CADTH Reimbursement Review

Pegvaliase (Palynziq)

Sponsor: BioMarin Pharmaceutical (Canada Inc.)

Therapeutic area: Phenylketonuria

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

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

Table 1: Submitted for Review

NOC = Notice of Compliance.

Source: CADTH review submission¹ and draft product monograph² for pegvaliase.

Introduction

Phenylketonuria (PKU) is a monogenic autosomal recessive disorder and 1 of the most common inborn errors of metabolism.³ Patients with PKU have mutations in both alleles of the PAH gene encoding phenylalanine hydroxylase (PAH), an enzyme that catalyzes the conversion of phenylalanine (Phe) to tyrosine using tetrahydrobiopterin as a cofactor.³ A wide variety of PAH gene mutations give rise to variations in clinical phenotype and disease severity.^3,4 Deficiency of PAH leads to uncontrolled blood Phe, which then crosses the blood-brain barrier, where it has neurotoxic effects.^3,4 Phenylketonuria is universally identified via newborn screening programs in Canada; uncontrolled Phe levels in untreated PKU during early childhood profoundly impair brain function and development.⁵ In adolescents and adults, uncontrolled Phe levels are associated with behavioural and psychiatric problems (inattentiveness and mood dysfunction, often collectively referred to as “executive dysfunction”).⁶ Symptoms of PKU, in conjunction with treatments, negatively affect the health-related quality of life (HRQoL) of patients with respect to employment, social relationships, and mental health.⁷

Phenylketonuria is rare, with an incidence of approximately 1 in 12,000 to 1 in 15,000 live births in Canada (equivalent to approximately 300 new cases per year). According to the sponsor, approximately 3,133 patients are living with PKU in Canada at present, of whom approximately ||||| are being managed and approximately ||| are 16 years of age or older and currently being treated with sapropterin.¹ PKU is diagnosed shortly after birth by newborn screening using biochemical and genetic tests. Physicians specializing in genetics and metabolic diseases are required to diagnose, treat, and monitor patients with PKU at hospital-based genetic or metabolic clinics that have support from dietitians trained in PKU management.

According to clinical experts consulted by CADTH for this review, the current cornerstone of PKU treatment is lifelong dietary control of Phe intake to curb blood Phe levels. This is principally accomplished by providing Phe-free foods and metabolic formulas with a small amount of “complete” Phe-containing protein allowed on top, sometimes collectively referred to as medical nutritional therapy (MNT). Adherence of adult patients with PKU to MNT is extremely challenging because low-protein medical food is very unpleasant to taste and smell. Other than dietary restriction, the only other approved medication is sapropterin, a cofactor of the deficient PAH enzyme in PKU. Approximately 25% of patients with milder PKU have a biochemically detectable response to sapropterin.⁸

Pegvaliase is a recombinant Phe ammonia lyase enzyme that converts Phe to ammonia and trans-cinnamic acid. The Health Canada indication for pegvaliase is “Palynziq (pegvaliase injection) is indicated to reduce blood phenylalanine concentrations in patients with phenylketonuria (PKU) aged 16 years and older who have inadequate blood phenylalanine control (blood phenylalanine levels greater than 600 micromol/L) despite dietary management.” Pegvaliase is supplied as a solution (2.5 mg/0.5 mL [5 mg/mL], 10 mg/0.5 mL [20 mg/mL], 20 mg/mL) in pre-filled syringes. Pegvaliase is self-administered at a titrated maintenance dose (following induction and titration) required to achieve a blood Phe level of 600 µmol/L or lower by subcutaneous injection. The objective of this report was to perform a systematic review of the beneficial and harmful effects of pegvaliase (self-administered subcutaneous injection, titrated to a maintenance dose required to achieve blood Phe levels of 600 μmol/L or lower; maximum dose 60 mg daily) for the treatment of patients with PKU aged 16 years and older who have inadequate blood Phe control (blood Phe levels greater than 600 μmol/L) on existing management.

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

One submission, from the Canadian PKU and Allied Disorders (CanPKU) association, was received for this review. Between November 30, 2021, and December 25, 2021, CanPKU conducted online surveys of 68 patients with PKU (46 patients in Canada and 14 in the US) and telephone interviews with 5 patients experienced with pegvaliase (1 patient in Canada and 4 in the US). Respondents described how PKU symptoms and the PKU protein-restricted diet had affected their physical and mental health, employment, and social relationships. Almost all respondents (≥ 95%) had experience with low-protein medical foods and formulas and 65% had experience with Kuvan (sapropterin), while only 21% had experience with pegvaliase. Respondents described barriers to existing therapies, including poor taste, lack of satiety, inconvenient preparation and administration, high cost, and limited availability.

The vast majority (≥ 85%) of respondents identified Phe control, reducing PKU symptoms, limiting long-term disease consequences, improving neurocognitive function, managing diet, reducing burden of treatment, improving HRQoL, and increasing natural protein intake as key outcomes of interest. Respondents experienced with pegvaliase reported that the drug limited long-term disease consequences; controlled Phe levels; reduced PKU symptoms; and had tolerable side effects, such as injection-site reactions, joint pain, and skin reactions.

Clinician Input

Input From Clinical Experts Consulted by CADTH

Two clinical specialists with expertise in the diagnosis and management of pediatric and adult patients with PKU who have inadequate Phe control provided input for this review. The clinical experts stated that currently available therapies (MNT with or without sapropterin) can in theory successfully meet treatment goals by decreasing Phe levels and preventing the neuropsychological complications of PKU. However, because adherence to MNT is generally low, MNT is not effective in most patients and only a small proportion of patients with milder PKU will respond to sapropterin. Pegvaliase would be used as last-line treatment following MNT and, if appropriate, sapropterin. Pegvaliase may shift the treatment paradigm for some adult patients with PKU by allowing liberalization of diets while maintaining Phe control.

The 2 clinical experts consulted by CADTH for this review differed on the subset of patients with PKU who would benefit most from pegvaliase. One expert reported that patients with high and uncontrolled Phe are most in need of an intervention to improve metabolic control that will lead to a decrease in their Phe levels and improved Phe tolerance. The clinical expert could not rule out the possibility that patients who are poorly compliant with MNT could become more compliant to therapies (including pegvaliase) over time as Phe levels decrease and their focus improves. The second clinical expert stated that patients with PKU who are highly compliant with MNT and other therapies and have the most severe forms of PKU would be the most suitable for treatment with pegvaliase. These patients are generally able to achieve Phe levels within the control range but have the most unpalatable diets and experience large deviations in Phe levels. These patients can be identified by assessing compliance with MNT and other therapy (using mean Phe values) and PKU severity (using PAH genotyping, variability in Phe levels, and/or degree of restriction of complete protein intake). This clinical expert indicated that patients who are noncompliant with therapy would be the least suitable for treatment with pegvaliase.

The clinical experts agreed that complete protein tolerance (or Phe tolerance) and blood Phe levels are the most convenient tests to assess response to treatment and are most often used in clinical trials. Clinically meaningful responses to treatment would be, in order of importance, increased complete protein tolerance (or Phe tolerance) and protein intake to levels in the general population, improvement in HRQoL, and improvement in psychological metrics (neurocognitive performance, mood, attention, and working memory). According to the clinical experts, high blood Phe levels can be used to show that pegvaliase treatment is ineffective, and stability of Phe levels with the treatment range can demonstrate improvements in protein tolerance in patients with low Phe who liberalize their diets to include natural foods. Patients who cannot maintain acceptable Phe levels (or whose levels are not monitored) with MNT and pegvaliase are noncompliant and should be discontinued from treatment, as should patients who experience significant adverse reactions.

Clinician Group Input

A group of 3 physicians who care for adult patients with PKU in Canada provided input for this review. Although the clinician group echoed the challenges in adhering to the PKU diet and the limited proportion of patients who can benefit from sapropterin, views contrasting with those of the clinical experts consulted by CADTH were presented on: the connection between blood Phe levels and neurologic symptoms, diet liberalization, and associated impacts on HRQoL in adult patients with PKU, which the clinician group reported were tightly and reversibly linked; the importance of Phe control as a treatment goal and marker of treatment response in and of itself; the patient subset most suitable for pegvaliase treatment, which the clinician group described as patients noncompliant with dietary restrictions who cannot benefit from sapropterin and therefore have poor or no Phe control; the patient subset least suitable for pegvaliase treatment, which the clinician group identified as patients able to maintain Phe levels within a target range on MNT with or without sapropterin; and the risks of Phe levels below 30 µmol/L resulting from overtreatment with pegvaliase in patients who do not comply with Phe monitoring, which the clinician group described as a potential concern. According to the clinician group, pegvaliase would be offered as last-line treatment for adult patients with PKU who have elevated Phe levels and neuropsychiatric symptoms and are able to self-administer the injection.

Drug Program Input

The Formulary Working Group identified the following jurisdictional implementation issues: relevant comparators, considerations for initiation of therapy, considerations for continuation or renewal of therapy, considerations for discontinuation of therapy, considerations for prescribing of therapy, generalizability, care provision, and system and economic issues. The clinical experts consulted by CADTH for this review weighed evidence from the included study and other clinical considerations to provide responses to the CADTH Provincial Advisory Group’s drug program implementation questions. Table 4 provides more details.

Clinical Evidence

Pivotal Studies and Protocol-Selected Studies

Description of Studies

PRISM-2 was a phase III, 4-part, 4-arm, double-blind, placebo-controlled randomized discontinuation trial (RDT) with an extension period of open-label treatment (N = 215).^9-16 The major feeder study for the PRISM-2 trial was the PRISM-1 trial,^16,17 a phase III, open-label study to assess the safety and tolerability of 2 pegvaliase dosage regimens (20 mg or 40 mg once daily; details are provided in the Other Relevant Evidence section). The main eligibility criteria for feeder studies were patients with PKU aged 16 years or older with blood Phe levels of greater than 600 µmol/L who were able to maintain a consistent diet. Dietary Phe control and adherence to MNT was not a requirement for participation in the feeder studies or the PRISM-2 trial. Following enrolment and screening at 29 centres in the US, patients either entered into part 1 (open-label Phe assessment) or directly into part 4 (open-label extension). In part 1, patients were randomized in a 1:1 ratio to receive open-label pegvaliase (20 mg or 40 mg once daily, vial and syringe) for up to 13 weeks; blood Phe levels were measured every 2 weeks. Patients who achieved a mean blood Phe reduction of 20% or greater from treatment-naive baseline and who were able to maintain their randomized pegvaliase dose were eligible for inclusion in the part 2 (RDT) modified intention-to-treat (mITT) set, while those who did not achieve this degree of Phe reduction or were unable to maintain their randomized pegvaliase dose due to adverse events (AEs) transitioned directly to part 4 (the open-label extension). In part 2, patients in each dose group (20 mg or 40 mg once daily, vial and syringe) were randomized 2:1 to either continue receiving their assigned dose of pegvaliase or to receive a matching-administration placebo over 8 weeks of double-blind treatment. In part 3, patients who completed part 2 received open-label pegvaliase (dose as assigned in part 1) in 2 formats (vial and syringe or pre-filled syringe) for 6 weeks and pharmacokinetics and pharmacodynamics were compared. Part 4 was an open-label extension in which patients received open-label pegvaliase (up to 60 mg once daily, pre-filled syringe) for up to 274 weeks. Only data for part 2 of the PRISM-2 RDT are described in the Systematic Review section of this report.

The primary objective of the PRISM-2 study was to evaluate the efficacy of pegvaliase in decreasing blood Phe levels by observing changes from part 2 baseline to part 2, week 8 in patients previously exposed to pegvaliase who received either pegvaliase (20 or 40 mg/day) or a matching-administration placebo in the RDT. Secondary objectives (all hierarchically tested) included comparing changes in the Attention Deficit Hyperactivity Disorder Response Scale (investigator-rated) (ADHD-RS-IV) inattention subscale scores (among participants with drug-naive baseline scores > 9 as well as all participants), Phenylketonuria-Specific Profile of Mood States (PKU POMS) (self-rated) confusion subscale scores and total mood dysfunction (TMD) scores, and Profile of Mood States (POMS) (self-rated) TMD scores from part 2 baseline to part 2, week 8 among patients previously exposed to pegvaliase who were randomized to receive either pegvaliase (20 mg/day or 40 mg/day) or a matching placebo in the RDT.

Almost all patients in the PRISM-2 trial were White (98.1%) and almost all were adults aged 18 years or older (94.9%); the average age was approximately 30 years. According to the clinical experts consulted by CADTH for this review, baseline blood Phe, mood and inattention symptoms, and protein intake in the PRISM-2 study population were as expected for adult patients with PKU with poor or no Phe control and limited adherence to MNT.

Efficacy Results

Key efficacy results of part 2 of the PRISM-2RDT are summarized in Table 2. A poolability assessment of the 2 placebo groups (20 mg/day and 40 mg/day) indicated that the magnitude of blood Phe increase from part 2 baseline to part 2, week 8 differed between the 2 placebo groups; the primary and secondary efficacy analyses were therefore conducted by comparing the pooled active group (patients who continued on their assigned dose of pegvaliase from part 1 in the part 2 RDT) versus the 20 mg/day placebo group and the 40 mg/day placebo group separately. At part 2, week 8 and in the mITT set, the least squares mean (LSM) change in blood Phe level from part 2 baseline was 26.50 µmol/L (95% confidence interval [CI], −68.26 to 121.26) in the pooled active group, 949.75 µmol/L (95% CI, 760.38 to 1,139.11) in the 20 mg/day placebo group, and 664.77 µmol/L (95% CI, 465.45 to 864.10) in the 40 mg/day placebo group. The difference in LSM change from baseline comparing the pooled active group to the 20 mg/day placebo group was −923.25 µmol/L (95% CI, −1,135.04 to −711.46; P < 0.0001). The difference in LSM change from baseline between the pooled active group and the 40 mg/day placebo group was −638.27 µmol/L (95% CI, −858.97 to −417.57; P < 0.0001). A cumulative distribution function analysis showed that at part 2, week 8 in the pooled active group, ||||| of patients had blood Phe of 120 µmol/L or lower while approximately ||| ||||||| had blood Phe between 600 µmol/L and 1,200 µmol/L, and approximately ||| ||||||| had blood Phe of 1,200 µmol/L or greater. By contrast, no patients in the placebo groups had blood Phe of 120 µmol/L or lower, while approximately ||| ||||||| had blood Phe between 600 µmol/L and 1,200 µmol/L, and approximately ||||| |||||||| had blood Phe of 1,200 µmol/L or greater.

No statistically significant differences were observed between treatment groups in ADHD-RS-IV inattention subscale scores among participants with drug-naive baseline scores of greater than 9, and further statistical testing for other neurocognitive or neuropsychiatric symptoms (ADHD-RS-IV inattention subscale scores among all participants, PKU POMS [self-rated] confusion subscale scores, PKU POMS [self-rated] TMD scores, and POMS [self-rated] TMD scores) was halted due to the hierarchical testing procedure. Changes in protein intake and HRQoL were not evaluated in part 2 of the PRISM-2 trial.

Harms Results

Key harms results of part 2 of the PRISM-2 RDT are summarized in Table 2. AEs were reported for the pooled active group (patients who continued to receive either 20 mg/day or 40 mg/day pegvaliase during the RDT) and the pooled placebo group (patients who received either 20 mg/day or 40 mg/day pegvaliase in part 1 and then switched to placebo during the RDT), as well as, in some cases, individual dose groups. In part 2 of the PRISM-2 trial, 83.3% of patients receiving active pegvaliase and 93.1% of patients receiving placebo experienced AEs. Common AEs in both the pooled active and pooled placebo groups were arthralgia (13.6% of those in the pooled active group and 10.3% of those in the pooled placebo group), headache (pooled active = 12.1% and pooled placebo = 24.1%), fatigue (pooled active = 10.6% and pooled placebo = 10.3%), anxiety (pooled active = 10.6% and pooled placebo = 6.9%), and injection-site bruising (pooled active = 4.5% and pooled placebo = 10.3%). Serious AEs (SAEs) occurred in 2 patients (3.0%) receiving active pegvaliase and 1 patient (3.4%) receiving placebo. AEs leading to dose reduction or interruption occurred in 1 patient (1.5%) receiving pegvaliase and 1 patient (3.4%) receiving placebo. No patients in part 2 of the PRISM-2 trial had AEs leading to discontinuation of the study drug. No deaths occurred during part 2 of the PRISM-2 trial.

Several study protocol-defined AEs of special interest (AESIs) occurred more frequently in patients receiving active pegvaliase than in those receiving placebo. These included hypersensitivity AEs (HAEs) (pooled active = 39.4% and pooled placebo = 13.8%), generalized skin reactions lasting 14 days or more (pooled active = 10.6% and pooled placebo = 0%), and injection-site skin reactions lasting 14 days or more (pooled active = 7.6% and pooled placebo = 3.4%). Arthralgia and injection-site reactions occurred at similar frequencies in patients receiving active pegvaliase (arthralgia = 13.6% and injection-site reaction = 24.2%) and in those receiving placebo (arthralgia = 10.3% and injection-site reaction = 24.1%). Among notable harms identified for this review, those occurring more frequently in patients receiving active pegvaliase than in those receiving placebo were rash (pooled active = 7.6% and pooled placebo = 3.4%), urticaria (pooled active = |||| and pooled placebo = ||), pruritis (pooled active = 7.6% and pooled placebo = 3.4%), injection-site pruritis (pooled active = |||| and pooled placebo = ||), diarrhea (pooled active = |||| and pooled placebo = ||), injection-site erythema (pooled active = |||| and pooled placebo = ||), and erythema (pooled active = |||| and pooled placebo = ||). No anaphylaxis events or systemic hypersensitivity reactions occurred during part 2 of the PRISM-2 trial.

Table 2: Summary of Key Results From PRISM-2 Part 2 RDT

AE = adverse event; AESI = adverse event of special interest; CI = confidence interval; FAAN = Food Allergy and Anaphylaxis Network; LSM = least squares mean; mITT = modified intention-to-treat; NA = not applicable; NIAID = National Institute of Allergy and Infectious Diseases; NR = not reported; Phe = phenylalanine; RDT = randomized discontinuation trial; SAE = severe adverse event; SD = standard deviation; WDAE = withdrawal due to adverse event.

^aP value based on a mixed model for repeated measures with study drug (pegvaliase or placebo), visit, and drug-by-visit interaction as factors adjusting for baseline blood Phe concentration. P values for comparisons between the pooled active group and each of the placebo groups were adjusted for multiple testing using a Hochberg procedure.

Source: PRISM-2 Clinical Study Report.⁹

Critical Appraisal

A major limitation of part 2 of the PRISM-2 RDT was the small size of the study and associated uncertainty. In addition, internal validity concerns included bias inherent to the RDT design (recruitment of a population of patients who did not discontinue treatment in feeder studies or part 1 of the PRISM-2 trial due to AEs or patient preference, who were able to achieve target dose in feeder studies, and who achieved a ≥ 20% decrease in blood Phe during part 1 of the PRISM-2 trial), baseline imbalances between treatment groups in gender, body mass index (BMI), mean blood Phe level, protein intake, and inattention and mood symptoms, uncertainty regarding the measurement properties or minimal important differences (MIDs) of any of the efficacy outcomes used in the study (and associated uncertainty regarding the connection between changes in blood Phe at part 2, week 8 and other outcomes, including inattention and mood symptoms, protein tolerance, diet liberalization, and HRQoL), and uncertainty in adherence to pegvaliase and consistency in dietary protein intake, both of which were self-reported.

There was some uncertainty regarding the target population of adult patients with PKU most appropriate for pegvaliase and the degree of generalizability of the PRISM-2 part 2 RDT results to this population. The study recruited patients with uncontrolled Phe who were willing and able to self-administer pegvaliase. Changes in blood Phe observed in the study would not be generalizable to patients with good Phe control, although the clinical experts consulted by CADTH for this review noted that these patients would be likely to benefit from treatment. The primary analysis of blood Phe may also not be generalizable to the general population of adult patients with PKU, which, according to the clinical experts, includes many patients who will not comply with any therapy, including pegvaliase. The specific relevance of pegvaliase-induced changes in blood Phe levels in the PRISM-2 RDT, measured at 1 or a few time points (e.g., week 4 and week 8 of the part 2 RDT), to improvements in dietary protein tolerance, neurocognitive and neuropsychiatric symptoms, and HRQoL, was uncertain. According to the clinical experts consulted by CADTH for this review, blood Phe measurements are highly variable in patients with PKU and the point estimate of Phe control associated with pegvaliase treatment at part 2, week 8 of the PRISM-2 trial provided no randomized trial evidence on duration or consistency of Phe control in patients.

Indirect Comparisons

No indirect evidence was identified for this review.

