Reimbursement Review

Vanzacaftor-Tezacaftor-Deutivacaftor (Alyftrek)

Sponsor: Vertex Pharmaceuticals (Canada) Incorporated

Therapeutic area: Cystic fibrosis, F508del or responsive CFTR mutation, in patients aged 6 years and older

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Clinical Review

Abbreviations

Executive Summary

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

Table 1: Background Information of Application Submitted for Review

NOC = Notice of Compliance.

Introduction

Cystic fibrosis (CF), an autosomal recessive condition, is the most common fatal genetic disease affecting children and young adults in Canada. It is caused by mutations in the CFTR gene, which is located on chromosome 7. A deletion of phenylalanine 508 in the first nucleotide binding domain (NBD1) (F508del) is the most common CFTR mutation that results in CF.¹ The Canadian Cystic Fibrosis Registry reported there were 4,513 people in Canada living with CF in 2023. Of these, 87.8% carried 1 or more F508del mutations.² CF results in airway obstruction, chronic endobronchial infection, and inflammation, which ultimately lead to destruction of lung tissue through the development of bronchiectasis and loss of lung function.³ Chronic endobronchial infection of the airways with bacterial pathogens, such as Pseudomonas aeruginosa, is associated with a more rapid loss of lung function.⁴ Acute or chronic endobronchial infections result in further destruction of lung tissue and are associated with respiratory morbidity. Lung disease accounts for the vast majority (> 80%) of deaths in patients with CF.^1,5 Gastrointestinal and pancreatic involvement results in pancreatic exocrine insufficiency in most individuals with CF, causing malabsorption of fats and fat-soluble vitamins, which leads to malnutrition.⁶ Almost all of these patients, approximately 90%, according to the clinical experts, will have pancreatic insufficiency and will need to take lifelong pancreatic enzyme replacement with every meal as well as fat-soluble vitamin therapy. With increasing age, these patients may develop CF-related diabetes and require treatment with insulin. In 2023, CF-related diabetes was reported in 32.5% of adults and 2.4% of children living with CF in Canada.²

CFTR genetic testing is standard of care and part of the routine diagnostic framework to confirm CF. No implementation barriers due to testing are anticipated from testing done as part of establishing treatment eligibility for vanzacaftor-tezacaftor-deutivacaftor (VNZ–TEZ–D-IVA).

Perspectives of Patients, Clinicians, and Drug Programs

The information in this section is a summary of input provided by the patient and clinician groups that responded to the CDA-AMC call for input and from the clinical experts consulted for the purpose of this review.

Patient Input

One patient group, Cystic Fibrosis Canada, submitted input for this review. Information was gathered through focus groups with people in Canada with CF and their caregivers. Other sources of data included surveys (including a 2021 survey with more than 1,200 responses), data from the Canadian Cystic Fibrosis Registry, and a multiphase burden of disease study and other relevant publications.

The patient group highlighted that patients not on a CFTR modulator (CFTRm) frequently have periods of infection and acute inflammation called exacerbations that require a hospital stay of at least 2 weeks and that may last up to 4 weeks. Some of these patients could benefit from a CFTRm, but there may not be any CFTRm therapies currently that have a Health Canada indication for their mutation. The group noted that approximately 33% of all adult patients with CF in Canada have CF-related diabetes. Any of the other drugs that patients need to take on a regular basis can also have negative side effects and long-term risks, such as chronic use of antibiotics that may lead to resistance. Moreover, as patients age, they may need to try multiple antibiotics to find one that works, making management of drug-to-drug interactions difficult. The patient group reiterated that, currently, only CFTRm therapies target the underlying pathophysiology of CF by improving the function of the CFTR gene, while the other therapies used to treat CF are for the management of complications and control of symptoms.

The patient group highlighted that too many patients lack access to any CFTRm, including those patients who have ultra-rare mutations that are not on the narrow list of 152 mutations indicated for elexacaftor-tezacaftor-ivacaftor and ivacaftor (ELX-TEZ-IVA) or who cannot tolerate it, or those who have experienced a response to only VNZ–TEZ–D-IVA. The patient group noted that the FDA has approved the use of ELX-TEZ-IVA in 272 mutations, compared with 152 mutations in Canada; the group suggested there are approximately 23 patients in Canada aged 2 years or older with 1 or more of the 119 mutations that are not approved in Canada for treatment with ELX-TEZ-IVA. Thus, they could access ELX-TEZ-IVA in the US but cannot access it in Canada. Additionally, the FDA approved VNZ–TEZ–D-IVA for an additional 31 unique mutations for which ELX-TEZ-IVA is not approved in Canada; this represents approximately 13 additional people in Canada without access to ELX-TEZ-IVA. The group pointed out that, if VNZ–TEZ–D-IVA gets approved in Canada, up to 36 individuals aged 6 or older could access it for the first time. The group specifically requested that the committee recommend access to VNZ–TEZ–D-IVA for the 152 CFTR mutations currently indicated in Canada for ELX-TEZ-IVA. The group further asked that the committee recommend access to VNZ–TEZ–D-IVA for patients in Canada with any of the 151 rare mutations or mutation combinations the FDA lists as indications for VNZ–TEZ–D-IVA, and to recommend the broadest possible access to VNZ–TEZ–D-IVA for individuals with other CFTR mutations when there is evidence, and when more evidence is generated over time (including in vitro laboratory evidence or clinical evidence), that shows these patients will or may experience a response to this therapy. The group highlighted that CF has significant financial implications for patients and their caregivers, health systems, and society.

No patients reported experience with VNZ–TEZ–D-IVA treatment. The patient group highlighted that VNZ–TEZ–D-IVA may improve quality of life and adherence as a 1-pill-a-day treatment, which would provide an option for parents who may have difficulty administering medications to their children, who already take several medicines a day to maintain their health.

Clinician Input

Input From Clinical Experts Consulted for This Review

The clinical experts described that ELX-TEZ-IVA is the current standard of care in patients with CF aged 6 years and older with eligible CF-causing mutations. However, there is an unmet need for more effective therapy that prolongs life, improves quality of life, and reduces the treatment burden of supportive medications in patients who have CF-causing mutations that are not eligible for treatment with ELX-TEZ-IVA, which represents approximately 10% of patients with CF in Canada and disproportionately represents racialized people. Additionally, even among patients eligible for treatment with ELX-TEZ-IVA, the clinical experts noted that most patients still have sweat chloride (SwCl) values in an abnormal range. Doses of ELX-TEZ-IVA are administered twice daily and require fat-containing meals for absorption; ergo, the experts noted that less frequent dosing is an unmet need that might improve adherence. Some patients may experience adverse events (AEs) or intolerance to ELX-TEZ-IVA requiring cessation of treatment, and there is an unmet need for alternative triple-combination CFTRm therapies in those circumstances. Finally, overall pill and treatment burden remains high in patients with CF, despite reductions after the advent of triple-combination CFTRm therapies (i.e., ELX-TEZ-IVA).

The clinical experts consulted by Canada’s Drug Agency (CDA-AMC) stated that VNZ–TEZ–D-IVA would occupy a place in therapy similar to that of ELX-TEZ-IVA, with the caveat that it may be applicable to a wider range of mutations, and that the current Health Canada indications begin at different ages (i.e., ELX-TEZ-IVA is currently indicated for patients aged at least 2 years, while VNZ–TEZ–D-IVA is indicated in patients aged at least 6 years). The experts noted that, given the once-daily dosing of VNZ–TEZ–D-IVA compared with ELX-TEZ-IVA, improvements in treatment adherence may be expected. However, patients with CF are typically taking numerous pills daily in addition to CFTRm therapies; ergo, this may not represent a substantial difference in overall treatment or pill burden. Nonetheless, the expert specialized in pediatric care said that a reduction in treatment or pill burden may be particularly significant for the treatment of pediatric patients with CF. The experts noted that it may be premature for every patient currently stable on ELX-TEZ-IVA to be switched to VNZ–TEZ–D-IVA due to the longer clinical safety record of ELX-TEZ-IVA; however, VNZ–TEZ–D-IVA is generally expected to become the new standard of care for eligible patients. Additionally, if there are clinical signals that achieving a lower SwCl than previously seen with ELX-TEZ-IVA results in less end-organ damage, it may further shift treatment paradigms and reduce the need for other supportive therapies.

The clinical experts indicated that the patient population appropriate for treatment with VNZ–TEZ–D-IVA is similar to that for ELX-TEZ-IVA, with the exception that there may be additional CF-causing mutations eligible for treatment with VNZ–TEZ–D-IVA. Aside from genotype, the patient population is expected to be similar.

The clinical experts noted that the assessment of CF-related treatment is based on lung function spirometry, number of days treated with oral and IV antibiotics for pulmonary exacerbations, number of CF-related hospitalizations, and monitoring of body mass index (BMI) or BMI z score (as appropriate by age) for signs of malnutrition. The clinical experts informed CDA-AMC that the assessments related to VNZ–TEZ–D-IVA should be similar to those expected for ELX-TEZ-IVA, although, in any reimbursement criteria, it is also important to consider that assessments for initiation, continuation, or discontinuation should differ for patients newly starting triple-combination CFTRm therapies (i.e., patients who are generally expected to experience greater improvements with therapy) versus switching between triple-combination CFTRm therapies (i.e., the new treatment would be expected to maintain the benefit previously observed with the other therapy). Assessment of response in the previously listed metrics should generally be performed after a year of treatment because, as the experts highlighted, some metrics (especially pulmonary exacerbations and BMI or BMI z score) may have fluctuations with seasonal illness and short-term instability in pediatric patients that may not be reflective of the effect of the drug.

The clinical experts noted that discontinuation should be considered in the case of intolerable AEs that cannot be sufficiently managed or, in the case of a lung transplant, similar to current practice with ELX-TEZ-IVA (although this does not reflect a contraindication specified by Health Canada).

The clinical experts noted that a physician practising within or under the directed supervision of a Canadian CF clinic should manage the prescription and renewal of VNZ–TEZ–D-IVA.

Clinician Group Input

One clinician group, the Cystic Fibrosis Canada Health Care Advisory Council, provided input for this review. Information for this submission was gathered through the following: experience gained by working with and delivering medical services to people with CF; consultations with people with CF not eligible for ELX-TEZ-IVA; a review of medical and scientific literature, including clinical trial results and real-world experience; and the Canadian Cystic Fibrosis Registry, a collection of patient data and other information regarding CF care and outcomes. A total of 2 clinicians contributed to the input submitted for this review.

The input from the clinician group was generally consistent with that of the clinical experts consulted for this review, emphasizing that about 10% of patients have mutations not currently approved for treatment with available CFTRm therapies in Canada, and that there are no alternative triple-combination CFTRm treatments for patients with intolerance to ELX-TEZ-IVA. The group agreed that the once-daily regimen of VNZ–TEZ–D-IVA may improve adherence and convenience compared with the twice-daily regimen of ELX-TEZ-IVA.

The group noted that rare mutations for which there is currently no CFTRm indication are disproportionately found in racialized communities, so expanded access to this therapy may improve equitable access to disease-modifying treatments in Canada.

The clinician group noted that, according to recent Canadian clinical guidelines, all patients with CF who have at least 1 CFTRm-responsive CFTR variant should be treated with a CFTRm. The clinician group noted that patients with CF best suited for the drug under review include patients already on ELX-TEZ-IVA who are experiencing challenges due to side effects or related to adherence to a twice-daily treatment regimen, and patients with rare mutations not approved for treatment with ELX-TEZ-IVA but who may experience a response to VNZ–TEZ–D-IVA. The group agreed with the clinical experts regarding the metrics and yearly time frame of reassessments for treatment efficacy, as well as factors to consider for treatment discontinuation and prescribing.

Drug Program Input

The following were identified as key factors that could potentially impact the implementation of a recommendation for VNZ–TEZ–D-IVA:

• Considerations for:

  • initiation of therapy

  • continuation or renewal of therapy

  • discontinuation of therapy

  • prescribing of therapy

• care provision issues

• system and economic issues.

The clinical experts provided advice on the potential implementation issues raised by the drug programs. Refer to Table 7 for more details.

Clinical Evidence

Systematic Review

Description of Studies

Two double-blind phase III randomized controlled trials (RCTs), SKYLINE 102 (VX21-121-102; N = 405) and SKYLINE 103 (VX21-121-103; N = 574), were included. One single-arm phase III study, RIDGELINE (VX21-121-105), was also included, specifically, cohort A1 (N = 17) and cohort B1 (N = 78).

The RCTs were randomized studies with a 52-week treatment phase and were similar in study design. Both SKYLINE studies compared VNZ–TEZ–D-IVA with ELX-TEZ-IVA in patients with CF aged at least 12 years with ELX-TEZ-IVA–responsive mutations. Both included patients with a percent predicted forced expiratory volume in 1 second (ppFEV₁) value between 40% and 90% of the predicted mean for age, sex, and height for patients currently receiving ELX-TEZ-IVA, or a ppFEV₁ of between 40% and 80% for patients not currently receiving ELX-TEZ-IVA. In terms of CF-related genotype, patients recruited to the SKYLINE 102 trial had to be heterozygous for F508del and a minimal function mutation, while patients recruited to the SKYLINE 103 trial had to have either homozygous F508del mutations, heterozygous F508del and a gating or residual function mutation, or at least 1 mutation identified as responsive to ELX-TEZ-IVA with no F508del mutation. All patients entered a 28-day run-in period receiving ELX-TEZ-IVA before being randomized 1:1 in a double-blind fashion to treatment with VNZ–TEZ–D-IVA or ELX-TEZ-IVA for 52 weeks, followed by an additional 28-day open-label follow-up for additional safety data.

In the RIDGELINE trial, patients aged between 1 and 11 years with a mutation responsive to ELX-TEZ-IVA were recruited. The RIDGELINE study included a part A, with the objective of evaluating the pharmacokinetics of VNZ–TEZ–D-IVA and relevant metabolites when dosed in therapeutic concentrations, and to evaluate the safety and tolerability of VNZ–TEZ–D-IVA. Part A preceded part B and informed the doses and weight-related dose thresholds for part B, in which the objectives were primarily to evaluate the safety and tolerability of VNZ–TEZ–D-IVA through week 24, and secondarily to evaluate the efficacy through week 24 as well as the pharmacokinetics. Parts A and B were each divided into 3 cohorts by age range, where the first cohort (A1 and B1) included patients aged 6 to 11 years (inclusive), the second (A2 and B2) included patients aged 2 to 5 years (inclusive), and the third (A3 and B3) included patients aged 1 year to younger than 2 years. Therefore, only cohorts A1 (N = 17) and B1 (N = 78) are relevant to the Health Canada indication considered here. The efficacy data presented in this report are informed by cohort B1, and the safety data are informed by both cohorts A1 and B1.

In the SKYLINE 102 trial, slightly more than 40% of patients were female and just under 60% of patients were male, whereas the sex distribution was closer to approximately 50% each in the SKYLINE 103 study. The mean age by treatment group in the 2 RCTs at baseline was 31 to 34 years across the SKYLINE studies, and the treatment groups were similar for these categories in each study. In cohorts A1 and B1 in the RIDGELINE study, 41.2% and 43.6% were female, 58.8% and 56.4% of patients were male, and the mean age was 9.5 years and 9.3 years, respectively. The baseline patient characteristics for ppFEV₁ demonstrated more progressed disease in the studies that recruited older patients (the SKYLINE 102 and SKYLINE 103 trials recruited patients aged 12 years or older) compared with the RIDGELINE study, which recruited patients aged 6 to 11 years for cohorts A1 and B1; this reflects the natural course of CF, and there were no concerning within-study differences between treatment groups. In all studies, most patients (range, 84.6% to 94.1%) had immediate prior experience with a CFTRm, most commonly, ELX-TEZ-IVA. The mean duration of prior ELX-TEZ-IVA use was approximately 2 years in the SKYLINE 102 and SKYLINE 103 RCTs, and approximately 1 year in cohort B1 of the RIDGELINE trial. The SKYLINE 102 study recruited patients with a genotype that was heterozygous for F508del and a minimal function mutation; ergo, this genotype was represented in 100% of patients, whereas in the other studies, the most common genotype was homozygous for F508del (range, 47.4% to 78.2%). The majority of patients in every study were white (range, 90.7% to 97.5%), whereas a minority were Asian (range, 0% to 0.5%), Black or African American (range, 0% to 2.0%), more than 1 race (range, 0% to 1.5%), or other (range, 0.% to 0.4%), or information about race was not collected due to local regulations.

Efficacy Results

Number of Pulmonary Exacerbations

In the SKYLINE 102 trial full analysis set (FAS), the estimated event rate per year was 0.42 in the ELX-TEZ-IVA group (n = 50 patients with events) versus 0.32 in the VNZ–TEZ–D-IVA group (n = 60 patients with events). The pulmonary exacerbation rate difference was −0.10 events (95% confidence interval [CI], −0.24 to 0.04) per year.

In the SKYLINE 103 trial FAS, the annual event rate was 0.29 in the VNZ–TEZ–D-IVA group (n = 61 patients with events) and 0.26 in the ELX-TEZ-IVA group (n = 59 patients with events). The rate difference was 0.03 events per year (95% CI, −0.07 to 0.13).

