usually good if the disease is recognized and if treatment and preventive measures are started before severe nerve damage has occurred.¹³ Hemin is the most effective treatment for acute attacks. As described earlier, it is administered intravenously. Intravenous glucose is usually given as a 10% solution while hemin is being prepared, but should not delay the administration of hemin.

Drug

Givosiran is a double-stranded, small, interfering RNA that causes the degradation of ALAS1 mRNA in hepatocytes through RNA interference, reducing elevated levels of liver ALAS1 mRNA. This leads to reduced circulating levels of neurotoxic intermediates ALA and PBG, 2 factors associated with attacks and other disease manifestations of AHP.⁵

In Canada, givosiran is indicated for the treatment of AHP in adults.⁵ It is available as a solution for subcutaneous injection (189 mg givosiran/mL) and does not require additional reconstitution or dilution before administration. The recommended dose is 2.5 mg/kg once monthly, based on body weight.

Givosiran underwent an expedited review at Health Canada and has been requested for reimbursement as per the approved Health Canada indication.

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

About the Patient Group and Information Gathered

CADTH received 2 patient group submissions for this review: 1 from CAP and 1 from APF.

CAP is a national, voluntary charity whose mission is to deliver evidence-based information and support to patients with porphyria, their families, health care providers, and the general public. The group also aims to achieve standards and evidence-based comprehensive care for all people with porphyria throughout their lifespans.

APF is dedicated to improving the health and well-being of individuals and families affected by porphyrias. The group provides programs to raise awareness and educate health care professionals and the general public in 76 countries around the world. APF also helps to establish support groups and mentorship of the next generation of experts and supports porphyria research. Of its international members, more than 300 are located in Canada. More information about APF can be found at http://www.porphyriafoundation.org/.

CAP created and distributed a survey to its members via email and social media platforms for a 2-week period in February 2021. The survey was available in English and French and was restricted to Canadian patients and caregivers who had experience with AHP. In total, 22 patients and 4 caregivers responded with diagnoses of AIP (n = 20), HCP (n = 3), and VP (n = 3). Of the patients represented in the survey, 23 were female and 3 were male; respondents were from Alberta (n = 10), British Columbia (n = 7), Ontario (n = 6), Manitoba (n = 2), and Quebec (n = 1). CAP also requested support from the British Porphyria Association, which shared interviews from 3 individuals who had received givosiran. While preparing this submission, CAP discussed and reviewed it with members of the Canadian Hemophilia Society and Network of Rare Blood Disorder Organizations.

APF used its social media platforms and online newsletters to connect with Canadian patients about their experiences with porphyria and collected responses by telephone and email. Some of the responses in the APF submission were collected during an Alnylam Patient Advisory Board meeting. Twelve individual patient submissions were collected from Canadians in Ontario (n = 6), British Columbia (n = 2), Quebec (n = 2), Manitoba (n = 1), and Nova Scotia (n = 1).

Disease Experience

Respondents in both submissions reported experiencing the following symptoms, among others: pain, fatigue, nausea, weakness, paralysis, neuropathy, seizures, anxiety, and depression. More than 80% of patients who responded to the CAP survey had experienced symptoms at least once a month, with many reporting that these symptoms occurred more than 20 days per month. The group also reported that 86% of respondents had at least 1 attack in the past year and 36% had at least 10. Furthermore, 55% of patients had gone to the emergency room at least once in the past year due to an attack, while 18% had gone at least 10 times. One of the most frequently reported symptoms is pain. An APF survey respondent stated:

I have to have a lot of pain meds. My desire is to free from them but I cannot be free from pain meds without being free of attacks. The pain is too much for a person to endure. A doctor with AIP said that it was not compatible with life to have attacks with no pain treatment.

Porphyria attacks can also prevent patients and caregivers from being able to work, lead to poorer quality of life, and negatively affect relationships. For instance, 59% of CAP respondents stated that it had affected their career and financial well-being. Respondents said, “It has destroyed and robbed me of my education, livelihood, finances and health” and “I usually have to make arrangements at work on a short notice when my wife is in attack and have to take her to the hospital and have to work from hospitals many times.”

A respondent who is both a patient and caregiver described their experience as follows:

Because of AIP, I had to give up my career as a Registered Nurse. The symptoms, including the anxiety, was so severe and debilitating that I could not longer function well in daily life. …now I am the caretaker advocate for my brother. He has been on disability for 15 years and is worsening. Two lives are unable to be productive because of porphyria.

Both groups emphasized the negative effect that porphyria had on their daily lives and mental health. In the CAP survey, 64% reported that the disease had significantly affected their mental health, while 82% said they experienced anxiety or depression monthly. Patients described their situations:

• “There is a lot of grieving, loss of work, social losses and having to accept the illness.”

• “The psychiatric symptoms are debilitating. Attacks start with a distinct foggy head difficulty processing thought, inability to make decisions, memory issues, high anxiety, restlessness withdrawal, obsessive thoughts paranoia unable to stop mind from racing causes insomnia.”

Moreover, 94% of patients and 100% of caregivers from the CAP group felt that AHP had affected their overall well-being, while significant impacts on family life and social life were noted by 50% and 41% of respondents, respectively. The patient input submissions described how experiencing symptoms and trying to avoid triggers could strain social relationships and make it difficult to care for families. Respondents were also concerned about the possibility of passing the condition on to their children. The APF submission stated: “Porphyrias are known as the ‘little imitator’ as it has a host of generic symptoms making it hard to identify, particularly because it cannot be diagnosed without porphyria specific tests. It will not show on normal tests.”

Consequently, patients often face difficulties when trying to obtain a diagnosis or access treatment for their attacks. One caregiver wished that emergency room doctors were more aware of the disease:

This could have helped doctors to come to right diagnosis earlier, and to avoid unnecessary tests and treatments. Especially in AIP, once an episode or attack starts, if the right medications/treatment are not given soon in the crises (and unsafe medications are not avoided), the consequences are devastating for the patient.

The caregiver also noted that the patient was initially misdiagnosed and given unnecessary tests and treatments.

Experiences With Currently Available Treatments

Most patients from the CAP survey have attempted to manage porphyria by avoiding attack triggers (86%), adopting a high-glucose diet (82%), using hematin (hemin, Panhematin) (27%), using GnRH analogues (9%), or taking additional medications to treat impacts like pain (50% opioid, 41% non-opioid), sleep problems (50%), anxiety (41%), and depression (32%). Although respondents felt that these treatments could help with symptoms, they noted that the treatments are not without side effects, such as weight gain, diabetes, iron overload, induced menopause, and others specific to the additional medications.

Both groups described hematin as being effective at reducing the length and/or severity of intermittent attacks, but they also acknowledged its drawbacks. For example, it has been associated with platelet aggregation, and frequent use can cause iron overload and damage to veins. Furthermore, it is often infused through a peripherally inserted central catheter line or port, which can become blocked, requiring surgical replacement. Other side effects that were mentioned include headaches and low-grade fever that resolve within 2 days. One individual said, “Panhematin saved my life. But as patients, it is much more than just saving our lives…There is not quality unless attacks are prevented.”

A caregiver described how weekly hemin treatments have helped their wife:

[S]he is in fact feeling better in terms of gaining her strength doing everyday activities. It has also reduced the amount of pain killer by 25% got rid of most of the nerve medications. It has also reduced the number of hospitalizations from 12 a years to 2-3 and has given her some independence….

This caregiver noted that while there were benefits, they felt as though their wife could be developing tolerance to the treatment and that the weekly schedule made working and planning around infusions somewhat challenging.

Patients described facing difficulties when trying to access treatment in hospitals. They reported that there is a lack of awareness of AIP among health care workers, and that the recommended treatment is not readily available. The patient groups also noted that delays and non-optimal treatments can exacerbate attacks and lead to permanent damage.

Improved Outcomes

Ideally, respondents would like a cure for porphyria. However, they believe a realistic goal is a treatment that prevents attacks and reduces symptoms, particularly pain, nerve damage, and paralysis. Patients and caregivers would like to see additional options that are more effective, cause fewer side effects, present an easier mode of administration, can be administered outside of a hospital, and lead to improvements in quality of life. Other limitations to treatment access that were identified include the need for travel, the requirement for venous access, and the lack of access to specialists and proper diagnostic testing. Respondents raised concerns about financial assistance and insurance coverage, and said they would like health care providers to have greater knowledge of the condition. They felt strongly that there should be financial assistance to help alleviate the burden of treatment costs.

A parent stated:

Hematin made a difference in the life of my daughter in that it stopped her attacks after they began, but it did not stop her attacks from happening. We need a treatment to stop the attacks from occurring, because attacks are so excruciating and can lead to paralysis and death.

When asked what would be important to discuss with their physicians about treatment options, APF survey respondents mentioned available treatments, benefits and drawbacks of treatments, safety and efficacy, side effects, patient monitoring, impact on quality of life, frequency of administration, and treatment cost.

Experience With Drug Under Review

Although none of the CAP survey respondents had received givosiran (Givlaari), with help from the British Porphyria Association, the group was able to share 3 patients’ experiences. Two had accessed the drug for 2 years through the ENVISION trial. All 3 patients described givosiran as “life-changing.” More specifically, the drug had been able to reduce or eliminate attacks, pain and pain medications, fatigue, nausea, and anxiety. It improved patients’ physical health, appetite, sleep, and concentration. In terms of social and work impacts, the respondents highlighted greater independence, the ability to attend and contribute more at school and work and toward family life, and the tendency to be more optimistic about planning their futures.

Because givosiran is a monthly injection, APF suggested that it is an easier treatment to receive compared to hemin infusions. The group noted a few side effects, such as injection-site reactions, allergic reactions, and nausea, while those reported by Alnylam include liver and kidney problems. APF also noted that patients may be able to access givosiran through the Alnylam Assist program. Successful treatment can reduce costly hospital admissions and the burden on patients and families. Respondents who had received the medication felt that their attacks were treated; they reported both fewer attacks and less severe attacks. Moreover, despite still experiencing attacks and side effects while receiving the drug, many wanted to continue with it. APF also stated that patients receiving givosiran may be prescribed hematin to stop an attack quickly as it is occurring. However, there may be exceptions to this approach that the patient group was not aware of.

The following quotes illustrate patients’ experiences with givosiran:

• “My greatest hope was to return to work. That dream is coming true as I’m almost finished school, a feat I could never have done without Givlaari. It gave me my life back.”

• “I was frightened for my life, because I was at the point that I no longer had veins for Panhematin infusions. Givlaari was my last chance. I was on the Givlaari trials and had immediate success. Some people say it takes a while to gain effect, but for me, I stopped having attacks after my first shot of Givlaari. Life changed. I could return to work. I could return to life.”

• “I was totally debilitated. I suffered terribly and was repeatedly hospitalized. I finally lost my job as I was unable to go to work. My life was a shambles until I was given Givlaari. Life has turned around for me. I want other people to have my experience.”

In its submission, APF stated that Alnylam provides free diagnostic DNA testing for many patients who fit the criteria, and that insurance may also cover testing.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise regarding the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol; assisting in the critical appraisal of clinical evidence; interpreting the clinical relevance of the results; and providing guidance on the potential place in therapy). In addition, as part of the givosiran review, a panel of 3 clinical experts from across Canada was convened to characterize unmet therapeutic needs, assist in identifying and communicating situations where there are gaps in the evidence that could be addressed through the collection of additional data, promote the early identification of potential implementation challenges, gain further insight into the clinical management of patients living with a condition, and explore the potential place in therapy of the drug (e.g., potential reimbursement conditions). A summary of this panel discussion is presented in this section.

Unmet Needs

One of the major goals in the management of AHP is to reduce the frequency of AHP attacks. According to the clinical experts consulted for this review, currently, most patients with recurrent attacks will continue to have these because there is no indicated treatment for prevention. The experts noted that while prophylactic hemin can be used to reduce the rate of AHP attacks (with case reports of improvement), the use of prophylactic hemin is outside of the Health Canada–approved indication and has not been studied well. The experts described further issues with prophylactic hemin. These include a serious side effect of iron overload, some minor but nuanced infusion reactions, and the need for a central venous catheter for regular prophylactic use. The experts stated that there is a need for treatments that are better tolerated due to the substantial risk and requirement for central venous access associated with the current IV treatments. Further, formulations are needed to improve convenience and compliance because IV access requires hospital-based care. The experts noted that this is particularly problematic for patients who do not live near large health care centres, which are often the only centres with access to specialized care. GnRH may also be used to prevent AHP attacks, but is not approved for prolonged use and is associated with side effects and loss of bone mineral density (given that it leads to a drug-induced menopause state).