Other Relevant Evidence

PRISM-1 Trial

The PRISM-1 trial was a phase III, open-label, randomized, multi-centre study of the safety and tolerability of pegvaliase among drug-naive patients with PKU (N = 261).^16,17 Of the 215 patients participating in the PRISM-2 trial, 203 (94.4%) entered from PRISM-1, making it the major feeder study for PRISM-2. PRISM-1 is briefly summarized in this section to provide context for the patient population enrolled in PRISM-2, as well as to contribute additional safety data. The primary objective of PRISM-1 was to characterize the safety and tolerability of induction, titration, and maintenance dosing in pegvaliase-naive patients with PKU who self-administered pegvaliase up to 20 mg/day or 40 mg/day. Patients with PKU aged 16 years or older were eligible to participate if they had blood Phe levels of greater than 600 µmol/L and had not been previously exposed to pegvaliase. Patients were randomized 1:1 to receive up to 20 mg/day or 40 mg/day pegvaliase for up to 36 weeks. Both randomized dose groups experienced reductions from baseline blood Phe levels. The mean blood Phe concentration at baseline was 1,232.7 µmol/L (standard deviation [SD] = 386.36) in the intention-to-treat (ITT) set and the mean reduction from baseline was ||||| |||||||| µmol/L at week 28 (n = 133) and ||||| |||||||| µmol/L at week 36 (n = 80). Almost all patients (99.6%) experienced AEs, most commonly arthralgia (65.1%), injection-site reactions (56.7%), injection-site erythema (45.2%), headache (31.4%), rash (25.7%), injection-site pruritis (24.9%), and injection-site pain (21.5%). SAEs occurred in 10.0% of patients; the most common SAE was anaphylaxis (3.1%). Anaphylaxis as defined by criteria established by the National Institute of Allergy and Infectious Diseases (NIAID) and the Food Allergy and Anaphylaxis Network (FAAN) occurred in 6.9% of patients and anaphylaxis as defined by NIAID-FAAN criteria meeting Brown’s severe criteria occurred in 1.5% of patients. Most patients (88.1%) experienced HAEs, including arthralgia (65.1%), generalized skin reaction lasting 14 days or more (22.6%), injection-site reactions (86.2%), injection-site skin reactions lasting 14 days or more (26.4%), serum sickness (3.1%), and angioedema (35.6%).

PRISM-2 Trial

Evidence from the non-RDT portions of the PRISM-2 trial,^9-16 including the part 4 open-label extension (N = 215), is briefly summarized in this section to provide insight into the long-term safety of pegvaliase treatment (including dosages of no more than 60 mg/day in the part 4 open-label extension). In the PRISM-2 trial, patients were treated with open-label pegvaliase in part 1 (20 mg/day or 40 mg/day, up to 13 weeks), part 3 (20 mg/day or 40 mg/day, 6 weeks), and part 4 (up to 60 mg/day, up to 274 weeks). In all parts of the study, self-reported adherence to pegvaliase was high with good exposure. Table 20 provides detailed harms data for PRISM-2 part 1, part 3, and part 4, and the overall study. In the overall PRISM-2 study, ||||| of patients receiving open-label pegvaliase experienced AEs and ||||| of patients experienced SAEs, the majority of which occurred during the open-label extension. No deaths occurred in the overall PRISM-2 study. Approximately |||| ||||||| of patients experienced AEs leading to pegvaliase dose reduction or interruption but only |||| of patients experienced AEs leading to pegvaliase discontinuation. Most patients ||||||| experienced HAEs. Approximately ||||| |||||||| of patients ||||||| experienced injection-site reactions, approximately ||| |||||| ||||||| experienced arthralgia, and nearly |||| |||||| each) experienced generalized skin reactions lasting 14 days or more and injection-site skin reactions lasting 14 days or more. Anaphylaxis reactions occurred in |||| of patients, acute systemic hypersensitivity reactions occurred in |||| of patients, and angioedema occurred in |||| of patients.

PRISM-3 Trial

PRISM-3 was an exploratory phase III substudy to evaluate executive function in adults with PKU participating in the PRISM-2 trial (N = 9).¹⁸ Although the study addressed outcomes (executive function and self-perception) that were not evaluated in the PRISM-2 trial, interpretation was limited by the small sample size.

Comparative Evidence With Sapropterin and MNT

Zori et al. (2019)¹⁹ conducted a retrospective observational cohort study of adolescent and adult patients with PKU receiving pegvaliase with or without MNT, sapropterin plus MNT, or MNT alone. A cohort of patients who received pegvaliase plus MNT in the phase II 165 to 205 trial or phase III PRISM studies (PRISM-1 and PRISM-2) were compared using a propensity score matching (PSM) approach with a historical control of patients who received sapropterin plus MNT or MNT alone who participated in the Phenylketonuria Demographics, Outcome, and Safety (PKUDOS) registry.²⁰ The outcomes evaluated in the study included change in blood Phe and natural protein intake after 1 and 2 years of treatment. Greater decreases in blood Phe levels and increases in protein intake from natural food were observed for patients treated with pegvaliase compared with patients receiving sapropterin plus MNT or MNT alone. However, because of numerous limitations in study design involving comparisons with a historical control cohort, potential bias due to the nonrandomized study design and PSM approach, and statistical limitations (exploratory analysis only), no clear conclusions could be drawn about the comparative effectiveness of pegvaliase, sapropterin plus MNT, and MNT alone.

Conclusions

Data from the PRISM-2 RDT suggested that continued self-administration of pegvaliase injections led to statistically significant and potentially clinically meaningful differences in blood Phe levels after 8 weeks compared with withdrawal of pegvaliase and injection of placebo. Low blood Phe (≤ 120 µmol/L) was observed in approximately half of patients receiving active pegvaliase. Durability and consistency of Phe control were not evaluated in the PRISM-2 RDT. Furthermore, the benefit in reducing blood Phe levels may have been overestimated relative to the general population of adult patients with PKU due to the enriched design of the RDT, which selected for patients more likely to adhere to and respond to pegvaliase. Despite significant differences in Phe at week 8 in patients receiving pegvaliase and placebo, no differences in inattention or mood symptoms were observed. Other outcomes important to patients, including HRQoL and protein tolerance, were not assessed in the PRISM-2 RDT. Efficacy data from nonrandomized studies, including the PRISM-2 open-label extension and an observational study comparing pegvaliase with sapropterin plus MNT and MNT alone, was limited by potential bias and/or confounding. The safety profile of pegvaliase, established through the phase III PRISM trials, including the open-label extension of PRISM-2, pointed to HAEs, arthralgia, injection-site reactions, generalized skin reactions lasting for 14 days or more, and generalized injection skin reactions lasting for 14 days of more as common side effects. Anaphylaxis and angioedema were less common but clinically important side effects. Other limitations of the available evidence included an unclear relationship between the magnitude of changes in blood Phe at a single time point (in the PRISM-2 RDT) and changes in other outcomes of importance to patients with PKU, as well as uncertainty regarding the target population of patients with PKU most appropriate for pegvaliase. The observed changes in blood Phe in the PRISM-2 RDT were aligned with 1 of the outcomes identified as important by patients with PKU, and there is clearly an unmet need for additional efficacious treatments for PKU with higher uptake and adherence rates compared with MNT.

Introduction

Disease Background

Phenylketonuria is a monogenic autosomal recessive disorder and 1 of the most common inborn errors of metabolism.³ Patients with PKU have mutations in both alleles of the PAH gene, which encodes an enzyme that catalyzes the conversion of Phe to tyrosine using tetrahydrobiopterin as a cofactor.³ A wide variety of PAH mutations give rise to variations in clinical phenotype and disease severity.^3,4 A deficiency of PAH leads to uncontrolled blood Phe, which then crosses the blood-brain barrier, where it has neurotoxic effects.^3,4 According to the clinical experts consulted by CADTH for this review, the subset of “brittle” patients who experience large fluctuations in Phe levels despite compliance with diet and some degree of Phe control is of particular concern. Phenylketonuria is universally identified via newborn screening programs in Canada; uncontrolled Phe levels in untreated PKU during early childhood profoundly impair brain function and development.⁵ In adolescents and adults, uncontrolled Phe levels are associated with behavioural and psychiatric problems (inattentiveness and mood dysfunction, often collectively referred to as “executive dysfunction”).⁶ Symptoms of PKU, in conjunction with treatments, negatively affect patient HRQoL with respect to employment, social relationships, and mental health.⁷

The disorder is rare, with an incidence of approximately 1 in 12,000 to 1 in 15,000 live births in Canada (equivalent to approximately 300 new cases per year). Applying the 1-in-15,000 figure to Canada’s 2018 population produces a rough estimate of 2,472 Canadian patients with PKU,²¹ of whom approximately 2,000 would be aged 16 years or older.²² According to the sponsor, approximately 3,133 patients with PKU are living in Canada at present, approximately ||||| of whom are being managed and ||| are aged 16 years or older and currently being treated with sapropterin¹; the sources of these estimates were not clear.

According to the clinical experts consulted by CADTH for this review, PKU is diagnosed shortly after birth by newborn screening using biochemical and genetic tests. Physicians specializing in genetics and metabolic diseases are required to diagnose, treat, and monitor patients with PKU at hospital-based genetic or metabolic clinics that have support from dietitians trained in PKU management.

Standards of Therapy

According to the clinical experts consulted by CADTH for this review, the current cornerstone of PKU treatment is lifelong dietary control of Phe intake to curb blood Phe levels. This is principally accomplished by providing Phe-free foods and metabolic formulas (generated from protein hydrolysate), with a small amount of “complete” Phe-containing protein allowed on top. The combination of dietary restriction and Phe-free medical foods is sometimes collectively referred to as MNT. Some Phe must be provided because it is an essential amino acid. American College of Medical Genetics and Genomics (ACMG) guidelines state that the goal of treatment is to maintain Phe levels of 120 µmol/L to 360 µmol/L.²³ European guidelines state that the primary goal of treatment is normal neurocognitive and psychosocial functioning through maintaining Phe concentrations between 120 µmol/L and 360 µmol/L up to the age of 12 years and up to 600 µmol/L thereafter.²⁴ Patients are monitored via blood sampling (either venous or dried blood spots) on a regular interval. Other than dietary restriction, the only other approved medication is sapropterin, a cofactor of the deficient PAH enzyme in PKU. Sapropterin coverage varies by province, and approximately 25% of patients with PKU have a biochemically detectable response (usually defined as a 30% drop in blood Phe on an equivalent Phe load) to sapropterin.⁸ Large neutral amino acid (LNAA) supplements are used to control neurologic symptoms in a tiny fraction of patients with PKU, primarily those who cannot control Phe levels through diet.

The clinical experts emphasized that current treatments work around the presence of disease but do not modify the underlying mechanisms and do not target PKU symptoms specifically, only Phe levels. The relationship between Phe levels and other outcomes is extremely clear for infants, young children, and pregnant women. Infants with unrestricted Phe have profound mental handicaps, and high Phe during pregnancy can cause birth defects. The relationship between Phe levels and symptoms is more tenuous in older age groups, and restricting adolescents and adults to the same Phe range (120 µmol/L to 360 µmol/L) used for children is a relatively recent innovation. Formerly restricted patients with poor Phe control have variable neurocognitive features as adults, and this group includes those who appear to have no symptoms and do not control their Phe for this reason. The temporal association between high Phe levels and complications is so weak that patients cannot truly sense elevation in Phe, nor does decreasing Phe rapidly alter any consequences of PKU.

The PKU treatment paradigm is based on evidence that long-term Phe elevation in early childhood profoundly affects brain function and development. The major goal of treatment in patients 16 years and older is to preserve mental capacity, usually collectively considered as executive function. Secondary goals are improvement of HRQoL through the reduction of symptoms such as inattentiveness, clouded cognition, anxiety, and depression as well as improving the ability to maintain employment and relationships. An ideal treatment would normalize Phe levels and eliminate these risks without the use of dietary restriction, which is cumbersome, requires modification based on Phe levels, and itself negatively affects HRQoL.

Drug

Key characteristics of pegvaliase are shown in Table 3. Pegvaliase is administered at a titrated maintenance dose (following induction and titration) required to achieve blood Phe levels of 600 µmol/L or lower by self-administered subcutaneous injection. The drug is not approved for use in Canada for other indications and has not been previously reviewed by CADTH. The mechanism of action of pegvaliase is replacement of the deficient PAH enzyme in patients with PKU with a polyethylene glycolylated recombinant Phe ammonia lyase enzyme that converts Phe to ammonia and trans-cinnamic acid.

Pegvaliase underwent a standard review process by Health Canada and received a Notice of Compliance on March 30, 2022. The proposed Health Canada indication for pegvaliase is: “Palynziq (pegvaliase injection) is indicated to reduce blood phenylalanine concentrations in patients with phenylketonuria (PKU) aged 16 years and older who have inadequate blood phenylalanine control (blood phenylalanine levels greater than 600 micromol/L) despite dietary management.” The FDA indication is: “Palynziq is a phenylalanine-metabolizing enzyme indicated to reduce blood phenylalanine concentrations in adult patients with phenylketonuria who have uncontrolled blood phenylalanine concentrations greater than 600 micromol/L on existing management.”²⁵ The European Medicines Agency indication is: “Palynziq is indicated for the treatment of patients with phenylketonuria (PKU) aged 16 years and older who have inadequate blood phenylalanine control (blood phenylalanine levels greater than 600 micromol/L) despite prior management with available treatment options.”²⁶ The sponsor’s reimbursement request is: “For the treatment of patients with PKU aged 16 years and older who have inadequate blood Phe control (blood Phe levels greater than 600 μmol/L) despite prior treatment with sapropterin.” This funding request differs from the proposed Health Canada indication by requiring prior treatment with sapropterin as a prerequisite for initiation of pegvaliase treatment. The sponsor clarified that nonresponse to sapropterin was not considered a condition for initiation of pegvaliase in its funding request.

Table 3: Key Characteristics of Pegvaliase, Sapropterin, and Medical Nutritional Therapy

HPA = hyperphenylalaninemia; NA = not applicable; PAH = phenylalanine hydroxylase; PEG = polyethylene glycol; Phe = phenylalanine.

^aHealth Canada–approved or proposed indication.

Source: CADTH review submission,¹ draft product monograph for pegvaliase,² and product monograph for sapropterin.²⁷

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups. The original patient group submission can be found in the Stakeholder Input section.

One submission, from CanPKU, was received for this review. Between November 30, 2021, and December 25, 2021, CanPKU conducted online surveys of 68 patients with PKU (46 in Canada and 14 in the US) and telephone interviews with 5 patients experienced with pegvaliase (1 in Canada and 4 in the US). Respondents explained how PKU symptoms and the PKU protein-restricted diet had affected their physical and mental health, employment, and social relationships. Almost all respondents (≥ 95%) had experience with low-protein medical foods and formulas, 65% had experience with Kuvan, and only 21% had experience with pegvaliase. Respondents described barriers to existing therapies, including poor taste, lack of satiety, inconvenient preparation and administration, high cost, and limited availability.

The vast majority (≥ 85%) of respondents identified Phe control, reducing PKU symptoms, limiting long-term disease consequences, improving neurocognitive function, managing diet, reducing burden of treatment, improving HRQoL, and increasing natural protein intake as key outcomes of interest. Respondents experienced with pegvaliase reported that the drug limited long-term disease consequences, controlled Phe levels, reduced PKU symptoms, and had tolerable side effects, such as injection-site reactions, joint pain, and skin reactions.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise in 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). The following input was provided by 2 clinical specialists with expertise in the diagnosis and management pediatric and adult patients with PKU who have inadequate Phe control.

Unmet Needs

According to the clinical experts consulted by CADTH for this review, all goals of treatment could potentially be met by available therapies (MNT with or without sapropterin). Delivery of MNT can prevent severe neuropsychological complications when supplied immediately once PKU diagnosis is established in infancy. However, because adherence of adolescent and adult patients with PKU to therapy is extremely challenging, treatment is not effective in most patients. Medical food is very unpleasant to taste and smell and, while it is available, coverage varies by province. This creates many barriers to its daily use, including enjoyment of food and socialization, especially through shared meals. Adults with PKU who were diagnosed early and treated continuously experience higher rates of comorbidities than the general population, including anxiety, depression, hyperactivity and inattentiveness, deficits in executive function, and social isolation; these factors may be due in part to the severely restrictive PKU diet.

Place in Therapy

The clinical experts stated that pegvaliase has a distinct mechanism of action through the direct elimination of circulating Phe. This allows it to complement dietary restriction. The rationale for pegvaliase complementing sapropterin is minimal, as sapropterin has a comparatively weak mechanism of action and is only effective in patients with the mildest PKU. Pegvaliase does not address the underlying disease process (loss of PAH activity) directly, but bypasses it. Pegvaliase is not a first-line treatment, and the sponsor is not seeking approval or reimbursement for younger children and infants. Patients cannot be “intolerant” to dietary restriction of Phe in the classical sense nor can dietary restriction be “contraindicated.” However, poor compliance severely limits the impact of dietary restriction in adults with PKU. Pegvaliase may shift the treatment paradigm for some adult patients with PKU by allowing liberalization of diet while maintaining Phe control.

According to the clinical experts consulted by CADTH, all patients are prescribed dietary restriction, while the use of sapropterin depends on the PAH genotype of the patient and provincial rules for sapropterin reimbursement. The use of pegvaliase should be properly viewed as a potentially more palatable choice for decreasing Phe levels. Recommending that patients try other forms of treatment to achieve Phe control before pegvaliase is equivalent to instructing them to adhere to the PKU diet. Making this a criterion for initiation of therapy could result in pegvaliase access only for patients who are medically noncompliant.

Patient Population

The 2 clinical experts consulted by CADTH for this review had different opinions regarding the subset of patients with PKU who would benefit most from pegvaliase. One clinical expert stated that patients with high and uncontrolled Phe are most in need of an intervention to improve metabolic control that will lead to a decrease in their Phe levels and improved Phe tolerance. The clinical expert could not rule out the possibility that patients who are poorly compliant with MNT could become more compliant to therapies (including pegvaliase) over time as Phe levels decrease and their focus improves. A second clinical expert described the target group of patients with PKU most in need of intervention as those who have an established track record of high compliance with diet and other therapies and who have the most severe forms of PKU. Patients who are able to follow dietary restrictions are the most likely to regularly inject themselves with pegvaliase and follow the instructions for its use. These patients are generally able to achieve PKU levels within the control range but have the most unpalatable diets and experience large deviations in Phe levels even with good control. They would therefore receive the largest benefit in improved access to complete protein and stabilization of Phe.

The clinical experts stated that patients most suitable for pegvaliase therapy can be identified by assessing patient compliance with diet and other therapies (as measured by mean Phe values over a period of approximately 12 months as well as adherence to a monthly monitoring schedule). Severity can be determined by PAH genotyping, assessing variability in Phe levels, and/or by degree of restriction of complete protein intake; the latter 2 can only be applied to patients generally in the Phe control range.

According to the clinical experts, there are few diagnostic issues for PKU, which is identified universally in Canada by screening newborns, and there is minimal potential for misdiagnosis. A tiny number of patients may have non–PAH-associated hyper-Phe syndromes (e.g., disorders of biopterin metabolism) but most are correctly identified through the existing diagnostic algorithm, which includes biochemical and genetic tests. It may be necessary to screen for disorders of biopterin metabolism in patients who have not been genotyped or where genotyping did not identify biallelic variants in PAH, as the optimal medication for pterin synthetic disorders is sapropterin, not pegvaliase.

The clinical experts emphasized that patients who are noncompliant with therapy (including diet) are the least suitable for treatment with pegvaliase. Noncompliance is an major problem in the PKU population. Noncompliant patients will receive no benefit from pegvaliase, and, if they are noncompliant with administration instructions, the medication could cause harm. Likelihood of response to pegvaliase treatment is probably a poor criterion for patient selection, and the use of this criterion for sapropterin has been a serious problem for providers. Few patients who do not respond to pegvaliase with reduction in Phe have been identified. Indeed, patients who are noncompliant (and least suitable for treatment) will appear to have the best response to a single dose of pegvaliase.

In their feedback to CADTH on this clinical review report, the sponsor noted that both patients with uncontrolled and controlled Phe with differing degrees of MNT compliance would be appropriate for pegvaliase treatment and emphasized that MNT compliance was not an eligibility criterion for the initiation of pegvaliase in the pivotal trial.

Assessing Response to Treatment

The clinical experts agreed that complete protein tolerance (or Phe tolerance) and blood Phe levels are the most convenient tests to assess response to treatment and are most often used in clinical trials. After starting on pegvaliase, some patients with PKU will show decreases in their Phe levels, which will allow for diet modification and increase in complete protein tolerance. Patients who are compliant with their diets and who switch to pegvaliase would not be expected to show improvements in Phe levels if these are already optimal, limiting the utility of Phe levels as an outcome. However, the clinical experts noted that Phe levels may drop below the treatment’s recommended range if dietary Phe intake is not adjusted. In addition, patients who manage to completely switch to pegvaliase and do not require dietary restriction will show no further improvement in Phe levels.

According to the clinical experts, clinically meaningful responses to treatment would be, in order of importance: increased complete protein tolerance (or Phe tolerance) and protein intake to levels comparable with those of the general population, improvement in HRQoL, and improvement in psychological metrics (neurocognitive performance, mood, attention, and working memory). The clinical experts stated that blood Phe levels can be used to show that pegvaliase treatment is ineffective. Patients who cannot maintain acceptable Phe levels (or whose levels are not monitored) within the treatment’s recommended range with dietary restriction and pegvaliase are noncompliant and should be discontinued from treatment.

The clinical experts noted that patients would be monitored annually (by reviewing Phe values on a monthly schedule and by an annual review of protein tolerance), but that response evaluation would be performed at a shorter interval (perhaps 4 months) as patients will need to adjust their diets based on Phe levels. Improvement in complete protein tolerance would not be expected to be observed after stabilization on pegvaliase but should be maintained.

Discontinuing Treatment

According to the clinical experts consulted by CADTH for this review, high Phe levels or significant adverse reactions indicate an absence of clinical benefit of pegvaliase.

Prescribing Conditions

The clinical experts stated that pegvaliase should be administered by specialized centres with expertise in the complex management of patients with PKU. These are usually hospital-based genetic or metabolic clinics that have support from dietitians trained in PKU management. In jurisdictions where it has been approved (e.g., the US), pegvaliase is subject to monitoring in an outpatient setting during initial administration and titration of doses. Centres also must be certified to administer the drug. Similar rules would probably have to be put in place in Canada.

According to the clinical experts, specialists in genetics and metabolic diseases are the appropriate physicians to diagnose, treat, and monitor patients with PKU. In some limited cases involving patients who are geographically dispersed, part of this role (e.g., monitoring of adverse reactions) could be assumed by a family physician with remote oversight. Telehealth communication is already used to manage and follow remotely located patients, and the same tools could be used to monitor patients treated with pegvaliase (including those performing self-injections at home).