In the RIDGELINE trial, there were 6 patients in the cohort B1 FAS who experienced an event, contributing to an annual event rate of 0.15.

No subgroup or sensitivity analyses were conducted for this end point in the included studies.

Time to First Pulmonary Exacerbation

In the SKYLINE 102 trial, based on a secondary analysis in the FAS of the time to first pulmonary exacerbation during the analysis period, the probability of event-free survival (Kaplan-Meier estimate) at 24 weeks was 0.820 (95% CI, 0.758 to 0.867) in the VNZ–TEZ–D-IVA group and 0.819 (95% CI, 0.758 to 0.866) in the ELX-TEZ-IVA group; at 52 weeks, it was 0.741 (95% CI, 0.673 to 0.797) and 0.695 (95% CI, 0.626 to 0.755), respectively.

In the SKYLINE 103 trial, based on a secondary analysis in the FAS of the time to first pulmonary exacerbation during the analysis period, the probability of event-free survival (Kaplan-Meier estimate) at 24 weeks was 0.857 (95% CI, 0.810 to 0.893) in the VNZ–TEZ–D-IVA group and 0.882 (0.838 to 0.914) in the ELX-TEZ-IVA group; at 52 weeks, it was 0.783 (95% CI, 0.729 to 0.827) and 0.792 (95% CI, 0.739 to 0.835), respectively.

Time to first pulmonary exacerbation was not reported in the RIDGELINE trial.

Pulmonary Exacerbations Requiring IV Antibiotics or Hospitalization

In the SKYLINE 102 trial FAS, 10 patients (5.1%) in the VNZ–TEZ–D-IVA group and 26 patients (12.9%) in the ELX-TEZ-IVA group had pulmonary exacerbations requiring hospitalization or IV antibiotic therapy during the analysis period. The probability of event-free survival (Kaplan-Meier estimate) in the 2 groups at 24 weeks was 0.969 (95% CI, 0.933 to 0.986) and 0.935 (95% CI, 0.890 to 0.962), respectively; at 52 weeks, it was 0.948 (95% CI, 0.906 to 0.972) and 0.868 (95% CI, 0.813 to 0.908), respectively. All patients with pulmonary exacerbations required IV antibiotic therapy; additionally, 8 patients in the VNZ–TEZ–D-IVA group and 17 patients in the ELX-TEZ-IVA group required hospitalization.

In the SKYLINE 103 trial FAS, 19 patients (6.7%) and 17 patients (5.9%) in the VNZ–TEZ–D-IVA group and ELX-TEZ-IVA group, respectively, had pulmonary exacerbations requiring hospitalization or IV antibiotic therapy. The probability of event-free survival (Kaplan-Meier estimate) in the 2 groups at 24 weeks was 0.961 (95% CI, 0.930 to 0.978) and 0.962 (95% CI, 0.932 to 0.979), respectively; at 52 weeks, it was 0.931 (95% CI, 0.894 to 0.956) and 0.940 (95% CI, 0.906 to 0.962), respectively. All patients with pulmonary exacerbations required IV antibiotic therapy; additionally, 14 patients in the VNZ–TEZ–D-IVA group and 9 patients in the ELX-TEZ-IVA group required hospitalization.

In cohort B1 of the RIDGELINE trial, 1 patient (1.3%) required both hospitalization and IV antibiotic therapy, for an observed event rate per year of 0.03.

Percent Predicted Forced Expiratory Volume in 1 Second

The primary outcome of ppFEV₁ was evaluated in the FAS through week 24 in the SKYLINE 102 (n = 196 in the VNZ–TEZ–D-IVA group and n = 202 in the ELX-TEZ-IVA group) and SKYLINE 103 trials (n = 284 in the VNZ–TEZ–D-IVA group and n = 289 in the ELX-TEZ-IVA group). In the mixed model for repeated measures (MMRM) analysis of noninferiority comparing VNZ–TEZ–D-IVA with ELX-TEZ-IVA, the least squares (LS) mean treatment difference was 0.2% (95% CI, −0.7% to 1.1%; 1-sided P < 0.0001) in the SKYLINE 102 trial and 0.2% (95% CI, −0.5% to 0.9%; 1-sided P for noninferiority < 0.0001) in the SKYLINE 103 trial. The prespecified noninferiority margin was −3.0%; ergo, the results of both studies met the preplanned primary analysis of noninferiority for ppFEV₁ because neither 95% CI crossed the noninferiority margin.

In cohort B1 of the RIDGELINE trial (n = 78), patients treated with VNZ–TEZ–D-IVA showed similar ppFEV₁ values compared with baseline use of ELX-TEZ-IVA, based on a within-group LS mean absolute change from baseline to 24 weeks of 0% (95% CI, −2.0% to 1.9%).

The secondary end point of ppFEV₁ through 52 weeks showed similar results in both studies, consistent with the 24-week primary analyses. In the SKYLINE 102 trial, the LS mean difference in absolute change from baseline was 0.1% (95% CI, −0.8% to 1.0%); in the SKYLINE 103 trial, it was 0.3% (95% CI, −0.4 to 1.0).

In the SKYLINE 102 and SKYLINE 103 trials, a supplementary sensitivity analysis was conducted using an alternative estimand where intercurrent events were addressed using the hypothetical strategy for the 24-week outcome of ppFEV₁, and the results were consistent with the primary analyses.

Subgroup analyses were conducted on the primary outcome at 24 weeks in the SKYLINE 102 and SKYLINE 103 trials based on age (aged younger than 18 years versus at least 18 years at screening) and ppFEV₁ at baseline (less than 70% versus at least 70%). Most of the results fell within the noninferiority margin of −3.0% and were similar to the primary analysis results. However, in the SKYLINE 102 trial in the subgroup of patients younger than age 18 years at screening, the 95% CI crossed this threshold, with an LS mean difference of −1.3% (95% CI, −3.9% to 1.3%). In the SKYLINE 103 trial, the value for this subgroup was 0.9% (95% CI, −1.2% to 3.0%).

No subgroup or sensitivity analyses were conducted in the RIDGELINE trial or for the week 52 outcomes in the SKYLINE studies.

Body Mass Index

In the RIDGELINE trial, the absolute change in BMI at week 24 was an LS mean of 0.22 kg/m² (95% CI, 0.05 kg/m² to 0.38 kg/m²).

In the SKYLINE 102 trial, the LS mean change from baseline in BMI in the VNZ–TEZ–D-IVA group (n = 179) and ELX-TEZ-IVA group (n = 187) at week 52 was 0.30 kg/m² (standard error [SE] = 0.10 kg/m²) and 0.08 kg/m² (SE = 0.10 kg/m²), respectively, with an LS mean difference of 0.22 kg/m² (95% CI, −0.05 kg/m² to 0.49 kg/m²).

In the SKYLINE 103 trial, the LS mean change from baseline in BMI in the VNZ–TEZ–D-IVA group (n = 248) and ELX-TEZ-IVA group (n = 266) at week 52 was 0.37 kg/m² (SE = 0.08 kg/m²) and 0.22 kg/m² (SE = 0.07 kg/m²), respectively, with an LS mean difference of 0.16 kg/m² (95% CI, −0.05 kg/m² to 0.36 kg/m²).

No subgroup or sensitivity analyses were reported.

BMI Z Score

In the RIDGELINE trial, the absolute change in BMI z score at week 24 was an LS mean of −0.05 (SE = 0.03; 95% CI, −0.12 to 0.02).

In the SKYLINE 102 trial at 52 weeks, the LS mean change from baseline in BMI z score among patients aged 20 years or younger at baseline was 0.25 (SE = 0.10) in the VNZ–TEZ–D-IVA group (n = 27) and −0.09 (SE = 0.09) in the ELX-TEZ-IVA group (n = 32), with an LS mean difference of 0.34 (95% CI, 0.07 to 0.61).

In the SKYLINE 103 trial, the LS mean change from baseline in BMI z score among patients aged 20 years or younger at baseline was 0.11 (SE = 0.06) in the VNZ–TEZ–D-IVA group (n = 42) and −0.13 (SE = 0.07) in the ELX-TEZ-IVA group (n = 35), with an LS mean difference of 0.24 (0.06 to 0.42) at 52 weeks.

No subgroup or sensitivity analyses were reported.

Cystic Fibrosis Questionnaire-Revised Respiratory Domain

Health-related quality of life (HRQoL) was measured using the Cystic Fibrosis Questionnaire-Revised (CFQ-R) respiratory domain.

In the SKYLINE 102 trial, the LS mean change in the CFQ-R respiratory domain score through week 24 was 0.5 points (SE = 1.1) in the VNZ–TEZ–D-IVA group (n = 186) and −1.7 points (SE = 1.0) in the ELX-TEZ-IVA group (n = 192). The LS mean difference was 2.3 points (95% CI, −0.6 to 5.2).

In the SKYLINE 103 trial, the LS mean change from baseline in the CFQ-R respiratory domain was −1.2 points (SE = 0.8) in the VNZ–TEZ–D-IVA group (n = 268) and the same value, i.e., −1.2 points (SE = 0.8) in the ELX-TEZ-IVA group (n = 270). The LS mean difference was −0.1 points (95% CI, −2.3 to 2.1).

Patients treated with VNZ–TEZ–D-IVA in cohort B1 of the RIDGELINE trial had a within-group LS mean absolute change from baseline through week 24 of 3.9 points (95% CI, 1.5 to 6.3).

Through week 52, the LS mean change in the SKYLINE 103 trial in the VNZ–TEZ–D-IVA group (n = 186) and ELX-TEZ-IVA group (n = 192) was 0.8 points (SE = 1.0) and −1.6 points (SE = 1.0), respectively, with an LS mean difference of 2.4 points (95% CI, −0.3 to 5.1).

In the SKYLINE 103 trial, the LS mean change in the VNZ–TEZ–D-IVA group (n = 268) and ELX-TEZ-IVA group (n = 270) through 52 weeks was −0.4 points (SE = 0.7) and −1.0 point (SE = 0.7), respectively, with an LS mean difference of 0.7 points (95% CI, −1.4 to 2.7).

No subgroup or sensitivity analyses were reported.

Sweat Chloride

In the SKYLINE 102 trial, the LS mean absolute change from baseline for SwCl through week 24 in the FAS was −7.5 mmol/L (standard deviation [SD] = 0.8 mmol/L) in the VNZ–TEZ–D-IVA group (n = 196) compared with 0.9 mmol/L (SD = 0.8 mmol/L) in the ELX-TEZ-IVA group (n = 202). The MMRM analysis of this secondary end point reported a reduction (i.e., improvement) in SwCl associated with VNZ–TEZ–D-IVA treatment compared with ELX-TEZ-IVA treatment, with an LS mean difference of −8.4 mmol/L (95% CI, −10.5 mmol/L to −6.3 mmol/L; P < 0.0001).

In the SKYLINE 103 trial, the LS mean absolute change from baseline in SwCl through week 24 (FAS; n = 294 and 289, respectively) was −5.1 mmol/L (SD = 0.7 mmol/L) in the VNZ–TEZ–D-IVA group compared with −2.3 mmol/L (SD = 0.7 mmol/L) in the ELX-TEZ-IVA group. The MMRM analysis reported a reduction (i.e., improvement) in SwCl associated with VNZ–TEZ–D-IVA treatment compared with ELX-TEZ-IVA treatment, with an LS mean difference of −2.8 mmol/L (95% CI, −4.7 mmol/L to −0.9 mmol/L; P = 0.0034).

In the single-arm study, RIDGELINE, the within-group LS mean absolute change from baseline through week 24 (averaging weeks 16 and 24) was −8.6 mmol/L (95% CI, −11.0 mmol/L to −6.3 mmol/L) in the cohort B1 FAS (n = 77).

In the SKYLINE 102 trial, the LS mean absolute change from baseline in SwCl through week 52 in the FAS was −7.5 mmol/L (SD = 0.7 mmol/L) in the VNZ–TEZ–D-IVA group (n = 196) compared with 0.5 mmol/L (SD = 0.7 mmol/L) in the ELX-TEZ-IVA group (n = 202). The MMRM analysis reported a reduction (i.e., improvement) in SwCl associated with VNZ–TEZ–D-IVA treatment with an LS mean difference of −8.0 mmol/L (95% CI, −9.9 mmol/L to −6.1 mmol/L).

In the SKYLINE 103 trial, the LS mean change in the VNZ–TEZ–D-IVA group (n = 284) was −5.0 mmol/L (SD = 0.6 mmol/L) and, in the ELX-TEZ-IVA group, it was −2.2 mmol/L (SD = 0.6 mmol/L). The MMRM analysis reported a reduction (i.e., improvement) in SwCl associated with VNZ–TEZ–D-IVA treatment, with an LS mean difference of −2.8 mmol/L (95% CI, −4.6 mmol/L to −1.0 mmol/L).

No subgroup or sensitivity analyses were reported.

Proportion of Patients With an SwCl Value of 30 mmol/L or Lower

The 24-week pooled outcome, including data from both the SKYLINE 102 and SKYLINE 103 studies, reported an odds ratio of 2.87 (95% CI, 2.00 to 4.12; P < 0.0001), indicating patients treated with VNZ–TEZ–D-IVA were more likely to have an SwCl of 30 mmol/L or lower at week 24 than patients treated with ELX-TEX-IVA.

In the RIDGELINE trial, 41 out of 78 patients (52.6%) had an SwCl of 30 mmol/L or lower at week 24 (95% CI, 40.9% to 64.0%).

The 52-week result in the SKYLINE 102 trial was an odds ratio of 5.77 (95% CI, 3.33 to 9.99) and in the SKYLINE 103 trial was 1.98 (95% CI, 1.36 to 2.88). While the magnitude of difference between the treatment arms appears to vary between the 2 studies, the trend was in the same direction and aligned with the 24-week outcome results reported in these studies.

No subgroup or sensitivity analyses were reported.

Harms Results

To contextualize the safety data from both the SKYLINE 102 and SKYLINE 103 studies, it is important to note that the majority of patients had received commercial ELX-TEZ-IVA before study enrolment, with a median exposure of approximately 2 years, and all patients received ELX-TEZ-IVA for 4 weeks during the run-in period. Additionally, patients who had a prior history of intolerance to ELX-TEZ-IVA were not eligible to enrol, and patients who developed intolerance to ELX-TEZ-IVA during the run-in period were discontinued from the study. Therefore, the safety data from the ELX-TEZ-IVA group are reflective of the experience in patients who have tolerated ELX-TEZ-IVA and who have continued on an existing treatment regimen, and the safety data from the VNZ–TEZ–D-IVA group are reflective of patients who initiated a new treatment regimen.

Similarly, in the RIDGELINE trial, all patients either had prior experience on ELX-TEZ-IVA or participated in a run-in period in which they received ELX-TEZ-IVA and then switched to VNZ–TEZ–D-IVA during the study period.

Overall Adverse Events

In the SKYLINE 102 and SKYLINE 103 trials, 94.4% and 96.5% of patients, respectively, who were treated with VNZ–TEZ–D-IVA experienced an AE, while 97% and 94.5% of patients, respectively, who were treated with ELX-TEZ-IVA experienced an AE. In both studies, the most common AEs were infective pulmonary exacerbations related to CF (range, 26.8% to 34.7% across the treatment groups in both studies), COVID-19 (range, 20.4% to 26.7%), cough (range, 20.3% to 23.0%), and nasopharyngitis (range, 17.3% to 23.0%).

In the RIDGELINE trial, the proportion of patients with at least 1 AE was 70.6% and 96.2% in cohorts A1 and B1, respectively.

Serious Adverse Events

In the SKYLINE 102 and SKYLINE 103 trials, among patients treated with VNZ–TEZ–D-IVA, 14.3% and 14.1%, respectively, experienced a serious adverse event (SAE), while 20.3% and 13.8% of patients, respectively, who were treated with ELX-TEZ-IVA experienced an SAE. The proportion of patients with any SAE appears elevated in the ELX-TEZ-IVA group of the SKYLINE 102 trial relative to the VNZ–TEZ–D-IVA group, but this was not the case in the SKYLINE 103 trial, where the values were similar across the 2 treatment groups. One life-threatening AE occurred, which was an infective pulmonary exacerbation of CF in a patient in the ELX-TEZ-IVA treatment group of the SKYLINE 103 trial.

In the 2 RCTs, the most common SAE was an infective pulmonary exacerbation of CF. In the SKYLINE 102 trial, this occurred in 5.6% of patients treated with VNZ–TEZ–D-IVA and 11.4% treated with ELX-TEZ-IVA. In the SKYLINE 103 trial, this occurred in 6.3% and 4.2% of patients, respectively. Other SAEs were individually uncommon (from 0 to approximately 2% of patients for each SAE) but included influenza, hemoptysis, pneumonia, suicidal ideation, syncope, COVID-19, alanine aminotransferase (ALT) increased, ALT or aspartate aminotransferase (AST) increased, constipation, distal intestinal obstruction syndrome, gamma-glutamyl transferase increased, cholelithiasis, and nephrolithiasis.