Place in Therapy

The clinical experts described givosiran as the first treatment approved for prevention or prophylaxis against acute attacks; however, givosiran is not the first treatment that addresses the underlying disease process, given that hemin also achieves this goal. As described, while hemin is used as a regular weekly treatment to prevent attacks, this is not an approved indication, as noted by the clinical experts. The mechanisms of action for givosiran and hemin are different, but both decrease ALA and PBG, which are the intermediates that cause AHP attacks and symptoms, as described by the clinical experts. The experts felt that the 2 treatments can complement each other; however, givosiran is used to prevent AHP attacks rather than to treat acute attacks, as hemin does. The clinical experts also noted that if an attack occurred despite treatment with givosiran, then hemin could still be used, and its effectiveness is unlikely to be diminished.

As per feedback from the clinical experts on this review, givosiran would be used in people who have recurrent attacks; there is no evidence to support its use in asymptomatic individuals or during acute attacks. Also, the experts stated that givosiran would not be used as a first-line treatment or to treat the first AHP attack. The clinical experts indicated that other approaches to treatment should be tried for patients with AHP before givosiran is recommended. These approaches include avoidance of triggers, such as smoking or fasting, administration of IV dextrose 10% in water, and/or hemin use; however, the experts also reiterated the limitations of hemin use described under unmet needs, particularly in the prevention of attacks. According to the clinical experts, most patients living with AHP will either have mild symptoms or 1 attack without frequent recurrence; prevention and avoidance of triggers is helpful for these cases.

Therefore, the experts expected givosiran to provide an alternative therapy for a small subset of patients with frequent or recurrent attacks who would otherwise require hospitalization and hemin administration. They recommended that givosiran be reserved for patients with recurrent symptoms or flares that consistently affect their HRQoL. Givosiran was also described by the experts as an appropriate treatment for patients who qualify for hemin prophylaxis but cannot adhere to treatment due to toxicity or lack of convenience.

Patient Population

The clinical experts described the diagnosis of AHP as relatively straightforward. It is obtained through biochemical tests for levels of urinary ALA, PBG, and other porphyrins. These tests were described by the experts as being very specific, but subject to false negatives, which makes clinician judgment an important factor in the diagnosis. The experts noted that AHP can be confirmed through genetic testing, but this testing is not required. They noted that the challenge with identifying patients with AHP is arriving at a suspicion of AHP because of the variable presentation of disease and symptoms that are not specific to the disease. The clinical experts also stated that it is highly unlikely that a clinician would incorrectly diagnose a patient with AHP. The greater concern is underdiagnosis of the disease due to the previously described challenges associated with identifying patients based on physiological and pre-analytical variables.

The clinical experts indicated that the group of patients most in need would be those with frequent or recurrent severe attacks, and that the frequency and severity of attacks are the most important factors when considering the use of givosiran. They noted that severity is heavily based on the number of severe attacks in a given period, but that it was challenging to further define frequency of attacks because the presentation of disease varies among patients. Therefore, the clinical experts felt it was most appropriate to assess the severity of disease on a case-by-case basis using clinical judgment. In addition to frequency and severity of attacks, hemin use and comorbidities were also considered important factors in the decision to use givosiran for patients. More specifically, the clinical experts noted that hemin use was an indicator of disease that was not well-controlled, and that the presence of comorbidities resulting from AHP would be considered a priority, with the goal of preventing further complications. According to the clinical experts, laboratory porphyria levels, subtypes and genetic testing do not predict disease severity because there is variable penetrance and variation in clinical course.

The clinical experts also stated that givosiran is least suitable — and would not be considered — for patients who are asymptomatic, are identified with AHP through family history and never develop clinical symptoms, have not been therapeutically challenged with initial lines of therapy, or whose clinical symptoms are mild or characterized by infrequent attacks. Additionally, the experts felt that givosiran would not be suitable in populations that have not been studied, including children, adults aged 65 years and older, pregnant or breastfeeding individuals, or patients with moderate to severe liver disease or significant elevation in transaminases.

Assessing Response to Treatment

The clinical experts stated that it is not possible to identify the patients most likely to exhibit a response to treatment with givosiran based on the current evidence, biochemical tests, or patient characteristics.

The clinical experts described the following outcomes as those that are used to determine response to treatment in clinical practice: reduced attack rate, reduced hospitalization, reduced need for hemin, frequency of neurovisceral flares, and improved patient-reported outcomes, such as daily symptoms, HRQoL, and work-life productivity.

An improvement in HRQoL and a reduction in frequency of attacks, hospitalizations, and attacks requiring hospitalization or hemin use would be considered clinically meaningful responses to treatment by the clinical experts consulted on this review, and would be considered when making a decision to continue treatment with givosiran. One of the experts also noted that the following would be characterized as a clinically meaningful response to treatment: a reduction in, improvement of, or stabilization of psychiatric manifestations and neurovisceral symptoms (abdominal or back pain, axonal neuropathy) in addition to the attainment of major motor milestones and the ability to perform activities of daily living.

The clinical experts suggested that patients be assessed for response to treatment every 6 months or annually. One suggested that patients be assessed more frequently (monthly) during the acute phase (described as the first 3 months), followed by evaluations every 3 months, then every 6 months once stabilization is achieved. All of the experts agreed that 1 year would be a sufficient amount of time in which to assess response to treatment; however, the variable presentation of disease (such as yearly fluctuations in attack frequency) was noted as a limitation to the certainty of this assessment.

Discontinuing Treatment

All of the clinical experts agreed that the frequency of attacks is the most important factor in the decision to continue treatment with givosiran. HRQoL and the frequency of attacks leading to hospitalization were also noted by the experts as being important for consideration. In general, the clinical experts felt that patients treated with givosiran would continue with treatment until they encountered a reason for discontinuation, such as safety concerns, including severe local reactions, anaphylaxis, severe gastrointestinal symptoms, severe fatigue, or increase in liver function tests and/or creatinine. Discontinuation of treatment may also be considered with deterioration of HRQoL, safety concerns that outweigh the perceived benefit due to reduced symptoms or attacks, or a lack of response to treatment. Of note, it was challenging for the clinicians to specifically define response to treatment due to the heterogeneous nature of AHP among patients, but an increased or similar rate of attacks while on treatment was noted as a sign that treatment may not be working. Feedback from the clinical experts also indicated that menopause would be a potential reason to trial treatment discontinuation in patients with stable disease. They also noted that if attacks recurred following discontinuation, restarting treatment with givosiran would be a possibility.

Prescribing Conditions

The clinical experts indicated that givosiran is administered by a subcutaneous injection and can be given in a community nursing clinic, hospital outpatient clinic, or any infusion clinic, whether it is a community or specialty clinic. The experts felt that ideally, administration of givosiran should be overseen by a specialist responsible for the diagnosis, treatment, and monitoring of patients receiving givosiran. According to the clinical experts, patients with AHP should be monitored by a clinic specialized in the management of porphyria, but because it is a rare condition, this is not always available. General internal medicine and hematology specialists would be most likely to manage these patients, but specialists in hepatology, nephrology, and neurology were also mentioned as relevant.

Clinician Group Input

CADTH did not receive any input from clinician groups for this review.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may affect their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 3.

Clinical Evidence

The clinical evidence included in the review of givosiran is presented in 2 sections. The first section, the Systematic Review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those studies that were selected according to an a priori protocol. The second section includes sponsor-submitted, long-term extension studies and additional relevant studies that were considered to address important gaps in the evidence included in the systematic review.

Systematic Review (Pivotal and Protocol-Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of givosiran 2.5 mg/kg once monthly for the treatment of AHP in adults.

Methods

Studies selected for inclusion in the systematic review include pivotal studies provided in the sponsor’s submission to CADTH and Health Canada as well as those meeting the selection criteria presented in Table 4. Outcomes included in the CADTH review protocol reflect those considered to be important to patients, clinicians, and drug plans.

The literature search was performed by an information specialist using a peer-reviewed search strategy. The literature search for clinical studies was performed by an information specialist using a peer-reviewed search strategy according to the PRESS Peer Review of Electronic Search Strategies checklist (https://www.cadth.ca/resources/finding-evidence/press).¹⁶

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946‒) through Ovid and Embase (1974‒) through Ovid. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concept was Givlaari (givosiran). Clinical trials registries were searched: the US National Institutes of Health’s clinicaltrials.gov, Health Canada’s Clinical Trials Database, and the European Union Clinical Trials Register.

No filters were applied to limit the retrieval by study type. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. See Appendix 1 for the detailed search strategies.

The initial search was completed on March 23, 2021. Regular alerts updated the search until the meeting of the CADTH Canadian Drug Expert Committee on July 21, 2021.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the CADTH Grey Matters: A Practical Tool For Searching Health-Related Grey Literature checklist (https://www.cadth.ca/grey-matters).¹⁷ Included in this search were the websites of regulatory agencies (US FDA and the European Medicines Agency). Google was used to search for additional internet-based materials. See Appendix 1 for more information on the grey literature search strategy.

These searches were supplemented through contacts with appropriate experts. In addition, the manufacturer of the drug was contacted for information regarding unpublished studies.

Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the predetermined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer were acquired. Reviewers independently made the final selection of studies to be included in the review, and differences were resolved through discussion.

Findings From the Literature

A total of 1 study was identified from the literature for inclusion in the systematic review (Figure 1). The included study is summarized in Table 5. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Description of Studies

Figure 2 outlines the study design. Study 003 was a multi-centre, placebo-controlled, DB phase III study designed to evaluate the efficacy and safety of givosiran administered once monthly in patients with AHP. Its primary objective was to evaluate the effect of givosiran compared to placebo in terms of the rate of porphyria attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home in patients with AIP. Study 003 was conducted internationally, and 1 of the 36 participating study centres was located in Canada.

An interactive response system was used to assign patients to a treatment and to maintain blinding; members of the study team did not have access to unblinded data until the end of the 6-month DB period. A total of 94 patients were randomized in a 1:1 ratio to receive 2.5 mg/kg givosiran or placebo administered subcutaneously. Randomization was stratified by AHP type: AIP with genetic evidence of mutation in the HMBS gene versus non-AIP (HCP, VP, ADP, or any AHP without an identified mutation in a porphyria-related gene). Patients with AIP were further stratified by prophylactic hemin use at screening as well as by each patient’s historical AAR. More specifically, patients who used hemin prophylactically before study entry were stratified by their historical porphyria AAR over the 12 months before randomization, based on having fewer than 7 attacks versus 7 attacks or more. Patients who were not using hemin prophylactically before study entry were stratified by their historical porphyria AAR in the previous 12 months up until randomization, based on fewer than 12 attacks versus 12 attacks or more.

Patients participated in a screening period for up to 2 months before randomization. During this period, patient history and ALA and PBG levels were collected for study eligibility purposes. Patients who did not meet the study eligibility criteria were able to undergo rescreening. The 6-month DB treatment period followed, with an OLE period of up to 29 months (summarized in Other Relevant Evidence later in this section). A 1-month follow-up period concluded the study.

The study was enriched for attack frequency to ensure the ability to measure a difference in treatment effect on the primary composite porphyria attack end point.

Patients who discontinued from the study drug or withdrew from the study were not replaced.

Figure 2: Design of Study 003

AHP = acute hepatic porphyria; Am ≤ 2 = original protocol, protocol amendment 1, and protocol amendment 2; Am 3 = protocol amendment 3.

Source: Clinical Study Report.⁶

Populations

Inclusion and Exclusion Criteria

To be eligible for inclusion in Study 003, patients had to be at least 12 years old with a documented diagnosis of AIP, HCP, VP, or ADP. The diagnosis was based on clinical features, evidence of elevated urinary or plasma PBG or ALA (at least 4 × the upper limit of normal [ULN]) within the year before screening, and genetic evidence of mutation in a porphyria-related gene (AIP = mutation in the HMBS gene; HCP = mutation in the CPOX gene; VP = mutation in the PPOX gene; ADP = mutation in the ALAD homozygous or compound heterozygous genes). Patients with genetic testing that did not identify a mutation in a porphyria-related gene were considered eligible if they had both clinical features and diagnostic biochemical criteria consistent with AHP, as outlined in Figure 3. Patients were also required to have active disease (at least 2 composite porphyria attacks; i.e., attacks requiring hospitalization, urgent health care visit, or treatment with IV hemin at home) within 6 months before screening and be willing to abstain from prophylactic use of hemin during the trial.

Patients were excluded from Study 003 if they had elevated levels of ALT (≥ 2 × ULN), total bilirubin (> 1.5 × ULN), or an international normalized ratio greater than 1.5; impaired renal function (eGFR < 30 mL/min/1.73 m²); or a history of allergies, infections, or malignancy. Additionally, patients were excluded if they had a history of recurrent pancreatitis or a major surgery planned within 6 months, or if they were pregnant, breastfeeding, or planning to become pregnant.

Figure 3: Biochemical Diagnosis of AHP in the Absence of Identified Mutation in a Porphyria-Related Gene

ADP = aminolevulinic acid dehydratase-deficient porphyria; AHP = acute hepatic porphyria; AIP = acute intermittent porphyria; HCP = hereditary coproporphyria; ULN = upper limit of normal; VP = variegate porphyria; WNL = within normal limits.