Diagnosis of PKU is typically established via newborn screening programs and confirmed with biochemical and genetic tests during the neonatal period. Apart from identifying disorders of biopterin metabolism, additional diagnostic testing for PKU as a requirement for pegvaliase would be an unnecessary burden as there is no realistic possibility that a physician and their patient would seek this medication or diagnosis.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups. The original clinician group submission can be found in the Stakeholder Input section.

A group of 3 physicians specialized in treating metabolic disorders who care for adult patients with PKU in Canada provided input for this review. Although the clinician group echoed the challenges in adhering to the PKU diet and the limited proportion of patients who can benefit from sapropterin, contrasting views were presented on: the connection between blood Phe levels and neurologic symptoms, diet liberalization, and associated impacts on HRQoL in adult patients with PKU, which the clinician group described as tightly and reversibly linked; the importance of Phe control as a treatment goal and marker of treatment response in and of itself; the patient subset most suitable for pegvaliase treatment, which the clinician group identified as patients noncompliant with dietary restrictions who cannot benefit from sapropterin and therefore have poor or no Phe control; the patient subset least suitable for pegvaliase treatment, which the clinician group described as patients able to maintain Phe levels within target range on MNT with or without sapropterin; and the risks of Phe levels below 30 µmol/L resulting from overtreatment with pegvaliase in patients who do not comply with Phe monitoring, which the clinician group reported was a potential concern. According to the clinician group, pegvaliase would be offered as last-line treatment to adult patients with PKU who have elevated Phe levels and neuropsychiatric symptoms and who are able to self-administer the injections.

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 4.

Table 4: Summary of Drug Plan Input and Clinical Expert Response

ACMG = American College of Medical Genetics and Genomics; AE = adverse event; BH₄ = tetrahydrobiopterin; CDEC = CADTH Canadian Drug Expert Committee; HRQoL = health-related quality of life; NIHB = non-insured health benefit; PAH = phenylalanine hydroxylase; PEG = polyethylene glycol; Phe = phenylalanine; PKU = phenylketonuria.

Clinical Evidence

The clinical evidence included in the review of pegvaliase 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. No indirect evidence was submitted by the sponsor or identified from the literature that met the selection criteria specified in the review. 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 pegvaliase (self-administered subcutaneous injection, titrated to a maintenance dose required to achieve a blood Phe level of 600 μmol/L or lower; maximum dose 60 mg daily) for the treatment of patients with PKU aged 16 years and older who have inadequate blood Phe control (blood Phe levels greater than 600 μmol/L) on existing management.

Methods

Studies selected for inclusion in the systematic review will 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 5. Outcomes included in the CADTH review protocol reflect those considered to be important to patients, clinicians, and drug plans.

The systematic review protocol presented was established before the granting of a Notice of Compliance from Health Canada.

Table 5: Inclusion Criteria for the Systematic Review

HRQoL = health-related quality of life; MNT = medical nutritional therapy; Phe = phenylalanine; PKU = phenylketonuria.

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.²⁸

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946—) via Ovid and Embase (1974—) via Ovid. All Ovid searches were run simultaneously as a multi-file search. Duplicates were removed using Ovid deduplication for multi-file searches, followed by manual deduplication in EndNote. 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 pegvaliase. Clinical trials registries searched included the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform search portal, 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. Appendix 1 provides the detailed search strategies.

The initial search was completed on February 3, 2022. Regular alerts updated the search until the meeting of the CADTH Canadian Drug Expert Committee on May 16, 2022.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool For Searching Health-Related Grey Literature checklist.²⁹ Included in this search were the websites of regulatory agencies (FDA and European Medicines Agency). Google was used to search for additional internet-based materials. Appendix 1 provides more information on the grey literature search strategy.

A focused literature search for network meta-analyses dealing with PKU was run in MEDLINE All (1946–) on February 2, 2022. No limits were applied to the search.

These searches were supplemented by reviewing bibliographies of key papers and through contacts with appropriate experts. Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the pre-determined 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

Eight reports^9-16 of a single study were identified from the literature for inclusion in the systematic review (Figure 1). The included study is summarized in Table 6.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Table 6: Details of the PRISM-2 Study

ADHD = attention-deficit/hyperactivity disorder; ADHD-RS-IV = Attention Deficit Hyperactivity Disorder Rating Scale (investigator-rated); AE = adverse event; MNT = medical nutritional therapy; PEG = polyethylene glycol; Phe = phenylalanine; PKU = phenylketonuria; PKUDOS = Phenylketonuria Demographics, Outcome, and Safety; POMS = Profile of Mood States; PKU DOS = Phenylketonuria-Specific Profile of Mood States; RDT = randomized discontinuation trial; TMD = total mood disturbance; ULN = upper limit of normal.

Note: One additional report was included (PRISM-2 Clinical Study Report).

Source: PRISM-2 Clinical Study Report.⁹

Description of Studies

PRISM-2 was a phase III, 4-part, 4-arm, double-blind placebo-controlled RDT with an extension period of open-label treatment (N = 215).^9-16 The study was funded by the sponsor. The major feeder study for the PRISM-2 trial was the PRISM-1 trial,^16,17 a phase III, open-label study to assess the safety and tolerability of 2 pegvaliase dosage regimens (20 mg or 40 mg once daily; Other Relevant Evidence section), while a small number of patients were enrolled from phase II studies (PAL-003³⁰ and 165 to 205³¹). The main eligibility criteria for all 3 feeder studies were patients with PKU aged 16 years or older with blood Phe levels of greater than 600 µmol/L (at screening and the prior 6 month average based on available data) who were able to maintain a consistent diet. Dietary Phe control and adherence to MNT were not requirements for participation in feeder studies. For entry into the PRISM-2 trial, patients with PKU aged 16 to 70 years had to have completed a pegvaliase feeder study, have a stable pegvaliase dose regimen for 2 weeks or longer before screening, be able to comprehend and complete or answer prompts for the ADHD-RS-IV, PKU POMS, and POMS, and have a competent observer available for injections. Patients using other investigational products or medications to treat PKU (e.g., LNAAs, sapropterin) were excluded.

A summary of the design of the PRISM-2 study is shown in Figure 2. Following enrolment and screening at 29 centres in the US, patients either first entered part 1 (open-label Phe assessment) or were directly enrolled in part 4 (open-label extension) if they were unable to achieve the target pegvaliase dose (20 mg/day or 40 mg/day) in the feeder study or due to closure of enrolment in part 2 after target enrolment was met. In part 1, patients were randomized 1:1 to receive open-label pegvaliase (20 mg or 40 mg once daily, vial and syringe) for up to 13 weeks; blood Phe levels were measured every 2 weeks. Randomization to 2 different dosages was based on the design of the PRISM-1 study (fixed maintenance dosages of 20 mg/day and 40 mg/day), which, unlike prior phase II studies in which the maintenance dose was titrated in individual patients to achieve a Phe level of 600 µmol/L or lower, was meant to support licensure of pegvaliase. The rationale for randomization to 20 mg/day and 40 mg/day, rather than selection of a single fixed maintenance dosage, was not stated. Patients who achieved a mean blood Phe reduction of 20% or greater (based on 2 consecutive assessments) from treatment-naive baseline and were able to maintain their randomized pegvaliase dose were eligible for part 2 (RDT), while those who did not achieve this degree of Phe reduction or were unable to maintain their randomized pegvaliase dose due to AEs transitioned directly to part 4 (open-label extension). In part 2, patients in each dosage group (20 mg or 40 mg once daily, vial and syringe) were randomized 2:1 to either continue receiving their assigned dosage of pegvaliase or to receive a matching-administration placebo over 8 weeks of double-blind treatment. Participants, investigators, site personnel, and the sponsor were blind to treatment allocation until all randomized participants in part 2 had completed or discontinued the RDT. Randomization was stratified by mean blood Phe level over the last 2 measurements (600 µmol/L or lower versus greater than 600 µmol/L) and ADHD-RS-IV inattention subscale score at PRISM-1 treatment-naive baseline (≤ 12 versus > 12 or missing). In part 3, patients who completed part 2 received open-label pegvaliase (dosage as assigned in part 1) in 2 formats (vial and syringe or pre-filled syringe) for 6 weeks and pharmacokinetics and pharmacodynamics were compared. Part 4 was an open-label extension in which patients received open-label pegvaliase (up to 60 mg once daily, pre-filled syringe) for up to 274 weeks. Only data for the part 2 RDT of the PRISM-2 study are described in the Systematic Review section of this report; the Other Relevant Evidence section provides non-comparative evidence from parts 1, 3, and 4. The data cut-off was February 5, 2019.

The primary objective of the PRISM-2 study was to evaluate the efficacy of pegvaliase in decreasing blood Phe levels by observing changes from part 2 baseline to part 2, week 8 in patients previously exposed to pegvaliase who received either pegvaliase (20 or 40 mg/day) or a matching-administration placebo in the RDT. Secondary objectives (all hierarchically tested) included comparing changes in ADHD-RS-IV inattention subscale scores (among participants with drug-naive baseline scores > 9 as well as all participants), PKU POMS (self-rated) confusion subscale scores, PKU POMS (self-rated) TMD scores, and POMS (self-rated) TMD scores from part 2 baseline to part 2, week 8 among patients previously exposed to pegvaliase who were randomized to receive either pegvaliase (20 mg/day or 40 mg/day) or a matching placebo in the RDT. Tertiary objectives in part 2 included change from part 2 baseline in protein intake from medical food and intact food at each scheduled visit, discontinuations from part 2 due to neuropsychiatric AEs, and change from part 2 baseline in various neurocognitive and neuropsychiatric parameters at each scheduled visit in part 2. Exploratory objectives analyzed in other parts of the study included studying the long-term efficacy of pegvaliase via changes in blood Phe levels, ADHD-RS-IV, PKU POMS, and POMS total scores and subscale scores, and protein intake medical food and intact food.

Figure 2: PRISM-2 Study Design

165 to 301 = PRISM-1; PD = pharmacodynamics; Phe = phenylalanine; PK = pharmacokinetics.

^aDuring part 4, open-label pegvaliase could have been administered at dosages of up to 60 mg/day.

Source: PRISM-2 Clinical Study Report.⁹

Populations

Inclusion and Exclusion Criteria

Key inclusion and exclusion criteria for the PRISM-2 study are summarized in Table 6. Adolescent and adult patients with PKU (aged 16 to 70 years) were eligible if they had completed a prior pegvaliase study (PRISM-1^16,17 or phase II studies PAL-003³⁰ and 165 to 205³¹; all feeder studies required blood Phe above 600 µmol/L at naive baseline), had a stable pegvaliase dose regimen for 2 weeks or longer, were able to maintain a consistent diet, were able to comprehend and complete or answer prompts for the ADHD-RS-IV, PKU POMS, and POMS, had a competent observer available for injections, and had stable doses of medications for ADHD, depression, anxiety, or other psychiatric disorders for 8 weeks or longer before enrolment. Patients using other investigational products or medications to treat PKU (e.g., LNAAs, sapropterin) were excluded, as were those using injectable drugs containing polyethylene glycol, including medroxyprogesterone.

Baseline Characteristics

The baseline demographic and disease characteristics of participants in the PRISM-2 study (from the drug-naive feeder study as well as PRISM-2 part 2) are shown in Table 7. The mean age for all participants was 29.22 years (SD = 8.74). Only 11 participants (5.1%) were adolescents aged 16 or 17 years. In both placebo groups of the RDT, 42.9% of participants were female compared with 53.4% of participants in the pooled active group. Nearly all participants (98.1%) were White, while 1 participant (0.5%) was American Indian or Alaska Native and 2 participants (0.9%) were Black/African-American. The mean weight and BMI in the 20 mg/day placebo group were 94.0 kg (SD = 27.17) and 32.6 kg/m² (SD = 7.75), respectively, compared with 73.1 kg (SD = 16.49) and 25.6 kg/m² (SD = 4.37), respectively, in the 40 mg/day placebo group, and 78.6 kg (SD = 21.55) and 27.8 kg/m² (SD = 6.85), respectively, in the pooled active group.

Mean blood Phe levels at drug-naive baseline were 1,459.1 µmol/L (SD = 354.71) in the 20 mg/day placebo group, 1,108.9 µmol/L (SD = 266.84) in the 40 mg/day placebo group, and 1,318.0 µmol/L (SD = 351.09) in the pooled active group. Mean blood Phe levels at PRISM-2 part 2 baseline were 563.9 µmol/L (SD = 504.62) in the 20 mg/day placebo group, 508.2 µmol/L (SD = 363.68) in the 40 mg/day placebo group, and 563.9 µmol/L (SD = 504.62) in the pooled active group. Mean daily protein intake from intact food at drug-naive baseline was 46.5 g (SD = 40.48) in the 20 mg/day placebo group, 35.2 g (SD = 21.02) in the 40 mg/day placebo group, and 42.6 g (SD = 25.39) in the pooled active group. Mean daily protein intakes from intact food at PRISM-2 part 2 baseline were 38.1 g (SD = 26.42) in the 20 mg/day placebo group, 39.4 g (SD = 22.69) in the 40 mg/day placebo group, and 49.0 g (SD = 23.84) in the pooled active group. Mean daily protein amounts from medical food at drug-naive baseline were 21.2 g (SD = 25.4) in the 20 mg/day placebo group, 29.8 g (SD = 25.79) in the 40 mg/day placebo group, and |||| ||||||| g in the pooled active group. No participants in the 20 mg/day placebo group, ||||| of participants in the 40 mg/day placebo group, and ||||| of participants in the pooled active group had restricted protein intake, defined as obtaining more than 75% of dietary protein from medical food. Imbalances at drug-naive baseline and at part 2 baseline between the pooled active group and 1 or more placebo groups were present to varying degrees in neurocognitive or neuropsychiatric symptoms (ADHD-RS-IV inattention subscale scores among all participants with baseline scores ≥ 9 and among all participants, PKU POMS confusion subscale scores, PKU POMS TMD scores, and POMS TMD scores).

Although rates of prior sapropterin use were not reported in the PRISM-2 study, 224 of 261 patients (85.8%) enrolled in PRISM-1 (the major feeder study for PRISM-2) were questioned and 196 (87.5%) reported prior sapropterin use.

Table 7: Summary of Drug-Naive Baseline and Part 2 Baseline Demographic and Disease Characteristics in the PRISM-2 Study

ADHD-RS-IV = Attention Deficit Hyperactivity Disorder Rating Scale (investigator-rated); mITT = modified intention-to-treat; NR = not reported; Phe = phenylalanine; PKU POMS = Phenylketonuria-Specific Profile of Mood States; POMS = Profile of Mood States; SD = standard deviation; TMD = total mood disturbance.

^aUnder amendment 1 (January 10, 2014) of the study protocol, participants 16 years of age and older were eligible.

^bOne participant in the 40 mg/day placebo group did not have weight and height measured at baseline in PRISM-1.

^cOne participant had a naive baseline Phe level of 600 µmol/L or lower, which was different from the blood Phe level assessed at the time of screening that met the eligibility criterion for entry into the study.

^dThe ADHD-RS-IV hyperactivity-impulsivity subscale and POMS tools were not used in PRISM-1 until the first protocol amendment; only participants who had baseline assessments were included. Possible scores for the ADHD-RS-IV inattention and hyperactivity-impulsivity subscales ranged from 0 to 27, with higher scores indicative of more severe symptoms. Possible scores for the POMS TMD ranged from −32 to 200, scores for the PKU POMS TMD ranged from −12 to 58, and scores for the PKU POMS confusion subscale ranged from 0 to 11, with higher scores indicative of more severe symptoms. Neurocognitive and neuropsychiatric tools were not administered in phase II feeder studies.

^eThe daily recommended allowance for total protein for adults in the general population is 0.75 g/kg. For an 80 kg individual, approximately 60 g of daily protein is recommended.

^fParticipants were considered to be on restricted protein intake if more than 75% of total daily protein intake was from medical food. Total daily protein intake was the sum of daily protein intake from medical food and daily protein intake from intact food.

Source: PRISM-2 Clinical Study Report.⁹

Interventions

During part 1, patients were randomized 1:1 to receive open-label pegvaliase (20 mg or 40 mg once daily, vial and syringe) for up to 13 weeks. In part 2, patients in each dose group (20 mg or 40 mg once daily, vial and syringe) were randomized 2:1 to either continue receiving their assigned dose of pegvaliase or to receive a matching placebo over 8 weeks of double-blind treatment. In part 3, patients who completed part 2 received open-label pegvaliase (dosage as assigned in part 1) in 2 formats (vial and syringe or pre-filled syringe) for 6 weeks and their pharmacokinetics and pharmacodynamics were compared. Part 4 was an open-label extension in which patients received open-label pegvaliase (up to 60 mg once daily, pre-filled syringe) for up to 274 weeks.

All pegvaliase injections were self-administered. Induction and titration to 20 mg/day or 40 mg/day was accomplished in the feeder studies for PRISM-2 (PRISM-1 or phase II studies PAL-003 and 165 to 205). For PRISM-2 part 2 (RDT), either pegvaliase or matching-administration placebo (a dextran solution of similar appearance and consistency) was provided in similarly labelled vials. Training on drug storage, self-injection (vial and syringe as well as pre-filled syringe), identification of AEs, and use of epinephrine was provided both in feeder studies and as a condition for participation in PRISM-2. Participants were given 2 epinephrine injectors at the beginning of the study and were instructed to always carry 1 injector with them. An observer was required for self-injections performed for 1 week after re-introduction of the study drug following an interruption of 4 days or longer. Moreover, during pegvaliase titration or dose increases (part 4) as well as during re-introduction of the study drug, patients received mandatory pre-medication with H1 and H2 blockers with or without nonsteroidal anti-inflammatory drugs or antipyretics (e.g., acetaminophen) before pegvaliase injections for 1 week.

The only PKU-related cointervention of relevance in PRISM-2 was dietary protein intake from intact and medical food (including MNT). Participants were asked to maintain daily protein intake from natural food and medical food through the duration of this study as changes in these parameters could confound efficacy assessments. A dietitian managed patient protein intake for the duration of the study. Throughout the study, all participants were instructed to take 500 mg of a tyrosine supplement 3 times per day with meals.

The study drug could be dose-reduced, interrupted, or permanently discontinued for patients who had HAEs (the Outcomes section provides details). For HAEs of grade 3 or higher related to study drug (and grade 4 HAEs irrespective of relationship to study drug) that were suspected by the investigator and sponsor medical monitor to meet Brown’s criteria for severe,³² the study drug could be permanently discontinued. For grade 3 HAEs related to the study drug that were not suspected to meet Brown’s criteria for severe, grade 3 HAEs not related to the study drug, and HAEs of grade 2 or lower, pegvaliase administration could be dose-reduced or interrupted. Dosages could be reduced from 60 mg/day to 40 mg/day (part 4 only), from 40 mg/day to 20 mg/day, from 20 mg/day to 10 mg/day, or from 10 mg/day to 5 mg/day. Once HAEs other than anaphylaxis improved to grade 1 or were resolved, dosing could be returned to the level administered before the onset of the AE. Suspected anaphylaxis was assessed in clinics, including laboratory evaluations. Following resolution of an anaphylaxis event, dosing could be resumed following the described incremental dose-reduction rules, or with further reduction. Table 8 lists the rules for dosing changes in response to HAEs in the PRISM-2 study.

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 9. These end points are further summarized in the following section. A detailed discussion and critical appraisal of the outcome measures is provided in Appendix 2.

Table 8: Dosing Changes in Response to HAEs in the PRISM-2 Study

CTCAE = Common Terminology Criteria for Adverse Events, version 4.03; HAE = hypersensitivity adverse event; NA = not applicable.

^aCTCAE grade determination was performed by the investigator either via telephone or clinic visit.

^bThe investigator instructed the participant to maintain the pegvaliase dose at the time of adverse event onset until improvement to grade 1 or resolution (according to the investigator assessment in the clinic or via telephone).

^cThe pegvaliase dose could have been reduced or interrupted if necessary as determined by the investigator. The investigator should have consulted with the sponsor’s medical monitor before performing dose reductions during PRISM-2 Part 2.

^dIf a participant had an HAE of grade 3 or higher that was related to the study drug and was suspected to meet Brown’s criteria for severe in the judgment of the investigator and the sponsor’s medical monitor, the participant could have been permanently discontinued from study drug.

^eIf the investigator determined that an HAE of grader 3 or higher was related to administration of pegvaliase, the participant was asked to return to the clinic within 24 hours of event onset for evaluation, including laboratory tests (chemistry, hematology, urinalysis, antidrug immunoglobulin E, antidrug- polyethylene glycol immunoglobulin E [sampling performed > 8 hours after event onset and before the next dose of study drug], urine albumin to creatinine ratio, urinary N-methyl histamine, high-sensitivity C-reactive protein, complement component, complement component 4, tryptase, and urinary N-methylhistamine).

Source: PRISM-2 Clinical Study Report.⁹

Table 9: Summary of Outcomes of Interest Identified in the CADTH Review Protocol

ADHD-RS-IV = Attention Deficit Hyperactivity Disorder Rating Scale (investigator-rated); AE = adverse event; AESI = adverse event of special interest; Phe = phenylalanine; PKU POMS = Phenylketonuria-Specific Profile of Mood States; POMS, Profile of Mood States; SAE = serious adverse event; TMD = total mood disturbance; WDAE = withdrawal due to adverse event.

^aScores for ADHD-RS-IV subscales, including the Inattention subscale, range from 0 to 27, with higher scores indicative of a greater degree of impairment.

^bScores for the PKU POMS confusion subscale range from 0 to 11, with higher scores indicative of a greater degree of impairment.