In the RIDGELINE trial, no patients in cohort A1 had an SAE, and 6 patients (7.7%) in cohort B1 had an SAE. In cohort B1, the SAEs included infective pulmonary exacerbations (n = 2); among these 2 patients, 1 also had an SAE of failure to thrive. Other SAEs that each occurred in 1 patient (1.3%) were adenovirus infection, constipation, pulmonary function test decreased, and cough. No life-threatening AEs occurred.

Withdrawals Due to Adverse Events

In the SKYLINE 102 trial, 2.0% of patients treated with VNZ–TEZ–D-IVA and 4.5% of patients treated with ELX-TEZ-IVA had an AE leading to treatment discontinuation. In the SKYLINE 103 trial, this occurred in 4.9% and 3.1% of patients, respectively. The AEs leading to treatment discontinuation were each individually uncommon (0% to approximately 2% of patients for each AE) and included a variety of types of AEs.

In the RIDGELINE trial, 1 patient (1.3%) had treatment interrupted due to a seizure, and 1 patient stopped treatment due to AEs of fatigue and coughing. Both patients belonged to cohort B1.

Mortality

No deaths occurred in the SKYLINE 102, SKYLINE 103, or RIDGELINE trials.

Notable Harms

Identified AEs of special interest included elevated aminotransferase levels, rash, elevated creatine kinase levels, cataracts, hypoglycemia, and neuropsychiatric events.

In the VNZ–TEZ–D-IVA and ELX-TEZ-IVA groups of the SKYLINE 102 trial, the proportion of patients with AEs of elevated aminotransferase levels was 8.2% and 6.4%, respectively. Rash occurred in 8.7% and 7.4%, elevated creatinine occurred in 9.2% and 11.9%, hypoglycemia occurred in 2.6% and 3.5%, and neuropsychiatric AEs occurred in 8.7% and 13.9%, respectively. No patients in either treatment group had AEs of █████████. No patients treated with VNZ–TEZ–D-IVA had serious ████████ ████████████, rash, creatine kinase elevation, ████████, or hypoglycemia events. Serious neuropsychiatric events occurred in 3 patients (1.5%) treated with VNZ–TEZ–D-IVA and in 2 patients (1.0%) treated with ELX-TEZ-IVA.

In the VNZ–TEZ–D-IVA and ELX-TEZ-IVA groups of the SKYLINE 103 trial, the proportion of patients with AEs of elevated aminotransferase levels was 9.5% and 7.6%, respectively. Rash occurred in 12.7% and 8.0%, elevated creatine kinase levels occurred in 8.8% and 5.9%, cataracts occurred in ████ ███ ████; hypoglycemia occurred in 1.1% and 3.8%, and neuropsychiatric AEs occurred in 13.4% and 10.7%, respectively. Serious neuropsychiatric events occurred in 1 patient (0.4%) treated with VNZ–TEZ–D-IVA and 2 patients (0.7%) treated with ELX-TEZ-IVA. Serious elevated aminotransferase events occurred in ███ ██████ patients treated with VNZ–TEZ–D-IVA and ████ treated with ELX-TEZ-IVA. Events of serious elevated creatine kinase occurred in 1 patient (0.4%) in the VNZ–TEZ–D-IVA group and in no patients in the ELX-TEZ-IVA group. No patients had serious rash, ████████, or hypoglycemia events.

In cohort A1 of the RIDGELINE trial, there were no events of ALT increased, AST increased, elevated creatine kinase levels, █████████, hypoglycemia, or neuropsychiatric AEs, while 3 patients (17.6%) had rash. In cohort B1, ALT increase occurred in 5.1% of patients, AST increase occurred in 2.6%, rash occurred in 5.1%, elevated creatine kinase levels occurred in 2.6%, cataracts occurred in ████, and neuropsychiatric events occurred in 5.1%; no patients in cohort B1 had hypoglycemia. No patients in either cohort had serious ████████ ████████████, rash, creatine kinase, ████████, hypoglycemia, or neuropsychiatric events.

Critical Appraisal

The SKYLINE 102 and SKYLINE 103 trials were double-blind, parallel-group, randomized studies with appropriate randomization, allocation concealment, blinding, and methodological approaches. Patient disposition and baseline characteristics were well balanced between the study groups and there were relatively few major protocol violations or discontinuations of treatment or study. The primary end point of change in ppFEV₁ at 24 weeks was a formal noninferiority assessment tested at a 1-sided alpha level of 0.025, which was met according to the prespecified noninferiority margin of −3%. This margin is consistent with the statistical approach selected and has been used in previous studies of CF.^7,8 It was considered to be appropriate by the clinical experts consulted by CDA-AMC, who noted that a between-group difference of approximately 5% in ppFEV₁ may be the minimal important difference (MID), although there is no published consensus. Other outcomes were assessed using 2-sided alpha levels of 0.05. All missing data were assumed to be missing at random and were not imputed, and there was a generally low number of patients (less than 10%) with missing data, and this was well balanced between the treatment groups for each end point as of the 52-week outcomes in both studies. Given the expected similarity in the treatment efficacy and safety profiles, this aligned with expectations, albeit there was a slight imbalance in discontinuations due to AEs in the SKYLINE 102 trial. The 52-week end points assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) were not included in the hierarchical testing procedure and were not adjusted for multiplicity. This may raise the risk of bias, but was not considered to be particularly significant in the context of these noninferiority studies.

The RIDGELINE study was a single-arm, noncomparative, nonrandomized, open-label study that assessed outcomes similar to the SKYLINE 102 and SKYLINE 103 studies in terms of within-group change from baseline after 24 weeks of therapy with VNZ–TEZ–D-IVA. Because this is a noncomparative, nonrandomized study, the risk of bias is high and the degree of certainty in the results in comparison with ELX-TEZ-IVA is very low. This is especially the case for the self-reported (or caregiver-reported) HRQoL metric of the CFQ-R respiratory domain.

The SKYLINE 102 and SKYLINE 103 trials reflect the patient population aged 12 years and older with CF, while patients aged 6 to 11 years (inclusive) were recruited for cohorts A1 and B1 of the RIDGELINE study. The eligibility criteria and baseline characteristics were considered appropriate and reflective of the patient population with CF in Canada within those age groups, and appropriate and reflective of the patients expected to be eligible for treatment with VNZ–TEZ–D-IVA. The comparator, treatment setting, concomitant medications, and patient eligibility criteria were relevant and generalizable to the treatment setting in Canada. Because all of the included patients had treatment experience with ELX-TEZ-IVA and/or participated in run-in periods with ELX-TEZ-IVA treatment, and patients intolerant to or ineligible for ELX-TEZ-IVA were not included, the efficacy and safety data during the treatment period of the studies reflect the experience of patients switching from one CFTRm to another (if assigned to VNZ–TEZ–D-IVA), or reflect the experience of patients continuing on a therapy to which they have demonstrated tolerance (if assigned to ELX-TEZ-IVA in the 2 RCTs). This represents a gap in the generalizability of these studies because no data were captured on patients naive to CFTRm therapy who were beginning treatment with VNZ–TEZ–D-IVA. However, this is expected to be a minority of patients.

GRADE Summary of Findings and Certainty of the Evidence

The selection of outcomes for GRADE assessment was based on the sponsor’s Summary of Clinical Evidence, consultation with clinical experts, and input received from patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with expert committee members:

• number of pulmonary exacerbations per year

• change from baseline to 52 weeks in:

  • lung function, as measured by ppFEV₁

  • BMI

  • BMI z score in pediatric patients

  • HRQoL, as measured by CFQ-R respiratory domain score

  • SwCl

• safety outcomes:

  • number of patients with AEs of elevated aminotransferase levels

  • number of patients with SAEs.

Table 2: Summary of Findings for VNZ–TEZ–D-IVA Versus ELX-TEZ-IVA for Patients Aged 12 Years or Older With CF

AE = adverse event; BMI = body mass index; CDA-AMC = Canada’s Drug Agency; CF = cystic fibrosis; CFQ-R = Cystic Fibrosis Questionnaire-Revised; CI = confidence interval; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; LS = least squares; MID = minimal important difference; NR = not reported; ppFEV₁ = percent predicted forced expiratory volume in 1 second; RCT = randomized controlled trial; SAE = serious adverse event; SwCl = sweat chloride; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

Note: Study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the table footnotes.

^aRated down 1 level for imprecision due to the small number of events. The null was used as the threshold for potential benefit or harms in the absence of a literature- or expert-informed MID, but this end point was not rated down an additional level because the narrow 95% CIs for the absolute between-group differences were judged by CDA-AMC to be unlikely to include important benefits or harms.

^bThe prespecified noninferiority margin for ppFEV₁ was 3%.

^cRated down 1 level for imprecision. The null was used as the threshold for potential benefit or harms in the absence of a literature- or expert-informed MID. This efficacy outcome was also exploratory and not included in the primary or secondary estimand, increasing uncertainty in interpretation.

^dRated down 1 level for imprecision. The null was used as the threshold for potential benefit or harms in the absence of a literature- or expert-informed MID. This efficacy outcome was also exploratory and not included in the primary or secondary estimand, increasing uncertainty in interpretation. Additionally, the sample size was small in this subgroup.

^eRated down 1 level for imprecision. The 95% CIs for the LS mean differences in both trials include values below the published MID threshold of 4 points for the CFQ-R respiratory domain. Although both the point estimates are below the MID, the upper 95% CI of the SKYLINE 102 trial is greater than the MID, which was not observed in the SKYLINE 103 trial. The outcome was rated down by only 1 level because the observed effects are consistent with the hypothesis of noninferiority between the treatments and do not suggest a meaningful difference in harm or benefit.

^fRated down 1 level for imprecision due to the lack of an established MID. Notably, the clinical importance of further reduction in SwCl (beyond that seen with ELX-TEZ-IVA) is unclear because the surrogacy of SwCl for clinical end points has not been established and was described by the clinical experts as being unreliable, the magnitude of reduction in SwCl was not considered to be meaningful by clinicians because a patient may experience greater changes in SwCl day to day, and the reduction in SwCl observed in these studies does not appear to correlate with significant improvement in any other outcomes on this timescale. If further reduction in SwCl is clinically important, it may take much longer studies to elucidate this relationship.

^gRated down 1 level for imprecision. The null was used as the threshold for potential benefit or harms in the absence of a literature- or expert-informed MID. Additionally, the small number of events increases uncertainty in the interpretation.

Sources: Details included in the table are from the sponsor’s Summary of Clinical Evidence and the submitted Clinical Study Reports.^9,10

Table 3: Summary of Findings for VNZ–TEZ–D-IVA Versus ELX-TEZ-IVA for Patients Aged 6 to 11 Years With CF

AE = adverse event; ALT = alanine aminotransferase; AST = aspartate aminotransferase; BMI = body mass index; CF = cystic fibrosis; CFQ-R = Cystic Fibrosis Questionnaire-Revised; CI = confidence interval; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; LS = least squares; MID = minimal important difference; NR = not reported; ppFEV₁ = percent predicted forced expiratory volume in 1 second; RCT = randomized controlled trial; SAE = serious adverse event; SwCl = sweat chloride; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

Note: Study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the table footnotes.

^aRated as very low certainty due to serious concerns about study design and imprecision. The evidence is derived from a single-arm noncomparative study, which limits the ability to attribute observed changes to the intervention and introduces a high risk of bias.

^bNo change from baseline was reported.

^cThe LS mean change from baseline was 3.9 points (95% CI, 1.5 to 6.3), just below the published MID target threshold of 4 points. Although the upper bound of the CI exceeds the MID, the lack of a control group and the proximity of the point estimate to the threshold preclude rating up for magnitude of effect. Additionally, the CFQ-R is a self-reported (or caregiver-reported) outcome measure, which contributes to uncertainty and potential bias in the context of an unblinded, single-arm study.

^dThe clinical importance of further reduction in SwCl (beyond that seen with ELX-TEZ-IVA) is unclear for several reasons; the surrogacy of SwCl for clinical end points has not been established and was described by the clinical experts as being unreliable, the magnitude of reduction in SwCl was not considered by the clinicians to be meaningful because a patient may experience greater changes in SwCl day to day, and the reduction in SwCl observed in these studies does not appear to correlate with significant improvement in any other outcomes on this timescale. If further reduction in SwCl is clinically important, it may take much longer studies to elucidate this relationship.

Sources: Details included in the table are from the sponsor’s Summary of Clinical Evidence and the submitted Clinical Study Report.¹¹

Long-Term Extension Studies

No long-term extension studies (with results) were submitted, although there are some presently ongoing for which results data are not yet available.

Indirect Comparisons

The sponsor submitted indirect comparisons between VNZ–TEZ–D-IVA and best supportive care; given the presence of direct evidence against ELX-TEZ-IVA, which is the standard of care, as previously discussed, this indirect comparison with supportive care was considered by CDA-AMC to have limited relevance; ergo, it was not summarized or critically appraised for the purpose of this review.

Studies Addressing Gaps in the Evidence

Four nonclinical studies evaluating the effect of VNZ–TEZ–D-IVA on cells in an in vitro setting were provided. For CFTR mutations with low prevalence, a conventional clinical trial approach to demonstrate efficacy is not always feasible.^2,12 To evaluate the potential clinical benefit of VNZ–TEZ–D-IVA in these circumstances, additional mutations were identified using the following:

• the in vitro Fischer rat thyroid (FRT) system (with a threshold of a 10% increase in chloride transport over baseline when expressed as a percentage of normal)

• the in vitro human bronchial epithelial primary cell culture model

• mechanistic data.¹³

There is precedent for using in vitro data in the approval of other CFTRm therapies in Canada (namely, ELX-TEZ-IVA) and other countries. A total of 303 CFTR mutations were proposed by the sponsor to be responsive to VNZ–TEZ–D-IVA. Based on the Health Canada Notice of Compliance (NOC) for VNZ–TEZ–D-IVA issued on July 21, 2025, 266 CFTR mutations were approved as part of the indication, of which 23 were based on clinical data, 227 were based on in vitro data, and 16 were based on extrapolation. The remaining CFTR mutations that were submitted by the sponsor were not included in the product monograph because they were excluded by Health Canada. Mutations were excluded by Health Canada if they were listed as not causing CF in both the CFTR2 and CFTR-France databases or if they were listed as not causing CF in 1 database and listed as variants of unknown significance or only causing CFTR-related disease in the other database and/or other sources.

Conclusions

Two noninferiority RCTs demonstrated with high-certainty evidence that VNZ–TEZ–D-IVA results in little to no difference in the change in lung function (measured using ppFEV₁) when compared with ELX-TEZ-IVA through 52 weeks of treatment in patients aged at least 12 years with CF who had at least 1 F508del mutation or another responsive CFTR mutation. Moderate- to high-certainty evidence was also demonstrated for little to no difference between the treatment groups for outcomes related to the rate of pulmonary exacerbations, BMI, BMI z score in patients aged 20 years or younger, HRQoL measured using the CFQ-R respiratory domain, the number of patients who experienced events of elevated aminotransferase levels, and the number of patients with SAEs.

One single-arm study of VNZ–TEZ–D-IVA in patients aged 6 to 11 years with CF and ELX-TEX-IVA–responsive CFTR mutation(s) reported little change from baseline in the outcomes of ppFEV₁, BMI, BMI z score, and CFQ-R respiratory domain, although the certainty of the evidence compared with ELX-TEZ-IVA or any comparator was very low owing to the noncomparative design of the study. Nonetheless, the efficacy and safety profile appeared to be consistent with that observed in the 2 RCTs and it was supportive of the extrapolation of the RCT results in this age group, in combination with the well-established precedent of ELX-TEZ-IVA treatment benefit in this age group and the shared mechanism of action between VNZ–TEZ–D-IVA and ELX-TEZ-IVA. The side effect profile was consistent with what was expected for triple-combination CFTRm therapies, and the rate of discontinuation due to AEs was very low.

There is precedent and support from the clinical experts for the practice of using in vitro studies to address gaps in the clinical study evidence related to rare CF-causing mutations for which it is infeasible to conduct standard clinical trials. Four nonclinical studies were provided evaluating the effect of VNZ–TEZ–D-IVA on cells in an in vitro setting. Based on the Health Canada NOC for VNZ–TEZ–D-IVA, 266 CFTR mutations were approved as part of the indication based on clinical data, in vitro data, and extrapolation. This represents an expansion from the 153 CFTR mutations approved for ELX-TEZ-IVA in Canada.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of VNZ–TEZ–D-IVA (Alyftrek) in the treatment of patients with CF who have a F508del mutation or another responsive CFTR mutation who are aged 6 years and older.

Disease Background

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the review team.