Source: Clinical Study Report.⁶

Baseline Characteristics

A summary of baseline characteristics for Study 003 is provided in Table 6. The data from Study 003 are presented using 2 datasets: 1 for patients with AIP and 1 for all patients with AHP. The dataset for patients with AIP included only patients who were identified with the AIP subtype. All patients with AHP, regardless of subtype (i.e., patients with AIP, ADP, HCP, VP, and patients with AHP without an identified mutation in a porphyria-related gene) were included in the “all patients with AHP” dataset. Where relevant, the data from Study 003 will be described using these 2 datasets for the remainder of the report.

Patients with AIP who were randomized in Study 003 were between the ages of 19 years and 65 years, with a mean age of 37.3 years to 40.7 years. Most were female (89% to 91%) and White (77% to 80%); 35% to 40% resided in North America. The mean number of years since diagnosis was 8 years to 11 years, and between 40% and 44% of patients had prior experience with prophylactic hemin. During the 6 months before screening for Study 003, the median number of porphyria attacks requiring hospitalization, urgent health care visit, or hemin use at home was between 3 and 4, with a range of 0 to 25. Using the composite definition of porphyria attacks, the median historical AAR was 8 attacks (range = 4 to 34) and 8 attacks (range = 0 to 46) in the givosiran and placebo treatment groups, respectively. Detail regarding the reason for including a patient with a minimum AAR of 0 in the 6 months before screening in the placebo treatment group was not provided. While not having a porphyria attack, between 48% and 56% of patients reported having chronic symptoms, and 28% to 30% reported chronic opioid use.

In general, the baseline characteristics were similar between treatment groups. Among patients with AIP, the givosiran treatment group had a greater number of years since their diagnosis, a greater proportion of patients with a prior hemin prophylaxis regimen, and a smaller proportion of patients with prior chronic symptoms when not having attacks.

Baseline characteristics in patients with AHP were similar to those reported for patients with AIP. Patient comorbidities are summarized in Table 7.

Interventions

Patients were randomized to givosiran or placebo. Those randomized to givosiran received 2.5 mg/kg once monthly during the 6-month DB period of Study 003. Givosiran was supplied as a sterile solution in water for subcutaneous injection and administered by a qualified and authorized health care professional into patients’ abdomens, upper arms, or thighs. Patients were observed for at least 20 minutes following the injection. Patients randomized to placebo received sodium chloride 0.9% weight/volume for subcutaneous injection. The placebo was administered using identical packaging and product volume. To maintain blinding, syringes were masked before the withdrawal of givosiran or placebo from a masked vial.

Modifications to the dose of givosiran were not permitted during the 6-month DB period, with the exception of changes due to ALT elevation. A downward titration to 1.25 mg/kg once monthly (from givosiran 2.5 mg/kg once monthly) for patients who were withheld from treatment due to elevated ALT was introduced in accordance with protocol amendment number 2.

Patients were permitted to miss 1 dose of the study drug within the 6-month treatment period. If more than 1 dose was missed, a decision regarding continuation of treatment was made at the discretion of the investigator and medical monitor.

Patients recorded concomitant medication use through an electronic case report form. A list of concomitant medications reported during the 6-month DB period of the ENVISION trial is available in Table 12. Hemin prophylaxis was not permitted during the study, but patients were allowed to receive hemin for the treatment of acute attacks if clinically indicated. Hemin used for acute attacks was reported as a concomitant medication. Patients who were receiving treatment with a GnRH analogue at screening were permitted to enrol if they remained on GnRH treatment throughout the 6-month DB period. The sponsor reported that initiation of treatment with GnRH during the study was discouraged, but the extent of GnRH use was not reported. Analgesic usage for the management of porphyria and porphyria attacks was permitted based on clinical judgment and either reported in the patient’s e-diary or as a concomitant medication, if taken at a health care facility. Lastly, standard vitamins and topical medications were permitted; however, topical steroids were not allowed to be applied anywhere near the injection site unless medically indicated.

Outcomes

A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trials included in this review is provided in Table 8. These end points are further summarized here. A detailed discussion and critical appraisal of the outcome measures is provided in Appendix 4.

Porphyria Attacks

In Study 003, the primary composite outcome was the annualized rate of porphyria attacks, which included attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home in patients with AIP during the 6-month DB period of the study. All porphyria attacks, which included all 4 definitions of porphyria-related attacks, were also reported.

Porphyria attacks were defined as acute episodes of neurovisceral pain in the abdomen, back, chest, extremities, or limbs with no other medically determined cause that required treatment with IV dextrose, hemin, carbohydrates, analgesics, or other medications (e.g., antiemetics at a dose frequency beyond that used in the patient’s usual daily management). Patients or caregivers were to record any potential porphyria attacks for the duration of the study when they occurred using the provided electronic diary, by telephone, or by email to the study site or investigator. Of note, patients or caregivers were trained on the use of the electronic diary at screening, and complete instructions were available at study sites in a study manual. Health care professionals were also able to notify the study centre of porphyria attacks. All potential porphyria attacks were adjudicated by the investigator. If an event was determined to be inconsistent with a protocol-defined attack, an alternative reason was recorded, such as duplicate entry, AE, or other.

Further, the ENVISION trial used the following definitions to identify non-overlapping components of events relating to porphyria attacks:

• attack requiring hospitalization: admission to an inpatient unit or a visit to an emergency department that resulted in a stay of at least 24 hours

• attack requiring urgent health care visit: urgent, unscheduled visit to a physician’s office or practice, an infusion centre, or an emergency department visit that did not meet the criteria for hospitalization

• attack requiring IV hemin administration at home: home was any location that did not meet the criteria for hospitalization or an urgent health care visit

• attack at home not requiring IV hemin.

No literature was identified that assessed the annualized rate of porphyria attack for validity, reliability, or responsiveness in patients with porphyria. No MID was identified in populations with AHP.

Health-Related Quality of Life

In the ENVISION trial, HRQoL was assessed through the SF-12, EQ-5D-5L, and PGIC.

12-Item Short Form Health Survey

The SF-12 is a generic, patient-reported measure of HRQoL based on the 36-item version of the survey (SF-36). Patients answer based on a 4-week recall period. Each of the 12 items fall into 1 of 8 health scales, including: physical functioning (PF) (2 items), RP (2 items), BP (1 item), general health (GH) (1 item), vitality (VT) (1 item), social functioning (SF) (1 item), role emotional (RE) (2 items), and mental health (MH) (2 items).²⁰ The first 4 scales (PF, RP, BP, and GH) make up the PCS, while the latter 4 (VT, SF, RE, and MH) fall under the MCS. The PCS and MCS correspond to the physical and psychological burden of disease, respectively. The component summaries are standardized to have a mean of 50 and an SD of 10 based on the general US population, and scores can range from 0 to 100, with higher scores reflecting better HRQoL.²⁰ Validity and reliability have been demonstrated in a diverse population; however, no literature was identified that assessed the SF-12 for validity, reliability, or responsiveness in patients with porphyria. An MID was not identified for this population, either; however, MIDs for the SF-12 PCS and MCS have been estimated based on a study of 458 patients with lower back pain.²¹ Overall, an improvement of at least 3.29 on the PCS and 3.77 on the MCS would be clinically meaningful in patients with low back pain.²¹ In the ENVISION trial, the component summary scores and individual item scores were reported as a change from baseline.

EuroQol 5-Dimensions 5-Levels Questionnaire

The EQ-5D-5L is a generic, self-reported, HRQoL instrument developed by the EuroQol Group that is applicable to a wide range of health conditions and treatments.²² The EQ-5D-5L consists of a descriptive system and the EuroQol VAS.

The descriptive system comprises 5 dimensions: mobility, self-care, usual activities, pain or discomfort, and anxiety or depression. Each dimension is answered based on 5 levels (1 = no problems; 2 = slight problems; 3 = moderate problems; 4 = severe problems; and 5 = extreme problems or unable to perform, the worst response in the dimension).²² Respondents choose the level that reflects their health state for each of the 5 dimensions. In total, there are 3,125 possible unique health states defined by the EQ-5D-5L, with 11111 and 55555 representing the best and worst health states, respectively. Results from the EQ-5D-5L descriptive system are converted into a single index score using a scoring algorithm that takes local patient and population preferences into account. Therefore, the index score is a country-specific value and a major feature of the EuroQol 5-Dimensions (EQ-5D) instrument.²³ The range of index scores will differ according to the scoring algorithm used; however, in all scoring algorithms of the EQ-5D-5L, a score of 0 represents the health state “dead” and 1.0 reflects “perfect health.”

The VAS records the respondent’s self-rated health on a vertical VAS where the end points are labelled 0 (“the worst health you can imagine”) and 100 (“the best health you can imagine”). Respondents are asked to mark an X on the point of the VAS that best represents their health on that day. The EQ-5D index and VAS scores can be summarized and analyzed as continuous data.^22,23

The validity and reliability of the EQ-5D-5L have been demonstrated in a diverse population, but no literature was identified that assessed the EQ-5D-5L for validity, reliability, or responsiveness in patients with porphyria. An MID was not identified in populations with AHP; however, the MID for the index score was estimated to range from 0.037 to 0.056 in the general Canadian population. Both the index score and VAS were reported in the ENVISION trial. Index scores were described as a change from baseline, and domain scores were reported categorically. The VAS was reported as a change from baseline.

Patient Global Impression of Change

The PGIC is a widely used, validated outcome measure for clinical pain trials.^24,25 It is a single question answered on a 7-point numerical scale to indicated perceived change in overall health status since the study began. Patients are asked to consider their overall health since the start of the trial by choosing a response (1 = very much improved; 2 = much improved; 3 = minimally improved; 4 = no change; 5 = minimally worse; 6 = much worse; 7 = very much worse).²⁶ The PGIC questionnaire has been recommended for use in chronic pain clinical trials by the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) as a core outcome measure of global improvement with treatment.²⁷ No literature was identified that assessed the PGIC for validity, reliability, or responsiveness in patients with porphyria. No MID was identified in populations with AHP. In the ENVISION trial, the PGIC was reported descriptively. It was also reported as 2 categories: “improved,” including responses for very much improved, much improved, and minimally improved; and “no change or worsening,” including responses for no change, minimally worse, much worse, and very much worse. The proportion of patients who improved was compared between treatment groups.

Management of Symptoms

Patients or caregivers were asked to record assessments of pain, nausea, and fatigue in the patient’s electronic diary daily throughout the ENVISION trial. As noted under the description of outcomes for porphyria attacks, patients or caregivers were trained in the use of the electronic diary, and instructions were provided in a study manual. The questionnaires used to evaluate each of these symptoms are described in this section. For each symptom, a weekly (7-day) score was derived from an average of daily symptom measurements reported in the electronic diary. The change from baseline in the weekly average score was reported in the ENVISION trial. The AUC of the change from baseline was also reported to account for peaks in symptoms during attacks and lingering chronic symptoms.

Pain

The Brief Pain Inventory (BPI) is a self-reported, 11-item questionnaire designed to provide information on 2 subscales: pain intensity (the sensory dimension, 4 items) and the degree to which pain interferes with functioning in daily living (the reactive dimension, 7 items). The ENVISION trial used a single question from the BPI Short Form (BPI-SF) to assess the severity of daily pain.⁶ Patients were asked to rate the worst level of pain they experienced during the past 24 hours on an NRS from 0 = “no pain” to 10 = “as bad as you can imagine.” As a single item, it has shown internal consistency (0.70 < Cronbach alpha < 0.90) and reliability (between 0.8 and 0.96) among various populations.²⁸ Construct, convergent, and discriminative validity, internal consistency, and test-retest reliability have been demonstrated for the BPI-SF^28,29; however, no literature was identified that assessed the BPI-SF for validity, reliability, or responsiveness in patients with porphyria. Although an overall MID of the BPI has not been identified from the literature, a 2-point change has been suggested as a reasonable estimate of the MID for worst pain among breast cancer patients with metastatic disease.³⁰ An MID specific to patients with AHP was not identified.

Fatigue

The Brief Fatigue Inventory (BFI) is a self-reported questionnaire to assess the severity and impact of fatigue on daily functioning. The items of the BFI are measured on a 0-to-10 NRS. In the ENVISION trial, a single question was used from the Brief Fatigue Inventory – Short Form (BFI-SF) to assess the severity of daily fatigue.⁶ Patients were asked to rate the worst level of fatigue they experienced during the past 24 hours on a scale from 0 = “no fatigue” to 10 = “as bad as you can imagine.”