^cScores for PKU POMS TMD range from −12 to 58, with higher scores indicative of a greater degree of impairment.

^dScores for POMS TMD range from −32 to 200, with higher scores indicative of a greater degree of impairment.

^eADHD-RS-IV total score ranges from 0 to 54, with higher scores indicative of a greater degree of impairment.

^fScores for POMS subscales depend on the number of items (tension, 0 to 28; depression, 0 to 60; anger, 0 to 48; fatigue, 0 to 28; confusion, 0 to 28; and vigour, 0 to 32). Higher scores are indicative of a greater degree of impairment.

^gScores for PKU POMS subscales depend on the number of items (anxiety, 0 to 16; depression, 0 to 16; anger, 0 to 12; activity, 0 to 12; and tiredness, 0 to 12). Higher scores are indicative of a greater degree of impairment.

Source: PRISM-2 Clinical Study Report.⁹

All primary and secondary efficacy outcomes in PRISM-2 were assessed in part 2 (RDT). The primary outcome was change in blood Phe level from part 2 baseline to part 2, week 8. Changes in inattention and mood symptoms from part 2 baseline to part 2, week 8 were assessed as secondary hierarchically tested outcomes; these included ADHD-RS-IV³³ inattention subscale scores (among participants with drug-naive baseline scores > 9 and among all participants), PKU POMS³⁴ confusion subscale scores, PKU POMS TMD scores, and POMS¹ TMD scores. Among secondary outcomes, the ADHD-RS-IV is investigator-rated while the PKU POMS and POMS are patient-reported. Tertiary efficacy outcomes planned to be evaluated in part 2 but not controlled for multiplicity were change in protein intake from medical food and intact food from part 2 baseline at each scheduled visit (weeks 1, 4, and 8), discontinuations due to neuropsychiatric AEs, and changes from part 2 baseline at each scheduled visit in various inattention and mood symptoms (ADHD-RS-IV total score, ADHD-RS-IV hyperactivity-impulsivity subscale score, POMS [observer-rated] TMD score and subscale scores, POMS [self-rated] subscale scores, and PKU POMS subscale scores). Protein intake was measured using 3-day patient-completed diet diaries that were reviewed at clinic visits by study dietitians.

A detailed discussion and critical appraisal of efficacy outcomes used in the PRISM-2 study is provided in Appendix 2. Aside from 1 study demonstrating acceptable internal consistency reliability of the PKU POMS,³⁴ no studies of the measurement properties of any of the efficacy outcomes used in the PRISM-2 study among adult patients with PKU were identified. According to the clinical experts consulted by CADTH for this review, blood Phe level is broadly understood to mechanistically drive the symptoms of PKU, has been associated with neurocognitive and neuropsychiatric symptoms in some studies, and is often used in clinical trials due to convenience. However, the clinical experts stated that the magnitude of blood Phe decreases as well as the duration and consistency of metabolic control required for adult patients with PKU to achieve improvements in other outcomes (such as dietary Phe tolerance, neurocognitive and neuropsychiatric symptoms, and HRQoL) is not known at present.

Harms outcomes included treatment-emergent AEs, SAEs, AEs requiring dose interruption or dose reduction, withdrawals due to AEs, and AESIs. The following were considered AESIs: anaphylaxis, angioedema, HAEs, injection-site reactions, injection-site skin reactions lasting for 14 days or more, generalized skin reactions lasting 14 days or more, arthralgia, and serum sickness. Hypersensitivity AEs, including anaphylaxis and angioedema, were identified using a broad algorithmic Standardized Medical Dictionary for Regulatory Activities (MedDRA) Query (SMQ) for anaphylactic reactions along with a modified version of the SMQ for hypersensitivity that included additional preferred terms. Anaphylaxis was defined by NIAID-FAAN criteria³⁵ and Brown’s severe criteria.³² AEs that began or worsened on or after the start of protocol therapy until 30 days after the last dose of study drug were captured. AEs were defined as any untoward medical occurrence and were coded according to MedDRA version 18.0³⁶ and graded according to version 4.03 of the National Cancer Institute Common Terminology Criteria for Adverse Events.³⁷

Statistical Analysis

Statistical analysis of efficacy outcomes in the PRISM-2 study is summarized in Table 10. No interim analyses were planned or conducted. The primary efficacy analysis was conducted after all participants had completed or discontinued part 2 (RDT).

Determination of Sample Size

For part 2 (RDT), the planned sample size (approximately 72 participants; 48 in the active groups and 24 in placebo groups) would provide 97% power to detect a statistically significant difference in the change in blood Phe from part 2 baseline to part 2, week 8 between the pooled active group and the pooled placebo group with a 2-sided type I error rate of 0.05. The assumptions in this power calculation were that, by part 2, week 8, participants in the pooled active group would maintain mean blood Phe levels of no more than 700 µmol/L with a common SD of 400 µmol/L, and participants in the pooled placebo group would have increased mean blood Phe levels to 1,100 µmol/L or greater, with a common SD of 400 µmol/L.

This planned sample size (N = 72) would provide approximately 70% power to detect a statistically significant difference in change from part 2 baseline to part 2, week 8 in ADHD-RS-IV inattention subscale scores (among participants with drug-naive baseline scores > 9 and among all participants) between the pooled active group and the placebo group using the sequential procedure for multiplicity adjustment within the secondary end points described in the following section. The assumptions in this power calculation were (i) a mean difference between the pooled active group and pooled placebo group among participants with a drug-naive baseline ADHD-RS-IV inattention subscale score above 9 of 5 (SD = 5.5) and (ii) 50% of participants in the mITT population had a drug-naive baseline ADHD-RS-IV inattention subscale score above 9. No power calculations for other secondary outcomes were provided.

Poolability Assessment

To test for potential differences between the 2 placebo groups (20 mg/day and 40 mg/day), change in blood Phe from part 2 baseline to part 2, week 8 was compared by a mixed model for repeated measures (MMRM) analysis with placebo group, visit, and group-by-visit interaction as factors, adjusting for baseline blood Phe level. If P was less than or equal to 0.1, the primary efficacy analysis was to be performed for the pooled active group versus the 20 mg/day placebo group and the 40 mg/day placebo group separately. No poolability assessment was performed for the 20 mg/day and 40 mg/day active pegvaliase treatment groups because it was assumed that there would be no carryover effects for these patients in part 2 (as opposed to patients randomized to receive placebo in part 2, in which carryover effects may differ based on the pegvaliase dose in part 1).

Control of type I Error

Type I error was controlled using a hierarchical testing strategy. Between the primary and secondary efficacy outcomes in part 2, and within the secondary outcomes, a sequential hypothesis testing procedures was used for multiplicity adjustment. The primary outcome (change in blood Phe level from part 2 baseline to part 2, week 8) was tested first, followed by change from part 2 baseline to part 2, week 8 in ADHD-RS-IV inattention subscale score among participants with drug-naive baseline scores above 9, change from part 2 baseline to part 2, week 8 in ADHD-RS-IV inattention subscale score among all participants, change from part 2 baseline to part 2, week 8 in PKU POMS confusion subscale score, change from part 2 baseline to part 2, week 8 in PKU POMS TMD score, and change from part 2 baseline to part 2, week 8 in POMS TMD score. Tertiary efficacy outcomes were not included in the hierarchical testing strategy and were not adjusted for multiplicity.

If poolability was not confirmed and the primary efficacy analysis was to be performed for the pooled active group versus the 20 mg/day and 40 mg/day placebo groups separately, the Hochberg method was used to adjust for multiple comparisons.

Analysis Methods

In the primary efficacy analysis, change in blood Phe concentration from part 2 baseline to part 2, week 8 was compared for the mITT set between the pooled active group versus the 20 mg/day placebo group and between the pooled active group versus the 40 mg/day placebo group using the MMRM method with the study drug group (pegvaliase, placebo), visit, and drug-by-visit interaction as factors adjusting for baseline blood Phe and without imputation for missing blood Phe concentrations. Analysis using MMRM was conducted under a missing-at-random assumption. The study was considered positive if both comparisons resulted in a P value of less than 0.05 or a comparison of either the 40 mg/day placebo group or 20 mg/day placebo group versus the pooled active group resulted in a P value of less than 0.025 favouring the pooled active group. In addition, a responder analysis using a cumulative distribution function approach was conducted, and descriptive and summary statistics were prepared.

Analyses of secondary efficacy outcomes were conducted according to the primary MMRM analysis adjusted for the appropriate baseline variable.

Sensitivity Analyses

Sensitivity analyses were conducted for the primary efficacy analysis of Phe levels as well as for secondary analyses of inattention and mood symptoms. Sensitivity analyses included MMRM in the mITT set, with replacement of missing values by multiple imputation and last observation carried forward imputation, MMRM analysis in the ITT set without imputation, and MMRM analysis in the per-protocol set without imputation. An additional sensitivity analysis was conducted by comparing individual dose groups (20 mg/day or 40 mg/day; MMRM, mITT set, no imputation). To assess the impact of protein intake on the primary efficacy outcome (blood Phe), an analysis of covariance model was used to evaluate change in blood Phe levels with study drug group (pegvaliase, placebo), baseline blood Phe, and change from baseline in protein intake from intact food at week 8 of part 2 in the model.

Subgroup Analyses

To explore the uniformity of the effect of study drug, analyses were performed to determine the potential interaction of subgroups with study drug using the MMRM method of the primary analysis with an additional subgroup covariate-by treatment interaction. The following groups were tested separately: baseline blood Phe categories at part 2 baseline visit (≤ 50% reduction versus > 50% from drug-naive baseline using the mean of 2 consecutive blood Phe measurements), baseline ADHD-RS-IV inattention subscale categories (≤ 9 versus > 9) for POMS-related secondary efficacy end points; sex (female versus male), BMI (< 25 kg/m², ≥ 25 kg/m², < 30 kg/m², ≥ 30 kg kg/m²), and immunogenicity.

Table 10: Statistical Analysis of Efficacy End Points in PRISM-2

ADHD-RS-IV = Attention Deficit Hyperactivity Disorder Rating Scale (investigator-rated); AE = adverse event; ANCOVA = analysis of covariance; CDF = cumulative distribution function; ITT = intention-to-treat; LOCF = last observation carried forward; LSM = least squares mean; MAR = missing at random; mITT = modified intention-to-treat; MMRM = mixed model for repeated measures; NA = not applicable; Phe = phenylalanine; PKU POMS = Phenylketonuria-Specific Profile of Mood States; POMS = Profile of Mood States; PP = per-protocol; SAE = serious adverse event; TMD = total mood disturbance; vs. = versus; WDAE = withdrawal due to adverse event.

Source: PRISM-2 Clinical Study Report.⁹

Analysis Populations

The ITT set was defined as all participants randomized to PRISM-2 part 2. The mITT set consisted of all participants who reached the randomized pegvaliase dose of 20 mg/day or 40 mg/day and were randomized into part 2 with a mean blood Phe reduction of 20% or greater (using the last 2 consecutive blood Phe assessments of part 1) from drug-naive baseline. The distinction between the ITT and the mITT sets occurred because some participants had already enrolled in part 2 when the blood Phe reduction criterion (≥ 20%) for inclusion in the mITT set was established by protocol amendment. The per-protocol set consisted of all subjects in the mITT set who were compliant with the study protocol and had no major protocol violations that were considered to have affected efficacy. The safety population consisted of all participants who enrolled into the study and all participants who enrolled into each study part for analysis by study part.

Results

Patient Disposition

Overall patient disposition in the PRISM-2 study is shown in Figure 3 and Table 11. A total of 215 participants were enrolled in the PRISM-2 study; numbers of patients screened and reasons for screen failure were not provided. Almost all patients (n = 203, 94.4%) had previously completed the PRISM-1 study. Fifty-one patients (23.7%) entered PRISM-2 Part 4 (open-label extension) directly, 33 (15.3%) because they did not achieve target pegvaliase dose in PRISM-1 and 18 (8.4%) because enrolment in part 2 was closed. Of the 164 patients participating in part 1, 12 (7.3%) discontinued the study early, while 57 (34.8%) transitioned directly to part 4 (open-label extension) due to failure to meet the blood Phe reduction criterion of 20% or greater for entry in part 2 imposed in protocol amendment 2 (n = 39; 23.7%), closure of part 2 enrolment (n = 9; 5.5%), or other reasons that were not stated (n = 9; 5.5%). The remaining 95 patients proceeded to part 2 (RDT), of whom 86 met the blood Phe reduction criterion of 20% or greater and made up the mITT set. Only 1 patient (1.1%) discontinued the study early during part 2. Of the 95 participants in part 2, 89 proceeded to part 3 (pharmacokinetic and/or pharmacodynamic analysis), while 5 (5.3%) transitioned directly to part 4 (open-label extension) due to AEs or other reasons that were not stated. Of the 202 patients in part 4 who received open-label pegvaliase, 30 (14.9%) discontinued the study early.

Figure 3: PRISM-2 Study Disposition

165 to 301 = PRISM-1; 165 to 302 = PRISM-2; mITT = modified intention-to-treat.

^aParticipants were not included in the primary efficacy analysis (mITT) if they did not have a mean blood Phe reduction of 20% or greater (using the last 2 consecutive blood Phe assessments of part 1) from baseline levels as established by amendment 2. Some participants had already enrolled in part 2 when this criterion for inclusion in the mITT was implemented.

^bOnly participants enrolled under amendment 2 were considered for the pharmacokinetic and pharmacodynamic analyses; 58 of the 89 participants who entered part 3 were enrolled under amendment 2 and were included for the pharmacokinetic and/or pharmacodynamic analyses.

^cA total of 51 participants entered part 4 directly from PRISM-1 because they did not achieve target dose in PRISM 1 (n = 33) or were affected by closure of enrolment into part 2 (n = 18) after the target enrolment had been met. Another 57 participants entered directly from part 1 because they did not meet the blood Phe reduction criterion to qualify for entry into part 2 (n = 39), due to closure of enrolment in part 2 (n = 9), or for other reasons as instructed by the sponsor (n = 9).

Source: PRISM-2 Clinical Study Report.⁹

The ITT set of PRISM-2 part 2 (RDT) consisted of all 95 patients enrolled in part 2 who were randomized to 1 of the 4 dose groups of pegvaliase or matching placebo (both 20 mg/day or 40 mg/day). All 95 patients received the study drug as determined by randomized allocation and were included in the safety population. The mITT set consisted of the subset of patients in the ITT (n = 86; 90.5%) who met the blood Phe reduction criterion of 20% or greater during part 1. The per-protocol set consisted of the subset of patients in the mITT (n = 80; 84.2%) who were compliant with the study protocol and had no major protocol violations that were considered to have affected the efficacy analyses.

Table 11: Disposition of Participants in PRISM-2

NR = not reported; Phe = phenylalanine.

^aUsed as the denominator to calculate percentages.

^bSome participants may have been enrolled in part 2 before the blood Phe eligibility criterion having been established with amendment 2. Participants were not included in the primary efficacy analysis (modified intention-to-treat) if they did not have a mean blood Phe reduction of 20% or greater (using the last 2 consecutive blood Phe assessments of part 1) from baseline levels.

^cParticipants who were unable to complete part 2 due to an adverse event transitioned directly into part 4. One participant transitioned from part 2 to part 4 and did not perform part 3 assessments; the reason for moving from part 2 directly to part 4 was not due to an adverse event.

^dOne participant who completed the week 8 visit of part 2 and then withdrew consent (due to participant decision) was considered to have discontinued early from the study because they did not subsequently enter part 3 or part 4.

^eParticipants who did not meet the part 2 eligibility criterion, were unable to complete part 1 due to an adverse event, did not complete PRISM-1 (165 to 301) with a dose of 20 mg/day or 40 mg/day, or could not maintain a dose of 20 mg/day or 40 mg/day during part 1 were to transition directly into part 4.

^fParticipants enrolled into part 4 directly from PRISM-1 (Study 165 to 301) because of the early closure of Study 165 to 301 by the sponsor.

^gOne participant withdrew from the study due to an investigator’s decision. However, the participant was captured in the list of participants who were withdrawn from the study due to an adverse event.

Source: PRISM-2 Clinical Study Report.⁹

Patient disposition for the PRISM-2 part 2 RDT by randomized study drug assignment is shown in Table 12. Four of 58 patients (6.9%) randomized to receive active pegvaliase and 1 of 28 patients (3.6%) randomized to receive placebo during part 2 transitioned directly to part 4 (open-label extension) due to AEs. Only 1 patient in the 40 mg/day active placebo group discontinued the study during part 2.

Important protocol deviations during the overall PRISM-2 study and during PRISM-2 part 2 are summarized in Table 13. The denominators for important protocol deviations and their occurrence by randomized treatment group were not provided. The most common important protocol deviation was self-reported departure from protein intake instructions outlined in the study protocol (n = 18 patients, representing 18.9% of the ITT). The degree of deviation from baseline protein intake required for classification as an important protocol deviation was not stated.

Table 12: Disposition of Participants by Randomized Study Drug Assignment in PRISM-2 Part 2 (mITT)

AE = adverse event; mITT = modified intention-to-treat.

^aUsed as the denominator to calculate percentages.

^bParticipants who were unable to complete part 2 due to an AE transitioned directly into part 4. One participant transitioned from part 2 to part 4 and did not perform part 3 assessments; the move from part 2 directly to part 4 was not due to an AE.

Source: PRISM-2 Clinical Study Report.⁹

Table 13: Important Protocol Deviations in PRISM-2

^aA total of 78 participants received 2 doses on the same calendar day, with 1,214 dose records (out of 211,101 total records) reporting 2 doses administered within 1 calendar day. Of these 78 participants, 31 were identified by a clinical review of the data as having had 2 doses less than 12 hours apart, with a total of 49 such dose events. In the clinical review of the data, none of these dose events were associated with a clinically important adverse event.

^bThree participants) received at least 1 dose of pre-filled syringe drug presentation rather than vial-and-syringe drug presentation during week 1 of part 3A due to site error.

^cParticipants who did not perform the last visit of part 2 (day 56) within the protocol-defined visit window (day 49 to day 56) affected the poolability testing for the primary and secondary efficacy analyses and were to be reported as major protocol deviations. Five additional participants in the pooled active group and 3 additional participants in the pooled placebo group performed the last assessment of week 8 on day 57 (i.e., 1 day after the protocol-defined visit window and before dosing in part 3).

^dTwo participants were dispensed the incorrect study drug during part 2 due to pharmacy error. One participant was dispensed an unblinded vial of the study drug (active), and another participant was dispensed a blinded vial of the study drug that did not match the study drug assigned by the interactive web response system. The study blind was not considered to have been broken due to the incorrect drug having been dispensed. Both participants were included in the primary and secondary efficacy analyses.

^eThree participants had changes to their medication during part 2, which were reported as major protocol deviations. The changes to medications were not reported until after the participants had completed part 2.

Source: PRISM-2 Clinical Study Report.⁹

Exposure to Study Treatments

Treatment exposure in the PRISM-2 study is summarized in Table 14. Adherence to pegvaliase treatment was self-reported by patients using workbooks and checked by study staff by counting used and unused vials and syringes. The mean daily dose received was similar to that planned in the allocated treatment group across all groups. The mean study drug use rate (defined as the total amount of study drug taken during part 2 divided by the planned dose) was 96.3% in the pooled active group and 97.6% in the pooled placebo group. Nearly all participants had study drug use rates of 80% or greater in part 2.

Table 14: Treatment Exposure in PRISM-2 Part 2 (Safety Population)

SD = standard deviation.

^aFor placebo groups, the dose was for amount of placebo received.

^bThe study drug use rate was calculated as the total amount of study drug taken during part 2 divided by the planned dose (number of days in part 2 multiplied by the randomized dose level). For participants randomized to receive placebo, the study drug usage rate was calculated as the number of days with a nonmissing dose divided by the duration of part 2. Study drug use rate was based on the intention-to-treat population.

^cExposure duration was calculated as last exposure date: first exposure date + 1. If a participant had no exposure data for more than 28 consecutive days, this missing exposure period was subtracted from the duration.

Source: PRISM-2 Clinical Study Report.⁹

Adherence to consistent dietary intake of protein from medical food and intact food was self-reported by patients using 3-day diet diaries that were reviewed at clinic visits by dietitians. A consistent diet was defined as 1 in which changes in intact food protein and medical food protein intake were less than 10% from baseline. Additional counselling by the dietitian was provided for patients who self-reported changes in intact food protein or medical food protein intake of 10% of greater. For patients who self-reported changes from baseline of 25% or greater in intake of intact food protein or medical food protein, the dietitian counselled resumption of the baseline diet; if changes of 25% or greater persisted at the next study visit (after 4 weeks), the sponsor’s medical monitor was contacted to discuss further actions related to nonadherence with diet. The proportions of patients who self-reported changes from study baseline and part 2 baseline in protein intake of 10% or greater and 25% or greater during part 2 were not stated.

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported in the following section. Appendix 3 provides detailed efficacy data (sensitivity and subgroup analyses of the primary outcome, change in blood Phe level from part 2 baseline to part 2, week 8).

Change From Part 2 Baseline in Protein Intake From Medical Food and Intact Food at Each Scheduled Visit in Part 2

Changes in protein tolerance were not directly addressed in PRISM-2 part 2. Participants were instructed to maintain a consistent level of protein intake during part 2 to ensure that changes in blood Phe concentrations were attributable to the study drug rather than to changes in protein intake.

Mean daily protein intake from intact food during PRISM-2 part 2 is shown in Table 15. At part 2 baseline, mean daily protein intake over the prior 3 days from intact food was 38.1 g (SD = 26.42) in the 20 mg/day placebo group, 39.4 g (SD = 22.69) in the 40 mg/day placebo group, and 49.0 g (SD = 23.84) in the pooled active group. At part 2, week 8, mean daily protein intake over the prior 3 days from intact food was |||| ||||||| g in the 20 mg/day placebo group, |||| ||||||| g in the 40 mg/day placebo group, and |||| ||||||| g in the pooled active group.