CF, an autosomal recessive condition, is the most common fatal genetic disease affecting children and young adults in Canada. It is caused by mutations in the CFTR gene, which is located on chromosome 7. The CFTR gene encodes a chloride channel that regulates ion and fluid transport across cell membranes. When the CFTR gene is dysfunctional, secretions become tenacious and sticky, resulting in pathology in multiple organs, including the lungs, large and small intestines, pancreatic and bile ducts, and the vas deferens. F508del is the most common CFTR mutation that results in CF.¹ The Canadian Cystic Fibrosis Registry reported there were 4,513 people in Canada living with CF in 2023. Of these, 87.8% carried 1 or more F508del mutations.²

More than 2,000 CFTR mutations have been identified among patients with CF.² The mutations are classified as impaired biosynthesis (class I), defective protein maturation and accelerated degradation (class II), defective regulation of the CFTR gene at the plasma membrane (class III), defective chloride conductance (class IV), diminished transcription of the CFTR gene (class V), and accelerated turnover at the cell surface (class VI).¹⁴ CFTR mutations within classes I to III are associated with severe CF because they are considered nonfunctional, while mutations in classes IV to VI may retain CFTR function.^14,15 Genotyping for mutations in the CFTR gene is performed routinely on almost all patients with CF in Canada and is also part of the newborn screening process.¹

CF results in airway obstruction, chronic endobronchial infection, and inflammation, which ultimately lead to destruction of lung tissue through the development of bronchiectasis and loss of lung function.³ Although chronic pulmonary therapies instituted early in the disease have reduced the decline in lung function over time, patients with a genotype that is homozygous for the F508del mutation will develop chronic infection with Pseudomonas and progressive bronchiectasis and airway obstruction. In a cohort of approximately 1,000 healthy young children with CF who did not have a Pseudomonas infection at enrolment, there was a greater annual decline in forced expiratory volume in 1 second (FEV₁) over the following 4 years in those patients with a genotype that was homozygous for the F508del mutation.¹⁶ Chronic endobronchial infection of the airways with bacterial pathogens, such as Pseudomonas aeruginosa,² is associated with a more rapid loss of lung function.⁴ Acute or chronic endobronchial infections result in further destruction of lung tissue and are associated with respiratory morbidity. Lung disease accounts for the vast majority (> 80%) of deaths in patients with CF.^1,5

Pulmonary exacerbations are associated with lung function decline and mortality and may require treatment with IV antibiotics and hospitalization. The Cystic Fibrosis Foundation has reported that approximately one-third of patients with CF will have at least 1 pulmonary exacerbation per year requiring IV antibiotics.¹⁷

Maintenance of pulmonary function (higher FEV₁) and fewer respiratory exacerbations are associated with increased survival.¹⁸ Pulmonary management of CF therefore aims to clear the airways of secretions and treat lung pathogens to minimize inflammation.

Patients typically have pancreatic, gastrointestinal, and nutritional disease as well as progressive pulmonary damage. Gastrointestinal and pancreatic involvement results in pancreatic exocrine insufficiency in most individuals with CF, causing malabsorption of fats and fat-soluble vitamins, which leads to malnutrition. Maintaining adequate nutrition is associated with an improved clinical outcome and longevity for patients with CF.⁶ Almost all of these patients, approximately 90% according to the clinical experts, will have pancreatic insufficiency and will need to take lifelong pancreatic enzyme replacement with every meal as well as fat-soluble vitamin therapy. With increasing age, these patients may develop CF-related diabetes and require therapy with insulin. In 2023, CF-related diabetes was reported in 32.5% of adults and 2.4% of children living with CF in Canada.²

The median age of survival in Canada for a child born with CF was estimated to be 62.3 years in 2023.² The Canadian Cystic Fibrosis Registry has reported an increase in the median age of death for patients with CF in Canada since the year 2000;¹ The median age of death was 41.2 years in 2023² compared with 38.7 years in 2021 and 26.6 years in 2002.¹⁹

Standards of Therapy

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the review team.

Before the advent of CFTRm therapies, the key goals of CF management were to treat the complications of CF and relieve symptoms because there were no therapies available that could treat the fundamental CF-causing defects.²⁰ Since CFTRm therapies became available, they have become the standard of care for eligible patients,^21,22 and the treatment goals now include maintaining normal lung function, reducing or possibly eliminating CF-related symptoms, reducing or eliminating hospitalizations or pulmonary exacerbations, and maintaining adequate nutrition; additionally, in the long term, the clinical experts consulted by CDA-AMC noted they expect a reduction in future lung transplants related to CF, and that the goal is to push closer to normal life expectancy and better HRQoL in people with CF. Although CFTRm therapies are not curative, they are the first and only class of therapy developed to target the underlying pathophysiology of CF by working to improve the function of the CFTR protein.²² The first CFTRm available commercially was ivacaftor (Kalydeco), which works to target defects in the “gating” of CFTR channels, and is currently indicated by Health Canada for 10 CF-causative mutations in patients aged as young as 2 months.²² Two combination therapies with ivacaftor, lumacaftor-ivacaftor (Orkambi) and tezacaftor-ivacaftor (Symdeko), were later developed for the treatment of patients with 2 copies of the most common CF-causing variant, F508del, which represents approximately 50% of patients.²² In contrast, the relatively newer triple-combination therapy, ELX-TEZ-IVA (Trikafta),^23-27 is indicated for patients aged 2 years and older with at least 1 F508del mutation or a mutation in 1 of the 152 non-F508del rare and responsive mutations, and it has become the current standard of care for most patients with CF in Canada.

There is an unmet need for patients with CF-causing mutations that are not eligible for treatment with ELX-TEZ-IVA, and for those who discontinue therapy due to intolerable side effects. Additionally, ELX-TEZ-IVA is administered twice daily and requires a fat-containing meal for absorption at each dose, which creates challenges for some patients around adherence. The clinical experts consulted by CDA-AMC noted that many patients while on therapy with ELX-TEZ-IVA will still have an SwCl level that is in an abnormal range, and that this represents an unmet need, i.e., for the improvement of SwCl levels into a normal range, although the long-term implications of this are still unclear. The experts also noted that non-F508del mutations are more likely to be diagnosed in racial minorities who may already be encountering systemic disadvantages within the health care system and who may be underrepresented in clinical trials; moreover, many non-F508del mutations are rare enough to preclude generating robust clinical evidence through a typical clinical trial. As a result, there is a disproportionate unmet need among racial minorities because a higher proportion of patients with CF who are racial minorities currently have no access to an approved CFTRm.

Patients receiving a CFTRm still receive symptom-relieving supportive care as needed, although the clinical experts consulted by CDA-AMC noted that treatment with a CFTRm may generally be expected to reduce (but not eliminate) the amount of other supportive care required.

Drug Under Review

VNZ–TEZ–D-IVA (Alyftrek) is a triple-combination CFTRm indicated for the treatment of CF in patients aged 6 years and older who have at least 1 F508del mutation or another responsive mutation in the CFTR gene; refer to Table 23 for the list of mutations acknowledged by Health Canada as responsive to VNZ–TEZ–D-IVA. The reimbursement request aligns with the Health Canada indication.

The sponsor proposed that VNZ–TEZ–D-IVA be reimbursed based on the criteria outlined in Table 4, which were judged by the clinical experts consulted by CDA-AMC to be reasonable and similar to those for ELX-TEZ-IVA.

Table 4: Sponsor’s Requested Reimbursement Criteria for VNZ–TEZ–D-IVA

BMI = body mass index; CF = cystic fibrosis; CFQ-R = Cystic Fibrosis Questionnaire-Revised; CFTRm = CFTR modulator; ppFEV₁ = percent predicted forced expiratory volume in 1 second; VNZ–TEZ–D-IVA vanzacaftor-tezacaftor-deutivacaftor.

Sources: Sponsor submission materials.

VNZ–TEZ–D-IVA is administered by mouth once daily. It is available in 2 different doses. Patients who are less than 40 kg in body weight take 3 tablets per day of vanzacaftor 4 mg, tezacaftor 20 mg, and deutivacaftor 50 mg. Patients who are at least 40 kg in body weight take 2 tablets per day of vanzacaftor 10 mg, tezacaftor 50 mg, and deutivacaftor 125 mg. This treatment must be taken with a fat-containing meal to facilitate absorption.²⁸

Vanzacaftor and tezacaftor are CFTR correctors that bind to different sites on the CFTR protein and have an additive effect in facilitating the cellular processing and trafficking of select mutant forms of the CFTR gene (including F508del) to increase the amount of CFTR protein delivered to the cell surface compared with either molecule alone. Deutivacaftor is a CFTR potentiator that increases the channel-open probability (or gating) of the CFTR protein at the cell surface. The half-life of deutivacaftor allows for once-daily dosing, in contrast with the twice-daily dosing of ivacaftor-containing therapies such as the other existing triple-combination CFTRm, ELX-TEZ-IVA (Trikafta). As with ELX-TEZ-IVA, the combination of multiple CFTR correctors and a CFTR potentiator in VNZ–TEZ–D-IVA is expected to be complementary and to increase both the quantity and function of CFTR protein and CFTR activity relative to monotherapies or 2-drug combination therapies.^29,30

Key characteristics of VNZ–TEZ–D-IVA (Alyftrek) are summarized in Table 5 with other CFTRm-based treatments currently reimbursed in Canada for the treatment of CF.

Table 5: Key Characteristics of VNZ–TEZ–D-IVA (Alyftrek), ELX-TEZ-IVA (Trikafta), and IVA (Kalydeco)

ALT = alanine aminotransferase; AST = aspartate aminotransferase; D-IVA = deutivacaftor; ELX = elexacaftor; IVA = ivacaftor; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; TEZ = tezacaftor; VNZ = vanzacaftor; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

^aHealth Canada–approved indication.

Sources: Product monographs for Alyftrek,²⁸ Trikafta,²⁷ and Kalydeco.³¹

Testing Procedure Considerations

CFTR genetic testing is the standard of care and part of the routine diagnostic work-up to confirm a CF diagnosis. An individual is diagnosed with CF when there is clinical presentation of disease and evidence of CFTR gene dysfunction through genetic testing for CFTR mutations.³⁵ CFTR genetic testing is a staged process. Patients who present with clinical features or a family history indicative of CF or are identified through newborn screening will receive CFTR genetic assay testing.³⁶ Initial genetic testing is used to identify the presence of 2 CF-causing mutations in the CFTR gene (1 inherited from each parent) to confirm a diagnosis of CF.³⁷ The CFTR genetic assay test result is assessed against a panel of the most common CF-causing mutations, including F508del, to determine a potential CFTR mutation match. According to the clinical experts, CFTR mutation panels may vary between jurisdictions but are typically based on mutations that have been identified by the Clinical and Functional Translation of CFTR (CFTR2) project.³⁸ In cases where none of the most common CFTR mutations are identified, only 1 mutation is identified, or no mutations are identified by the initial panel assay, follow-up genetic testing using an extended panel of CFTR mutations or full gene sequencing of the CFTR coding region can be used to confirm the mutation status of an individual.³⁵

The clinical experts noted that standard CFTR gene panels, such as those based on the CFTR2 project, are often optimized for common mutations that are likely to be identified in white individuals, while rare or unclassified CFTR mutations may be more prevalent in racialized individuals.³⁹ Based on the Cystic Fibrosis Canada Patient Registry, 91.7% of the currently identified population in Canada with CF is white, and 87.8% of the total number of patients with CF in Canada possess at least 1 F508del mutation.⁴⁰ This raises important equity considerations about the mutations included in CFTR gene panels, and the potential impacts of the underrepresentation of racialized individuals. The patient group input highlighted that individuals with rare CFTR mutations may face significant challenges related to being diagnosed late with more advanced disease, which may also impact access to standard of care treatments. Initial CFTR genetic testing may be inconclusive if an individual has a rare mutation that has not yet been identified;³⁶ however, according to the clinical experts, additional follow-up testing using full gene sequencing will likely capture patients who may have mutations not included in the panel. As previously mentioned, the Health Canada NOC approved 266 CFTR mutations as part of the indication for VNZ–TEZ–D-IVA, which may not capture individuals with rare or ultra-rare CFTR mutations.

CDA-AMC considered the potential impacts of CFTR genetic testing to confirm eligibility for VNZ–TEZ–D-IVA, including impacts to health systems, patients (including families and caregivers), and costs. No implementation barriers or only minimal barriers are anticipated because CFTR genetic testing is currently performed as the standard of care for a CF diagnosis across jurisdictions in Canada. The review team validated key considerations and the relevant information available from the materials submitted by the sponsor, input from the clinical experts consulted by the review team, patient group input, sources from the literature, and a scan of jurisdictional testing availability, when possible, which are summarized in Table 6.

Table 6: Considerations for CFTR Genetic Testing for Establishing Treatment Eligibility With VNZ–TEZ–D-IVA in CF

CF = cystic fibrosis; CI = confidence interval; NBS = newborn screening; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

^aSource: Sponsor-submitted budget impact assessment.

^bCystic Fibrosis Canada did not provide testing information for Nunavut.

^cCanada’s Drug Agency has not evaluated or critically appraised this evidence to determine its validity or reliability.

Perspectives of Patients, Clinicians, and Drug Programs

The full patient and clinician group submissions received for this review are available in the consolidated patient and clinician group input document on the project website.

Patient Group Input

This section was prepared by the review team based on the input provided by a patient group.

One patient group, Cystic Fibrosis Canada, submitted input for this review. Information was gathered through focus groups with people in Canada with CF and their caregivers. Other sources of data included surveys (including a 2021 survey with more than 1,200 responses), data from the Canadian Cystic Fibrosis Registry, and a multiphase burden of disease study and other relevant publications.

The patient group highlighted that patients not on a CFTRm frequently have periods of infection and acute inflammation called exacerbations that require a hospital stay of at least 2 weeks and that may last up to 4 weeks. Some of these patients could benefit from a CFTRm, but there may not be any CFTRm therapies currently that have a Health Canada indication for their mutation. The group noted that approximately 33% of all adult patients with CF in Canada have CF-related diabetes. Any of the other drugs that patients need to take on a regular basis can also have negative side effects and long-term risks, such as chronic use of antibiotics that may lead to resistance. Moreover, as patients age, they may need to try multiple antibiotics to find 1 that works, making management of drug-to-drug interactions difficult. The patient group reiterated that currently, only CFTRm therapies target the underlying pathophysiology of CF by improving the function of the CFTR gene, while other therapies used to treat CF are for the management of complications and control of symptoms.

The patient group highlighted that too many patients lack access to any CFTRm, including those patients who have ultra-rare mutations that are not on the narrow list of 152 mutations indicated for ELX-TEZ-IVA or who cannot tolerate it, or those who have experienced a response only to VNZ–TEZ–D-IVA. The patient group noted the FDA has approved the use of ELX-TEZ-IVA in 272 mutations, compared with 152 mutations in Canada; the group suggested there are approximately 23 patients in Canada aged 2 years or older with 1 or more of the 119 mutations that are not approved in Canada for treatment with ELX-TEZ-IVA. Thus, they could access ELX-TEZ-IVA in the US, but cannot access it in Canada. Additionally, the FDA approved VNZ–TEZ–D-IVA for an additional 31 unique mutations for which ELX-TEZ-IVA is not approved in Canada; this represents approximately 13 additional people in Canada without access to ELX-TEZ-IVA. The group pointed out that, if VNZ–TEZ–D-IVA gets approved in Canada, up to 36 individuals aged 6 years or older could access it for the first time. The group specifically requested that the committee recommend access to VNZ–TEZ–D-IVA for the 152 CFTR mutations currently indicated in Canada for ELX-TEZ-IVA. The group further asked that the committee recommend access to VNZ–TEZ–D-IVA for patients in Canada with any of the 151 rare mutations or mutation combinations the FDA lists as indications for VNZ–TEZ–D-IVA that currently are not indicated in Canada for any CFTRm, and to recommend the broadest possible access to VNZ–TEZ–D-IVA for individuals with other CFTR mutations when there is evidence, and when more evidence is generated over time (including in vitro laboratory evidence or clinical evidence), that shows these patients will or may experience a response to this therapy. The group highlighted that CF has significant financial implications for patients and their caregivers, health systems, and society.

No patients reported experience with VNZ–TEZ–D-IVA treatment. The patient group highlighted that VNZ–TEZ–D-IVA may improve quality of life and adherence as a 1-pill-a-day treatment, which would provide an option for parents who may have difficulty administering medications to their children, who already take several medicines a day to maintain their health.

Clinician Input

Input From Clinical Experts Consulted for This Review

All CDA-AMC review teams include at least 1 clinical specialist with expertise regarding the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 2 clinical specialists with expertise in the diagnosis and management of CF.

Unmet Needs

The clinical experts described that ELX-TEZ-IVA is the current standard of care in patients with CF aged 6 years and older with eligible CF-causing mutations, but there is an unmet need for more effective therapy that prolongs life, improves quality of life, and reduces the treatment burden of supportive medications in patients who have CF-causing mutations that are not eligible for treatment with ELX-TEZ-IVA, which represents approximately 10% of patients with CF in Canada, and disproportionately represents racialized people. Additionally, even among patients eligible for treatment with ELX-TEZ-IVA, the clinical experts noted that most patients still have SwCl levels in an abnormal range. Doses of ELX-TEZ-IVA are administered twice daily and require fat-containing meals for absorption; ergo, the experts noted that less frequent dosing is an unmet need that might improve adherence. Some patients may experience AEs or intolerance to ELX-TEZ-IVA, requiring dose adjustments or cessation, and there is an unmet need for alternative triple-combination CFTRm therapies in those circumstances. Finally, overall pill and treatment burden remains high in patients with CF, despite reductions after the advent of triple-combination CFTRm therapies (i.e., ELX-TEZ-IVA).