Validity and reliability of the whole scale was assessed in samples of patients with cancer, rheumatoid arthritis, and community-dwelling adults and older adults.³¹ Construct validity, concurrent validity, and discriminant validity of the BFI have been demonstrated in cancer patients, as well as reliability, which was acceptable (Cronbach alpha values were 0.95 to 0.96).³² Further, both the Chinese and Taiwanese versions of the BFI were previously validated in patients with cancer and provided evidence of validity of the single item for worst fatigue compared to the Chinese version of the SF-36 and Eastern Cooperative Oncology Group (ECOG) performance status, as well as the Taiwanese Profile of Mood States (POMS) short-form questionnaire.³³ An MID for the worst fatigue item was estimated to be 1.5 using the 0.5 SD method,³³ but no MID was identified in populations with AHP. Moreover, no literature was identified that assessed the BFI-SF for validity, reliability, or responsiveness in patients with porphyria. Nausea

The nausea NRS was used to assess nausea in Study 003. It is a self-reported, single question used to assess a patient’s level of nausea in the past 24 hours on a scale from 0 to 10, where 0 represents no nausea and 10 represents “nausea as bad as you can imagine.” Construct validity and reliability were demonstrated for the instrument.³⁴ No literature was identified that assessed the Nausea Numerical Rating Scale for validity, reliability, or responsiveness in patients with porphyria. Further, an MID was not identified in populations with AHP.

Rescue Medication Use

Measures of rescue medication in Study 003 included an evaluation of hemin use for acute attacks. This was reported as the annualized rate of hemin use over the 6-month DB period.

Opioid Use

Opioid usage was reported under analgesic usage in Study 003. Analgesic usage at home by patients throughout the study was self-reported through the electronic diary that was also used to report porphyria attacks and symptoms. Analgesic usage was reported using daily questions regarding type, dose, and frequency. Analgesic medications that were taken at a health care facility, such as during an attack, were captured as concomitant medication usage. Analgesic medication usage was reported descriptively by medication category (opioid, non-opioid, either opioid or non-opioid). This included the proportion of patients reporting analgesic medication use and the proportion of days with analgesic use during the 6-month DB period.

Activities of Daily Living

In the ENVISION trial, activities of daily living were reported using the PPEQ. The PPEQ was described by the sponsor as a custom tool consisting of 8 questions used to assess a patient’s ability to perform daily living activities (questions 1 to 5), treatment experience (questions 6 and 7), and functional status (question 8). Questions are answered on a 5-point global rating of change scale.⁶ More specifically, patients select from the following options for questions 1 to 7: much better, minimally better, no change, minimally worse, or much worse, based on their current experience compared to before the start of the study. Question 8 is answered by choosing 1 of the following: always, most of the time, sometimes, rarely, or never, based on the last 4 weeks. Responses for each of the 8 questions of the PPEQ were reported descriptively in the ENVISION trial. No literature was identified that assessed the PPEQ for validity, reliability, or responsiveness in patients with porphyria. No MID was identified in populations with AHP.

Hospitalization and Health Care Use

Hospitalization and health care use were reported indirectly through reporting of porphyria attacks, namely those leading to hospitalizations or urgent health care visits. These outcomes were previously described under frequency of porphyria attacks.

Ability to Work or Attend School

In Study 003, the ability to work or attend school was reported as the number of missed days of work or school due to porphyria symptoms or attacks in the past 4 weeks and summarized descriptively.

ALA, PBG, and Porphyrin Levels

Study 003 reported urinary ALA levels and PBG levels during the study. Spot urine samples for ALA and PBG measurement were collected at study visits before treatment dosing; however, if hemin was used for an attack, scheduled urinary ALA and PBG were collected 4 days after the patient’s last hemin dose. If feasible and permitted by local regulations, urine samples for ALA and PBG assessments may have been collected by a home health care professional, sent to the study centre by mail, or brought to the study centre at the next visit. For analysis in the trial, urinary ALA levels were reported at 3 months and 6 months, and PBG was reported at 6 months. Measurements of ALA or PBG that were taken up to 3 days following hemin use were treated as missing and excluded from analysis.

Mortality

Mortality was not reported as an efficacy outcome in Study 003.

Harms

Safety outcomes reported included the incidence, severity, seriousness, and relatedness of AEs over the 6-month DB period in patients with any AHP. Porphyria attacks were reported as efficacy outcomes; therefore, they were not treated as AEs or SAEs. However, non–porphyria-related AEs during porphyria attacks were reported. Other AEs of special interest were reported, and these included ALT elevations greater than 3 times the ULN, lipase or amylase greater than 3 times the ULN, severe or serious injection-site reactions, and anaphylactic reactions.

Statistical Analysis

Primary Outcome

The primary outcome of the pivotal trial was annualized rate of porphyria attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home in patients with AIP over the 6-month DB period. The definitions used for the 3 types of attacks included in the primary outcome are summarized in the Outcomes section. The AAR was calculated using the following formula:

AAR = (total number of porphyria attacks ÷ total number of days in the treatment period) × 365.25

Power Calculation

The study planned to enrol at least 74 patients, including 70 with AIP. It was estimated that 70 patients would yield 90% power to detect a 45% reduction in the AAR at a 2-sided 5% significance level, with the following assumptions: a mean AAR of 8, an SD of 5 in the placebo group, and a mean AAR of 4.4 with an SD of 3 in the givosiran group. The study was also designed to have at least 80% power with a dropout rate of up to 15% under the same assumptions that have been described.

Statistical Test or Model

A summary of the statistical testing methodology is presented in Table 9. The primary analysis was conducted in the mFAS, which included only patients with AIP from the FAS. A negative binomial regression model was used for the primary analysis, which included stratification factors for patients with AIP as fixed effects — namely, the use of a hemin prophylaxis regimen before the study (yes versus no) and historical AAR (high versus low). The logarithm of the amount of time (in units of years) that each patient spends in the 6-month DB period was included in the model as an offset variable to account for the different lengths of follow-up time among patients. An estimated ratio of mean AARs between the givosiran and placebo groups with corresponding 95% CI based on the negative binomial regression model was reported. Data were not imputed for the primary analysis of AAR. Descriptive statistics for the AAR were also reported.

Each of the 3 components of the primary outcome or AAR were analyzed similarly to the primary analysis. A zero-inflated negative binomial regression model was also used to analyze each component to address the potential issue of an “excessive” number of patients with zero attacks. A model-based analysis would not be conducted if fewer than 10 patients experienced an attack.

Secondary Outcomes

Secondary outcomes were conducted in the mFAS unless otherwise specified.

Annualized event rate type end points were analyzed in the same way as the primary analysis of AAR. This included the annualized rate of administered hemin doses in patients with AIP and AAR in patients with any type of AHP (in the FAS) over the 6-month DB treatment period. Data were not imputed for any of these outcomes.

The biomarker outcomes (urinary ALA and PBG) were analyzed and reported as the LS mean difference between treatment groups at 3 months and 6 months for urinary ALA, and at 6 months for urinary PBG. The analysis of biomarker end points involved a comparison between treatment groups using a mixed-effects model for repeated measures (MMRM), with the baseline biomarker level as a continuous covariate, stratification factors, visit, treatment, and visit by treatment as fixed effects, and patient as a random effect. Measurements of urinary ALA or PBG within 3 days after hemin use were treated as missing and excluded from analysis. The SF-12 scores were also analyzed as a change from baseline at month 6 using an MMRM model with baseline score as a continuous covariate, and treatment group, stratification factors, visit (month 3 or month 6), and visit by treatment interaction as fixed effects. Missing data for biomarker outcomes and the SF-12 were implicitly imputed through the use of the MMRM model.

Daily worst scores for pain, fatigue, and nausea were measured using the BPI-SF NRS, BPI-SF NRS, and an NRS, respectively, and reported through electronic patient diaries. The changes from baseline in weekly mean scores were reported, as well as the AUC from baseline over 6 months. An ANCOVA model was used to compare the AUC change from baseline between treatment groups. The ANCOVA model included fixed effects of treatment arms and the 2 stratification factors. Weekly mean scores were derived from an average of daily scores for a completed week, defined by the completion of at least 4 daily diary entries. A missing week was defined as a week with at least 4 missing entries. If an attack occurred during the missing week, a mean weekly score was computed if there was at least 1 entry. If no entries were reported, the weekly mean score was imputed using the average of all non-missing attack-week weekly mean scores if there were available data; if not, the weekly mean score was imputed by the worst weekly mean scores within the same treatment period for the patient. In a missing week without any attack days, the weekly mean score was imputed using the last observed non-missing week without any attack days, if available; if not, the baseline observation carried forward was imputed.

Exploratory Outcomes

All secondary end points defined for the patients with AIP were analyzed for the patients with AHP using similar methods. Continuous exploratory end points in patients with AIP included the EQ-5D-5L index score, which was analyzed using an MMRM model similar to the model used for the PCS of the SF-12. The remainder of the outcomes were reported descriptively, which included the PGIC at month 6 and month 12, missed days of work or school at month 6, PPEQ, patient responses to each of the EQ-5D domains, and analgesic medication use.

Testing Strategy

The primary analysis was conducted at a 2-sided significance level of 0.049, reflecting a penalty of 0.001 for an unblinded interim analysis. To control for the overall type I error rate, a fixed-sequence, hierarchical testing strategy was implemented. If the primary analysis was statistically significant, then the secondary end points were tested at the same significance level of 0.049 in the order-specified testing hierarchy. Statistical testing was not conducted following failure to reject the null hypothesis at an early end point within the hierarchy; however, nominal P values were reported. Statistical tests were conducted according to the following hierarchy:

• urinary ALA levels in patients with AIP at 3 months

• urinary ALA levels in patients with AIP at 6 months

• urinary PBG levels in patients with AIP at 6 months

• annualized rate of administered hemin doses (evaluated by annualized days of hemin use) in patients with AIP over the 6-month DB period

• annualized rate of porphyria attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home in patients with any AHP over the 6-month DB period

• daily worst pain score as measured by BPI-SF NRS in patients with AIP over the 6- month DB period

• daily worst fatigue score as measured by BFI-SF NRS in patients with AIP over the 6-month DB period

• daily worst nausea score as measured by NRS in patients with AIP over the 6-month DB period

• change from baseline in the PCS of the SF-12 in patients with AIP at 6 months.

Interim Analysis

A preplanned, unblinded interim analysis of patients’ ALA levels at 3 months was conducted when approximately 30 patients with AIP had completed at least 3 months of the treatment period. The significance level for the comparison at the interim was 0.001 (2-sided).

Subgroup Analyses

The following subgroups were included in the CADTH systematic review protocol: AHP subtype (AIP, ADP, HCP, VP) and frequency and history of attacks. Study 003 included pre-specified subgroup analyses of the primary outcome, using similar methods as the primary analysis, for the following subgroups:

• age at screening (< or ≥ median age in the overall population)

• race (White or non-White)

• sex (female or male)

• region group 1: North America (including the US and Canada) or other (outside North America)

• region group 2: Europe or other (outside Europe)

• baseline body mass index (< 25 or ≥ 25)

• prior hemin prophylaxis status (yes or no)

• historical attack rates before randomization based on the hemin prophylaxis status before the study (high or low)

• for patients on a hemin prophylaxis regimen at the time of screening, if AAR ≥ 7, the patient is considered having high attack rates before the study; for patients who were not on a hemin prophylaxis regimen at screening, AAR ≥ 12 is considered with high attack rates

• prior chronic opioid use when not having attacks (yes or no)

• prior chronic symptoms when not having attacks (yes or no).

Anticipating that there would be only a few non-AIP patients, the sponsor planned to report only descriptive statistics for these patients. The sponsor also reported that other subgroups, or subgroup analyses of other secondary end points, might be conducted.

For the purpose of this review, subgroup analyses by historical attack rates were presented as per the systematic review protocol. Additionally, all outcomes presented in this review were reported in the mFAS where available, in alignment with the subgroup for AHP subtypes outlined in the systematic review protocol.

Sensitivity Analyses

The following sensitivity analyses were conducted on the primary end point:

• To account for undercounting attacks, a sensitivity analysis counting all discrete attacks was conducted. For investigator-confirmed attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home, all discrete attacks were counted even if they overlapped during a day.

• To account for overcounting attacks, a sensitivity analysis was conducted where attacks were counted using a 2-day window. More specifically, the attack-counting window was extended from a 1-day to a 2-day window for investigator-confirmed attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home. This meant that attacks that occurred on the same calendar day, or were separated by 1 calendar day, were counted as 1 attack.

• A sensitivity analysis was conducted using all investigator-confirmed attacks, which included all attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home in addition to attacks treated at home not requiring hemin use.

• A sensitivity analysis was conducted that included all porphyria attacks, counting potential attacks. More specifically, both investigator-confirmed attacks and attacks that were not confirmed by the investigator but were deemed to be potential attacks were included. Duplicates and attacks due to entry errors were excluded.