Table 15: Mean Daily Protein Intake From Intact Food at PRISM-2 Part 2 Baseline and Part 2 Week 8 (mITT)

mITT = modified intention-to-treat; NA = not applicable; NR = not reported; SD = standard deviation.

Source: PRISM-2 Clinical Study Report.⁹

Health-Related Quality of Life

HRQoL was not evaluated as an outcome in the PRISM-2 study.

Change in Blood Phe Concentration From Part 2 Baseline to Part 2, Week 8

The results of an MMRM analysis of changes in blood Phe levels from part 2 baseline to part 2, week 8 of the PRISM-2 study are shown in Table 16, Figure 4, and Figure 5. Poolability of the 2 placebo groups (20 mg/day and 40 mg/day), as assessed by MMRM analysis, indicated a difference (P = 0.0424, pre-specified significance level of 0.1) in the magnitude of blood Phe increase from part 2 baseline to part 2, week 8 between the 2 placebo groups. The primary and secondary efficacy analyses were therefore conducted by comparing the pooled active group versus the 20 mg/day placebo group and the 40 mg/day placebo group separately.

At part 2, week 8, the LSM change in blood Phe level from part 2 baseline was 26.50 µmol/L (95% CI, −68.26 to 121.26) in the pooled active group, 949.75 µmol/L (95% CI, 760.38 to 1,139.11) in the 20 mg/day placebo group, and 664.77 µmol/L (95% CI, 465.45 to 864.10) in the 40 mg/day placebo group. The difference in LSM change from baseline between the pooled active group and the 20 mg/day placebo group was −923.25 µmol/L (95% CI, −1,135.04 to −711.46; P < 0.0001). The difference in LSM change from baseline between the pooled active group and the 40 mg/day placebo group was −638.27 µmol/L (95% CI, −858.97 to −417.57; P < 0.0001).

Sensitivity analyses imputing missing data and in other analysis sets (ITT and per-protocol), as well as an analysis of covariance model adjusting for changes in protein intake, were consistent with the primary efficacy analysis results (Appendix 3). To test for potential subgroup covariates by treatment interactions, MMRM analysis with baseline blood Phe (≤ 50% versus > 50% reduction from drug-naive baseline using the mean of 2 consecutive measurements) as a covariate was performed and produced results similar to those of the primary analysis (Appendix 3).

Table 16: MMRM of Change From Part 2 Baseline in Blood Phe Concentration (µmol/L) at Part 2, Week 8 (mITT)

CI = confidence interval; LSM = least squares mean; mITT = modified intention-to-treat; Phe = phenylalanine; SD = standard deviation.

^aP value based on a mixed model for repeated measures with the study drug (pegvaliase or placebo), visit, and drug-by-visit interaction as factors adjusting for baseline blood Phe concentration. P values for comparisons between the pooled active group and each of the placebo groups were adjusted for multiple testing using a Hochberg procedure.

Source: PRISM-2 Clinical Study Report.⁹

Figure 4: Least Squares Mean (SE) Change From Baseline in Blood Phe Concentration (µmol/L) to Week 8 of Part 2 (mITT)

LS = least squares; mITT = modified intention-to-treat; Phe = phenylalanine; SE = standard error of the mean.

Source: PRISM-2 Clinical Study Report.⁹

Figure 5: Mean (SE) Observed Blood Phe Concentration From Drug-Naive Baseline Through PRISM-2 Part 2 (mITT)

mITT = modified intention-to-treat; Phe = phenylalanine; SE = standard error of the mean.

Note: Numbers indicate sample size.

Source: PRISM-2 Clinical Study Report.⁹

Descriptive statistics for the percentage of participants in part 2 who met blood Phe reduction thresholds at part 2, week 8 are shown in Table 17. In the pooled active group, ||||| of patients sustained the previously achieved reductions of 20% of greater in blood Phe required for inclusion in the mITT at part 2, week 8 compared with |||| of patients in the 20 mg/day placebo group and ||||| of patients in the 40 mg/day placebo group. In the pooled active group, || patients ||||||| had blood Phe levels of 120 µmol/L or lower at part 2, week 8 while only ||| patients |||||| had blood Phe between 120 µmol/L and 360 µmol/L. The remaining || patients ||||||| had blood Phe levels of less than 600 µmol/L.

Table 17: Percentage of Participants Who Met Blood Phe Reduction Thresholds at Week 8 of Part 2 (mITT)

mITT = modified intention-to-treat; Phe = phenylalanine.

Source: PRISM-2 Clinical Study Report.⁹

A cumulative distribution function analysis of blood Phe levels at part 2, week 8 is shown in Figure 6. In the pooled active group, approximately |||| of patients had blood Phe levels of 120 µmol/L or lower while approximately ||| ||||||| had blood Phe between 600 µmol/L and 1,200 µmol/L and approximately ||| ||||||| had blood Phe of 1,200 µmol/L or greater. By contrast, in the placebo groups, no patients had blood Phe of 120 µmol/L or lower, while approximately ||| ||||||| had blood Phe between 600 µmol/L and 1,200 µmol/L and approximately ||||| |||||||| had blood Phe of 1,200 µmol/L or greater.

Figure 6: CDF Plot of Blood Phe Concentration at Part 2, Week 8 (mITT With Available Data) — Redacted

Figure redacted at the sponsor's request.

Source: PRISM-2 Clinical Study Report.⁹

Change in Neurocognitive and Neuropsychiatric Symptoms From Part 2 Baseline to Part 2, Week 8

The results of an MMRM analysis of changes in neurocognitive and neuropsychiatric symptoms (ADHD-RS-IV inattention subscale score among participants with PRISM-1 baseline score > 9 and among all participants, PKU POMS confusion subscale score, PKU POMS TMD score, and POMS TMD score) from part 2 baseline to part 2, week 8 of the PRISM-2 study are shown in Table 18, Figure 7, and Figure 8. No clear differences were observed between treatment groups. Because of the hierarchical testing procedure, P values in Table 18 following the nonsignificant results of testing for differences in ADHD-RS-IV inattention subscale scores among participants with a PRISM-1 baseline score of greater than 9 (P = 0.0591 and P = 0.2447) were considered descriptive.

Table 18: MMRM of Change From Part 2 Baseline in Neurocognitive and Neuropsychiatric Symptom Scores at Part 2, Week 8 (mITT)

ADHD-RS-IV = Attention Deficit Hyperactivity Disorder Rating Scale (investigator-rated); CI = confidence interval; LSM = least squares mean; mITT = modified intention-to-treat; MMRM = mixed model for repeated measures; PKU POMS = Phenylketonuria-Specific Profile of Mood States; POMS = Profile of Mood States; SD = standard deviation; TMD = total mood disturbance.

^aSome participants did not have part 2 neurocognitive and neuropsychiatric assessment data collected. The ADHD-RS-IV inattention subscale, PKU POMS, and POMS tools were not performed in PRISM-1 until a protocol amendment; only participants who had baseline assessments were included. Participants who were included in the mITT population from a phase II study were not included because neurocognitive and neuropsychiatric tools were not administered in the phase II studies.

^bNegative values indicate a decline in symptom score (toward improvement) while positive values indicate an increase in symptom score (toward decline).

^cP values were adjusted for multiple comparisons among primary and secondary outcomes using a hierarchical step-down procedure. P values for comparisons between the pooled active group and each of the placebo groups were adjusted for multiple testing using a Hochberg procedure.

^dStatistical testing for these end points followed a prior failed end point in the testing hierarchy and therefore these P values should be interpreted descriptively.

Source: PRISM-2 Clinical Study Report.⁹

Figure 7: Least Squares Mean (SE) Change From Baseline in ADHD-RS-IV Inattention Subscale Score for Participants With Baseline Score Above 9 to Week 8 of Part 2 (mITT) — Redacted

Figure redacted at the sponsor's request.

Source: PRISM-2 Clinical Study Report.⁹

Figure 8: Least Squares Mean (SE) Change From Baseline in ADHD-RS-IV Inattention Subscale Score to Week 8 of Part 2 (mITT) — Redacted

Figure redacted at the sponsor's request.

Source: PRISM-2 Clinical Study Report.⁹

Discontinuation From Part 2 Due to Neuropsychiatric Adverse Events

One participant receiving 40 mg/day pegvaliase was unblinded during part 2 due to a SAE of grade 3 anxiety. The participant discontinued part 2 and moved into part 4 and continued the study drug.

Other Tertiary Efficacy Outcomes

Analyses of other tertiary efficacy outcomes (change from part 2 baseline in ADHD-RS-IV total score, ADHD-RS-IV hyperactivity-impulsivity subscale scores, POMS [observer-rated] TMD and subscale scores, POMS [self-rated] subscale scores, and PKU POMS subscale scores at each scheduled visit in part 2) were not performed given the nonsignificant results of the secondary analysis of inattention and mood symptoms.

Harms

Only those harms identified in the review protocol are reported in the following section. Table 19 provides detailed harms data from PRISM-2 part 2.

Adverse Events

In PRISM-2 part 2, 83.3% of patients receiving active pegvaliase and 93.1% of patients receiving placebo experienced AEs. Common AEs in both the pooled active and pooled placebo groups were arthralgia (13.6% in the pooled active group and 10.3% in the pooled placebo group), headache (pooled active = 12.1% and pooled placebo = 24.1%), fatigue (pooled active = 10.6% and pooled placebo = 10.3%), anxiety (pooled active = 10.6% and pooled placebo = 6.9%), and injection-site bruising (pooled active = 4.5% and pooled placebo = 10.3%).

Serious AEs

In PRISM-2 part 2, SAEs occurred in 2 patients (3.0%) receiving active pegvaliase and 1 patient (3.4%) receiving placebo.

AEs Leading to Dose Reduction or Interruption

In PRISM-2 part 2, AEs leading to a dose reduction or interruption occurred in 1 patient (1.5%) receiving pegvaliase and 1 patient (3.4%) receiving placebo.

Withdrawals Due to AEs

No patients in PRISM-2 part 2 had AEs leading to discontinuation of study drug.

Mortality

No deaths occurred during PRISM-2 part 2.

Notable Harms

In PRISM-2 part 2, several study protocol–defined AESIs occurred more frequently in patients receiving active pegvaliase than in those receiving placebo. These included HAEs (39.4% of pooled active patients and 13.8% of pooled placebo patients), generalized skin reactions lasting 14 days or more (pooled active = 10.6% and pooled placebo = 0%), and injection-site skin reactions lasting 14 days or more (pooled active = 7.6% and pooled placebo = 3.4%). Arthralgia and injection-site reactions occurred at similar frequencies in patients receiving active pegvaliase (arthralgia = 13.6%; injection-site reactions = 24.2%) and in those receiving placebo (arthralgia = 10.3%; injection-site reactions = 24.1%).

Among notable harms identified for this review, those occurring more frequently in patients receiving active pegvaliase than in those receiving placebo were rash (pooled active = 7.6% and pooled placebo = 3.4%), urticaria (pooled active = |||| and pooled placebo = ||), pruritis (pooled active = 7.6% and pooled placebo = 3.4%), injection-site pruritis (pooled active = |||| and pooled placebo = ||), diarrhea (pooled active = |||| and pooled placebo = ||), injection-site erythema (pooled active = |||| and pooled placebo = ||), and erythema (pooled active = |||| and pooled placebo = ||).

No anaphylaxis events or systemic hypersensitivity reactions occurred during PRISM-2 part 2.

Table 19: Summary of Harms in PRISM-2 Part 2 (Safety Population)

AE = adverse event; AESI = adverse event of special interest; FAAN = Food Allergy and Anaphylaxis Network; NIAID = National Institute of Allergy and Infectious Diseases; NR = not reported; SAE = serious adverse event.

Note: Treatment-emergent AEs reported in this table were defined as any untoward medical occurrence occurring after administration of the first dose of study drug and within 30 days of the last dose of study drug. AEs were coded using version 18.0 of the Medical Dictionary for Regulatory Affairs and graded according to version 4.03 of the National Cancer Institute’s Common Terminology Criteria for Adverse Events.

^aAdverse events with a frequency of 10% or greater overall in the overall study and a frequency of 10% or greater in the pooled active or pooled placebo study group during part 2 are reported.

^bAll SAEs occurring during part 2 are reported.

Source: PRISM-2 Clinical Study Report.⁹

Critical Appraisal

Internal Validity

PRISM-2^9-16 was a phase III, 4-part, 4-arm, double-blind, placebo-controlled RDT with an extension period of open-label treatment conducted in adolescent and adult patients with PKU who completed a prior pegvaliase study (N = 215). Only evidence from the part 2 RDT (ITT n = 95; mITT n = 86) is included in the Systematic Review section of this report. The relatively small size of the PRISM-2 study was expected due to the rarity of PKU. According to the sponsor, the RDT design of the study was selected because of its efficiency in demonstrating therapeutic efficacy in a small study population, due to concerns of unblinding resulting from HAEs over a longer treatment period, and to minimize long-term exposure to placebo for ethical reasons. However, the RDT design has limitations in its ability to estimate the magnitude of absolute treatment effects and harms in the overall population of adult patients with PKU.

The RDT design of PRISM-2 part 2 was associated with an unavoidable but notable risk of bias. Losses to follow-up in PRISM-2 part 2 were minimal: only 1 patient discontinued part 2 early. However, participants in the major feeder trial, PRISM-1, were pre-screened for their ability to maintain a consistent diet as an eligibility criterion. Ability to maintain a consistent diet was also applied as an eligibility criterion for PRISM-2. Only patients who completed a feeder study were eligible, and those who discontinued feeder studies early due to AEs, noncompliance, or patient and/or physician decision were not included. For example, only 203 of 261 participants in PRISM-1 entered PRISM-2, while the remaining 58 patients (22.2%) did not. Upon entry into PRISM-2, patients who did not achieve the target dosage (20 mg/day or 40 mg/day) in PRISM-1 (33 of 203; 16.3%) transitioned directly to part 4 (open-label extension). Of the 164 patients who entered PRISM-2 part 1, 12 (7.3%) discontinued the study early due to AEs or patient and/or physician decision, while 39 (23.8%) did not meet the blood Phe reduction criterion of 20% or greater for the part 2 mITT set. The clinical expert agreed that the PRISM-2 RDT population was enriched for patients who would be more likely to tolerate and adhere to pegvaliase and therefore would be more likely to respond to treatment than the general adult PKU population with blood Phe levels above 600 µmol/L.

Randomization in PRISM-2 part 2 appeared generally successful in balancing baseline demographic and disease characteristics between study groups. However, because of the small number of participants and the fact that the 2 placebo groups could not be pooled for efficacy analyses, some baseline imbalances between groups were present and of potential prognostic significance. In PRISM-2 part 2, 53.4% of patients in the pooled active group were female compared with 42.9% of participants in the placebo groups. According to the clinical experts consulted by CADTH for this review, gender is a factor in compliance with PKU treatment, with women having higher rates of adherence. In the 20 mg/day placebo group, the mean BMI was 32.6 (SD = 7.75) kg/m²_. According to the clinical experts consulted by CADTH for this review, BMI is associated with diet and Phe control in adult patients with PKU. There were also baseline imbalances (in drug-naive, part 2, or both baselines) of varying degrees between the pooled active group and 1 or both placebo groups in blood Phe level, mood and inattention symptoms evaluated as secondary outcomes, and protein intake from medical food and intact food. According to the clinical experts consulted by CADTH for this review, these imbalances would not limit interpretation of the study results.

The psychometric properties of the outcomes evaluated in primary and secondary efficacy analyses in PRISM-2 (blood Phe level, ADHD-RS-IV inattention subscale score, PKU POMS confusion subscale score, PKU POMS TMD, POMS TMD) have not been studied in adult patients with PKU apart from a study demonstrating acceptable internal consistency reliability of the PKU POMS.³⁴ No information on MIDs for any of the outcomes evaluated was available. According to the clinical experts consulted by CADTH for this review, blood Phe can be used to show that PKU treatment is ineffective, while consistent decreases in Phe levels during treatment can potentially indicate treatment effects in conjunction with improvements in protein tolerance, inattention and mood symptoms, and HRQoL. However, input from a group of 3 clinicians challenged this assessment and emphasized the importance of Phe control as a treatment goal and marker of treatment response in and of itself. All clinicians agreed that blood Phe levels mechanistically drive PKU symptoms. However, according to the clinical experts consulted by CADTH for this review, the magnitude of blood Phe decreases, as well as the duration and consistency of metabolic control required for adult patients with PKU to achieve improvements in other outcomes (such as dietary Phe tolerance, neurocognitive and neuropsychiatric symptoms, and HRQoL), is not known at present. However, the clinical experts emphasized that diet liberalization enabled by Phe control is the goal of PKU treatment and they expected that sustained decreased Phe levels induced by treatment with pegvaliase of sufficient magnitude, consistency and duration could in theory lead to improvements in neurocognitive and neuropsychiatric symptoms, protein tolerance, and HRQoL.

Several statistical issues should be considered when interpreting the results of the PRISM-2 study. Statistical tests were appropriate, and multiplicity was controlled using a hierarchical testing strategy. Some data were missing for the primary efficacy analysis of change in Phe levels from part 2 baseline to part 2 week 8 (15 of 86 patients in the mITT population; 17.4%). Missing data were not replaced in the primary efficacy analysis, but were accounted for under a missing-at-random assumption, which was further evaluated in sensitivity analyses using a variety of imputation techniques (multiple imputation and last observation carried forward), all of which showed similar results. The potential for heterogeneity among the 4 dose groups in the RDT (20 mg/day and 40 mg/day active pegvaliase and placebo) complicated interpretation of results. Heterogeneity between the placebo groups was accounted for by separate testing; however, heterogeneity between the pegvaliase groups was not addressed as it was assumed that there were no carryover effects in either group. This assumption was not formally evaluated. Although evidence was limited for secondary efficacy outcomes (inattention and mood symptoms) identified as important to patients, the clinical experts consulted by CADTH for this review agreed that differences in neurocognitive and neuropsychiatric function were unlikely to be detectable over the 8-week period of the PRISM-2 RDT using the available instruments because of the relatively short duration and high variability in the symptoms of adult patients with established PKU. No efficacy analyses of pre-specified subgroups were planned and the impact of baseline blood Phe levels on the primary outcome was evaluated only by testing for potential baseline blood Phe (≤ 50% versus > 50% reduction from drug-naive baseline) treatment interactions in MMRM analysis. This analysis was exploratory, not specifically powered to evaluate individual strata, and not adjusted for multiplicity.

Another major concern in the PRISM-2 study was adherence to pegvaliase and to consistent protein intake, both of which were patient-reported (although pegvaliase adherence was also confirmed by vial counting). Despite high self-reported adherence to pegvaliase injections that was confirmed by collection and counting of empty drug vials and syringes, a clear dichotomy was observed in the blood Phe levels of patients randomized to continue receiving active pegvaliase in part 2: roughly half of participants had Phe levels of 120 µmol/L or lower, while the other half had either poor Phe control (Phe levels of 600 µmol/L to 1,200 µmol/L; approximately 1-quarter) or no Phe control (Phe levels of 1,200 µmol/L or greater). With the high treatment-compliance rates reported in the pivotal trial, the clinical experts consulted by CADTH were unable to provide an explanation for the relatively high proportion of patients who did not achieve normalization of blood Phe levels (defined as 120 µmol/L or lower) in the pooled active group. Based on the kinetics of blood Phe following a single pegvaliase injection,³⁸ the clinical experts stated that the most plausible reason for this pattern could be low compliance to pegvaliase in the pooled active group. Uncertainty in adherence to the study protocol for study drug administration limited interpretation of the relationship between pegvaliase administration and changes in blood Phe levels in the RDT. Also of potential concern was adherence to diet and maintenance of stable protein intake; patients who self-reported deviations from baseline protein intake from medical food and intact food faced escalating pressure from study dietitians to self-report compliant values. Changes in protein intake, if they were imbalanced by study group and especially considering the small sample size of the individual placebo groups, could have confounded the primary efficacy analysis of blood Phe levels either in favour of or against pegvaliase.

External Validity

According to the clinical experts consulted by CADTH for this review, the demographic and disease characteristics of patients enrolled in the PRISM-2 study were reflective of the Canadian population of adolescent and adult patients with PKU they would see in their clinical practice. Similarly, the clinical experts expected that the study eligibility criteria would result in recruitment of a patient population reflective of Canadian practice, albeit a cross-section more likely to tolerate and adhere to pegvaliase than the general PKU population (as discussed earlier in the Internal Validity section). The clinical experts confirmed that eligibility criteria related to neurocognitive and linguistic capacity and concurrent conditions would eliminate only a small proportion of patients with PKU who would not be good candidates for pegvaliase. The clinical experts stated that, although nearly all patients in the PRISM-2 study were White (98.1%) and 18 years of age or older (94.9%), and despite the enrichment of the study population for treatment-compliant patients who were able to tolerate pegvaliase, the study results would be generalizable to most adult and adolescent patients with PKU, including adolescent patients (aged 16 and 17 years), pegvaliase-naive patients, and patients with good or limited Phe control who are compliant with MNT.

Planned dosing of pegvaliase in the PRISM-2 study was aligned with Health Canada–approved dosing. Although adherence to the dosing regimen was assessed by counting empty drug vials and unused syringes, because of the self-reported nature of reporting adherence to treatment (i.e., injections were not monitored) some level of uncertainty remains around the actual dosing received by the study participants. As the PRISM-2 study population was enriched for pegvaliase tolerance, compliance, and responsiveness, and was poorly or not adherent to therapy, this could limit generalizability of the primary analysis of blood Phe levels to all adult patients with PKU. According to the clinical experts consulted by CADTH, the general population of adult patients with PKU would probably be less compliant than the trial population, and therefore mean reductions in blood Phe level would likely be smaller in real-world clinical practice than those observed in the PRISM-2 study.