Place in Therapy

The clinical experts consulted by CDA-AMC stated that VNZ–TEZ–D-IVA would occupy a place in therapy similar to that of ELX-TEZ-IVA, with the caveat that it may be applicable to a wider range of mutations, and that the current Health Canada indications begin at different ages (i.e., ELX-TEZ-IVA is currently indicated for patients aged at least 2 years, while VNZ–TEZ–D-IVA is indicated in patients aged at least 6 years). The experts noted that, given the once-daily dosing of VNZ–TEZ–D-IVA compared with ELX-TEZ-IVA, improvements in treatment adherence may be expected. However, patients with CF are typically taking numerous pills daily in addition to CFTRm therapies; ergo, this may not represent a substantial difference in overall treatment or pill burden. Nonetheless, the expert specialized in pediatric care said that a reduction in treatment or pill burden may be particularly significant for the treatment of pediatric patients with CF. The experts noted that it may be premature for every patient currently stable on ELX-TEZ-IVA to be switched to VNZ–TEZ–D-IVA due to the longer clinical safety record of ELX-TEZ-IVA; however, VNZ–TEZ–D-IVA is generally expected to become the new standard of care for eligible patients.

Additionally, if there are clinical signals that achieving a lower SwCl than previously seen with ELX-TEZ-IVA results in less end-organ damage, it may further shift treatment paradigms and reduce the need for other supportive therapies.

Patient Population

The patient population appropriate for treatment with VNZ–TEZ–D-IVA is similar to that for ELX-TEZ-IVA, with the exception that there may be additional CF-causing mutations eligible for treatment with VNZ–TEZ–D-IVA. Aside from genotype, the patient population is expected to be similar.

Assessing the Response to Treatment

The clinical experts noted that the assessment of CF-related treatment is based on lung function spirometry, number of days treated with oral and IV antibiotics for pulmonary exacerbations, number of CF-related hospitalizations, and monitoring of BMI or BMI z score (as appropriate by age) for signs of malnutrition. The clinical experts informed CDA-AMC that the assessments related to VNZ–TEZ–D-IVA should be similar to those expected for ELX-TEZ-IVA, although, in any reimbursement criteria, it is also important to consider that assessments for initiation, continuation, or discontinuation should differ for patients newly starting triple-combination CFTRm therapies (i.e., patients who are generally expected to experience greater improvements with therapy) versus switching between triple-combination CFTRm therapies (i.e., the new treatment would be expected to maintain the benefit previously observed with the other therapy). Assessment of response in the previously listed metrics should generally be performed after a year of treatment because, as the experts highlighted, some metrics (especially pulmonary exacerbations and BMI or BMI z score) may have fluctuations with seasonal illness and short-term instability in pediatric patients that may not be reflective of the effect of the drug.

Discontinuing Treatment

The clinical experts noted that discontinuation of VNZ–TEZ–D-IVA should be considered in the case of intolerable AEs that cannot be sufficiently managed or, in the case of a lung transplant, similar to current practice with ELX-TEZ-IVA (although this does not reflect a contraindication specified by Health Canada).

Prescribing Considerations

The clinical experts noted that a physician practising within or under the directed supervision of a Canadian CF clinic should manage the prescription and renewal of VNZ–TEZ–D-IVA.

Clinician Group Input

This section was prepared by the review team based on the input provided by clinician groups.

One clinician group, the Cystic Fibrosis Canada Health Care Advisory Council, provided input for this review. Information for this submission was gathered through the following: experience gained by working with and delivering medical services to people with CF; consultations with people with CF not eligible for ELX-TEZ-IVA; a review of medical and scientific literature, including clinical trial results and real-world experience; and the Canadian Cystic Fibrosis Registry, a collection of patient data and other information regarding CF care and outcomes. A total of 2 clinicians contributed to the input submitted for this review.

The input from the clinician group was generally consistent with that of the clinical experts consulted for this review, emphasizing that about 10% of patients have mutations not currently approved for treatment with available CFTRm therapies in Canada, and that there are no alternative triple-combination CFTRm treatments for patients with AEs or intolerance to ELX-TEZ-IVA. The group agreed that the once-daily regimen of VNZ–TEZ–D-IVA may improve adherence and convenience compared with the twice-daily regimen of ELX-TEZ-IVA.

The group noted that rare mutations for which there is currently no CFTRm indication are disproportionately found in racialized communities, so expanded access to this therapy may improve equitable access to disease-modifying treatments in Canada.

The clinician group noted that, according to recent Canadian clinical guidelines, all patients with CF who have at least 1 CFTRm-responsive CFTR variant should be treated with a CFTRm. The clinician group noted that patients with CF best suited for the drug under review include patients already on ELX-TEZ-IVA who are experiencing challenges due to side effects or related to adherence to a twice-daily treatment regimen, and patients with rare mutations not approved for treatment with ELX-TEZ-IVA but who may experience a response to VNZ–TEZ–D-IVA. The group agreed with the clinical experts regarding the metrics and yearly time frame of reassessments for treatment efficacy, as well as factors to consider for treatment discontinuation and prescribing.

Drug Program Input

The drug programs provide input on each drug being reviewed through the reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted for this review are summarized in Table 7.

Table 7: Summary of Drug Plan Input and Clinical Expert Response

CF = cystic fibrosis; CFTRm = CFTR modulator; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; ppFEV₁ = percent predicted forced expiratory volume in 1 second; SwCl = sweat chloride; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

Clinical Evidence

The objective of this Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of VNZ–TEZ–D-IVA (Alyftrek) in the treatment of patients with CF who have a F508del mutation or another responsive CFTR mutation and are aged 6 years and older. The focus will be placed on comparing VNZ–TEZ–D-IVA with relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of VNZ–TEZ–D-IVA is presented in 4 sections, with the review team’s critical appraisal of the evidence included at the end of each section. The first section, the systematic review, includes pivotal studies and RCTs that were selected according to the sponsor’s systematic review protocol. The review team’s assessment of the certainty of the evidence in this first section using the GRADE approach follows the critical appraisal of the evidence. The second section typically includes sponsor-submitted long-term extension studies, although there are no completed long-term extension studies at this time. The third section typically includes indirect evidence from the sponsor, but there was no indirect treatment comparison relevant to the review. The fourth section includes additional studies that were considered by the sponsor to address important gaps in the systematic review evidence, which in this case pertain to in vitro assays of potentially responsive non-F508del mutations that may not have (and, due to rarity, may never have) robust clinical trial evidence.

Included Studies

Clinical evidence from the following is included in the review and appraised in this document:

• 2 RCTs and 1 single-arm pivotal study identified in the systematic review

• 4 additional studies addressing potentially responsive non-F508del mutations in an in vitro nonclinical setting.

Systematic Review

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

Description of Studies

Characteristics of the included studies are summarized in Table 8. Two double-blind phase III RCTs, SKYLINE 102 (VX21-121-102, N = 405) and SKYLINE 103 (VX21-121-103; N = 574), and 1 single-arm phase III study, RIDGELINE (VX21-121-105), were included.

Both RCTs compared VNZ–TEZ–D-IVA with ELX-TEZ-IVA in patients with CF aged at least 12 years with ELX-TEZ-IVA–responsive mutations. These were randomized (1:1) studies with a 52-week treatment phase and were similar in study design. Randomization was done using an interactive web response system and was stratified by the patient’s age at screening visit (aged younger than 18 years versus at least 18 years), ppFEV₁ during the run-in period (less than 70% versus at least 70%), SwCl during the run-in period (less than 30 mmol/L versus at least 30 mmol/L), and prior CFTRm use (yes versus no). All patients entered a 28-day run-in period receiving ELX-TEZ-IVA before being randomized in a double-blind fashion to treatment with VNZ–TEZ–D-IVA or ELX-TEZ-IVA for 52 weeks, followed by an additional 28-day open-label follow-up for additional safety data. In the SKYLINE 102 trial, there were no study sites in Canada, and no patients were located in Canada. In the SKYLINE 103 trial, there were 29 patients (4.9%) in Canada across 8 sites in various provinces. In both studies, the majority of patients were located in the US (44.7% and 33.8%, respectively), while the remainder were located in Australia, Europe, Israel, and New Zealand.

In the RIDGELINE trial, patients aged between 1 and 11 years with a mutation responsive to ELX-TEZ-IVA were recruited. The RIDGELINE study included a part A with the objective of evaluating the pharmacokinetics of VNZ–TEZ–D-IVA and relevant metabolites when dosed in therapeutic concentrations, and to evaluate the safety and tolerability of VNZ–TEZ–D-IVA. Part A preceded part B and informed the doses and weight-related dose thresholds for part B, in which the objectives were primarily to evaluate the safety and tolerability of VNZ–TEZ–D-IVA through week 24, and secondarily to evaluate the efficacy through week 24 as well as the pharmacokinetics. Parts A and B were each divided into 3 cohorts by age range, where the first cohort (A1 and B1) included patients aged 6 to 11 years (inclusive), the second (A2 and B2) included patients aged 2 to 5 years (inclusive), and the third (A3 and B3) included patients aged 1 year to younger than 2 years. Therefore, only cohorts A1 (N = 17) and B1 (N = 78) are relevant to the Health Canada indication considered here, and only cohort B1 is associated with results data for efficacy outcomes that are considered in this review, although both cohort A1 and cohort B1 contributed to the safety data that will be assessed.

Patients in cohort A1 of the RIDGELINE trial received VNZ–TEZ–D-IVA once daily for 22 days and were permitted to participate in cohort B1 if they met eligibility criteria. Patients in cohort B1 who were receiving stable ELX-TEZ-IVA treatment had the run-in period waived and entered the treatment period (VNZ–TEZ–D-IVA), which was 24 days, followed by a safety follow-up visit 28 days after the last dose. Patients entering cohort B1 who were not already receiving stable ELX-TEZ-IVA underwent a 28-day run-in period in which they received ELX-TEZ-IVA before the VNZ–TEZ–D-IVA treatment period. Dosing of ELX-TEZ-IVA during the run-in period and dosing of VNZ–TEZ–D-IVA during the treatment period were based on weight categories, as described in more detail in the subsequent Interventions section of this report. No sites or patients in Canada were included. The majority of sites and patients were in the US (100% of cohort A1 and 60.3% of cohort B1), and the remainder were located in Australia or Europe.

Table 8: Details of Studies Included in the Systematic Review

AE = adverse event; BMI = body mass index; CF = cystic fibrosis; CFB = change from baseline; CFQ-R = Cystic Fibrosis Questionnaire-Revised; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; CFTRm = CFTR modulator; F/F = homozygous for F508del; F/G = heterozygous for F508del and a gating mutation; F/MF = heterozygous for F508del and a minimal function mutation; F/other = heterozygous for F508del and a non-F508del mutation that does not fit into any of the other specified mutation categories (i.e., gating, residual function, minimal function, triple combination–responsive); F/RF = heterozygous for F508del and a residual function mutation; GLI = Global Lung Function Initiative; NA = not applicable; ppFEV₁ = percent predicted forced expiratory volume in 1 second; RD = respiratory domain; SwCl = sweat chloride; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

^aThe trial was conducted in patients aged 1 to 11 years and included several cohorts (A1 and B1, A2 and B2, A3 and B3). The part B cohorts will not initiate until results are available from the corresponding part A cohort for the same age group. Cohorts A1 and B1 evaluated patients aged 6 years through 11 years, inclusive, while cohorts A2 and B2 included patients aged 2 to 5 years, and cohorts A3 and B3 included patients aged 1 year to younger than 2 years. Of the cohorts, only A1 and B1 were conducted in the age group evaluated in this study; ergo, the age group 2 and group 3 cohorts are not further described here. The purpose of cohort A1 was to determine doses and weight cut-offs for cohort B1; therefore, for efficacy outcomes, the focus is on cohort B1, while cohort A1 is also included in safety analyses.

^bExclusion and inclusion criteria sourced from clinicaltrials.gov records.

Sources: Clinical Study Reports.^9-11

Populations

Inclusion and Exclusion Criteria

The RCTs, SKYLINE 102 and SKYLINE 103, both included patients aged at least 12 years with a ppFEV₁ value between 40% and 90% of the predicted mean for age, sex, and height for patients receiving ELX-TEZ-IVA, or a ppFEV₁ between 40% and 80% for patients not receiving ELX-TEZ-IVA therapy. Patients recruited to the SKYLINE 102 trial had to be heterozygous for F508del and a minimal function mutation, while patients recruited to the SKYLINE 103 trial had to have a genotype that was either homozygous for F508del mutations, heterozygous for F508del and a gating or residual function mutation, or with at least 1 mutation identified as responsive to ELX-TEZ-IVA with no F508del mutation. Patients had to have stable CF, as determined by the investigator. Inadequate organ function or a recent (within 28 days of the run-in period) pulmonary exacerbation, respiratory infection, or changes in therapy were reasons for exclusion. Prohibited medications included moderate and strong CYP3A inducers or inhibitors (except ciprofloxacin) within 14 days of the first dose in the run-in period and during the study, and CFTRm therapies produced by other drug manufacturers taken within 28 days or 5 terminal half-lives (whichever is longer) before screening and during the study.

The single-arm RIDGELINE study included patients aged 1 to 11 years with stable CF and at least 1 triple combination–responsive mutation (including F508del) in the CFTR gene and ppFEV₁ of at least 60% at screening. Cohorts A1 and B1 captured patients aged 6 to 11 years, inclusive, which is the focus of this review; the younger age groups were evaluated in cohorts A2, B2, A3, and B3, which will not be discussed in depth. Prohibited medications were the same as those listed for the RCTs described previously.

Key exclusion criteria for all trials were patients with a history of solid organ or hematological transplant, hepatic cirrhosis with portal hypertension, moderate hepatic impairment (Child-Pugh score of 7 to 9), severe hepatic impairment (Child-Pugh score of 10 to 15), lung infection with organisms associated with a more rapid decline in pulmonary status, or who were pregnant or breastfeeding.

Interventions

Intervention and Comparators

VNZ–TEZ–D-IVA was orally administered once daily as a fixed-dose combination tablet with a fat-containing meal or snack.

In the double-blind RCTs (SKYLINE 102 and SKYLINE 103), VNZ–TEZ–D-IVA, ELX-TEZ-IVA, ivacaftor, and identical placebos were supplied as tablets of similar size and appearance, and all patients received the same number of tablets per day to maintain blinding; i.e., the ELX-TEZ-IVA regimen included a second dose later in the day of ivacaftor only, so patients assigned to VNZ–TEZ–D-IVA received a matched placebo for that second dose to maintain blinding. The daily dose of VNZ–TEZ–D-IVA was 20 mg of vanzacaftor, 100 mg of tezacaftor, and 250 mg of deutivacaftor in the combination tablet. In the ELX-TEZ-IVA groups, the dose of elexacaftor was 200 mg daily, tezacaftor was 100 mg daily, and ivacaftor was 150 mg every 12 hours (i.e., twice daily for a total of 300 mg per day). There was a run-in period from day −28 to day −1 (after which the treatment period began), during which patients received ELX-TEZ-IVA at the same dose as previously described. Dose modifications were not allowed, although treatment interruptions were allowed or required to resolve toxicity.

In the single-arm open-label RIDGELINE study, because patients were younger (aged 6 to 11 years in cohorts A1 and B1), the dose was weight-dependent. In patients in cohort B1 who were under 30 kg in body weight at screening, the dosage for ELX-TEZ-IVA during the 28-day run-in period was 100 mg of elexacaftor once daily, 50 mg of tezacaftor once daily, and 75 mg of ivacaftor twice daily (i.e., a total of 150 mg of ivacaftor per day); in patients in this cohort who were at least 30 kg in body weight or greater, the dosage of ELX-TEZ-IVA in the 28-day run-in period was 200 mg of elexacaftor once daily, 100 mg of tezacaftor once daily, and 150 mg of ivacaftor twice daily (i.e., 300 mg daily over 2 doses). During the treatment period, the dosage of VNZ–TEZ–D-IVA in patients under 40 kg was 12 mg of vanzacaftor once daily, 60 mg of tezacaftor once daily, and 150 mg of deutivacaftor once daily. The dosage of VNZ–TEZ–D-IVA in patients 40 kg or greater was 20 mg of vanzacaftor once daily, 100 mg of tezacaftor once daily, and 250 mg of deutivacaftor once daily.

Concomitant Medications and Cointerventions

Patients with CF are typically managed with a variety of treatments to improve their symptoms, quality of life, and longevity. Patients could receive other medications during the studies, including CF medications consistent with standards of care (e.g., pancreatic enzymes), although some concomitant medications with potential interactions (such as products that are sensitive substrates of p-glycoprotein, products that are CYP2C9 substrates, or products that are substrates of OATP1B1 or OATP1B3, e.g., statins, among others) required additional caution and monitoring. As described in the inclusion and exclusion criteria listed previously, some medications were prohibited immediately preceding and during the studies if there was potential for them to affect the exposure to study therapies or to confound the study results, such as moderate and strong CYP31 inducers and inhibitors (except ciprofloxacin) or other CFTRm therapies.

Outcomes

A list of efficacy end points assessed in this Clinical Review Report is provided in Table 9, followed by descriptions of the outcome measures. Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence as well as any outcomes identified as important to this review according to the clinical experts consulted for this review, and input from patient and clinician groups and public drug plans. Using the same considerations, the review team selected end points that were considered to be most relevant to inform expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. All summarized efficacy end points were assessed using GRADE. Select notable harms outcomes considered important for informing expert committee deliberations were also assessed using GRADE.