• A sensitivity analysis was conducted in which attacks were treated as recurrent events through the Andersen-Gill model. The primary porphyria attack composite end point was analyzed using an Andersen-Gill model by treatment group and stratification factors (prior hemin prophylaxis status and historical attack rates).

• Lastly, the primary porphyria attack composite end point was analyzed based on the per-protocol analysis set.

Protocol Deviations

Major protocol deviations were deviations that were considered to have potentially had a significant impact on the data collected in the study, or to have significantly affected a patient’s rights, safety, or well-being. The sponsor reported a designated plan for handling protocol deviations. Briefly, whether or not to exclude a patient from the per-protocol set was based on the sponsor’s judgment of the potential impact on the primary efficacy results.

Analysis Populations

The FAS included all randomized patients who received at least 1 dose of study drug. Patients were grouped by the treatment group to which they had been randomized.

An mFAS was used in Study 003, which followed the same definition as the FAS except that only randomized patients with AIP identified by a mutation in the HMBS gene were included.

The per-protocol set included all randomized patients with AIP (with identified mutation in the HMBS gene) who received at least 4 doses (> 60%) of the study drug during the 6-month DB period, were followed for the collection of attack data through 6 months (≥ 162 days) and did not experience major protocol deviations that could affect the primary efficacy results (e.g., not meeting the key inclusion or exclusion criteria). Patients were analyzed according to their randomly assigned treatment groups.

The safety analysis set included all patients who received at least 1 dose of the study drug, grouped according to the treatment actually received. Of note, patients who received any amount of givosiran during the 6-month DB period were included in the givosiran arm.

Results

Patient Disposition

A summary of the patient disposition in Study 003 is available in Table 10. A total of 109 patients were screened, and 94 were randomized to receive either givosiran or placebo. A total of 89 (95%) of randomized patients had AIP, and 5 (5%) had non-AIP types of AHP. One patient had HCP and was randomized to givosiran; 2 patients had VP, with 1 randomized to givosiran and 1 randomized to placebo; and 2 patients had AHP without an identified mutation. Both were randomized to placebo. No patients with ADP were enrolled in the study. All randomized patients completed the 6-month DB treatment period. Overall, 1 patient discontinued from study due to an AE. This patient was randomized to the placebo treatment group and had non-AIP AHP.

Nineteen major protocol deviations among 6 patients were reported. The study drug was administered with unmasked syringes at 1 study site, resulting in protocol deviations for 4 patients. Two patients did not meet the inclusion criteria, despite enrolment in the study, which also resulted in protocol deviations.

Exposure to Study Treatments

Exposure to study treatments is summarized in Table 11. The mean duration of exposure to treatment was 5.5 months for patients with AHP in the givosiran and placebo treatment groups. For patients with AIP, the mean durations of exposure to treatment were 5.6 months and 5.5 months for patients in the givosiran and placebo treatment groups, respectively. In patients with AIP, 8% and 2% of patients in the givosiran and placebo treatment groups, respectively, missed 1 treatment dose, and 1 patient in the givosiran treatment group missed 2 doses; the rest of the patients with AIP did not miss any doses. Reasons for missed doses were summarized by the sponsor. Three patients in the givosiran treatment group and 1 patient in the placebo treatment group missed their doses because the study visit could not be scheduled within the dose window. The fourth patient in the givosiran treatment group missed 1 dose because their week 2 study visit was documented as month 1. One patient in the givosiran treatment group missed 2 doses because givosiran was withheld due to transaminase elevations after they had received 4 doses.

The proportion of patients with missing doses was similar for patients with AHP.

Concomitant medication use during the 6-month DB period is summarized in Table 12. Concomitant medication use was reported by almost all patients with AHP (i.e., 95% of patients receiving givosiran and 100% of patients receiving placebo). Among the most commonly reported concomitant medications used during the study (by therapeutic class), a difference in concomitant use of at least 10% between treatment groups was reported for the following classes of medications (in patients treated with givosiran versus placebo, respectively): natural opium alkaloids (46% versus 70%), other heme products (46% versus 74%), electrolyte solutions (10% versus 24%), fluoroquinolones (8% versus 20%), other opioids (8% versus 22%), solutions affecting the electrolyte balance (6% versus 17%), corticosteroids (4% versus 15%), vitamin D and analogues (25% versus 13%), and other antidepressants (19% versus 9%). Overall and differential concomitant medication use among patients with AIP was similar to the overall patient population (patients with AHP). Additionally, concomitant use of GnRH was reported by 2 patients in the givosiran treatment group and 2 patients in the placebo treatment group. All patients who used GnRH concomitantly were patients with AIP.

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported in this section. See Appendix 3 for detailed efficacy data.

Frequency of AHP Attacks

A summary of outcomes related to the frequency of porphyria attacks is provided in Table 13.

The results described herein are for patients with AIP using the mFAS unless otherwise stated. Based on the definition of porphyria attacks used for the primary end point, which included attacks that required hospitalization, an urgent health care visit, or IV hemin administration at home, patients with AIP reported a total of 83 and 284 porphyria attacks in the givosiran group and placebo group, respectively, over the 6-month treatment period. Among patients in the givosiran group, 50.0% did not report a porphyria attack with the 6-month treatment period, compared to 16.3% of patients receiving placebo. The mean AAR based on the composite end point was 3.22 (95% CI, 2.25 to 4.59) and 12.52 (95% CI, 9.35 to 16.76) for patients in the givosiran group and placebo group, respectively. This corresponded to a rate ratio of 0.26 (95% CI, 0.16 to 0.41; P < 0.001) or a 74% reduction in the rate of porphyria attacks for patients in the givosiran group relative to patients receiving placebo.

When only considering attacks requiring hospitalization, the mean AAR was 1.65 (95% CI, 0.98 to 2.78) and 3.21 (95% CI, 0.98 to 5.20) for patients in the givosiran and placebo groups, respectively. This corresponded to a rate ratio of 0.51 (95% CI, 0.25 to 1.04) or a 49% rate reduction in porphyria attacks requiring hospitalization. The mean AAR for attacks requiring an urgent health care visit was 1.22 (95% CI, 0.73 to 2.05) and 7.53 (95% CI, 5.13 to 11.05) for patients in the givosiran and placebo groups, respectively, corresponding to an 84% rate reduction (rate ratio = 0.16, 95% CI, 0.09 to 0.31). Lastly, a total of 3 attacks required IV hemin administration at home for patients in the givosiran group, while 32 attacks required this for patients in the placebo group.

The AAR for patients with at least 1 attack and the number of porphyria attacks treated at home without IV hemin were also reported (Table 13). Porphyria attacks that occurred at home that did not require IV hemin accounted for 19 of the 102 porphyria attacks reported in the givosiran group and for 14 of the 298 attacks reported in the placebo group for patients with AIP. The numbers of porphyria attacks that occurred at home that did not require IV hemin were similar between treatment groups for all patients with AHP (19 for givosiran, 20 for placebo).

All of the outcomes described previously were also measured for all patients with AHP (FAS), which yielded results similar to those reported for patients with AIP.

The average number of attacks based on the composite end point, per patient per month over the 6-month treatment period, is presented in Figure 4 and Figure 5 for all patients with AHP and for patients with AIP, respectively. Briefly, the average number of attacks per patient per month was approximately 1 for patients receiving placebo and fewer than 0.6 for patients receiving givosiran over the 6-month treatment period in patients with AIP (Figure 4). The results for all patients with AHP were similar to the results reported for patients with AIP (Figure 5).

Figure 4: Average Number of Attacks (Composite End Point) per Patient per Month (6-Month Double-Blind Period, mFAS)

mFAS = modified full analysis set.

Source: Clinical Study Report.⁶

Figure 5: Average Number of Attacks (Composite End Point) per Patient per Month (6-Month Double-Blind Period, FAS)

FAS = full analysis set.

Source: Clinical Study Report.⁶

Sensitivity Analyses

Sensitivity analyses were only conducted on the primary analysis, annualized rate of porphyria attack in patients with AIP (Table 14).

|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||

|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Subgroup Analyses

The AAR based on the composite end point was reported by historical attack rates as a subgroup (Appendix 3, Figure 21 and Figure 22). Based on the mFAS, for patients with AIP and a high historical attack rate (n = 43), the rate ratio for givosiran compared to placebo was 0.27 (95% CI, 0.16 to 0.46). For patients with AIP and a low historical attack rate (n = 46), the rate ratio for givosiran compared to placebo was 0.23 (95% CI, 0.09 to 0.56). The comparison of givosiran to placebo in the FAS corresponded to a rate ratio of 0.29 (95% CI, 0.18 to 0.47) for patients with a high historical attack rate (n = 45) and a rate ratio of 0.23 (95% CI, 0.09 to 0.56) for patients with a low historical attack rate (n = 49).

Health-Related Quality of Life

Study 003 evaluated HRQoL using the SF-12, EQ-5D-5L, and PGIC. The results for these outcomes are available in Table 15; categorical domain-level results for the EQ-5D-5L are available in Appendix 3.

The PCS of the SF-12 was the only HRQoL outcome included as a secondary outcome in Study 003; the rest were exploratory outcomes. In patients with AIP, the change from baseline in the PCS score at month 6 was an LS mean of 5.37 (95% CI, 3.05 to 7.69) for the givosiran treatment group and 1.43 (95% CI, –1.00 to 3.86) for the placebo treatment group. The between-groups difference in the LS mean for givosiran compared to placebo was 3.94 (95% CI, 0.59 to 7.29; P = 0.0216); however, the P value cannot be interpreted due to a prior failure in the hierarchical testing strategy. The change from baseline in the MCS was reported descriptively. At month 6 in patients with AIP in the givosiran treatment group, the mean change from baseline was 3.66 (SD = 10.08), and for patients in the placebo treatment group, it was 1.30 (SD = 8.54). A between-groups comparison was not reported. The change from baseline to month 6 in each of the domains of the SF-12 is provided in patients with AIP in Figure 6. The PF, RP, BP, and GH domains contribute to the PCS. The LS mean differences between givosiran and placebo was 1.4 (95% CI, –2.0 to 4.7), 4.4 (95% CI, 1.3 to 7.5), 7.2 (95% CI, 3.2 to 11.2), and 3.3 (95% CI, –0.7 to 7.2) for each of these domains, respectively. VT, SF, RE, and MH contribute to the MCS, and had an LS mean difference of 17 (95% CI, –2.0 to 5.5), 5.1 (95% CI, 1.6 to 8.7), 1.4 (95% CI, –2.5 to 5.2), and 2.8 (95% CI, –0.9 to 6.4), respectively.

In terms of the EQ-5D-5L index for patients with AIP, the LS mean change from baseline at month 6 was |||||||||||||||||||||||||| |||| for the givosiran treatment group, and |||||||||||||||||||||||||||||||| for the placebo treatment group, which corresponded to a between-groups difference of ||||||||||||||||||||||||||||||||||. Using the VAS of the EQ-5D-5L, the LS mean change from baseline at month 6 was |||||||||||||||||||||||||||| and |||||||||||||||||||||||||||||| for the givosiran and placebo treatment groups, respectively. The between-groups difference at month 6 in the EQ-5D-5L VAS was ||||||||||||||||||||||||||||||.

Lastly, the PGIC was assessed at month 6 as the change in patient status since the start of the study. For patients with AIP, 88.9% of those in the givosiran group and 37.1% of those in the placebo group reported their status as improved since the start of the study. Further, in the givosiran treatment group, 1 (2.8%) patient reported no change, 2 (5.6%) patients reported that their status was minimally worse, and 1 (2.8%) patient reported that their status was very much worse. In the placebo treatment group, 42.9% of patients reported no change in their status, 6 (17.1%) reported that their status was minimally worse, and 1 (2.9%) reported that their status was much worse. The proportion of patients who reported an improvement in their status was compared between the 2 treatment groups. An odds ratio of 13.5 (95% CI, 3.90 to 47.03) for the givosiran group compared to placebo was reported for patients with AIP.

All of the outcomes described previously were also reported for all patients with AHP and were consistent with the results for patients with AIP.

Figure 6: Difference Between Groups in Change From Baseline to Month 6 in SF-12 Domain Scores (6-Month Double-Blind Period, mFAS)

CI = confidence interval; LS = least squares; MCS = Mental Component Summary; mFAS = modified full analysis set; PCS = Physical Component Summary; SF-12 = 12-item Short Form Health Survey.

Source: Clinical Study Report.⁶

Management of Symptoms

The change in self-reported assessments of pain, fatigue, and nausea based on an NRS were secondary outcomes in Study 003. The results are presented in Figure 16.