There was substantial disagreement in the input received by CADTH from clinical experts and clinician groups on the target population of patients with PKU most and least appropriate for pegvaliase treatment. The overall goal of the phase III PRISM studies was to evaluate the efficacy of pegvaliase in adult patients with PKU with poorly controlled blood Phe (above 600 µmol/L). One clinical expert consulted by CADTH for this review stated that that patients compliant with MNT, who would generally have Phe levels above 600 µmol/L, would be the most suitable candidates for pegvaliase; by contrast, another clinical expert as well as clinician group input suggested that patients noncompliant with dietary restriction who cannot benefit from sapropterin and therefore have poor or no Phe control (Phe above 600 µmol/L) would be the targeted population. Nevertheless, the clinical experts agreed that the results of the study would be generalizable to patients with some or good Phe control (600 µmol/L or lower) who are compliant with MNT. In this group of patients, pegvaliase treatment could permit liberalization of diets while maintaining Phe control.

Several of the outcomes in PRISM-2 part 2 (blood Phe levels, inattention and mood symptoms) were identified as clinically important by patients and clinicians alike. The specific relevance of pegvaliase-induced changes in blood Phe levels in the PRISM-2 RDT, measured at 1 or a few time points, to improvements in dietary protein tolerance, neurocognitive and neuropsychiatric symptoms, and HRQoL, was uncertain. According to the clinical experts consulted by CADTH for this review, the duration of follow-up in the part 2 RDT was adequate for point estimation of the impact of pegvaliase discontinuation on blood Phe level but potentially insufficient for assessment of inattention and mood symptoms in patients with PKU.

Patients in the PRISM-2 study were likely followed up more frequently by clinicians and dietitians than the average adult Canadian PKU patient. According to clinical experts consulted by CADTH for this review, a large proportion of adult patients with PKU are averse to monitoring and tend to exhibit weak treatment compliance. Patients in PRISM-2 may also have received better training on pegvaliase injections than would be expected in Canadian practice. However, the clinical experts consulted by CADTH for this review did not anticipate that these factors would limit generalizability of the study findings.

Pegvaliase dosing in the PRISM-2 RDT (20 mg/day or 40 mg/day pegvaliase) differed from the Health Canada–recommended induction, titration, and maintenance dosing (maximum 60 mg/day) in terms of the maximum possible dose. Part 2 of the PRISM-2 trial had a relatively short duration of 8 weeks, and potentially longer titration periods could establish the precise maintenance doses required for individual patients with PKU to maintain blood Phe levels within a safe window while permitting liberalization of diet. However, the clinical experts consulted by CADTH for this review did not expect that these factors would limit generalizability of the study findings.

One important generalizability concern in the PRISM-2 study was that, given the acknowledged high variability between measurements of Phe levels in individual adult patients with PKU, a statistically significant decrease in mean blood Phe at a single time point (week 8 of the PRISM-2 part 2 RDT) provided limited evidence to substantiate sustained decreases in blood Phe levels of sufficient magnitude to realize improvements in PKU patient inattention and mood symptoms, HRQoL, and protein intake. At week 8 of the PRISM-2 part 2 RDT, there was no direct evidence to support changes in any important end points (inattention and mood symptoms, HRQoL, or protein intake) beyond Phe level. Placebo-controlled longitudinal analyses of per-patient blood Phe control over time, and correlations with improvements in protein tolerance and neurocognitive and/or neuropsychiatric symptoms, were not reported in PRISM-2. The clinical experts consulted by CADTH for this review acknowledged that, given the high variation in the blood Phe measurements of patients with PKU and potential changes over time in pegvaliase adherence, the point estimate of Phe control at part 2 week 8 of PRISM-2 provided no randomized trial evidence on duration or consistency of Phe control.

Indirect Evidence

No indirect evidence was identified for this review.

Other Relevant Evidence

This section includes submitted long-term extension studies and additional relevant studies included in the sponsor’s submission to CADTH that were considered to address important gaps in the evidence included in the systematic review.

PRISM-1 Trial

The PRISM-1 trial was a phase III, open-label, randomized, multi-centre study to assess the safety and tolerability of pegvaliase among drug-naive patients with PKU (N = 261).^16,17 The PRISM-1 trial was the major feeder study for the PRISM-2 trial. As 203 of 215 (94.4%) of patients participating in the PRISM-2 trial entered from PRISM-1 trial, the study is briefly summarized here to provide context for the patient population enrolled in the PRISM-2 trial, as well as to contribute additional safety data. The primary objective of the PRISM-1 trial was to characterize the safety and tolerability of induction, titration, and maintenance dosing in pegvaliase-naive patients with PKU who self-administered pegvaliase up to 20 mg/day or 40 mg/day. Patients with PKU aged 16 years or older were eligible to participate if they had blood Phe levels above 600 µmol/L and had not been previously exposed to pegvaliase. Patients were randomized 1:1 to receive up to 20 mg/day or 40 mg/day pegvaliase for up to 36 weeks. Both randomized dose groups experienced reductions from baseline blood Phe levels. The mean blood Phe concentration at baseline was 1,232.7 µmol/L (SD = 386.36) in the ITT set and the mean reduction from baseline was ||||| |||||||| µmol/L at week 28 (n = 133) and ||||| |||||||| µmol/L at week 36 (n = 80). Almost all patients (99.6%) experienced AEs, most commonly arthralgia (65.1%), injection-site reactions (56.7%), injection-site erythema (45.2%), headaches (31.4%), rash (25.7%), injection-site pruritis (24.9%), and injection-site pain (21.5%). Serious AEs occurred in 10.0% of patients; the most common SAE was anaphylaxis (3.1%). Anaphylaxis according to the NIAID-FAAN criteria occurred in 6.9% of patients and anaphylaxis as defined by NIAID-FAAN criteria meeting Brown’s severe criteria occurred in 1.5% of patients. Most patients (88.1%) experienced HAEs, including arthralgia (65.1%), generalized skin reactions lasting 14 days or more (22.6%), injection-site reactions (86.2%), injection-site skin reactions lasting 14 days or more (26.4%), serum sickness (3.1%), and angioedema (35.6%).

PRISM-2 Trial

Evidence from part 1, part 3, and part 4 of the PRISM-2 trial,^9-16 including the part 4 open-label extension, is briefly summarized here to provide insight into the long-term safety and efficacy of pegvaliase treatment (including doses up to 60 mg/day in the part 4 open-label extension). The Systematic Review section includes an overall description of the PRISM-2 study (N = 215).

Safety

In the PRISM-2 trial, patients were treated with open-label pegvaliase in part 1 (20 mg/day or 40 mg/day, up to 13 weeks), part 3 (20 mg/day or 40 mg/day, 6 weeks), and part 4 (up to 60 mg/day, up to 274 weeks). In all parts of the study, self-reported adherence to pegvaliase was high with good exposure. Table 20 lists detailed harms data in PRISM-2 part 1, part 3, part 4, and the overall study. In the overall PRISM-2 study, ||||| of patients receiving open-label pegvaliase experienced AEs and ||||| of patients experienced SAEs, the majority of which occurred during the open-label extension. No deaths occurred in the overall PRISM-2 study. Approximately |||| ||||||| of patients experienced AEs leading to pegvaliase dose reduction or interruption but only |||| of patients experienced AEs leading to pegvaliase discontinuation. Most patients ||||||| experienced HAEs. Approximately ||||| |||||||| of patients ||||||| experienced injection-site reactions, approximately ||| |||||| ||||||| experience arthralgia, and nearly |||| |||||| ||||| experienced generalized skin reactions lasting 14 days or more and injection-site skin reactions lasting 14 days or more. Anaphylaxis reactions occurred in |||| of patients, acute systemic hypersensitivity reactions occurred in |||| of patients, and angioedema occurred in |||| of patients.

Table 20: Summary of Harms in PRISM-2 Part 1, Part 3, and Part 4 (Safety Population)

AE = adverse event; AESI = adverse event of special interest; FAAN = Food Allergy and Anaphylaxis Network; NIAID = National Institute of Allergy and Infectious Diseases; NR = not reported; SAE = serious adverse event.

Note: Treatment-emergent AEs reported in this table were defined as any untoward medical occurrence occurring after administration of the first dose of study drug and within 30 days of the last dose of study drug. Adverse events were coded using version 18.0 of the Medical Dictionary for Regulatory Affairs, and graded according to version 4.03 of the National Cancer Institute’s Common Terminology Criteria for Adverse Events.

^aThe part 3 safety population included only the 60 participants who enrolled in part 3 after amendment 2 and qualified for inclusion in pharmacokinetic assessments with 2 different drug presentations. Safety data from the remaining 29 participants in part 3 of the study before amendment 2 is not included in this separate part 3 summary but the data were captured as part of the overall PRISM-2 safety totals.

^bThe analysis population is all participants who entered PRISM-2.

^cAdditional AEs (8 events in 4 participants) were upgraded by the sponsor to SAEs and were not factored into the incidence reported in summary tables.

^dOne participant withdrew from the study due to an investigator’s decision but was captured in the list of participants who were withdrawn due to an AE.

^eData were assessed for part 4 and overall only.

^fBased on sponsor’s clinical adjudication of angioedema. The change in search strategy and definition was applied for part 4 and overall data only.

^gBased on a broad standard Medical Dictionary for Regulatory Affairs query of selected preferred terms that also included arthralgia, back pain, musculoskeletal pain, pain in extremity, and neck pain. The change in search strategy and definition was applied for part 4 and overall data only.

Source: PRISM-2 Clinical Study Report.⁹

Efficacy

Among patients who remained on open-label pegvaliase during PRISM-2 part 4, within 24 months, ||| of participants achieved a blood Phe level of 600 μmol/L or lower, ||| of participants achieved blood Phe of 360 μmol/L or lower, and ||| (increasing to ||| after 48 months of treatment) achieved blood Phe of 120 μmol/L or lower.¹ These figures were based on point measurements (i.e., patients had to meet these Phe thresholds once only to be counted). Among patients who remained on open-label pegvaliase during PRISM-2 part 4, long-term (48 months) assessment of neurocognitive and neuropsychiatric symptoms showed improvements in inattention and mood symptoms that were correlated with blood Phe reduction.¹ Among patients who remained on open-label pegvaliase during PRISM-2 part 4, protein intake from natural protein (and therefore dietary Phe) increased from a mean of 1,700.2 (SD = 1,194.4) mg dietary Phe at baseline to 2,123.2 mg (SD = 1,302.2) at 1 year and 2,679.7 mg (SD = 1,285.7) at 2 years.¹ At month 48, mean daily protein intake increased by 7.1 g (SD = 32.2) from baseline levels, with an overall increase in natural protein and decrease in protein from medical food.¹

Critical Appraisal

According to the clinical experts consulted by CADTH for this review, long-term trends over time in blood Phe levels, inattention and mood symptoms, and protein intake in PRISM-2 part 4 may be partially attributed to selection bias: patients who remained on treatment were potentially more treatment-compliant, may have had lesser degrees of inattention and mood symptoms, and may have had better protein tolerance than those who discontinued from open-label treatment. The long-term efficacy evidence from PRISM-2 part 4 was noncomparative and nonrandomized. In addition, longitudinal analyses of per-patient Phe control were not reported.

PRISM-3 Trial

The PRISM-3 trial was an exploratory phase III substudy to evaluate executive function in adults with PKU participating in PRISM-2 (N = 9).¹⁸ The study is briefly summarized here because it addressed outcomes (executive function and self-perception) that were not evaluated in PRISM-2. Although targeted enrolment was approximately 100 patients, only 9 were enrolled in PRISM-3 by the closure of enrolment in PRISM-2 part 2 (RDT). The reasons for the low enrolment were unclear as all adult participants in PRISM-2 (age 18 to 70 years) were eligible for PRISM-3. Patients participated in the PRISM-3 substudy for up to 63 weeks. The primary outcome was executive function assessed using a selected set of 3 tasks from the Cambridge Neuropsychological Test Automated Battery tool: rapid visual processing, spatial working memory, and stop-signal task. Patient self-perception of their current state was measured using a Subject Global Assessment (SGA) questionnaire consisting of 7 questions about current state perception of attention, energy level, tiredness, confusion, sadness, anger, and tension over the previous 7 days. The SGA and the Cambridge Neuropsychological Test Automated Battery tasks were administered at PRISM-2 part 2 baseline and part 2, week 8. At part 2 week 8, patients receiving active pegvaliase (n = 6) had numeric improvements compared with patients receiving placebo (n = 3) in rapid visual processing mean response latency (mean change = 24.23 ms versus −18.03 ms), spatial working memory between errors 4 to 8 Boxes (mean change = −4.2 versus 3.0), and stop-signal task reaction time (mean change = 8.42 ms versus 58.93 ms). No differences in SGA questionnaire responses were observed between patients treated with pegvaliase and those treated with placebo.

Comparative Evidence With Sapropterin and MNT

One additional retrospective observational cohort study by Zori et al. (2019)¹⁹ did not meet the selection criteria for inclusion in the systematic review but is summarized here because it was designed to generate comparative evidence of the efficacy of pegvaliase plus MNT, sapropterin plus MNT, and MNT in adolescent and adult patients with PKU.¹⁹ The results of this study were used to inform the transitions between health states in the sponsor’s economic model (CADTH Pharmacoeconomic Report).

Methods

Zori et al. conducted a retrospective observational cohort study of adult patients with PKU receiving pegvaliase with or without MNT, sapropterin plus MNT, or MNT alone. A cohort of patients who received pegvaliase plus MNT in the phase II 165 to 205 trial or phase III PRISM studies (PRISM-1 and PRISM-2) were compared using a PSM approach with a historical control of patients who received sapropterin plus MNT or MNT alone and participated in the PKUDOS registry.²⁰ The outcomes evaluated in the study included change in blood Phe and natural protein intake after 1 and 2 years of treatment. The source studies (pegvaliase trials and the PKUDOS registry) took place in the US.

Populations

Patient data from participants in pegvaliase clinical trials and matched historical cohorts selected by PSM from the PKUDOS registry were used to select the study cohort. Pegvaliase patients were those who were originally enrolled in the parent studies 165 to 205 (24 patients) and the phase III PRISM studies (261 patients) and participated in extension studies (PAL-003 and PRISM-2 part 4).¹ Inclusion criteria for the pegvaliase cohort included adolescents or adult patients with PKU (age ≥ 16 years) with a baseline Phe greater than 600 µmol/L who received induction, titration, and maintenance dosing of pegvaliase (5 mg to 60 mg daily self-administered injection). Patients included in the sapropterin plus MNT and MNT-alone groups were identified from the PKUDOS registry.²⁰ This registry is a phase IV voluntary observational study of patients with PKU who are or have been treated with sapropterin, or who intend to take sapropterin within 90 days of registry enrolment, with data (from 2007 onwards) on 1,867 patients at the time of the study by Zori et al. Patients in the sapropterin plus MNT group included those in the PKUDOS registry who intended to initiate sapropterin within 90 days, had 1 or more pre-treatment blood Phe value measurements, baseline blood Phe (last available measurement before initiating sapropterin) greater than 600 µmol/L, were 18 years or older at sapropterin initiation, and available information on sapropterin dosing (5 mg to 20 mg daily). Although patients in the PKUDOS registry were not required nor monitored for dietary restriction, the product monograph for sapropterin²⁷ states that sapropterin is indicated in conjunction with a Phe-restricted diet to reduce blood Phe levels. This group is referred to as sapropterin plus MNT. Patients in the MNT-alone group included those in the PKUDOS registry who received a Phe-restricted diet alone who had previously received sapropterin before enrolling in the registry or who discontinued sapropterin, had baseline blood Phe (last available measurement before initiating sapropterin) greater than 600 µmol/L, and who were 18 years or age or older.

Propensity scores for PSM were estimated using logistic regression with probability of treatment as the outcome (pegvaliase, sapropterin plus MNT, or MNT alone). Baseline blood Phe concentration, age, and gender were included in the PSM model. A propensity score including baseline dietary Phe was considered in a sensitivity analysis due to smaller sample size. Patients in the sapropterin plus MNT or MNT-alone groups were randomly ordered according to propensity score and then sequentially matched (1:1) using the nearest neighbour method) to a pegvaliase patient. Of the 1,867 patients who participated in the PKUDOS registry, 221 intended to initiate sapropterin; 64 met the inclusion criteria and were propensity-score matched to 1 of the 285 patients who received pegvaliase. A total of 557 patients had previously received sapropterin, of whom 125 who were on MNT alone met the inclusion criteria and were propensity score matched to 1 of the 285 patients who received pegvaliase.

As the PKUDOS registry (N = 1,867; 52.8% female; mean age of 16 years [SD = 13]; mean blood Phe of 585 µmol/L [SD = 407]; mean baseline daily natural protein intake of 20 g [SD = 21] included patients of all ages, the mean age was lower for this group than for the pegvaliase-treated patients, who were adults only (N = 285; 50.2% female; mean age of 29 years [SD = 9]; mean blood Phe of 1,227 µmol/L [SD = 379]; mean baseline natural protein intake of 38 g [SD = 28]). Baseline blood Phe was the last available measurement before starting pegvaliase or sapropterin; for the MNT-alone group, baseline blood Phe was the measurement closest to the enrolment date within 90 days in case sapropterin was discontinued before enrolment or the value closest to the discontinuation date within 90 days of discontinuation if sapropterin was discontinued after enrolment. Because the propensity score was calculated based on age, sex, and baseline blood Phe concentration, the groups were well balanced with respect to these characteristics. Sixty-four patients who received pegvaliase (59.4% female; mean age of 32 years [SD = 9]; mean blood Phe of 1,172 µmol/L [SD = 329], mean baseline natural protein intake of 33 g [SD = 19]) were matched to 64 patients who received sapropterin plus MNT [57.8% female; mean age of 33 years [SD = 10]; mean blood Phe of 1,176 µmol/L [SD = 383]; mean baseline natural protein intake of 36 g [SD = 31]). A total of 125 patients who received pegvaliase (44.8% female; mean age of 30 years [SD = 8]; mean blood Phe of 1,085 µmol/L [SD = 294]; mean baseline natural protein intake of 34 g [SD = 24]) were matched to 125 patients who received MNT alone (44.8% female; mean age of 31 years [SD = 11]; mean blood Phe of 1,089 µmol/L [SD = 302]; mean baseline natural protein intake of 25 g [SD = 19]). Baseline mean daily natural protein intake was roughly similar across groups.

Interventions

Patients in the pegvaliase group received daily self-administered subcutaneous injections (5 mg to 60 mg). Patients who were in the pegvaliase trials were instructed to maintain consistent protein intake from natural foods and MNT during the trial unless Phe decreased to less than 30 µmol/L. Dietary control of Phe and adherence to MNT were not requirements for entering the pegvaliase trials. Patients in the sapropterin plus MNT group received 5 mg to 20 mg sapropterin orally, and would have been encourage to consume a low Phe diet (MNT) according to prescribing information in the sapropterin product monograph.²⁷ Patients in the MNT alone cohort did not receive pegvaliase or sapropterin. As both historical cohorts were derived from the PKUDOS registry, participants did not receive any specific trial-related interventions such as study visits, questionnaires, education or bloodwork, or any specific dietary interventions.

Outcomes

Outcomes were based on available data from the PKUDOS registry patients (and matched pegvaliase patients) who had at least 1 measurement of Phe levels after treatment for 1 or 2 years. Outcomes included: mean blood Phe concentration at 1 and 2 years; change in blood Phe from baseline; percentages of patients achieving blood Phe of 600 µmol/L or lower, 60 µmol/L or lower, and 120 µmol/L or lower; percentages of patients achieving reductions of 20% or greater, 30% or greater, or 50% or greater from baseline in blood Phe; and natural intact protein (g/day). Reduction in blood Phe was calculated as: baseline blood Phe minus the year 1 or year 2 blood Phe value divided by baseline blood Phe. Follow-up blood Phe concentrations at 1 and 2 years were the levels recorded within 365 days ± 45 days or 730 days ± 90 days. If there was more than 1 value for blood Phe, the median was used.

For patients in the sapropterin plus diet and diet-alone groups, natural intact protein (g/day) was defined as total protein intake minus medical food protein intake. For patients receiving pegvaliase, natural intact protein (g/day) was defined as average dietary protein intake from intact food.

Statistical Analysis

To evaluate differences in blood Phe reductions between treatment groups in the patients with PKU and uncontrolled Phe (blood Phe above 600 μmol/L), mean change in blood Phe reduction from baseline was modelled with treatment as the primary factor and propensity score as the secondary factor. There was no formal statistical hypothesis testing or power calculations, and the results were presented as descriptive analyses with no control of type I error.

Patient Disposition

Although there were 1,867 patients in the PKUDOS registry, only 64 patients receiving sapropterin plus MNT and 125 patients receiving MNT alone were matched to patients receiving pegvaliase. Follow-up data were available for only a subset of these patients. At year 1, data were available for 43 pegvaliase patients and 25 sapropterin plus MNT patients of the 64 matched pairs, and for 87 pegvaliase patients and 51 MNT-alone patients of the 125 matched pairs. At year 2 follow-up data were available for 40 pegvaliase patients and 25 sapropterin plus MNT patients (of 64 matched pairs), and 80 pegvaliase patients and 42 MNT-alone patients (of the 125 matched pairs). There was a far higher rate of missing data for Phe levels among patients from the PKUDOS registry, particularly those in the MNT-alone group.

Exposure to Study Treatments

Adherence to pegvaliase in the phase III PRISM trials was self-reported as very high, leading to good exposure. No information was provided on adherence for the patients receiving sapropterin plus MNT or MNT alone participating in the PKUDOS registry.