Table 9: Outcomes Summarized From the Studies Included in the Systematic Review

AE = adverse event; BMI = body mass index; CF = cystic fibrosis; CFQ-R = Cystic Fibrosis Questionnaire-Revised; FAS = full analysis set; GRADE = Grading of Recommendations Assessment, Development and Evaluation; PEx = pulmonary exacerbation(s); ppFEV₁ = percent predicted forced expiratory volume in 1 second; SAE = serious adverse event; SwCl = sweat chloride.

^aIn the RIDGELINE trial, which has no comparator arm, all outcomes were reported as within-group change only and reflect patients aged 6 to 11 years in cohort B1. Additionally, no 52-week outcomes were captured because this was a shorter study, so all outcomes from the RIDGELINE trial are approximately 24-week outcomes, unless otherwise specified.

^bThe PEx events requiring IV treatment and/or hospitalization are also discussed in this Clinical Review, as well as hospitalizations overall, but have not been assessed using GRADE.

^cThe 52-week end point has been preferentially assessed for this outcome, where available, based on clinical expert input that the longer duration is appropriate for assessment. End points of shorter duration (e.g., 24 weeks) for this outcome have been assessed using GRADE when they provide the only data available, which is the case for the RIDGELINE trial (which is the only source of data for patients aged 6 to 11 years) due to the shorter duration of the study.

^dStatistical testing for this end point was adjusted for multiple comparisons (e.g., hierarchical testing) in the SKYLINE 102 and SKYLINE 103 trials (when the hierarchy included, in order, absolute change in ppFEV₁ at 24 weeks, absolute change in SwCl through week 24, proportion of patients with an SwCl level of ≤ 60 mmol/L at 24 weeks, and proportion of patients with an SwCl level of ≤ 30 mmol/L at 24 weeks). No hierarchical testing or multiplicity adjustment was conducted for other outcomes, nor for any outcomes in the RIDGELINE trial.

^eIn the SKYLINE 102 and SKYLINE 103 trials, the absolute change in BMI was assessed in the FAS (model adjusted for baseline BMI), while the absolute change in BMI z score was assessed in patients aged younger than 20 years. In the RIDGELINE trial, because all patients were aged 11 years at most, BMI and BMI z score were both assessed in the FAS.

^fAll safety-related outcomes (including overall AEs, withdrawals due to AE, deaths, and so forth) have been reported, as is standard; however, per clinical expert input, only these 2 outcome measures (number of patients with SAEs and number with events of elevated aminotransferase levels) were assessed using GRADE.

Sources: Clinical Study Reports.^9-11

Number of Pulmonary Exacerbations

A pulmonary exacerbation was defined as a new or change in antibiotic therapy (IV, inhaled, or oral) for at least 4 sinopulmonary signs or symptoms. Pulmonary exacerbation is a critical contributor to the clinical and economic burden experienced by people with CF. Not only do pulmonary exacerbations have an acute negative effect on patient health and HRQoL, they are also associated with several long-term consequences, including a faster rate of lung function decline, increased risk of future exacerbations and hospitalizations, increased likelihood of a future lung transplant, and increased risk of mortality.^47-50 An empirically derived MID for pulmonary exacerbation rate has yet to be established, and the clinical experts consulted by CDA-AMC agreed there is no clear threshold. The threshold used in the GRADE assessments was therefore null.

The SKYLINE 102 and SKYLINE 103 trials both measured the change in the number of pulmonary exacerbations from baseline to the week 52 evaluation period, whereas the RIDGELINE trial measured from baseline to week 24.

Although not assessed using GRADE, related outcomes such as the time to first pulmonary exacerbation and the number of pulmonary exacerbation events requiring hospitalization or IV antibiotic therapies have also been summarized.

Percent Predicted Forced Expiratory Volume in 1 Second

Lung function in people with CF is typically reported using ppFEV₁, which determines the maximum volume of air exhaled during the first second of a forced breath. The ppFEV₁ value is calculated by dividing the patient’s FEV₁ score by the normal FEV₁ Global Lung Initiative reference value for healthy, nonsmoking individuals based on age, height, and sex and multiplying by 100.^2,51 For every 1% decline in ppFEV₁, 5-year mortality risk increases by 4%, highlighting the clinical importance of maintaining lung function at the highest possible ppFEV₁ level.^18,52 Although no empirically validated MID was identified, and there is no clinician consensus on an MID for ppFEV₁, a 5% threshold was previously accepted by CDA-AMC and Cystic Fibrosis Canada as a meaningful improvement.^22,24

The primary outcome of the 2 included RCTs (SKYLINE 102 and SKYLINE 103) was the absolute change from baseline in ppFEV₁ at 24 weeks, which was also a secondary outcome in the single-arm study, RIDGELINE. The 2 RCTs also reported the 52-week outcome for ppFEV₁ as a secondary outcome. In the SKYLINE 102 and SKYLINE 103 trials, the between-group findings were assessed based on the prespecified noninferiority margin of 3%.

BMI and BMI Z Score

BMI is a measure of a patient’s nutritional status and is based on their weight (in kilograms) and height (in metres), and BMI z score is a standardized measure of BMI that takes into account age group and sex.² Malnutrition leads to a low BMI and failure to thrive (i.e., insufficient weight gain and growth), which increases susceptibility to lung infections.^53,54 BMI has also been established as a predictor of survival in patients with CF.⁵⁵ All 3 included studies measured weight-related parameters, including BMI and BMI z score as a change from baseline to 24 weeks and, in the 2 RCTs, 52 weeks. In the SKYLINE 102 and SKYLINE 103 studies, which included both adult and pediatric patients older than 12 years, the change in BMI z score was measured only in patients who were aged 20 years or younger, and the change in BMI score was measured in all patients. No MID was identified through the literature or clinical expert input for these end points, so the threshold used in the GRADE assessment was null.

Sweat Chloride

The SwCl test is a standard diagnostic tool for CF because elevated concentrations of chloride in sweat indicate abnormality of CFTR protein activity.^22,35 The CFTR protein normally functions as a chloride ion transport channel. SwCl alone is not always sufficient to diagnose or exclude a diagnosis of CF because some patients with CF may have only mildly elevated SwCl at assessment, which is impacted in part by the specific CF-causing mutations carried by the patient.^22,35 A normal concentration level of SwCl is typically considered to be less than 30 mmol/L, while values from 30 mmol/L to 59 mmol/L are diagnostically unclear, and values greater than 60 mmol/L indicate CF.^22,35 Continued monitoring of SwCl is not common in real-world clinical practice after diagnosis.^22,35

Despite some heterogeneity between patients and the relationship between specific CF-causing mutations, changes in SwCl shortly after treatment initiation are an indicator of the restoration of some CFTR ion channel function; as such, it may be a useful mechanistic biomarker of response to a CFTRm.^22,35,56 The clinical experts consulted by CDA-AMC noted that SwCl can be a more sensitive indicator of mechanistic response to treatment initiation than clinical outcomes such as ppFEV₁ in patients whose lung function is still within the normal range at baseline (wherein improvements beyond baseline are unlikely), as is typically the case in most children. Also, changes in SwCl can be evaluated sooner than clinical outcomes such as lung function, which will typically require a longer duration of treatment to show meaningful changes. Patients commonly continue to have elevated SwCl concentrations after treatment with CFTRm therapies, even if there have been significant reductions compared with their before-treatment SwCl concentration results.⁵⁷ The clinical experts consulted by CDA-AMC also highlighted that there can be differences (5 mmol/L to 10 mmol/L) in SwCl levels in the same patient, depending on which arm is sampled (even when taken at the same time), or between different testing dates.

The experts consulted by CDA-AMC noted that, in their experience, SwCl is not strongly correlated with clinical outcomes owing to other sources of between-patient heterogeneity that more greatly impact clinical outcomes. They highlighted that even siblings with the same CF-causing genotype may appear to have a different relationship between changes in SwCl and clinical outcomes. The experts described that SwCl may not always be an appropriate biomarker, depending on the specific CF-causing mutation carried by a patient, because some CFTR mutations are associated with higher or lower SwCl elevations than others and, in some cases, such as the N1303K CFTR mutation, SwCl is not expected to substantially reduce with CFTRm therapy, despite objective improvement in clinical symptoms, indicating that the typical mechanistic reduction in SwCl may not always be directly relevant to the clinical improvements seen with a CFTRm. They also indicated there is currently no well-established MID for the end point of SwCl when comparing between groups treated with CFTRm therapies, and that it remains to be seen whether there is long-term clinical value with regard to reductions in SwCl beyond those seen with ELX-TEZ-IVA treatment.

In concordance with the clinical expert opinion presented to CDA-AMC, the magnitude and reliability of the correlation between SwCl and specific clinical outcomes has not been thoroughly established in published literature. Modern clinical studies of treatments for CF often include the assessment of SwCl, and patients treated with CFTRm therapies (compared with patients not treated with CFTRm therapies or compared with within-patient changes before and after CFTRm therapy) typically demonstrate reduced SwCl as well as improved clinical outcomes;^56,58-60 however, a causal relationship or reliable prediction of clinical benefit of SwCl as a surrogate end point has not been reported.

Several studies were submitted by the sponsor as evidence of support for this relationship. This review will examine several of them here briefly, excluding those studies that did not conduct any analysis of correlation or causality.

In a phase II study of ivacaftor (NCT00457821) in patients with CF and at least 1 G551D mutation,⁵⁶ within-patient SwCl was evaluated as a biomarker for changes in CFTR activity. There was a large amount of variability seen between patients in the ivacaftor treatment groups, which the study authors suggested may be due to differences in CFTR activity correction between individuals, variability in exposure to ivacaftor, differences in the mutation of the other allele, the relative quantity of CFTR channels at the cell surface, and/or the role of modifier genes impacting ion balance. Although this study demonstrated that CFTR activity increases after therapy with ivacaftor and that SwCl was also generally reduced, thereby demonstrating the effect of ivacaftor as a CFTRm that partially restores CFTR function, a clear connection or quantification of any potential correlation between changes in SwCl and specific clinical outcomes was not evaluated.⁵⁶

In an analysis of the US Cystic Fibrosis Foundation Patient Registry data that used Cox regression to evaluate the correlation of SwCl with clinical end points, there was no independent association of SwCl with survival, lung function, height, or BMI in patients with a known CFTR genotype functional class. However, among patients with an unclassified functional class of CFTR genotype, SwCl was an independent predictor of survival.⁶¹

In a study by Zemanick et al. (2025),⁵⁹ 6 phase III placebo-controlled studies (and their open-label extensions) of CFTRm in patients with CF with a homozygous F508del mutation or a heterozygous F508del and a minimal function mutation genotype were pooled and evaluated to examine the relationship between attained values of SwCl and several clinical end points: lung function (ppFEV₁), BMI, patient-reported outcomes, pulmonary exacerbations, and change in lung function over time. The studies included EVOLVE/EXTEND, AURORA F/MF, AURORA F/F, KEPLER, NCT03447249 (parent study) and NCT03447262 (open-label extension of NCT03447249), and NCT03460990 (parent study) and NCT03447262 (open-label extension of NCT03460990). The CFTR modulators used as interventions in these studies included tezacaftor-ivacaftor, ELX-TEZ-IVA, and VX-659-TEZ-IVA, and the median duration of the studies ranged from 0.5 years to 4.0 years. In total, the studies included 1,314 patients treated with a CFTRm and 642 patients treated with placebo. The change from baseline outcomes was analyzed using an analysis of covariance model; the number of pulmonary exacerbations through week 24 was analyzed using a negative binomial regression model, and the rate of change in ppFEV₁ was analyzed using a linear random coefficients model. End points were analyzed for the association between SwCl and the stated clinical outcomes using 4 SwCl-based groupings: less than 30 mmol/L, at least 30 mmol/L to less than 60 mmol/L, at least 60 mmol/L to less than 80 mmol/L, and at least 80 mmol/L. The results of the study suggest a correlation between attainment of the lower categories of SwCl (< 30 mmol/L or ≥ 30 mmol/L to < 60 mmol/L) and numerically better clinical outcomes. However, the 95% CIs of the results for most end points across the SwCl-based subgroups were overlapping to a large degree, suggesting these differences were generally not statistically significant.

The results of the Zemanick et al. (2025)⁵⁹ study provide support for the biologic plausibility of a correlation between SwCl and clinical end points, and it pools data across several trials in populations that are mostly relevant to the target population. However, there are limitations related to temporal ambiguity and causality, magnitude and clinical relevance of differences, and transferability across CFTR-causing mutation variants. These limitations will be discussed further subsequently.

Temporal ambiguity and causality: The Zemanick et al. (2025)⁵⁹ study adjusted baseline values for patients who had run-in periods with Trikafta, using expected untreated baselines derived from the phase III trials. While this helps standardize comparisons, it introduces model-based assumptions rather than direct observation, which weakens causal inference. These adjustments are necessary for harmonizing data but inherently limit the ability to draw causal conclusions. The primary analyses assessed change from baseline in both SwCl and clinical outcomes over the same time period. This concurrent measurement means that temporal precedence is not established. It is difficult to determine whether changes in SwCl preceded, followed, or occurred simultaneously with clinical improvements. The analysis of covariance, negative binomial regression, and linear random coefficients models are appropriate for assessing associations and adjusting for covariates; however, they do not account for unmeasured confounding, nor do they establish a directional pathway from SwCl to clinical outcomes.

Magnitude and clinical relevance of differences: While lower SwCl levels were associated with greater mean improvements in ppFEV₁, CFQ-R respiratory domain, BMI, and exacerbation rates, the 95% CIs overlapped across the 3 lower SwCl groups (< 80 mmol/L) for several outcomes. For example, the LS mean change in ppFEV₁ was:

• 15.65% in the less than 30 mmol/L group

• 14.01% in the 30 mmol/L to less than 60 mmol/L group

• 12.74% in the 60 mmol/L to less than 80 mmol/L group.

Using an MID of 5% for ppFEV₁, as mentioned previously, these differences (each under 2%) are unlikely to be clinically meaningful. This undermines the strength of the association and further limits the ability to infer causality. The article by Zemanick et al. (2025)⁵⁹ notes this point as a limitation.

Transferability across CFTR variants: The Zemanick et al. (2025)⁵⁹ article notes that clinical benefit has been observed in patients with the N1303K variant despite no consistent improvements in SwCl, suggesting that SwCl may not be informative across all genotypes. This limits its generalizability and further challenges its use as a surrogate end point because, mechanistically, it is apparent that these patients are experiencing a clinical benefit with CFTRm without necessarily also experiencing the reductions in SwCl that are typically seen in patients with CF caused by non-N1303K mutations.

While the Zemanick et al. study (2025)⁵⁹ is well conducted and provides important data, it is not definitive in establishing a causative correlation between SwCl as a surrogate for clinical end points and is insufficient to be relied on alone for decision-making in the context of health technology assessment. Moreover, regulatory agencies have not accepted SwCl as a surrogate for clinical benefit, reinforcing the need for caution in its interpretation within a health technology assessment framework.

In summary, despite the uncertainty about its prognostic utility, SwCl is an important biomarker that provides information about the mechanistic effect of CFTRm therapies and could be supportive in assessing their comparability. Nonetheless, the clinical importance or surrogacy of SwCl reductions beyond that previously seen in studies of ELX-TEZ-IVA has not been clearly established to predict clinical outcomes and, accordingly, there is no established MID for SwCl for comparisons between treatment groups; therefore, the GRADE assessment was conducted using null as the threshold of interest.

In the SKYLINE 102, SKYLINE 103, and RIDGELINE trials, SwCl was collected before study drug dosing at each time point (2 samples, 1 from each arm) and sent to a central laboratory for testing and interpretation of results. In addition to the absolute change from baseline in SwCl, this review will also report (although not assessed with GRADE) the proportion of patients with an SwCl level of 30 mmol/L or lower at week 24 and week 52.

CFQ-R Respiratory Domain

The CFQ-R is the most widely used and validated patient-reported outcome measure in CF that assesses the clinical benefit of a given treatment from the patient or caregiver’s perspective.⁶² It includes 12 domains, each of which is scored from 0 to 100, with higher numbers representing better HRQoL or fewer symptoms. The respiratory domain includes items that measure patient- (or caregiver-) reported improvements in the following respiratory symptoms: waking up from coughing, wheezing, coughing, congestion, difficulty breathing, and mucus production.^29,63,64 The MID, which corresponds to the smallest clinically important change a patient can detect, is estimated to be a 4-point change in CFQ-R respiratory domain scores.⁶² The SKYLINE 102, SKYLINE 103, and RIDGELINE trials measured the absolute change in CFQ-R respiratory domain score from baseline to the week 24 evaluation period. The measurement properties of the CFQ-R respiratory domain are detailed in Table 10.