Based on the pre-specified ANCOVA model, the LS mean for the AUC of change from baseline in weekly mean daily worst pain scores at 6 months in patients with AIP corresponded to a treatment-group difference of –12.7 (95% CI, –25.5 to 0.2; P = 0.053). A post hoc reanalysis of this outcome using a non-parametric test was conducted following a demonstration of deviation from normality. In patients with AIP, the medians of the AUC of change from baseline in weekly mean score for daily worst pain over the 6-month treatment period were –11.5 (IQR = –29.2 to 3.0) and 5.3 (IQR = –23.1 to 11.2) for the givosiran and placebo treatment groups, respectively. The median of the treatment-group difference for givosiran compared to placebo was –10.1 (95% CI, –22.8 to 0.9; P = 0.0455). The average change from baseline in weekly mean score for daily worst pain over the 6-month treatment period was –0.5 (95% CI, –1.0 to 0.1). Both the median of the AUC and the median of the average change from baseline in the weekly mean score for daily worst pain correspond to a reduction in pain for the givosiran treatment group and an increase in pain for the placebo treatment group.

For the assessment of daily worst fatigue score over the 6-month treatment period in patients with AIP, the means of the AUC of the change from baseline were –11.2 (95% CI, –20.1 to –2.2) and –4.2 (95% CI, –13.5 to 5.1) for patients in the givosiran and placebo treatment groups, respectively. These corresponded to a treatment-group difference for givosiran compared to placebo of –6.4 (95% CI, –19.8 to 6.0; P = 0.2876). The treatment-group difference in LS mean for the average change from baseline in weekly mean score was –0.32 (95% CI, –0.90 to 0.25; P = 0.2698).

In patients with AIP, the AUC of the change from baseline in weekly mean scores for daily worst nausea over the 6-month treatment period were an LS mean of 1.5 (95% CI, –5.1 to 8.1) in the givosiran group and –4.0 (95% CI, –10.9 to 2.9) in the placebo group. The between-groups difference in LS mean for givosiran compared to placebo was 5.5 (95% CI, –4.0 to 15.0; P = 0.2532). Based on the average change from baseline in weekly mean scores for daily worst nausea, a treatment-group difference of 0.3 (95% CI, –0.2 to 0.7; P = 0.2459) was reported for the givosiran treatment group compared to placebo.

Assessments of pain, fatigue, and nausea over the 6-month treatment period were also reported for all patients with AHP and were consistent with the results for patients with AIP.

Reduced Complications of AHP

Outcomes related to reduced complications of AHP were not reported in the included studies.

Rescue Medication Use

In Study 003, hemin was only permitted as a rescue medication for the treatment of acute porphyria attacks and was reported as days of hemin use (Table 17). In patients with AIP, 54% of patients in the givosiran group and 23% of patients in the placebo group reported zero days of hemin use over the 6-month treatment period. The means of annualized days of hemin use in the givosiran and placebo groups were 6.8 days (95% CI, 4.2 to 10.9) and 29.7 days (95% CI, 18.4 to 47.9), respectively. This corresponded to a 77% rate reduction in days of hemin use for givosiran compared to placebo based on a rate ratio of 0.23 (95% CI, 0.11 to 0.45; P < 0.001).

In patients with AHP, the reported use of hemin for the treatment of acute porphyria attacks was similar to that for patients with AIP. In the givosiran and placebo groups, 54% and 26% of patients had zero days of hemin use in the 6-month treatment period. The rate ratio for annualized days of hemin use for givosiran compared to placebo was 0.26 (95% CI, 0.13 to 0.52; P = 0.0002).

Opioid Use

Opioid use was reported descriptively under analgesic medication use in Study 003 and has been summarized in Table 18. In patients with AIP, the mean proportion of days with opioid use was 23.3 days (SD = 35.3 days) for patients in the givosiran treatment group and 38.0 days (SD = 39.4 days) for patients in the placebo treatment group. The median proportion of days with opioid use was 3.0 (Q1 to Q3: 0 to 38.5) and 10.8 (Q1 to Q3: 2.4 to 83.3) for patients in the givosiran and placebo groups, respectively.

Over the 6-month treatment period, 67% and 88% of patients in the givosiran and placebo groups, respectively, reported opioid use. Opioid use was also reported as the number of patients with analgesic opioid use during months 1 to 3 and months 4 to 6. In the givosiran treatment group, 63% of patients used opioids for pain between months 1 and 3, and 52% used opioids for pain between months 4 and 6. In the placebo treatment group, 81% of patients used opioids for pain between months 1 and 3, and 84% used opioids for pain between months 4 and 6.

Analgesic use of opioids in patients with AHP were similar to those for patients with AIP (Table 18).

Activities of Daily Living

Data for the PPEQ were reported descriptively.

Responses to the PPEQ at month 6 were reported in patients with AIP only (Figure 7) and in all patients with AHP (Figure 8). Questions 1 through 7 of the PPEQ reflect the patient’s response compared to before the start of study, and question 8 is based on the previous 4 weeks. Overall, a greater proportion of patients receiving givosiran answered “much better” or “always” to all questions included the PPEQ compared to those receiving placebo. More specifically, 33% to 36% of patients with AIP in the givosiran group (versus 6% to 14% of patients in the placebo group) rated the following activities as being much better at month 6 compared at to the start of the study: travelling for more than 1 day for work or pleasure, participating in social activities, planning future events, doing household chores, and exercising moderately. The convenience of current porphyria treatment and patients’ overall satisfaction with porphyria treatment were both rated as much better at month 6 compared to the start of study by 74% of patients in the givosiran group compared to 9% and 15% of patients in the placebo group for each of the 2 questions, respectively. For question 8, 43% and 6% of patients in the givosiran and placebo groups, respectively, responded “always” to whether the study drug helped them live more normal lives.

The proportion of patients who responded “much better” or “always” was similar between patients with AIP and AHP.

Figure 7: Response to the PPEQ at Month 6 (6-Month Double-Blind Period, mFAS)

mFAS = modified full analysis set; PPEQ = Porphyria Patient Experience Questionnaire; Q = question.

Note: Questions 1 to 7 contained response options of much better, minimally better, no change, minimally worse, and much worse. Question 8 had response options of always, most of the time, sometimes, rarely, and never.

Source: Clinical Study Report.⁶

Figure 8: Response to the PPEQ at Month 6 (6-Month Double-Blind Period, FAS)

FAS = full analysis set; PPEQ = Porphyria Patient Experience Questionnaire; Q = question.

Source: Clinical Study Report.⁶

Hospitalization and Health Care Use

Attacks requiring hospitalization or an urgent health care visit were reported in the Frequency of AHP Attacks section.

Ability to Work or Attend School

Study 003 reported on days of missed work or school. The results for the number of missed days in the past 4 weeks are presented in Table 19. In patients with AIP at baseline, the mean reported numbers of missed days of work were 3.3 days (SD = 3.5 days) and 6.4 days (SD = 6.5 days) for patients in the givosiran and placebo groups, respectively. At month 3, the mean numbers of missed days of work were 2.4 days (SD = 3.9 days) and 4.8 days (SD = 6.8 days) for givosiran and placebo, respectively. At month 6, a mean of 2.4 days (SD = 6.8 days) of missed work was reported in the givosiran treatment group, and a mean of 6.9 days (SD = 8.0 days) was reported for placebo.

Among patients with AIP, 5 contributed to the evaluation of missed days of school. For all patients with AHP, 6 patients contributed to this evaluation. In the givosiran group, the mean numbers of missed days of school at baseline, month 3, and month 6 were 7.3 days (SD = 11.8 days), 0 days, and 2.8 days (SD = 4.9 days), respectively, for patients with AIP. In the corresponding placebo group, the mean numbers of missed days of school at baseline, month 3, and month 6 were 3.5 days (SD = 0.7 days), 0.5 days (SD = 0.7 days), and 0 days, respectively.

The results for the overall population of all patients with AHP were similar to those for patients with AIP (Table 19).

ALA, PBG, and Porphyrin Levels

Biomarkers for AHP, namely ALA and PBG, were analyzed in Study 003; the results are presented in Table 20.

For patients with AIP at baseline, the mean levels of urinary ALA were 19.97 mmol/mol Cr (SD = 16.80 mmol/mol Cr) and 17.52 mmol/mol Cr (SD = 10.89 mmol/mol Cr) in the givosiran and placebo groups, respectively. At month 3, the LS mean change from baseline in urinary ALA levels were 1.76 mmol/mol Cr (95% CI, –1.05 to 4.57) for givosiran and 19.97 mmol/mol Cr (95% CI, 17.03 to 22.90) for placebo, which corresponded to a between-groups difference of –18.21 (95% CI, –22.26 to –14.16; P < 0.001) in favour of givosiran. At month 6, the LS mean of the change from baseline in urinary ALA levels was 4.01 mmol/mol Cr (95% CI, –0.69 to 8.72) for givosiran and 23.15 mmol/mol Cr (95% CI, 18.09 to 28.21) for placebo, corresponding to a between-groups difference of –19.14 mmol/mol Cr (95% CI, –26.04 to –12.24; P < 0.001) in favour of givosiran as well.

In patients with AIP, urinary levels of PBG at baseline were 50.36 mmol/mol Cr (SD = 34.33 mmol/mol Cr) and 46.80 mmol/mol Cr (SD = 24.32 mmol/mol Cr) for patients in the givosiran and placebo groups, respectively. At month 6, the LS mean of the change from baseline in urinary PBG was 12.91 mmol/mol Cr (95% CI, 3.66 to 22.15) for the givosiran group and 49.11 mmol/mol Cr (95% CI, 39.24 to 58.98) for the placebo group. The between-groups difference in urinary PBG at month 6 for givosiran compared to placebo was –36.20 mmol/mol Cr (95% CI, – 49.71 to – 22.70; P < 0.001) in favour of givosiran.

Post hoc non-parametric tests were conducted for urinary ALA and PBG levels following demonstration of a significant deviation from the normal distribution. The median urinary ALA levels and PBG levels for patients in the givosiran and placebo groups are presented by month in patients with AIP (Figure 9 and Figure 10). Based on the median measured at each month, levels of urinary ALA and PBG were consistent over the 6-month treatment period following month 1. The results of the non-parametric tests were consistent with the results of the pre-specified analyses presented in Table 20.

Overall, the results of urinary ALA and PBG levels in patients with AIP were similar for all patients with AHP.

Figure 9: Median (IQR) Urinary ALA Levels Over Time During the 6-Month Double-Blind Period in Patients With AIP (mFAS)

AIP = acute intermittent porphyria; ALA = aminolevulinic acid; IQR = interquartile range; mFAS = modified full analysis set.

Source: Source: Clinical Study Report.⁶

Figure 10: Median (IQR) Urinary PBG Levels Over Time During the 6-Month Double-Blind in Patients With AIP (mFAS)

AIP = acute intermittent porphyria; IQR = interquartile range; mFAS = modified full analysis set; PBG = porphobilinogen.

Source: Source: Clinical Study Report.⁶

Subgroup Analyses

Assessments of urinary ALA levels were analyzed by subgroups reflecting the historical frequency of attacks, or by patients with a high AAR versus a low AAR.

In patients with AIP and a high AAR, baseline levels of ALA were 24.35 mmol/mol Cr (SD = 21.47 mmol/mol Cr) and 19.67 mmol/mol Cr (SD = 8.10 mmol/mol Cr) in the givosiran and placebo groups, respectively. The changes from baseline at months 3 and 6 were 2.47 mmol/mol Cr (95% CI, –1.74 to 6.68) and 6.43 mmol/mol Cr (95% CI, –2.37 to 15.22), respectively, for the givosiran group and 23.58 mmol/mol Cr (95% CI, 19.06 to 28.09) and 28.84 mmol/mol Cr (95% CI, 18.81 to 38.87), respectively, for the placebo group. This corresponded to a between-groups difference of –21.11 (95% CI, –27.31 to –14.90) at month 3 and–22.41 (95% CI, –35.87 to –8.96) at month 6.

In patients with AIP and a low AAR, baseline levels of ALA were 15.60 mmol/mol Cr (SD = 8.74 mmol/mol Cr) and 15.65 mmol/mol Cr (SD = 12.71 mmol/mol Cr) for patients in the givosiran and placebo groups, respectively. The changes from baseline at month 3 and month 6 were 1.32 mmol/mol Cr (95% CI, –2.27 to 4.91) and 1.84 mmol/mol Cr (95% CI, –0.84 to 4.53), respectively, for the givosiran group and 15.50 mmol/mol Cr (95% CI, 11.84 to 19.17) and 14.06 mmol/mol Cr (95% CI, 11.25 to 16.87), respectively, for the placebo treatment group. This corresponded to a between-groups difference of –14.19 (95% CI, –19.32 to –9.06) at month 3 and –12.22 (95% CI, –16.10 to –8.33) at month 6.

The subgroup results for patients with AHP were similar to those discussed for patients with AIP only (Table 21).

Mortality

Mortality was not reported as an efficacy outcome in Study 003.

Harms

Only those harms identified in the review protocol are reported here. See Table 22 for detailed harms data.