Efficacy

Overall, a greater number of patients in the pegvaliase groups achieved lower blood Phe levels compared with the other groups. Among the 43 patients in the pegvaliase group matched to the sapropterin plus MNT group with data available, mean Phe decreased from 1,180 µmol/L (SD = 317) to 505 µmol/L (SD = 509) at 1 year. For the 25 patients in the sapropterin plus MNT group with data available, mean Phe decreased from 1,075 µmol/L (SD = 419) to 807 µmol/L (SD = 389). The LSM difference between Phe levels in the pegvaliase and sapropterin plus MNT groups was −399.4 µmol/L (95% CI, −660.2 to −138.7) at 1 year. At year 2, mean blood Phe levels decreased from a baseline of 1,195 µmol/L (SD = 323) to 427 µmol/L (SD = 527) (n = 40 pegvaliase patients) and from a baseline of 1,060 µmol/L (SD = 337) to 891 µmol/L (SD = 381) (n = 25 sapropterin plus MNT group) with an LSM difference between groups of −647.6 µmol/L (95% CI, −910.0 to −385.3).

For the 87 patients in the pegvaliase group matched to MNT alone group with data available, mean Phe decreased from 1,089 µmol/L (SD = 289) to 473 µmol/L (SD = 451) at 1 year. For the 51 patients in the MNT-alone group with data available, mean blood Phe levels decreased from 1,037 µmol/L (SD = 271) to 1,022 µmol/L (SD = 322) at 1 year. The LSM difference between the pegvaliase and MNT-alone groups was −567.8 µmol/L (95% CI, −708.3 to −427.4) at 1 year. At year 2, mean blood Phe levels decreased from baseline of 1,107 µmol/L (SD = 293) to 302 µmol/L (SD = 392) (n = 80 pegvaliase patients) and from a baseline of 1,051 µmol/L (SD = 302) to 965 µmol/L (SD = 359) (n = 42 MNT alone patients) with an LSM difference between groups of −670.9 µmol/L (95% CI, −824.1 to −517.7).

Higher proportions of patients in the pegvaliase group achieved Phe reductions at various thresholds compared with matched patients receiving sapropterin plus MNT or MNT alone. In addition, patients in the pegvaliase group had increased natural protein intake from baseline at follow-up compared with matched patients receiving sapropterin plus MNT or MNT alone. However, little data on protein intake were available for some groups at follow-up, particularly in the sapropterin plus MNT group (4 patients at 1 year and 7 patients at 2 years). Data for the 4-factor PSM cohort (including baseline dietary Phe) analysis demonstrated similar findings in terms of reduction in blood Phe and increased protein intake.

Harms

Comparisons of harms between patients receiving pegvaliase in the phase III PRISM trials and patients receiving sapropterin plus MNT or MNT alone participating in the PKUDOS registry were uninformative. AE were not actively solicited in the PKUDOS registry.²⁰

Critical Appraisal

Internal Validity

The study by Zori et al. was a post hoc exploratory analysis of patients participating in phase III clinical trials of pegvaliase compared with a historic control of patients participating in the PKUDOS registry who received sapropterin plus MNT or MNT alone. Due to the nonrandomized design, there was the potential for confounding of treatment-effect estimates by known and unknown confounders or by natural fluctuations over the course of PKU that cannot be adjusted for by the PSM analysis.

A major limitation associated with assessing the results of any analyses from a registry is the potential for data to be incomplete. In addition, data are not verified (at the source) as they would be in the context of a clinical trial. Although there were 1,867 patients in the PKUDOS registry, few patients receiving MNT alone and sapropterin plus MNT could be matched to patients receiving pegvaliase; follow-up data were collected for only a subset of these participants at 1 and 2 years. There are many potential sources of bias that could affect measurements of blood Phe levels that were not measured as part of the PKUDOS registry and would therefore be missing from the information used for PSM in the study by Zori et al. Because the PSM was based on baseline blood Phe, age, and gender, the groups were well balanced for these factors. However, many factors that may have affected levels of blood Phe were not included in the PSM approach, including but not limited to: PAH genotype, PKU severity, race or ethnicity, education, socioeconomic status, insurance status, measured Phe tolerance, other medical conditions, mental health conditions, neuropsychological deficits, dietary intervention or education, rationale for prescribing sapropterin, type and dose of MNT, BMI, drug dosing, compliance or adherence, or differences in clinical approach. These factors were not captured and their impact on blood Phe measurements in different groups could not be accounted for in the current analysis.

The relatively rigorous conditions of a clinical trial (mandated bloodwork, follow-up visits, and dietary reminders and interventions) imply monitoring and clinical care that are fundamentally distinct from those of a voluntary observational registry under real-world conditions. These different conditions may have affected blood Phe in the study by Zori et al. For example, variability in blood Phe data could be due to differences in analytical measurements between centres and inconsistency in the number of blood Phe measurements acquired per patient. In the PKUDOS registry, assessments were based on current medical practice at each separate study centre, and variability in sample methods differed by site.²⁰ Patients in the PKUDOS registry were not required nor followed to ensure adherence to MNT; no information about patient adherence with sapropterin or MNT was reported. Information about MNT intake was not described.²⁰ Diet and MNT intake may have varied between the pegvaliase, sapropterin plus MNT, and MNT-alone groups. Potential differences in these factors and their effects on blood Phe limit interpretation of the findings of the study by Zori et al.

The limitations of blood Phe as a study outcome are discussed in the main report. Point observations of blood Phe levels was measured at year 1 and 2 in the study by Zori et al. and could have been affected by numerous factors, mostly adherence to therapy, including diet and MNT. Although the number of blood Phe values used for the outcome was not stated, this was probably a very low number of observations for comparison purposes. As information on diet or MNT was not captured in the PKUDOS registry, and patients were not required to follow any dietary restrictions nor track protein intake, dietary Phe, or MNT, these factors may have affected single blood Phe level measurements, limiting the conclusions that can be drawn about the relative impacts of pegvaliase, sapropterin plus MNT, or MNT alone on blood Phe.

Only a small number of patients were included in the analysis set for efficacy outcomes. The analysis was not pre-specified and lacked control for type I error, and all results are therefore exploratory. There was a lack of planned hypothesis testing (and lack of sample-size considerations related to hypothesis testing). Time effects may have affected the study outcome as the historical control took place in a different time frame. These statistical limitations affect the conclusions that can be drawn from the study.

External Validity

The retrospective observational cohort study design with a historical control makes it difficult to generalize beyond the population of the study by Zori et al. Selection of small numbers of patients from the PKUDOS registry who had blood Phe greater than 600 µmol/L limits generalizability to patients with similar features. However, the clinical experts consulted by CADTH for this review stated that, in Canada, patients would not continue on sapropterin if they maintained high blood Phe levels. The patient population derived from the PKUDOS registry had either recently initiated sapropterin or had discontinued sapropterin; these groups may not be representative of the general population of patients with PKU. Changes in the natural history of the disease, dietary interventions, education or availability or uptake of MNT may have been affected by the use of historic control data.

Discussion

Summary of Available Evidence

One phase III, 4-part, 4-arm, double-blind placebo-controlled RDT with an extension period of open-label treatment (PRISM-2,^9-16 N = 215 adolescent and adult patients with PKU) contributed evidence to this report. In addition, the PRISM-1 trial,^16,17 a phase III, open-label, randomized, multi-centre study to assess the safety and tolerability of pegvaliase among drug-naive patients with PKU (N = 261), contributed additional evidence related to the safety of pegvaliase, and the PRISM-3 trial, an exploratory phase III substudy to evaluate executive function in adults with PKU participating in PRISM-2 (N = 9),¹⁸ addressed outcomes (executive function and self-perception) that were not evaluated in the PRISM-2 trial. Finally, an observational retrospective cohort study by Zori et al.¹⁹ was intended to assess the comparative efficacy of pegvaliase, sapropterin plus MNT, and MNT alone in adult patients with PKU.

The pivotal PRISM-2 study provided the only randomized controlled trial evidence of pegvaliase against an appropriate comparator (placebo with or without MNT) for this review. According to the clinical experts consulted by CADTH for this review, the baseline characteristics of the PRISM-2 study population were broadly representative of adolescent and adult patients with PKU who would be candidates for pegvaliase. Almost all patients were White adults 18 years or age or older; the average age was approximately 30 years. According to the clinical experts, baseline blood Phe, mood and inattention symptoms, and protein intake in the PRISM-2 study were as expected for adult patients with PKU with poor or no Phe control and limited adherence to MNT. However, the experts acknowledged that the PRISM-2 study population was enriched for patients (those who did not discontinue treatment in feeder studies or PRISM-2 part 1 due to AEs or patient preference, who were able to achieve target dose in feeder studies, and who achieved a decrease in blood Phe of 20% or greater during PRISM-2 part 1). The major limitations of the PRISM-2 RDT were uncertainty in adherence to pegvaliase and to dietary protein intake, both of which were self-reported, and unclear relevance of pegvaliase-induced decreases in blood Phe of the reported magnitude at a single time point (week 8) to consistency and durability of long-term Phe control, improved inattention and mood symptoms, improved protein tolerance and diet liberalization, and improved HRQoL.

Interpretation of Results

Efficacy

The PRISM-2 RDT did not address several key outcomes that were identified by clinical experts consulted by CADTH for this review as the goal of treatment in adult patients with PKU (improved protein tolerance and HRQoL). These outcomes were also identified as important by patients. The clinical experts consulted by CADTH for this review stated that blood Phe levels can be used to show that pegvaliase treatment is ineffective, and in patients with low Phe who liberalize their diets to include natural foods, stability of Phe levels with the treatment range can demonstrate improvements in protein tolerance. A clinician group had contrasting views and viewed blood Phe as an important marker of treatment response. However, all clinicians agreed that blood Phe is an acceptable surrogate measure that is widely used in clinical trials for reasons of convenience, and is well established to mechanistically drive the symptoms of PKU.

Withdrawal of pegvaliase from patients in the placebo groups of the PRISM-2 RDT led to increases in blood Phe at part 2, week 8 (the difference in LSM change from baseline between the pooled active group and the 20 mg/day placebo group was −923.25 µmol/L, and the difference between the pooled active group and the 40 mg/day placebo group was −638.27 µmol/L). This difference was statistically and clinically meaningful, according to the clinical experts consulted by CADTH for this review, and was driven by the approximately half of patients in the pooled active group with very low Phe (≤ 120 µmol/L). The clinical experts indicated that the very low Phe levels in the subset of patients (roughly half) whom they speculated were taking pegvaliase injections had the potential to lead to improvements in other important outcomes, such as inattention and mood symptoms, HRQoL, and protein tolerance. However, the clinical experts acknowledged that no MID is known for the degree of reduction of blood Phe and its duration that would lead to improvements in other outcomes. In addition, the questionable accuracy of self-reported adherence data for pegvaliase and protein intake (which could be a major confounder of changes in blood Phe levels) was an important source of uncertainty in the primary analysis of Phe levels. In PRISM-2, blood Phe levels were studied at only a single time point in the primary efficacy analysis, and there was no randomized controlled trial evidence regarding the durability and consistency in Phe control associated with pegvaliase.

The clinical experts consulted by CADTH for this review acknowledged that, outside of a trial setting, compliance with pegvaliase would likely be lower among adult patients with PKU and blood Phe levels above 600 µmol/L than that observed in the PRISM-2 RDT. The proportion of patients in a real-world clinical setting who would comply with pegvaliase injections, and the resulting degree of Phe control that would be achieved in the overall adult PKU population, was uncertain.

Despite drastic decreases in blood Phe at part 2, week 8 in pegvaliase-treated patients compared with placebo-treated patients, no statistically significant differences were observed in ADHD-RS-IV inattention subscale scores among participants with a drug-naive score of 9 or greater, and further conclusions could not be drawn for other measures of inattention and mood symptoms (ADHD-RS-IV inattention subscale among all participants, PKU POMS confusion subscale, PKU POMS TMD, or POMS TMD) due to halting of the testing procedure. According to the clinical experts consulted by CADTH for this review, this was not surprising given the relatively short observation period and challenges in measuring inattention and mood symptoms in adult patients with PKU with established disease, who can experience differing degrees of neuropsychiatric and neurocognitive deficits irrespective of blood Phe level.

There was disagreement in the input received by CADTH from clinical experts and clinician groups on the target population of patients with PKU most and least appropriate for pegvaliase treatment (patients with good compliance to MNT and blood Phe of 600 µmol/L or less versus patients with uncontrolled blood Phe of greater than 600 µmol/L who are poorly compliant with MNT). The subgroup of patients with PKU for whom the findings of the PRISM-2 RDT in terms of blood Phe control are most relevant was uncertain. Presumably these would be patients with uncontrolled Phe and poor compliance with diet who are nevertheless willing and able to adhere to pegvaliase injections. This may represent a niche population, as the clinical experts consulted by CADTH for this review expected that adherence to MNT and pegvaliase would be correlated in many patients. Because of the enrichment design of the PRISM-2 RDT, the magnitudes of treatment effects (blood Phe decreases) may be overestimated for a large proportion of adult patients with PKU who would be less compliant and less able to tolerate pegvaliase. Long-term trends in adherence to pegvaliase and evaluation of patients who achieved durable Phe control was not evaluated in the PRISM-2 RDT. Nevertheless, the clinical experts agreed that the findings of the PRISM-2 RDT (reduced blood Phe following pegvaliase injection in adult patients with uncontrolled Phe at baseline) could be generalized to patients who are compliant with diet and have some degree of Phe control.

Analyses of long-term open-label pegvaliase treatment in PRISM-2 part 4 (open-label extension) were suggestive of continued Phe control over time in some patients. However, the absence of a comparison group, attrition bias, and lack of per-patient longitudinal Phe data limited interpretation of long-term efficacy. Analysis of executive function in the PRISM-3 substudy was limited by a small sample size (N = 9). Because of numerous limitations in the study design involving comparisons with a historical control cohort from the PKUDOS registry, potential bias due to the nonrandomized study design and PSM approach, missing data, and statistical limitations (exploratory analysis only), no clear conclusions could be drawn concerning the comparative effectiveness of pegvaliase, sapropterin plus MNT, and MNT based on the observational retrospective cohort study by Zori et al.

Harms

The phase III PRISM studies, including PRISM-1 and PRISM-2 with its open-label extension, provide a consistent picture of the safety profile of pegvaliase injections. Safety data from the placebo-controlled RDT (PRISM-2 part 2) over a treatment period of 8 weeks suggested that HAEs, generalized skin reactions lasting 14 days or more, injection-site skin reactions lasting 14 days or more, rash, urticaria, pruritis, and erythema were potentially associated with pegvaliase rather than placebo injections. During longer-term open-label treatment with pegvaliase in the PRISM-2 study (in which it was not possible to separate the impacts of pegvaliase and the self-injection procedure), HAEs, arthralgia, and injection-site reactions occurred in |||| patients, while generalized skin reactions lasting 14 days or more and injection-site skin reactions lasting 14 days or more occurred in |||||| |||| of patients. Anaphylaxis occurred in |||| of patients and angioedema occurred in |||| of patients. Pegvaliase received a black-box warning from the FDA due to the risk of anaphylaxis.

Because treatment adherence in the PRISM-2 trial was self-reported and these data were potentially flawed, according to the clinical experts consulted for this review (based on high adherence but lack of Phe control in approximately half of patients), AE rates may have been underestimated in the study. In addition, patients with PKU participating in the PRISM-2 study took mandatory pre-medication during re-introduction of study drug, and were likely medically followed more frequently and received additional training on AE detection and management compared with the general population of adult PKU population. According to the clinical experts consulted by CADTH for this review, these factors would not be major impediments to generalizing the safety profile of pegvaliase in the PRISM-2 study to real-world clinical practice.

Conclusions

Data from the PRISM-2 RDT suggest that continued self-administration of pegvaliase injections led to statistically significant and potentially clinically meaningful differences in blood Phe levels after 8 weeks compared with withdrawal of pegvaliase and injection of placebo. Low blood Phe (≤ 120 µmol/L) was observed in approximately half of patients receiving active pegvaliase. Durability and consistency of Phe control were not evaluated in the PRISM-2 RDT. Furthermore, the estimated benefit in reducing blood Phe levels may have been overestimated relative to the general population of adult patients with PKU due to the enriched design of the RDT, which selected for patients more likely to adhere and respond to pegvaliase. Despite significant differences in Phe at week 8 in patients receiving pegvaliase and placebo, no conclusions could be drawn regarding differences in inattention or mood symptoms resulting from continued treatment with pegvaliase. Other outcomes important to patients, including HRQoL and protein tolerance, were not assessed in the PRISM-2 RDT. Efficacy data from nonrandomized studies, including the PRISM-2 open-label extension and an observational study comparing pegvaliase with sapropterin plus MNT and MNT alone, was limited by potential bias and/or confounding. The safety profile of pegvaliase, established through the phase III PRISM trial, including the open-label extension of PRISM-2, pointed to HAEs, arthralgia, injection-site reactions, generalized skin reactions lasting 14 days or more, and generalized injection skin reactions lasting 14 days or more as common side effects. Anaphylaxis and angioedema were less common but clinically important serious adverse effects. Other limitations of the available evidence included an unclear relationship between the magnitude of changes in blood Phe at a single time point (in the PRISM-2 RDT) and changes in other outcomes of importance to patients with PKU, as well as uncertainty regarding the target population of patients with PKU most appropriate for pegvaliase. The observed changes in blood Phe in the PRISM-2 RDT were aligned with 1 of the outcomes identified as important by patients with PKU, and there is clearly an unmet need for additional efficacious treatments for PKU with higher uptake and adherence rates compared with MNT.

References

1.Drug Reimbursement Review sponsor submission: Palynziq (pegvaliase injection), 2.5 mg/0.5 ml (5 mg/ml), 10 mg/0.5 ml (20 mg/ml), 20 mg/ml, solution for injection in a prefilled syringe [internal sponsor's package]. Woodridge (ON): Biomarin Pharmaceutical (Canada) Inc.; 2021 Dec 22.

2.PALYNZIQ® (pegvaliase injection): solution 2.5 mg/0.5 ml (5 mg/ml), 10 mg/0.5 ml (20 mg/ml), 20 mg/1.0 ml (20 mg/ml) subcutaneous injection [product monograph]. Toronto (ON): BioMarin Pharmaceutical (Canada) Inc.; 2022 Mar 30.

3.Blau N, van Spronsen FJ, Levy HL. Phenylketonuria. Lancet. 2010;376(9750):1417-1427. PubMed

4.van Spronsen FJ, Blau N, Harding C, Burlina A, Longo N, Bosch AM. Phenylketonuria. Nat Rev Dis Prim. 2021;7(1):36. PubMed

5.Surtees R, Blau N. The neurochemistry of phenylketonuria. Eur J Pediatr. 2000;159 Suppl 2:S109-113. PubMed

6.Trefz F, Maillot F, Motzfeldt K, Schwarz M. Adult phenylketonuria outcome and management. Mol Genet Metab. 2011;104 Suppl:S26-30. PubMed

7.Simon E, Schwarz M, Roos J, et al. Evaluation of quality of life and description of the sociodemographic state in adolescent and young adult patients with phenylketonuria (PKU). Health Qual Life Outcomes. 2008;6:25. PubMed

8.Burton BK, Grange DK, Milanowski A, et al. The response of patients with phenylketonuria and elevated serum phenylalanine to treatment with oral sapropterin dihydrochloride (6R-tetrahydrobiopterin): a phase II, multicentre, open-label, screening study. J Inherit Metab Dis. 2007;30(5):700-707. PubMed

9.Clinical Study Report: 165-302. A four-part, phase 3, randomized, double-blind, placebo-controlled, four-arm, discontinuation study to evaluate the efficacy and safety of subcutaneous injections of BMN 165 self administered by adults with phenylketonuria [internal sponsor's report]. Novato (CA): BioMarin Pharmaceutical Inc.; 2019 Oct 9.

10.Aryal M, Lau K, Boyer R, et al. Achieving efficacy in subjects with sustained pegvaliase-neutralizing antibody responses. Mol Genet Metab. 2021;134(3):235-242. PubMed

11.Bilder DA, Arnold GL, Dimmock D, et al. Improved attention linked to sustained phenylalanine reduction in adults with early-treated phenylketonuria. Am J Med Genet A. 2022;188(3):768-778. PubMed

12.Gupta S, Lau K, Harding CO, et al. Association of immune response with efficacy and safety outcomes in adults with phenylketonuria administered pegvaliase in phase 3 clinical trials. EBioMedicine. 2018;37:366-373. PubMed

13.Harding CO, Amato RS, Stuy M, et al. Pegvaliase for the treatment of phenylketonuria: A pivotal, double-blind randomized discontinuation Phase 3 clinical trial. Mol Genet Metab. 2018;124(1):20-26. PubMed

14.Larimore K, Nguyen T, Badillo B, et al. Depletion of interfering IgG and IgM is critical to determine the role of IgE in pegvaliase-associated hypersensitivity. J Immunol Methods. 2019;468:20-28. PubMed

15.Qi Y, Patel G, Henshaw J, et al. Pharmacokinetic, pharmacodynamic, and immunogenic rationale for optimal dosing of pegvaliase, a PEGylated bacterial enzyme, in adult patients with phenylketonuria. Clin Transl Sci. 2021;14(5):1894-1905. PubMed

16.Thomas J, Levy H, Amato S, et al. Pegvaliase for the treatment of phenylketonuria: Results of a long-term phase 3 clinical trial program (PRISM). Mol Genet Metab. 2018;124(1):27-38. PubMed

17.Clinical Study Report: 165-301. A phase 3, open-label, randomized, multi-center study to assess the safety and tolerability of an induction, titration, and maintenance dose regimen of BMN 165 self-administered by adults with phenylketonuria not previously treated with BMN 165 [internal sponsor's report]. Novato (CA): Biomarin Pharmaceutical Inc.; 2017 Mar 17.