Elevated Aminotransferase Levels

Elevated aminotransferase is an AE associated with the treatment of CF using CFTRm therapies that can indicate issues with the liver.22 This AE was identified by the clinical experts consulted by CDA-AMC as one that is particularly important to monitor. The magnitude of elevation can indicate the need to interrupt, reduce, or discontinue therapy.²²

Serious Adverse Events

In the included studies, SAEs are defined as any AE that meets the following criteria:

• fatal (death, regardless of cause, that occurs during participation in the study or occurs after participation and is suspected of being a delayed toxicity due to the administration of the study drug)

• life-threatening, such that the patient was at immediate risk of death from the reaction as it occurred

• inpatient hospitalization or prolongation of hospitalization

• persistent or significant disability or incapacity (disability is defined as a substantial disruption of a person’s ability to conduct normal life functions)

• congenital anomaly or birth defect

• important medical event that, based on appropriate medical judgment, may jeopardize the patient or may require medical or surgical intervention to prevent 1 of the outcomes listed previously (e.g., an allergic bronchospasm requiring intensive treatment in an emergency department or at home).

Table 10: Summary of Outcome Measures and Their Measurement Properties

CF = cystic fibrosis; CFQ-R = Cystic Fibrosis Questionnaire-Revised; F = F statistic; HRQoL = health-related quality of life; ICC = intraclass correlation coefficient; MID = minimal important difference; r = correlation coefficient; rs = Spearman correlation; SF-36 = Short Form (36) Health Survey.

Statistical Analysis

Sample Size and Power Calculation

For the SKYLINE 102 trial, assuming a within-group SD of 8, a 10% dropout rate at week 24, and a treatment difference of zero between VNZ–TEZ–D-IVA and ELX-TEZ-IVA, a sample size of 200 patients in each group for a total of 400 patients provided more than 90% power to test the primary hypothesis for the primary end point, based on a 1-sided, 2-sample t test at a significance level of 0.025.

For the SKYLINE 103 trial, assuming a within-group SD of 8, a 10% dropout rate at week 24, and a treatment difference of zero between VNZ–TEZ–D-IVA and ELX-TEZ-IVA, a sample size of 275 patients in each group for a total of 550 patients provided more than 95% power to test the primary hypothesis for the primary end point, based on a 1-sided, 2-sample t test at a significance level of 0.025.

For the RIDGELINE trial, between 12 and 20 patients were planned to be enrolled in cohort A1. Sample size calculations were determined based on VNZ–TEZ–D-IVA pharmacokinetic parameters, such as clearance and volume of distribution. Based on the variability observed in adults, data from 12 patients allowed 80% power to target a 95% CI within 60% and 140% of the geometric mean estimate of clearance for VNZ–TEZ–D-IVA. No formal power calculation was performed for cohort B1. Approximately 65 patients were planned for enrolment in cohort B1, which was deemed adequate to meet the primary safety objective. The occurrence of treatment-emergent AEs was a safety end point. With approximately 55 patients expected to complete cohort B1, the study had a 94% chance of observing an AE in at least 1 patient if the true incidence rate was 5%, and a greater than 99% chance of observing an AE in at least 1 patient if the true incidence rate was 10%. The probabilities were calculated by assuming a binomial distribution for the number of AEs.

End Points

Primary End Point Analysis (ppFEV₁)

In the SKYLINE 102 and SKYLINE 103 trials, the primary noninferiority analysis was performed using an MMRM with change from baseline at day 15, week 4, week 8, week 16, and week 24 as the dependent variable. The model included treatment group, visit, and treatment-by-visit interaction as fixed effects, with continuous baseline ppFEV₁, age at screening (aged < 18 years versus ≥ 18 years) and mutation group as covariates. The MMRM included data from all available visits up to week 24. A Kenward-Roger approximation was used for denominator degrees of freedom. Treatment effect through week 24 was estimated by averaging weeks 16 and 24. Intercurrent events of prohibited medication and treatment discontinuation were addressed using the treatment policy strategy (observed data following the event were included in the model). A hierarchical testing procedure was used to control the overall type I error at an alpha of 0.05. The primary end point of absolute change from baseline in ppFEV₁ through week 24 was tested for noninferiority at a 1-sided alpha level of 0.025. Subgroups assessed for ppFEV₁ included age (younger than 18 years, 18 years or older), ppFEV₁ at baseline (less than 70%, at least 70% and greater), SwCl at baseline (less than 30 mmol/L, at least 30 mmol/L), sex (female, male), and region (North America, the rest of the world). A supplementary analysis was conducted in which intercurrent events were handled using the hypothetical strategy (observed data after the event were set to missing) instead of the treatment policy strategy; no other sensitivity analyses were described.

The noninferiority margin selected in the SKYLINE 102 and SKYLINE 103 trials was 3 percentage points. A statistical approach using the Rothmann method recommends that the noninferiority margin preserve at least 50% of the treatment effect of the active control (ELX-TEZ-IVA) compared with placebo, where the treatment effect is estimated by the lower bound of the 95% CI.^68,69 The selected noninferiority margin of 3 percentage points is consistent with this statistical method and the noninferiority margin for ppFEV₁ that is used in clinical studies evaluating symptomatic CF treatments.^7,8

The analysis of the 52-week end point for change in ppFEV₁ was conducted in a manner similar to that of the 24-week end point described previously, with the end points estimated by averaging weeks 16, 24, 36, and 52.

In the RIDGELINE trial, no primary efficacy analysis was reported, but the ppFEV₁ analysis used an MMRM to estimate the within-group mean absolute change through week 24 (estimated by averaging weeks 16 and 24). The end point was the dependent variable, visit was a fixed-effect variable, and baseline value and genotype group were covariates. If there was nonconvergence due to there being only a small number of patients in the genotype group, then it was removed from the model. An unstructured covariance structure was used to model the within-patient errors. If model estimation did not converge, a compound symmetry covariance structure was used instead. No multiplicity adjustments, sensitivity analyses, or subgroup analyses were conducted in the RIDGELINE trial.

Secondary Continuous End Point Analysis (SwCl, CFQ-R Respiratory Domain, BMI, and Other Weight Metrics)

In all 3 trials, the analysis of absolute change from baseline in SwCl, CFQ-R respiratory domain, BMI, and BMI z score was based on an MMRM that was similar to the one described previously for ppFEV₁, with absolute change from baseline at outcome assessment (e.g., week 24) as the dependent variable. For SwCl and the CFQ-R respiratory domain, the primary result obtained from the model was the estimated treatment difference through week 24, estimated by averaging weeks 16 and 24. No subgroup or sensitivity analyses were conducted for secondary outcomes in any of the 3 trials.

For the RCTs, a hierarchical testing procedure was used to control the overall type I error at an alpha of 0.05 for the primary outcome and the key secondary end points (the absolute change from baseline in SwCl through week 24, the proportion of patients with an SwCl level of 60 mmol/L or less at week 24, and the proportion of patients with an SwCl level of 30 mmol/L or less, pooled between the SKYLINE 102 and SKYLINE 103 trials). The primary end point of absolute change from baseline in ppFEV₁ through week 24 was tested for noninferiority at a 1-sided alpha level of 0.025. The key secondary end points were formally tested at an alpha level of 0.05 only if the primary analysis of absolute change from baseline in ppFEV₁ through week 24 was statistically significant, i.e., if the null hypothesis of inferiority was rejected. For a test at any step to be considered statistically significant within the testing hierarchy, it must have been statistically significant, and all previous tests (if any) within the hierarchy must have been statistically significant at the 0.05 level (1-sided 0.025 level for the primary end point). In the 2 RCTs, the order of the hierarchy after the primary outcome was as follows: absolute change from baseline in SwCl through week 24, proportion of patients with an SwCl level of less than 60 mmol/L through week 24 (pooled FAS in the SKYLINE 102 and SKYLINE 103 trials), and proportion of patients with an SwCl level of less than 30 mmol/L through week 24 (pooled FAS).

The analyses of the 52-week ppFEV₁- and SwCl-related end points were conducted in a manner similar to that of the 24-week end points previously described, but these end points were not included in a hierarchical testing procedure and were estimated by averaging weeks 16, 24, 36, and 52.

No hierarchical testing or multiplicity adjustments were conducted in the RIDGELINE trial.

Secondary End Point Analysis: Number of Pulmonary Exacerbations (Rate)

In the SKYLINE 102 and SKYLINE 103 trials, pulmonary exacerbations through 52 weeks were analyzed by comparing differences in the annual exacerbation rates between treatment groups, with 95% CIs providing statistical context.

In the RIDGELINE study, the number of pulmonary exacerbations during the 24-week analysis period was assessed using descriptive statistics, including annualized event rate and duration of events. The annualized duration of events for each patient was the total number of days with the event multiplied by 48 weeks and divided by the total number of weeks in the study up to the end of the analysis period. The annualized duration was descriptively summarized.

No other details on the methodology of assessing this outcome were reported in any of the included studies, such as how missing data were handled, although the RIDGELINE study stated that, in general, missing data were not imputed unless otherwise specified.

No subgroup or sensitivity analyses were conducted for the secondary outcomes in any of the included studies. No hierarchical testing or multiplicity adjustment was conducted in the RIDGELINE study, and pulmonary exacerbations were not included in the hierarchical testing structure for either the SKYLINE 102 or SKYLINE 103 trial.

BMI and BMI Z Scores

Across all 3 trials, the BMI and BMI z scores were calculated by using Centers for Disease Control and Prevention growth charts.⁷⁰

In the SKYLINE 102 and SKYLINE 103 RCTs, BMI and BMI z score were assessed as an absolute change from baseline at week 52 using an MMRM similar to the one described for ppFEV₁. BMI was assessed in the FAS, while BMI z score was assessed in patients aged 20 years or younger at baseline, and the covariate of age at screening was not included in the BMI z score model for this reason. Data obtained from the day 15, week 4, week 8, week 16, and week 24 visits were included in the model.

In the RIDGELINE trial, because all patients in the study were aged 11 years at most, both BMI and BMI z score were assessed in the FAS (within the cohort in question, in this case, cohort B1). This end point was also assessed through an MMRM similar to the one described for SwCl.

No subgroup or sensitivity analyses were conducted for secondary outcomes in any of the 3 trials, and no BMI or BMI z scores were included in the hierarchical testing structure of either the SKYLINE 102 or SKYLINE 103 trials. No hierarchical testing or multiplicity adjustment was conducted in the RIDGELINE trial.

Safety Analyses

In all 3 trials, safety analyses were assessed during the treatment-emergent period in the safety analysis set, including follow-up time after the last day of the treatment period (approximately 1 month). Patients were analyzed according to the actual treatment received during the treatment period. Patients who received the study drug while in more than 1 treatment group were allocated to the VNZ–TEZ–D-IVA group.

For the safety analyses, baseline was defined as the most recent nonmissing measurement (scheduled or unscheduled) collected before the first dose of the study drug in the treatment period (i.e., the day 1 visit).

Subgroup Analyses

In both the SKYLINE 102 and SKYLINE 103 trials, the MMRM used for the primary analysis was used for the subgroup analysis of ppFEV₁, where a similar model was applied to each category of the subgroup: age at screening (aged < 18 years, ≥ 18 years), ppFEV₁ (< 70%, ≥ 70%), and sex (female, male). Model-based estimates for a given category are displayed if the analysis converges in that category. For subgroup analyses based on age, the term “age at screening” was removed from the MMRM model. The adjusted means with 2-sided 95% CIs were provided.

The clinical experts consulted by CDA-AMC suggested there may be differences in expected treatment outcomes in pediatric patients and patients with highly preserved lung function compared with others, so these subgroup results have been summarized in the ppFEV₁ section of the efficacy results. However, sex is not expected to have an influence on prognosis or treatment response in CF, according to the clinical experts, so it will not be addressed. The studies are not powered to analyze treatment differences in the smaller sample sizes of subgroup analyses, so the findings may be influenced by variability and may lack precision, increasing the risk of false-positive or false-negative results. These analyses are exploratory and should be considered hypothesis-generating rather than confirmatory.

In the RIDGELINE trial, no subgroup analyses were performed due to the small sample size.

Table 11: Statistical Analysis of Efficacy End Points

BMI = body mass index; CFQ-R = Cystic Fibrosis Questionnaire-Revised; CI = confidence interval; F/F = homozygous for F508del; F/MF = heterozygous for F508del and a minimal function mutation; FAS = full analysis set; LS = least squares; MMRM = mixed model for repeated measures; ppFEV₁ = percent predicted forced expiratory volume in 1 second; RCT = randomized controlled trial; RD = respiratory domain; SAE = serious adverse event; SAS = safety analysis set; SwCl = sweat chloride; vs. = versus.

Note: Only ppFEV₁ at 24 weeks in the SKYLINE 102 and SKYLINE 103 trials, and key secondary outcomes (change from baseline through week 24 in SwCl, proportion of patients with an SwCl level of 60 mmol/L or less at week 24 and proportion of patients with an SwCl level of 30 mmol/L or less at week 24) were part of the hierarchical testing procedure. No 52-week outcomes were controlled for multiplicity. No outcomes from the RIDGELINE trial were controlled for multiplicity.

Sources: Clinical Study Reports.^9-11

Analysis Populations

The analysis populations of the SKYLINE 102, SKYLINE 103, and RIDGELINE trials are described in Table 12.

Table 12: Analysis Populations of the Included Studies

FAS = full analysis set.

Sources: Clinical Study Reports.^9-11

Results

Patient Disposition

Patient disposition in the SKYLINE 102, SKYLINE 103, and RIDGELINE trials is summarized in Table 13.

In the SKYLINE 102 trial, 200 patients in the VNZ–TEZ–D-IVA group and 205 patients in the ELX-TEZ-IVA group were randomized, although 4 and 3 patients in each group, respectively, did not receive any dose. Of the patients who received any dose, 12 (6.1%) and 11 (5.4%) patients, respectively, discontinued from the study, while 15 (7.7%) and 16 (7.9%) patients, respectively, discontinued from treatment (but not necessarily the study). The most common reasons for discontinuation from the study were AEs (2% in each group) and withdrawal of consent not due to AEs (2.6% and 2.5%, respectively). Discontinuation from treatment was also due most commonly to AEs (2% in the VNZ–TEZ–D-IVA group and 5% in the ELX-TEZ-IVA group), pregnancy (1.5% and 0.5%, respectively), or refusal of further dosing not due to AEs (2.6% and 1.0%, respectively). Patient disposition was generally well balanced between the groups, with some relatively small differences.

In the SKYLINE 103 trial, 285 patients were randomized to VNZ–TEZ–D-IVA and 289 were randomized to ELX-TEZ-IVA. One patient randomized to VNZ–TEZ–D-IVA did not receive any dose of the study drug. Discontinuations from the study occurred in 10 (3.5%) and 20 (7.0%) patients, respectively, and discontinuations from treatment occurred in 25 (8.8%) and 16 (5.5%) patients, respectively. The most common reasons for study discontinuation were again AEs (3.2% and 1.0%, respectively), followed by withdrawal of consent not due to AEs (1.1% and 0.3%, respectively). The most common reasons for discontinuation from treatment were AEs (4.9% and 3.1%, respectively), pregnancy (1.4% and 1.0%, respectively), and refusal of further dosing not related to AEs (0.7% and 0.3%, respectively). Patient disposition was generally well balanced between the groups, with some relatively small differences.

In the RIDGELINE trial, 17 patients were recruited for cohort A1, and 78 patients were recruited for cohort B1. No patients discontinued from cohort A1, and only 1 patient discontinued treatment (but not the study) in cohort B1, which was due to an AE.

Table 13: Summary of Patient Disposition From Studies Included in the Systematic Review (Set of All Patients)

AE = adverse event; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; FAS = full analysis set; NA = not applicable; NR = not reported; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

Sources: Clinical Study Reports.^9-11

Important Protocol Deviations

In the SKYLINE 102 trial, 20 patients (5.0%) had important protocol deviations, which were defined as any protocol deviation that may have significantly impacted the completeness, accuracy, and/or reliability of key study data or that may have significantly affected a patient’s rights, safety, or well-being. Of these, 8 patients (5 in the VNZ–TEZ–D-IVA group and 3 in the ELX-TEZ-IVA group) had events related to investigational products, such as study drug temperature excursions, study drug dispensation errors, and study drug dosing outside of the relevant treatment period. The deviation was related to prohibited concomitant medications in 7 patients (3 in the VNZ–TEZ–D-IVA group and 4 in the ELX-TEZ-IVA group), including solumedrol, commercial ELX-TEZ-IVA (Trikafta), cimetidine, fluconazole, and dexamethasone. Deviations related to eligibility criteria occurred in 3 patients (2 in the VNZ–TEZ–D-IVA group and 1 in the ELX-TEZ-IVA group) due to acute respiratory infection or antibiotics received during the run-in period. Two patients had deviations related to study conduct or procedures: 1 in the VNZ–TEZ–D-IVA group (a subinvestigator performed a physical examination without documented training) and 1 in the ELX-TEZ-IVA group (morning and evening doses were exchanged at the week 36 visit). Finally, 1 patient had a deviation due to delay in SAE reporting in the VNZ–TEZ–D-IVA group.