Adverse Events

Over the 6-month DB treatment period in patients with AIP, 89% of patients receiving givosiran reported at least 1 AE, as did 81% of patients receiving placebo. Overall, the most frequently reported AEs in patients receiving givosiran versus placebo were nausea (28% versus 12%), injection-site reaction (17% versus 0), headache (13% versus 16%), and urinary tract infection (7% versus 14%). The following AEs were reported at least 5% more frequently in patients treated with givosiran: nausea, injection-site reaction, chronic kidney disease, fatigue, increase in ALT, decrease in glomerular filtration rate, and rash. The following AEs were reported at least 5% more frequently in patients in the placebo group: urinary tract infection, vomiting, back pain, pyrexia, and hypoesthesia. Reporting of AEs was comparable between patients with AIP and AHP.

Serious Adverse Events

In patients with AIP, at least 1 SAE was reported by 17% (n = 8) of patients in the givosiran group and 9% (n = 4) of patients in the placebo group over the 6-month treatment period. For patients with AHP, 21% (n = 10) and 9% (n = 4) of patients in the givosiran and placebo groups, respectively, reported at least 1 SAE. Among patients with AIP, infections and infestations were reported more frequently in the placebo group (n = 3) than in the givosiran group (n = 1). Chronic kidney disease was reported by 2 patients in the givosiran group and by no patients in the placebo treatment group. Additionally, hypoglycemia, major depression, asthma, and a surgical or medical procedure for pain management were each reported by 1 patient in the givosiran group and by no patients in the placebo treatment group.

The frequencies of specific SAEs in patients with AHP was similar to those reported in patients with AIP.

Withdrawals Due to Adverse Events

One patient from the givosiran group withdrew from treatment due to an AE. The patient |||||||||||| discontinued treatment because of ALT elevation.

Mortality

No deaths were reported during the 6-month DB period of Study 003.

Notable Harms

Motor neuropathy, hepatocellular carcinoma, injection-site reactions, transaminase elevation, and progression of renal impairment were included in the CADTH systematic review protocol as notable harms. In patients with AIP, injection-site reactions and progression of renal impairment were reported in 8 (17%) and 4 (9%) patients, respectively, in the givosiran group, and by 0 patients in the placebo group. Increased ALT was reported by 4 (9%) patients receiving givosiran and by 1 (2%) patient receiving placebo, while increased AST was reported in 3 (7%) patients and 1 (2%) patient receiving givosiran and placebo, respectively. Three (7%) patients in the placebo group reported motor neuropathy as an AE; none reported this in the givosiran group. There were no cases of hepatocellular carcinoma reported during the 6-month treatment period. The frequency of notable harms was the same for patients with AHP, with the exception of progression of renal impairment, which was reported in 5 (10%) patients in the givosiran group.

Critical Appraisal

Internal Validity

Randomization was conducted using an interactive response system, which is an adequate method. Randomization was stratified by type of AHP (AIP versus non-AIP). Patients with AIP were further stratified by prophylactic hemin use and high or low historical attack rate in the 12 months before randomization. More specifically, patients who used hemin prophylactically before study entry were stratified by a historical AAR of fewer than 7 attacks versus 7 attacks or more. Patients who were not using hemin prophylactically before study entry were stratified by a historical AAR of fewer than 12 attacks versus 12 attacks or more. Although the sponsor reported that the cut-offs used for high and low historical AAR were based on the EXPLORE study, the criteria used to define high and low historical annualized attack rates were fairly arbitrary, according to the clinical experts; therefore, it is unclear how meaningful the stratification was. Study 003 was a DB study, and efforts were made to maintain blinding; however, injection-site reactions (which were reported solely in the givosiran group) offered the potential for unblinding, as did higher rates of nausea. Additionally, a major protocol deviation was reported for 4 patients who experienced a total of 17 instances of study drug administration (givosiran or placebo) with a syringe that was not masked with a blinding strip before study drug administration. However, this affected the givosiran and placebo groups equally and is not expected to bias the results.

Despite a small sample size, the treatment groups were well balanced overall. Differences to note include a slight difference in race, although feedback from the clinical experts indicated that AHP does not present differently by race. On average, patients in the placebo group had received a diagnosis of AHP more recently than patients in the givosiran group, by 3 years. A greater proportion of patients receiving givosiran had a history of prophylactic hemin use compared to those in the placebo group. Whether there would be a difference in the level of disease control between the 2 treatment groups is uncertain. The treatment groups differed in their history of medical conditions, including their history of peripheral neuropathy, gastrointestinal disorders (constipation, gastroesophageal reflux disease, and nausea), psychiatric disorders (depression and anxiety), and anemia. Feedback from the clinical experts indicated that patients’ history of medical conditions may not be related to AHP. The impact of these on outcomes such as daily worst nausea and HRQoL is unknown.

Overall dropout rates were extremely low. One patient with a non-AIP type of AHP who was randomized to placebo withdrew from the study due to an AE (elevated ALT). Missing data were not an issue for outcomes, and adherence was strong throughout the study. Among patients with AIP, 89% and 98% of those randomized to the givosiran and placebo groups did not miss a dose during the 6-month treatment period; adherence was similar among patients with AHP. All of the reported missed doses were due to administrative and visit scheduling issues, with the exception of 2 doses missed by 1 patient in the givosiran group, whose treatment was withheld due to transaminase elevations after 4 doses.

The primary outcome in Study 003 was the annualized rate of porphyria attacks, which was a composite end point that included attacks requiring hospitalization, urgent health care visit, or IV hemin administration at home in patients with AIP during the 6-month DB period. Of note, the sponsor indicated that “the analysis of the primary end point in patients with AIP was to allow evaluation of efficacy in a homogenous population and because of the scarcity of non-AIP subtypes compared to AIP.”⁶ The 3 components are objective measures of attacks; therefore, they are unlikely to contribute to the misidentification of a porphyria attack. Further, the sponsor noted that the 3-component primary end point was supported by results from the EXPLORE natural history study,²⁵ which demonstrated that the majority of porphyria attacks were aligned with this definition.³⁵ The primary end point excludes porphyria attacks that occur at home that do not require IV hemin; however, a sensitivity analysis was conducted that included all porphyria attacks, which was consistent with the primary analysis.

The components of the composite end point were reported descriptively. According to the clinical experts, the definition used for a porphyria attack requiring hospitalization was suggestive of an attack that is fairly severe. However, the experts also noted that hospitalization of a patient and options for are determined by a physician and clinical practice, which may have varied across study sites. The possibility of unblinding (previously described) also had the potential to influence patient or physician beliefs around treatment. Despite the observed variation in the rate reduction of attacks for each component, the ability to interpret the individual components of the composite end point is limited by potential bias. However, the estimates of effect for each of the components were in the same direction and not expected to have substantially affected the overall composite outcome.

A number of patient-reported outcomes were included in Study 003, including the SF-12, EQ-5D-5L, and PGIC to evaluate HRQoL; a single item of the BPI-SF, a single item of the BFI-SF, and NRS for daily worst nausea to evaluate symptoms of AHP; and the PPEQ to evaluate activities of daily living. With the exception of the PPEQ, patient-reported outcomes were not validated in patients with AHP, and disease-specific MIDs were not available for most of the outcomes; thus, the ability to appropriately evaluate the treatment effect based on these outcomes was limited. The patient-reported outcomes are broadly used in clinical trials and well validated for other populations, but evidence of validity of these for the use of the single item for daily worst pain from the BPI-SF was not identified. The PPEQ is specific to patients with porphyria; however, evidence of validity, reliability, responsiveness, or an MID was not identified for this outcome. The remaining outcomes included in this review — namely, the assessments of rescue medication use, opioid use, ability to attend work or school, and urinary ALA and PBG — were objective and subject to few limitations.

Study 003 was adequately powered to detect a change in the primary outcome, and a hierarchical testing procedure was employed to control for multiplicity among the secondary end points. Most of the secondary end points included in the testing hierarchy were specific to patients with AIP, with the exception of AAR in patients with AHP. Secondary outcomes were also evaluated in all patients with AHP, but included as exploratory outcomes and subject to substantial uncertainty. The statistical testing strategy stopped testing of the remaining end points in the hierarchy following a test that was not statistically significant. In Study 003, the results for daily worst pain were not statistically significant; therefore, nominal P values whose significance cannot be interpreted were reported for daily worst fatigue, daily worst nausea, and the change in PCS of the SF-12, which are clinically relevant outcomes considered important to patients.

There were no missing data for the following outcomes: composite AAR, annualized rate of hemin use, opioid usage. For ALA and PBG levels, 6% to 7% of data were missing; less than 3% of data were missing for the SF-12 and EQ-5D-5L. Two patient-reported exploratory outcomes, PGIC and PPEQ, were missing data for 20% of patients; according to the sponsor, this was because these outcomes were added through an amendment to the protocol. Data were imputed for missing patient-reported outcomes of pain, fatigue, and nausea following a pre-specified algorithm describing the methods used to account for missing data. Of note, missing data for up to 8% of patients for most of the DB period were unlikely to bias the results, but data were missing for 30% of patients at the week 24 visit (month 6). A post hoc reanalysis of end points using a non-parametric Wilcoxon test was conducted for outcomes that demonstrated a significant deviation from a normal distribution, namely daily worst pain, urinary ALA levels, and urinary PBG levels. Although this was performed post hoc, a non-parametric test was considered appropriate. Most of the outcomes included in the trials were reported as a rate ratio or mean change from baseline. The assessments of pain, fatigue, and nausea were conducted using the AUC to account for variability in symptoms for patients with AHP, which was considered appropriate.

Additionally, a preplanned, unblinded interim analysis was conducted to ensure that the study was adequately powered for the primary end point comparison at the final analysis. The primary analysis and subsequent secondary analyses were conducted at a 2-sided significance level of 0.049, reflecting a penalty of 0.001 for an unblinded interim analysis.

Subgroup analyses of the primary end point and urinary ALA secondary end point were pre-specified. As noted, patients with AIP were stratified by historical AAR at baseline, although the cut-offs used were arbitrary, according to the clinical experts. Rate ratios were calculated for the subgroup analyses, but statistical tests were not conducted. Subgroup analyses were not adjusted for multiplicity, and the trial was not powered to detect a difference based on the subgroups. CIs were wide, but did not cross 0.

External Validity

Givosiran is indicated for the treatment of adults with AHP in Canada. The diagnostic criteria used in the clinical trial were consistent with clinical practice, according to feedback from the clinical experts. The pivotal trial was composed of patients with AIP and non-AIP types of AHP, and was analyzed as such. In total, 95% of the study population were patients with AIP. However, feedback from the clinical experts on this review indicated that the trial data applied to all patients with AHP because of the mechanisms of action of givosiran and disease pathology, which were not expected to result in a differential treatment effects by AHP type. In terms of the study population being reflective of Canadian patients with AHP, the mean age of patients in the pivotal trial was higher than that of patients seen in clinical practice; the clinical experts described the average age of their patient population as between 20 years and 30 years. Of note, patients at least 12 years of age were eligible to enrol in Study 003; however, the patients who actually participated in Study 003 were 19 years or older and, as a result, generalizable to the patient population under review. Another consideration regarding the generalizability of the study population was that 41% to 48% of patients resided in Europe, where there is a higher prevalence (and therefore greater awareness) of this rare disease. The clinical experts relayed that because of the lack of available treatments for AHP, standard of care for a patient with a diagnosis of AHP is unlikely to vary by region. However, this is something to consider. The additional baseline characteristics and medical histories of the study population were typical of Canadian patients living with AHP.

Based on the composite definition of porphyria attacks used for the primary end point, at baseline, the median numbers of porphyria attacks in the 6 months before screening were 4 (range = 2 to 24) and 3 (range = 0 to 25) for patients with AIP randomized to givosiran and patients with AIP randomized to placebo, respectively. According to the clinical experts, patients in the study had a history of more frequent porphyria attacks than most patients seen in clinical practice in Canada. The sponsor acknowledged this, stating that “the study was enriched for attack frequency to ensure the ability to measure a difference in treatment effect on the primary composite porphyria attack end point.”⁶ As a result, the change in the rate of porphyria attacks observed in each of the treatment groups in the trial may not be generalizable to what would be expected in clinical practice.

The use of givosiran in Study 003 was aligned with the Health Canada–approved indication. Although the indication for givosiran is broad, the clinical experts anticipated that it would be used to prevent porphyria attacks rather than to treat acute attacks in clinical practice; this aligns with its use in Study 003. Patients receiving prophylactic hemin were required to discontinue its use before the study; screening for this occurred at least 4 days following discontinuation. Patients who were receiving treatment with a GnRH analogue at screening were permitted to enrol if they remained on GnRH treatment throughout the 6-month DB period, but the number of patients using GnRH before enrolment was not reported. The use of hemin (to treat acute attacks) and GnRH in the trial was consistent with standard of care in clinical practice. The duration of treatment evaluated in the DB treatment period was 6 months, followed by an open-label treatment period of up to 29 months. The clinical experts indicated that they would expect patients to remain on treatment unless there was a specific safety or biological reason to discontinue it; therefore, a 6-month evaluation period is not representative of the long-term use of givosiran in clinical practice. The duration of the trial may limit the ability to evaluate the impact of givosiran on HRQoL and safety outcomes.