18.Clinical Study Report: 165-303. A phase 3 substudy to evaluate executive function in adults with phenylketonuria who are participating in the phase 3 study, 165-302 [internal sponsor's report]. Novato (CA): BioMarin Pharmaceutical Inc.; 2018 Jun 19.

19.Zori R, Ahring K, Burton B, et al. Long-term comparative effectiveness of pegvaliase versus standard of care comparators in adults with phenylketonuria. Mol Genet Metab. 2019;128(1-2):92-101. PubMed

20.Longo N, Arnold GL, Pridjian G, et al. Long-term safety and efficacy of sapropterin: the PKUDOS registry experience. Mol Genet Metab. 2015;114(4):557-563. PubMed

21.Hillert A, Anikster Y, Belanger-Quintana A, et al. The genetic landscape and epidemiology of phenylketonuria. Am J Hum Genet. 2020;107(2):234-250. PubMed

22.Statistics Canada. Population and demography statistics. 2022; https://statcan.gc.ca/en/subjects-start/population_and_demography. Accessed 2022 Mar 5.

23.Vockley J, Andersson HC, Antshel KM, et al. Phenylalanine hydroxylase deficiency: diagnosis and management guideline. Genet Med. 2014;16(2):188-200. PubMed

24.van Wegberg AMJ, MacDonald A, Ahring K, et al. The complete European guidelines on phenylketonuria: diagnosis and treatment. Orphanet J Rare Dis. 2017;12(1):162. PubMed

25.U. S. Food and Drug Administration (FDA). Drug Approval Package: PALYNZIQ (pegvaliase-pqpz). 2018; https://www.accessdata.fda.gov/drugsatfda_docs/nda/2018/761079Orig1s000TOC.cfm. Accessed 2022 Mar 5.

26.European Public Assessment Report (EPAR): Palynziq (pegvaliase). Amsterdam (NL): European Medicines Agency; 2019: https://www.ema.europa.eu/en/medicines/human/EPAR/palynziq. Accessed 2022 Mar 5.

27.Kuvan® (sapropterin dihydrochloride): 100 mg tablets [product monograph]. Toronto (ON): BioMarin Pharmaceutical (Canada) Inc.; 2016 Oct 14: https://pdf.hres.ca/dpd_pm/00036799.PDF. Accessed 2022 Mar 5.

28.McGowan J, Sampson M, Salzwedel DM, Cogo E, Foerster V, Lefebvre C. PRESS Peer Review of Electronic Search Strategies: 2015 guideline statement. J Clin Epidemiol. 2016;75:40-46. PubMed

29.Grey matters: a practical tool for searching health-related grey literature. Ottawa (ON): CADTH; 2019: https://www.cadth.ca/grey-matters. Accessed 2022 Jan 19.

30.Longo N, Zori R, Wasserstein MP, et al. Long-term safety and efficacy of pegvaliase for the treatment of phenylketonuria in adults: combined phase 2 outcomes through PAL-003 extension study. Orphanet J Rare Dis. 2018;13(1):108. PubMed

31.Burton BK, Longo N, Vockley J, et al. Pegvaliase for the treatment of phenylketonuria: Results of the phase 2 dose-finding studies with long-term follow-up. Mol Genet Metab. 2020;130(4):239-246. PubMed

32.Brown SG. Clinical features and severity grading of anaphylaxis. J Allergy Clin Immunol. 2004;114(2):371-376. PubMed

33.Wyrwich KW, Auguste P, Yu R, et al. Evaluation of neuropsychiatric function in phenylketonuria: psychometric properties of the ADHD Rating Scale-IV and Adult ADHD Self-Report Scale inattention subscale in phenylketonuria. Value Health. 2015;18(4):404-412. PubMed

34.Bacci ED, Wyrwich KW, Gries KS, et al. An adaptation of the Profile of Mood States for use in adults with phenylketonuria. Journal of Inborn Errors of Metabolism and Screening. 2016;4:2326409816669373.

35.Loprinzi Brauer CE, Motosue MS, Li JT, et al. Prospective validation of the NIAID/FAAN criteria for emergency department diagnosis of anaphylaxis. J Allergy Clin Immunol Pract. 2016;4(6):1220-1226. PubMed

36.International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH). Medical Dictionary for Regulatory Activities (MedDRA) version 21.1. 2018; https://www.meddra.org. Accessed 2022 Jan 1.

37.National Cancer Institute (NCI). Common Terminology Criteria for Adverse Events (CTCAE) version 4.03. 2010; https://ctep.cancer.gov/protocoldevelopment/electronic_applications/ctc.htm. Accessed 2022 Jan 1.

38.Longo N, Harding CO, Burton BK, et al. Single-dose, subcutaneous recombinant phenylalanine ammonia lyase conjugated with polyethylene glycol in adult patients with phenylketonuria: an open-label, multicentre, phase 1 dose-escalation trial. Lancet. 2014;384(9937):37-44. PubMed

39.Cleary M, Trefz F, Muntau AC, et al. Fluctuations in phenylalanine concentrations in phenylketonuria: a review of possible relationships with outcomes. Mol Genet Metab. 2013;110(4):418-423. PubMed

40.Hood A, Grange DK, Christ SE, Steiner R, White DA. Variability in phenylalanine control predicts IQ and executive abilities in children with phenylketonuria. Mol Genet Metab. 2014;111(4):445-451. PubMed

41.Hood A, Antenor-Dorsey JA, Rutlin J, et al. Prolonged exposure to high and variable phenylalanine levels over the lifetime predicts brain white matter integrity in children with phenylketonuria. Mol Genet Metab. 2015;114(1):19-24. PubMed

42.Viau KS, Wengreen HJ, Ernst SL, Cantor NL, Furtado LV, Longo N. Correlation of age-specific phenylalanine levels with intellectual outcome in patients with phenylketonuria. J Inherit Metab Dis. 2011;34(4):963-971. PubMed

Appendix 1: Literature Search Strategy

Note that this appendix has not been copy-edited.

Clinical Literature Search

Overview

Interface: Ovid

Databases:

• MEDLINE All (1946–)

• Embase (1974–)

Note: Subject headings and search fields have been customized for each database. Duplicates between databases were removed in Ovid.

Date of search: February 3, 2022

Alerts: Bi-weekly search updates until project completion

Search filters applied: No filters were applied to limit the retrieval by study type.

Limits:

• No date or language limits were used

• Conference abstracts: excluded

Table 21: Syntax Guide

Multi-Database Strategy

1. Phenylalanine Ammonia-Lyase/ad, tu

2. (palynziq* or pegvaliase* or BMN165 or BMN 165 or PEGPAL or PEG-PAL or rAvPAL* or phenylase or N6UAH27EUV).ti,ab,kf,ot,hw,rn,nm.

3. ((pegylated or recombinant) adj4 phenylalanine ammonia lyase).ti,ab,kf,ot,hw,rn,nm.

4. 1 or 2 or 3

5. Enzyme Replacement Therapy/ or (enzyme adj3 (replac* or substitut*) adj2 (treat* or therap*)).ti,ab,kf.

6. exp Phenylketonurias/ or (Phenylketonuria* or PKU or “261600” or McKusick 26160 or ((phenylalanine hydroxylase or PAH or Tetrahydrobiopterin or BH4 or Dihydropteridine reductase or DHPR) adj2 deficien*) or hyperphenylalanin?emia* or HPA or Folling* disease or Foelling* disease).ti,ab,kf.

7. 5 and 6

8. 4 or 7

9. 8 use medall

10. *pegvaliase/

11. (palynziq* or pegvaliase* or BMN165 or BMN 165 or PEGPAL or PEG-PAL or rAvPAL* or phenylase).ti,ab,kf,dq.

12. ((pegylated or recombinant) adj4 phenylalanine ammonia lyase).ti,ab,kf,dq.

13. 10 or 11 or 12

14. enzyme replacement/ or (enzyme adj3 (replac* or substitut*) adj2 (treat* or therap*)).ti,ab,kf,dq.

15. Phenylketonuria/ or (Phenylketonuria* or PKU or “261600” or McKusick 26160 or ((phenylalanine hydroxylase or PAH or Tetrahydrobiopterin or BH4 or Dihydropteridine reductase or DHPR) adj2 deficien*) or hyperphenylalanin?emia* or HPA or Folling* disease or Foelling* disease).ti,ab,kf,dq.

16. 14 and 15

17. 13 or 16

18. 17 not (conference abstract or conference review).pt.

19. 18 use oemezd

20. 9 or 19

21. remove duplicates from 20

Clinical Trials Registries

ClinicalTrials.gov

Produced by the US National Library of Medicine. Targeted search used to capture registered clinical trials.

• [Search — palynziq OR pegvaliase OR BMN165 OR “BMN 165” OR PEGPAL OR “PEG-PAL” OR rAvPAL OR ravpalpeg]

WHO ICTRP

International Clinical Trials Registry Platform, produced by the WHO. Targeted search used to capture registered clinical trials.

• [Search terms — palynziq OR pegvaliase OR BMN165 OR “BMN 165” OR PEGPAL OR “PEG-PAL” OR rAvPAL OR ravpalpeg]

Health Canada’s Clinical Trials Database

Produced by Health Canada. Targeted search used to capture registered clinical trials.

• [Search terms — palynziq, pegvaliase, BMN165, BMN 165, PEGPAL, PEG-PAL, rAvPAL, ravpalpeg]

EU Clinical Trials Register

European Union Clinical Trials Register, produced by the European Union. Targeted search used to capture registered clinical trials.

• [Search terms — palynziq OR pegvaliase OR BMN165 OR “BMN 165” OR PEGPAL OR “PEG-PAL” OR rAvPAL OR ravpalpeg]

Grey Literature

Search dates: January 17, 2021 – February 3, 2022

Keywords: palynziq*, pegvaliase*, BMN165, BMN 165, PEGPAL, PEG-PAL, rAvPAL*

Limits: None

Updated: Search updated before the completion of stakeholder feedback period

Relevant websites from the following sections of the CADTH grey literature checklist Grey Matters: A Practical Tool for Searching Health-Related Grey Literature were searched:

• Health Technology Assessment Agencies

• Health Economics

• Clinical Practice Guidelines

• Drug and Device Regulatory Approvals

• Advisories and Warnings

• Drug Class Reviews

• Clinical Trials Registries

• Databases (free)

• Internet Search

Appendix 2: Description and Appraisal of Outcome Measures

Note that this appendix has not been copy-edited.

Aim

To describe the outcome measures summarized in Table 22, and review their measurement properties including validity, reliability, responsiveness to change, and the MID.

Table 22: Outcome Measures Included in PRISM-2 Part 2

ADHD-RS-IV = Attention Deficit Hyperactivity Disorder Rating Scale (Investigator-Rated); Phe = phenylalanine; PKU POMS = PKU-Specific Profile of Mood States; POMS = Profile of Mood States.

Source: PRISM-2 Clinical Study Report.⁹

Findings

The validity, reliability, responsiveness, and the MID of each outcome measure is summarized and evaluated in Table 23.

Table 23: Summary of Outcome Measures and Their Measurement Properties

ADHD-RS-IV = Attention Deficit Hyperactivity Disorder Rating Scale (Investigator-Rated); CGI-S = Clinical Global Impression Severity; MID = minimal important difference; Phe = phenylalanine; PKU = phenylketonuria; PKU POMS = Phenylketonuria-Specific Profile of Mood States; POMS = Profile of Mood States.

Blood Phe Concentration

It is well known that deleterious effects on neurocognition, intelligence, and executive functioning are associated with high blood and brain concentrations of Phe in patients diagnosed with PKU.³⁹ The purpose of dietary Phe restriction is to lower blood and brain Phe levels; thus, reducing the risk of damage to the brain.

There is speculation that fluctuations in Phe levels are of potential significance in their relation to intelligence and neurocognition; however, there remains no current definition regarding Phe fluctuations, and these have been measured in many different ways (e.g., SD, regression analysis of Phe concentrations, standard error of the estimate, and mean and its accompanying SD of index dietary control [IDC] measured by 6-month mean Phe values).³⁹ In the review by Cleary et al.³⁹ it was noted that a number of studies have reported that the highest concentrations of Phe were reported in the morning in patients with PKU (children and adults) when observing diurnal variation of blood Phe levels. Other studies summarized in this review have also reported that there may be up to 400% variation in day-to-day blood Phe levels in adults with well-controlled PKU, that blood Phe levels fluctuate to a larger extent in patients with PKU when compared to their healthy counterparts, that blood Phe concentrations increase with age while there is uncertainty surrounding whether fluctuations decrease with age, and that fluctuations may be influenced by PAH genotype, rates of growth, dietary adherence, diet, and illness.³⁹

With regard to the impact of blood Phe fluctuations on the brain, Cleary et al.³⁹ reported that healthy individuals have approximately equal concentrations of blood and brain Phe, while increases in Phe concentrations in the blood is higher than increases in the brain in patients with PKU. In addition, they noted that peaks of Phe levels last longer, are not as steep, and occur later in the brain when compared to blood in these patients.³⁹ There is conflicting evidence regarding the effects of Phe fluctuations on neurocognition and other measures of brain activity in patients with PKU, with some studies showing correlations between Phe fluctuations and deficits in executive functioning, cognition, and intelligence while others finding no associations.³⁹ Hood et al.⁴⁰ reported that Phe variability was a better predictor of cognitive performance when compared to the various other aforementioned indices of Phe control, along with being a better predictor of executive functioning in children aged 5 years and older when compared to patients under 5 years of age. Hood et al.⁴¹ retrospectively analyzed microstructural white matter integrity using mean diffusivity from diffusion tensor imaging and various Phe indices to measure blood Phe concentrations in early and continuously treated children with PKU to determine if prolonged exposure to both high and variable levels of blood Phe correlated with white matter compromise. The authors reported that microstructural white matter integrity compromise was correlated with mean Phe, the IDC, mean exposure, and the SD of exposure, indicating that high and variable blood Phe concentrations were predictors of white matter compromise in children with PKU.⁴¹ Viau et al.⁴² suggested that there was an association between measures of intelligence and the quality of metabolic blood Phe control during certain developmental periods in a study of mostly children with PKU. Perceptual reasoning appeared to be strongly associated with proper blood Phe control during ages 0 to 6 and 7 to 12 years, with specific areas of verbal comprehension being affected by increases of blood Phe levels in children aged 0 to 6 years.⁴² However, the evidence obtained by Viau et al.⁴² was not supportive that blood Phe level variability was a good predictor of intelligence. That being said, all of these results indicate that continual blood Phe concentrations should be monitored and controlled in childhood and that this should continue throughout the life of patients with PKU.^39-41

According to the ACMG, relaxation of Phe control can lead to the development of neurocognitive deficits and psychiatric symptoms, which can ultimately have an impact on quality of life. Hence, ACMG recommends that blood Phe levels be monitored and controlled throughout the life of a patient with PKU.²³ For patients with early-treated PKU who have discontinued therapy, ACMG recommends that treatment be reinitiated to lower blood Phe levels because this group of patients may see benefit, including improvement in neuropsychological symptoms. For patients with untreated or late-treated PKU, ACMG recommends that treatment be offered to lower blood Phe levels because this group of patients may still see benefit, including improvement in behaviour and psychiatric symptoms.²³ However, the magnitude of decrease in blood Phe levels as well as the duration and consistency of metabolic control required to see improvements in outcomes, such as Phe tolerance, neurocognitive and neuropsychiatric symptoms, and HRQoL, are not known. Further, studies determining the psychometric properties, including the MID, of blood Phe concentration in the adult PKU population, were not found.

Profile of Mood States (POMS)

The original POMS is a 65-item, self-administered questionnaire that assesses a patient’s transient and variable mood states. Six mood domains are evaluated: tension-anxiety (9 items), depression-dejection (15 items), anger-hostility (12 items), vigour-activity (8 items), fatigue-inertia (7 items), and confusion-bewilderment (7 items). The friendliness domain (7 items) was later determined by the original POMS developers to be too weak to be scored, resulting in a 58-item POMS. The full list of items can be found in the POMS user manual. Each item is rated on a 5-point Likert scale (0 = not at all; 1 = a little; 2 = moderately; 3 = quite a bit; and 4 = extremely). The POMS has been widely used in different therapeutic areas and as a result, the instrument has been modified a number of times, including the removal of, addition of, and changes in the items and domains, to adapt the questionnaire for its use in a targeted patient population or for a specific culture.³⁴

Extensive validation studies have not been conducted for any version of the POMS.³⁴ No studies regarding the psychometric properties of the POMS in adults with PKU were found.

PKU-Specific Profile of Mood States (PKU POMS)

Bacci et al.³⁴ adapted the original POMS to better assess relevant mood states in adult patients with PKU. Qualitative and quantitative assessments were conducted to determine the comprehensibility, acceptability, relevance, and the performance of each item and domain in the PKU population. The result was the development of the PKU POMS, a 20-item, self-administered questionnaire that assesses 6 renamed mood domains (items): anxiety (panicky, uneasy, nervous, anxious), depression (unhappy, sad, discouraged, lonely), anger (angry, grouchy, annoyed), activity (lively, active, energetic), tiredness (worn out, exhausted, sluggish), and confusion (confused, unable to concentrate, forgetful). Each item is rated on a 5-point Likert scale (0 = not at all; 1 = a little; 2 = moderately; 3 = quite a bit; and 4 = extremely).³⁴

Psychometric Properties of the PKU POMS

Bacci et al.³⁴ used the PKU POMS to ascertain its validity, reliability, and responsiveness in adult patients with PKU from an open-label study (PRISM-1) who were undergoing treatment with pegvaliase and completed the POMS on day 1. Any correlations with plasma Phe were exploratory because according to the authors, it was uncertain if a change in plasma Phe would correspond to a change in mood states.

Validity

For convergent validity, each domain of the PKU POMS was compared to ADHD-RS-IV (modified version with adult prompts) and plasma Phe values using Pearson correlations (n = 114). The Pearson correlation between ADHD-RS-IV (adult version) and activity, anxiety, confusion, tiredness, anger, and depression domains were −0.04, 0.33, 0.60, 0.28, 0.31, and 0.15, respectively. The correlation between ADHD-RS-IV (adult version) and anxiety, confusion, and anger domains were statistically significant (P < 0.05). The Pearson correlation between plasma Phe and all 6 domains ranged from 0 to 0.17, none of which were statistically significant.

Reliability

For internal consistency reliability, the Cronbach alpha ranged from 0.75 to 0.87, indicating each domain measured the same construct.

Responsiveness

For responsiveness, the change scores between day 1 and end of study for each PKU POMS domain was compared to the ADHD-RS-IV (adult version) and plasma Phe using Pearson correlation. The Pearson correlation between ADHD-RS-IV (adult version) and activity, anxiety, confusion, tiredness, anger, and depression domains were –0.40, 0.43, 0.52, 0.34, 0.32, and 0.21, respectively (n = 65). The correlation between ADHD-RS-IV (adult version) and activity, anxiety, confusion, tiredness, and anger domains were statistically significant (P < 0.05). The Pearson correlation between plasma Phe and activity, anxiety, confusion, tiredness, anger, and depression domains were –0.25, 0.34, 0.47, 0.15, 0.26, and 0.32, respectively (n = 52). The correlation between plasma Phe values and anxiety, confusion, and depression domains were statistically significant (P < 0.05).

Clinical Relevance

Studies determining the minimally MID for adults with PKU on the PKU POMS were not found.

Dietary Protein Intake

Dietary therapy with the restriction of dietary Phe intake is defined as the reduced intake of natural protein that is replaced with a protein source that is free of Phe to lower blood Phe levels. The recommended daily dietary intake of Phe, tyrosine, and protein for patients with PKU can be found in the ACMG practice guidelines for PKU.²³ Dietary Phe intake is determined by blood Phe levels but it can also be influenced by various factors including residual PAH activity, age, rate of growth, and responsiveness to pharmacological therapy. Reduced intake of natural protein can lead to an inadequate consumption of protein, nutrients, and calories that are necessary for maintenance of health and growth. Hence, modified low-protein foods and medical foods, which are a mixture of amino acids and free of Phe, are medically necessary to meet daily nutritional requirements in patients with PKU on dietary therapy. However, medical foods may not always contain an appropriate amount of nutrients because nutritional requirements can vary depending on a patient’s unique needs at different stages of life and in the presence of comorbidities. For patients who are able to maintain metabolic control with respect to blood Phe levels using dietary therapy alone, the benefits of adding on a pharmacologic agent would be an improvement in Phe tolerance, liberalization of diet, and enhanced HRQoL. An improvement in Phe tolerance is defined as an increase in dietary Phe intake while maintaining Phe control, thereby increasing natural protein intake.²³ However, studies determining the psychometric properties, including the MID, of Phe tolerance and natural protein intake in adult patients with PKU were not found.

Attention Deficit Hyperactivity Disorder Rating Scale (Investigator-Rated) (ADHD-RS IV)

The ADHD-RS-IV is based on the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders criteria used to assess ADHD symptoms in children and is completed by a parent or guardian.³³ The full scale comprises 18 items that are separated into 2 subscales: an Inattention subscale and a Hyperactivity/Impulsivity subscale. Each of the subscales is composed of 9 items that assesses the frequency of ADHD symptoms, with each item rated on a 4-point Likert frequency scale (0 = never or rarely; 1 = sometimes; 2 = often; and 3 = very often).³³ The recall period is 1 month. A higher score corresponds with worse severity of ADHD. This scale has been observed to be valid and reliable in patients with ADHD.³³ In the Inattention subscale, scores range from 0 to 27, with greater inattentive severity measured with higher scores.³³