The outcomes of greatest importance to clinicians and patients were frequency of AHP attacks, management of symptoms, and HRQoL, in addition to a reduction in side effects from treatment. All of these outcomes were included in the pivotal trial in some capacity. The clinical experts indicated that formal evaluation of HRQoL— through the use of questionnaires, for example — is not done in clinical practice, where HRQoL is evaluated informally. In the study, patient-reported symptoms, such as pain, fatigue, and nausea, were recorded using an electronic diary and measured using an NRS, which is also unlikely in clinical practice but important to patients. The definitions used for AHP attacks were consistent with those used in clinical practice.

Additionally, the clinical experts indicated that patients with a history of attacks related to AHP are accustomed to facing scarce treatment options for an acute attack, and often tend to treat their attacks symptomatically at home. Therefore, it is plausible that participation in the clinical trial would increase patients’ access and attention to care and, in turn, their likelihood of seeking care for acute attacks. This could affect the generalizability of the trial results to clinical practice and has the potential to further inflate the reported attack rate.

Indirect Evidence

A focused literature search for network metanalyses dealing with porphyrias was run in MEDLINE All (1946–) and Embase (1974‒) through Ovid on March 24, 2021. No limits were applied to the search. Indirect treatment evidence for givosiran was not identified in this review.

Other Relevant Evidence

A phase I study, a phase I/II study, a phase III OLE study, and an expanded access program (EAP) were included in the sponsor’s submission to CADTH and were considered to address important gaps in the evidence included in the systematic review.

Study ALN-AS1-001 (001) and Study ALN-AS1-002 (002) provided additional information and long-term data on the safety and tolerability of givosiran in adult patients with AIP. The ENVISION ALN-AS1-003 study (Study 003 OLE) provides additional long-term evidence on the safety and efficacy of givosiran in adult patients with AHP. Study ALN-AS1-005 (005) is an EAP of givosiran for patients with AHP. Study 002 and Study 003 OLE were ongoing at the time of this review.

Phase I (Completed) and Phase I/II (Ongoing) Clinical Trials: Study ALN-AS1-001 and Study ALN-AS1-002

Methods

Study 001 was a 3-part, multi-centre, placebo-controlled, phase I study of the safety and tolerability of subcutaneous givosiran for the treatment of adults with AIP. Parts A, B, and C were single-ascending dose, multiple-ascending dose, and multidose in design, respectively. The adaptive design allowed for different dosing regimens and dose levels to be assessed based on new safety, tolerability, and PD data. Study 001 was conducted at 6 sites in Great Britain, Sweden, and the US. The study was initiated on May 6, 2015, and completed on September 6, 2017. In total, 40 patients with AIP who are CHEs were randomized to parts A and B (n = 23), while those with AIP who had recurrent attacks were randomized to part C (n = 17).

In all 3 parts, patients were randomized 3:1 to receive givosiran or placebo. Parts A and B had 60-day screening periods before treatment was administered. Part A was single-blind, single-ascending dose with 5 possible cohorts, where each cohort received a different dose of givosiran. Patients in a cohort received a single administration of 1 dose of givosiran or placebo. The following doses were assessed: 0.035 mg/kg, 0.10 mg/kg, 0.35 mg/kg, 1.0 mg/kg, and 2.5 mg/kg, each administered to a different cohort. Patients who completed part A could take part in a subsequent part A cohort or in part B if they met the eligibility criteria. Part B was a single-blind, multiple-ascending dose, and patients received 2 injections over a 28-day period of 1 of the following doses: 0.35 mg/kg givosiran, 1.0 mg/kg givosiran, or placebo. Patients in parts A and B were followed for up to 6 weeks after the treatment period. Part C was DB and multidose, and after the 168-day run-in period, patients received doses of 2.5 mg/kg givosiran, 5.0 mg/kg givosiran, or placebo once monthly (4 injections total) or once every 3 months (2 injections total) for a 3-month period. Patients in part C were followed for up to 3 months after the treatment period. Patients who completed part C were able to participate in Study 002, an OLE.

Study 002 was an ongoing, multi-centre, open-label, phase I/II study of the long-term safety and tolerability of subcutaneous givosiran for the treatment of adults with AIP who completed part C of Study 001. Study 002 (N = 16) was conducted at 5 sites in Great Britain, Sweden, and the US.³⁶ The study began on October 24, 2016³⁶; the data presented in this summary are based on the interim report, with a cut-off date of December 13, 2018. Patients received givosiran 2.5 mg/kg monthly or 5.0 mg/kg monthly or every 3 months until the safety review committee assessed safety, tolerability, and efficacy data and agreed that all patients would transition to receive the 2.5 mg/kg dose. Treatment duration was estimated to be up to 36 months, with the estimated total time in study with screening and baseline up to 44 months.

Figure 11, Figure 12, Figure 13, and Figure 14 outline the study designs for parts A, B, and C of Study 001 and for Study 002, respectively.

Figure 11: Design of Study 001, Part A

ALA = aminolevulinic acid; ALN-AS1 = givosiran; D = day; ET = early termination; PBG = porphobilinogen; Q2W = every 2 weeks; Q4W = every 4 weeks; W = week.

Source: Clinical Study Report for Study ALN-AS1-001.³⁷

Figure 12: Design of Study 001, Part B

ALA = aminolevulinic acid; ALN-AS1 = givosiran; D = day; ET = early termination; PBG = porphobilinogen; Q2W = every 2 weeks; Q4W = every 4 weeks; W = week.

Source: Clinical Study Report for Study ALN-AS1-001.³⁷

Figure 13: Design of Study 001, Part C

ALA = aminolevulinic acid; ALN-AS1 = givosiran; D = day; ET = early termination; PBG = porphobilinogen; Q4W = every 4 weeks; W = week.

Source: Clinical Study Report for Study ALN-AS1-001.³⁷

Figure 14: Design of Study 002

ALN-AS1-001 = Study 001.

Source: Clinical Study Report of ALN-AS1-002.³⁸

Populations

The inclusion and exclusion criteria for studies 001 and 002 are outlined in Table 23. In short, adults between 18 years and 65 years were eligible if they had a diagnosis of AIP. More specifically, patients in parts A and B were CHEs (and had experienced no porphyria attacks within 6 months of the study), while those in part C had to have experienced at least 2 porphyria attacks in the 6 months before the run-in period. Patients in Study 002 must have completed and tolerated the study drug dosing in Study 001, part C. Furthermore, patients were to stop receiving prophylactic hemin during either Study 001 or Study 002. Table 24 summarizes the baseline characteristics of patients who received either placebo or givosiran 2.5 mg/kg in Study 001 and all patients in Study 002.

In Study 001, part A, 3 patients received givosiran 2.5 mg/kg and 6 received placebo, while in part C, 3 patients received givosiran 2.5 mg/kg and 4 received placebo. Briefly, patients who were treated with givosiran 2.5 mg/kg had mean ages of 42 years (SD = 11.6 years) and 30 years (SD = 10.7 years) for Study 001, parts A and C, respectively. The majority of patients in the 2.5 mg/kg cohort of Study 001 were female and White, with a mean body mass of 82.5 kg (SD = 10.45 kg) in part A and 84.5 kg (SD = 16.86 kg) in part C. Patients in the 2.5 mg/kg cohort of part C had a mean of 14.7 attacks (SD = 18.9 attacks) in the 12 months before the study. One-third of patients in the study were on prophylactic hemin.

Initially, patients could receive 2.5 mg/kg or 5.0 mg/kg of givosiran. Based on a review of the safety data, it was later decided that all 16 patients were to receive 2.5 mg/kg. Baseline characteristics for Study 002 were similar. Patients had a mean age of 37.4 years (SD = 12.0 years); 87.5% were female; 81.3% were White; and patients had a mean body mass of 75.84 kg (SD = 18.46 kg). Nearly all patients (93.8%) in Study 002 had experienced at least 1 porphyria attack in the 12 months preceding the study, with a mean of 13.0 attacks (SD = 13.1 attacks) during that time period. All had used hemin during an acute attack, and half had used it prophylactically.

Interventions

In Study 001, patients received subcutaneous givosiran or placebo on an inpatient basis at the study centre. For part A, a single administration of givosiran (0.035 mg/kg, 0.1 mg/kg, 0.35 mg/kg, 1.0 mg/kg, or 2.5 mg/kg) or placebo was given. In part B, patients received 2 monthly doses of givosiran (0.35 mg/kg or 1.0 mg/kg) or placebo. For part C, patients received doses of 2.5 mg/kg givosiran, 5.0 mg/kg givosiran, or placebo once monthly (4 injections total) or once every 3 months (2 injections total) for a 4-month period.

In Study 002, patients received subcutaneous givosiran 2.5 mg/kg monthly or 5.0 mg/kg monthly or every 3 months. Data from Study 001 indicated that the 2.5 mg/kg and 5.0 mg/kg doses both demonstrated acceptable safety and more stability in lowering delta-ALA when administered monthly versus every 3 months. It was also noted that the higher dose of givosiran did not confer any additional benefit in terms of lowering delta-ALA levels. As a result, all patients receiving 5.0 mg/kg (monthly or every 3 months; n = 7) were transitioned to the lower dose, such that all study patients were receiving 2.5 mg/kg monthly. This was the dose used in the phase III ENVISION study and the intended commercial dose.

Prophylactic hemin use was not permitted during either study, though hemin use was allowed during acute attacks.

Outcomes

The primary outcomes of Study 001 were the safety and tolerability of givosiran for the treatment of patients with AIP who are CHEs or have recurrent attacks. Secondary end points included pharmacokinetic (PK) and PD evaluations. Exploratory end points included additional PD investigations and, for part C only, the number of porphyria attacks and hemin doses, among other clinical activity measures. Porphyria attacks were reported as exploratory efficacy end points rather than as AEs. Attacks were analyzed based on 2 definitions: all attacks (comprising attacks requiring hospitalization, urgent health care visit, or hemin administration at home, or attacks at home that did not require hemin administration); and composite attacks (comprising attacks requiring hospitalization, urgent health care visit, or hemin administration at home). The individual components of the attack rate were also analyzed. The EQ-5D-5L was captured as an exploratory outcome, and harms data for AEs, SAEs, WDAEs, and deaths were collected.

The primary outcomes of Study 002 were the long-term safety and tolerability of givosiran for the treatment of patients with AIP and recurrent attacks. Secondary end points included PD evaluations and clinical activity assessments. Exploratory end points included PK analyses, HRQoL assessments, and biochemical investigations. The clinical activity of givosiran was assessed by the frequency and characteristics of porphyria attacks and doses of hemin administered. HRQoL was assessed using the EQ-5D-5L and BPI-SF.

Statistical Analysis

The sample size for Study 001 was not based on power calculations. The study planned to enrol up to 64 patients. Summary statistics were presented for continuous data, and event count data were summarized as annualized rates and SEM estimates. No inferential statistical analyses were planned for safety assessments in the study. No imputations were made for missing data. The safety population was the main analysis set used for evaluating safety and clinical activity.

The sample size for Study 002 was not based on power calculations. It was made up of patients who had completed Study 001, part C. No formal statistical analyses were performed. Descriptive statistics were presented for continuous variables, while frequencies and percentages were presented for categorical and ordinal variables. In Study 002, missing continuous and categorical variables were not imputed unless otherwise specified, and no imputations were made for missing or partially missing dates. Interim analyses were descriptive and did not involve formal hypothesis testing. The safety population was the main analysis set used for evaluating safety and clinical activity.

Patient Disposition

The patient disposition for studies 001 and 002 is summarized in Table 25.

In Study 001, 49 individuals were screened, 40 of whom were randomized. Parts A and B had a total of 23 unique patients where 20 patients and 16 patients made up the givosiran and placebo groups, respectively. Three patients participated in both parts A and B, and 2 patients took part in 2 different dose cohorts of part A. Each unique patient was only counted once in each column of the summaries. Part C consisted of 17 patients, where 13 patients and 4 patients made up the givosiran and placebo groups, respectively. All patients completed treatment in parts A and B, while 94.1% of patients in part C completed treatment. In the latter, there was 1 WDAE when a patient in the givosiran 5.0 mg/kg group had a fatal SAE.

All 16 patients who completed Study 001, part C took part in Study 002. At the time of the interim report, no patients had completed treatment, and 1 patient (6.3%) had discontinued due to an AE.

The safety populations for studies 001 and 002 included all patients who received at least 1 dose of the study drug.