In the SKYLINE 103 trial, 27 patients (4.7%) had important protocol deviations, which were defined in the same manner as in the SKYLINE 102 trial. Of these, 16 had deviations related to investigation products (11 in the VNZ–TEZ–D-IVA group and 5 in the ELX-TEZ-IVA group) such as study drug dispensation errors, compliance, study drug kits not being returned, and incorrect study drug dosing. Five had deviations related to eligibility criteria (1 in the VNZ–TEZ–D-IVA group and 4 in the ELX-TEZ-IVA group) due to medical history, acute illness, or antibiotics received during the run-in period. Three patients had a deviation due to delay in SAE reporting, 2 in the VNZ–TEZ–D-IVA group and 1 in the ELX-TEZ-IVA group. Two patients had deviations related to study conduct: 1 patient in the VNZ–TEZ–D-IVA group had delayed the interruption of the study drug and resumed the study drug without meeting resumption criteria, and 1 patient in the ELX-TEZ-IVA group performed day 1 assessments after study drug dosing. Finally, 1 patient’s parent signed the incorrect version of the consent form, and 1 patient took commercial CFTRm during an interruption in the study drug.

In the RIDGELINE trial, 2 patients in cohort A1 had protocol deviations (defined in the same manner as stated previously): 1 enrolled without a documented review of the screening laboratory results, and 1 did not have a baseline hemoglobin value confirmed before study drug dosing. In cohort B1, 2 patients had deviations related to incorrect study drug dispensing (the dose for higher-weight patients) on day 1 and were dispensed the correct kits (the dose for lower-weight patients) on day 15 and thereafter.

Baseline Characteristics

The baseline characteristics outlined in Table 14 are limited to those that are most relevant to this review or were felt to affect the outcomes or interpretation of the study results.

The SKYLINE 102 (N = 398) and SKYLINE 103 (N = 573) trials were similarly designed RCTs. In the SKYLINE 102 trial, slightly more than 40% of patients were female and just under 60% of patients were male, whereas the sex distribution was closer to approximately 50% each in the SKYLINE 103 trial. The mean age by treatment group in the 2 RCTs at baseline was 31 to 34 years across the SKYLINE studies, and the treatment groups were similar for these categories in each study. In general, the characteristics appeared to be well balanced between the treatment groups in each study.

In cohorts A1 and B1 of the RIDGELINE trial, 41.2% and 43.6% were female, 58.8% and 56.4% of patients were male, and the mean age was 9.5 years and 9.3 years, respectively.

The majority of patients in every study were white (range, 90.7% to 97.5%), while a minority were Asian (range, 0% to 0.5%), Black or African American (range, 0% to 2.0%), more than 1 race (range, 0% to 1.5%), or other (range, 0.% to 0.4%), or information about race was not collected due to local regulations. The baseline data for ppFEV₁ demonstrated more progressed disease in the studies that recruited older patients (the SKYLINE 102 and SKYLINE 103 trials recruited patients aged 12 years or older) compared with the RIDGELINE trial, which recruited patients aged 6 to 11 years for cohorts A1 and B1; this reflects the natural course of CF, and there were no concerning within-study differences between treatment groups. In all studies, most patients (range, 84.6% to 94.1%) had prior experience with a CFTRm, most commonly, ELX-TEZ-IVA. The mean duration of prior ELX-TEZ-IVA use was approximately 2 years in the SKYLINE 102 and SKYLINE 103 RCTs, and approximately 1 year in cohort B1 of the RIDGELINE trial. The SKYLINE 102 trial recruited patients with a genotype that was heterozygous for F508del and a minimal function mutation; ergo, this genotype was represented in 100% of patients while, in the other studies, the most common genotype was homozygous for F508del (range, 47.4% to 78.2%).

Table 14: Summary of Baseline Characteristics From the Studies Included in the Systematic Review (Full Analysis Sets)

BMI = body mass index; CFQ-R RD = Cystic Fibrosis Questionnaire-Revised respiratory domain; CFTRm = CFTR modulator; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; F/F = homozygous for F508del; F/G = heterozygous for F508del and a gating mutation; F/MF = heterozygous for F508del and a minimal function mutation; F/other = heterozygous for F508del and a non-F508del mutation that does not fit into any of the other specified mutation categories (i.e., gating, residual function, minimal function, triple combination–responsive); F/RF = heterozygous for F508del and a residual function mutation; NR = not reported; ppFEV₁ = percent predicted forced expiratory volume in 1 second; SD = standard deviation; SwCl = sweat chloride; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

Note: Racial categories used in the table are as reported in the source and may not align with Canada's Drug Agency inclusive language guidelines.

^aIn the SKYLINE 102 trial, North America includes patients from the US, and rest of the world includes patients from Australia, Europe, Israel, and New Zealand. In the SKYLINE 103 trial, North America includes patients from the US and Canada, and rest of the world includes patients from Australia, Europe, Israel, and New Zealand. In the RIDGELINE trial, North America includes patients from the US, and rest of the world includes patients from Australia and Europe.

^bIn the RIDGELINE trial, the CFQ-R RD score is based on the child version and was assessed in 75 of the 78 patients in cohort B1.

^cThe SKYLINE 102 trial included this genotype only, based on eligibility criteria.

Sources: Clinical Study Reports.^9-11

Exposure to Study Treatments

The exposure to study treatments is summarized in Table 15.

Table 15: Summary of Patient Exposure in Studies Included in the Systematic Review (FAS)

ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; FAS = full analysis set; NA = not applicable; NR = not reported; SD = standard deviation; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

^aFor the RIDGELINE trial only, this row represents an exposure of > 24 weeks instead of > 24 weeks to ≤ 36 weeks.

Sources: Clinical Study Reports.^9-11

Concomitant Medications

A summary of concomitant medications in at least 20% of patients is presented in Table 16.

Table 16: Summary of Concomitant Medications in at Least 20% of Patients in Any Group During the Treatment Phase of the Included Studies (FAS)

ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; FAS = full analysis set; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

Sources: Clinical Study Reports.^9-11

Efficacy

The results for key end points from the SKYLINE 102 and SKYLINE 103 studies are presented in Table 17, while results from the single-arm study, RIDGELINE, are presented in Table 18. These tables include the longest-duration end points available for the included studies and capture the results of any end points chosen to be assessed using GRADE.

Number of Pulmonary Exacerbations

Annual Event Rate

In the SKYLINE 102 trial FAS, the estimated event rate per year was 0.42 in the ELX-TEZ-IVA group (n = 50 patients with events) versus 0.32 in the VNZ–TEZ–D-IVA group (n = 60 patients with events). The pulmonary exacerbation rate difference was −0.10 events (95% CI, −0.24 to 0.04) per year.

In the SKYLINE 103 trial FAS, the annual event rate was 0.29 in the VNZ–TEZ–D-IVA group (n = 61 patients with events) and 0.26 in the ELX-TEZ-IVA group (n = 59 patients with events). The rate difference was 0.03 events per year (95% CI, −0.07 to 0.13).

In the RIDGELINE trial, there were 6 patients in the cohort B1 FAS who experienced an event, contributing to an annual event rate of 0.15.

No subgroup or sensitivity analyses were conducted for this end point in the included studies.

Time to First Pulmonary Exacerbation

In the SKYLINE 102 trial, based on a secondary analysis in the FAS of the time to first pulmonary exacerbation during the analysis period, the probability of event-free survival (Kaplan-Meier estimate) at 24 weeks was 0.820 (95% CI, 0.758 to 0.867) in the VNZ–TEZ–D-IVA group and 0.819 (95% CI, 0.758 to 0.866) in the ELX-TEZ-IVA group; at 52 weeks, it was 0.741 (95% CI, 0.673 to 0.797) and 0.695 (95% CI, 0.626 to 0.755), respectively.

In the SKYLINE 103 trial, based on a secondary analysis in the FAS of the time to first pulmonary exacerbation during the analysis period, the probability of event-free survival (Kaplan-Meier estimate) at 24 weeks was 0.857 (95% CI, 0.810 to 0.893) in the VNZ–TEZ–D-IVA group and 0.882 (0.838 to 0.914) in the ELX-TEZ-IVA group; at 52 weeks, it was 0.783 (95% CI, 0.729 to 0.827) and 0.792 (95% CI, 0.739 to 0.835), respectively.

Time to first pulmonary exacerbation was not reported in the RIDGELINE trial.

Pulmonary Exacerbations Requiring IV Antibiotics or Hospitalization

In the SKYLINE 102 trial FAS, 10 patients (5.1%) in the VNZ–TEZ–D-IVA group and 26 patients (12.9%) in the ELX-TEZ-IVA group had pulmonary exacerbations requiring hospitalization or IV antibiotic therapy during the analysis period. The probability of event-free survival (Kaplan-Meier estimate) in the 2 groups at 24 weeks was 0.969 (95% CI, 0.933 to 0.986) and 0.935 (95% CI, 0.890 to 0.962), respectively; at 52 weeks, it was 0.948 (95% CI, 0.906 to 0.972) and 0.868 (95% CI, 0.813 to 0.908), respectively. All patients with events required IV antibiotic therapy; additionally, 8 patients in the VNZ–TEZ–D-IVA group and 17 patients in the ELX-TEZ-IVA group required hospitalization.

In the SKYLINE 103 trial FAS, 19 patients (6.7%) and 17 patients (5.9%) in the VNZ–TEZ–D-IVA group and ELX-TEZ-IVA group, respectively, had pulmonary exacerbations requiring hospitalization or IV antibiotic therapy. The probability of event-free survival (Kaplan-Meier estimate) at 24 weeks was 0.961 (95% CI, 0.930 to 0.978) and 0.962 (95% CI, 0.932 to 0.979); at 52 weeks, it was 0.931 (95% CI, 0.894 to 0.956) and 0.940 (95% CI, 0.906 to 0.962), respectively. All patients with pulmonary exacerbations required IV antibiotic therapy; additionally, 14 patients in the VNZ–TEZ–D-IVA group and 9 patients in the ELX-TEZ-IVA group required hospitalization.

In cohort B1 of the RIDGELINE trial, 1 patient (1.3%) required both hospitalization and IV antibiotic therapy, for an observed event rate per year of 0.03.

Percent Predicted Forced Expiratory Volume in 1 Second

Twenty-Four Weeks

The primary outcome of ppFEV₁ was evaluated in the FAS through week 24 in both the SKYLINE 102 trial (n = 196 in the VNZ–TEZ–D-IVA group and n = 202 in the ELX-TEZ-IVA group) and SKYLINE 103 trial (n = 284 and 289, respectively). In the MMRM analysis of noninferiority comparing VNZ–TEZ–D-IVA with ELX-TEZ-IVA, the LS mean treatment difference was 0.2% (1-sided P < 0.0001; 95% CI, −0.7% to 1.1%) in the SKYLINE 102 trial and was 0.2% (95% CI, −0.5% to 0.9%; 1-sided P for noninferiority < 0.0001) in the SKYLINE 103 trial. The prespecified noninferiority margin was −3.0%; ergo, the results of both studies met the preplanned primary analysis of noninferiority for ppFEV₁ because neither 95% CI crossed the noninferiority margin.

In cohort B1 of the RIDGELINE trial (n = 78), patients treated with VNZ–TEZ–D-IVA showed similar ppFEV₁ values compared with baseline use of ELX-TEZ-IVA, based on a within-group LS mean absolute change from baseline to week 24 of 0% (95% CI, −2.0% to 1.9%).

Fifty-Two Weeks

The secondary end point of ppFEV₁ through 52 weeks showed similar results in both studies, consistent with the 24-week primary analyses. In the SKYLINE 102 trial, the LS mean difference in absolute change from baseline was 0.1% (95% CI, −0.8% to 1.0%); in the SKYLINE 103 trial, it was 0.3% (95% CI, −0.4 to 1.0).

Subgroup and Sensitivity Analyses

In the SKYLINE 102 and SKYLINE 103 trials, a supplementary sensitivity analysis was conducted using an alternative estimand where intercurrent events were addressed using the hypothetical strategy for the 24-week outcome of ppFEV₁, and the results were consistent with the primary analyses.

Subgroup analyses were conducted on the primary outcome at 24 weeks in the SKYLINE 102 and SKYLINE 103 trials based on age (aged younger than 18 years versus at least 18 years at screening) and ppFEV₁ at baseline (less than 70% versus at least 70%). Most of the results fell within the noninferiority margin of −3.0% and were similar to the primary analysis results. However, in the SKYLINE 102 trial in the subgroup of patients younger than age 18 years at screening, the 95% CI crossed this threshold, with an LS mean difference of −1.3% (95% CI, −3.9% to 1.3%). In the SKYLINE 103 trial, the value for this subgroup was 0.9% (95% CI, −1.2% to 3.0%).

No subgroup or sensitivity analyses were conducted in the RIDGELINE trial or for the week 52 outcomes in the SKYLINE studies.

Body Mass Index

Twenty-Four Weeks

In the RIDGELINE trial, the absolute change in BMI at week 24 was an LS mean of 0.22 kg/m² (95% CI, 0.05 kg/m² to 0.38 kg/m²).

Fifty-Two Weeks

In the SKYLINE 102 trial, the LS mean change from baseline in BMI in the VNZ–TEZ–D-IVA group (n = 179) and ELX-TEZ-IVA group (n = 187) at week 52 was 0.30 kg/m² (SE = 0.10 kg/m²) and 0.08 kg/m² (SE = 0.10 kg/m²), respectively, with an LS mean difference of 0.22 kg/m² (95% CI, −0.05 kg/m² to 0.49 kg/m²).

In the SKYLINE 103 trial, the LS mean change from baseline in BMI in the VNZ–TEZ–D-IVA group (n = 248) and ELX-TEZ-IVA group (n = 266) at week 52 was 0.37 kg/m² (SE = 0.08 kg/m²) and 0.22 kg/m² (SE = 0.07 kg/m²), respectively, with an LS mean difference of 0.16 kg/m² (95% CI, −0.05 kg/m² to 0.36 kg/m²).

Subgroup and Sensitivity Analyses

No subgroup or sensitivity analyses were reported.

BMI Z Score

Twenty-Four Weeks

In the RIDGELINE trial, the absolute change in BMI z score at week 24 was an LS mean of −0.05 (95% CI, −0.12 to 0.02).

Fifty-Two Weeks

In the SKYLINE 102 trial at 52 weeks, the LS mean change from baseline in BMI z score among patients aged 20 years or younger at baseline was 0.25 (SE = 0.10) in the VNZ–TEZ–D-IVA group (n = 27) and −0.09 (SE = 0.09) in the ELX-TEZ-IVA group (n = 32), with an LS mean difference of 0.34 (95% CI, 0.07 to 0.61).

In the SKYLINE 103 trial, the LS mean change from baseline in BMI z score among patients aged 20 years or younger at baseline was 0.11 (SE = 0.06) in the VNZ–TEZ–D-IVA group (n = 42) and −0.13 (SE = 0.07) in the ELX-TEZ-IVA group (n = 35), with an LS mean difference of 0.24 (0.06 to 0.42) at 52 weeks.

Subgroup and Sensitivity Analyses

No subgroup or sensitivity analyses were reported.

CFQ-R Respiratory Domain

Twenty-Four Weeks

In the SKYLINE 102 trial, the LS mean change in the CFQ-R respiratory domain score through week 24 was 0.5 points (SE = 1.1) in the VNZ–TEZ–D-IVA group (n = 186) and −1.7 points (SE = 1.0) in the ELX-TEZ-IVA group (n = 192). The LS mean difference was 2.3 points (95% CI, −0.6 to 5.2).

In the SKYLINE 103 trial, the LS mean change from baseline in the CFQ-R respiratory domain was −1.2 points (SE = 0.8) in the VNZ–TEZ–D-IVA group (n = 268) and the same value, i.e., −1.2 points (SE = 0.8) in the ELX-TEZ-IVA group (n = 270). The LS mean difference was −0.1 points (95% CI, −2.3 to 2.1).

Patients treated with VNZ–TEZ–D-IVA in cohort B1 of the RIDGELINE study had a within-group LS mean absolute change from baseline through week 24 of 3.9 points (95% CI, 1.5 to 6.3).

Fifty-Two Weeks

In the SKYLINE 102 trial, the LS mean change in the VNZ–TEZ–D-IVA group (n = 186) and ELX-TEZ-IVA group (n = 192) at 52 weeks was 0.8 points (SE = 1.0) and −1.6 points (SE = 1.0), respectively, with an LS mean difference of 2.4 points (95% CI, −0.3 to 5.1).

In the SKYLINE 103 trial, the LS mean change in the VNZ–TEZ–D-IVA group (n = 268) and ELX-TEZ-IVA group (n = 270) at 52 weeks was −0.4 points (SE = 0.7) and −1.0 points (SE = 0.7), respectively, with an LS mean difference of 0.7 points (95% CI, −1.4 to 2.7).

Subgroup and Sensitivity Analyses

No subgroup or sensitivity analyses were reported.

Sweat Chloride

Twenty-Four Weeks

In the SKYLINE 102 trial, the LS mean absolute change from baseline for SwCl through week 24 in the FAS was −7.5 mmol/L (SD = 0.8 mmol/L) in the VNZ–TEZ–D-IVA group (n = 196) compared with 0.9 mmol/L (SD = 0.8 mmol/L) in the ELX-TEZ-IVA group (n = 202). The MMRM analysis of this secondary end point reported a reduction (i.e., improvement) in SwCl associated with VNZ–TEZ–D-IVA treatment compared with ELX-TEZ-IVA treatment, with an LS mean difference of −8.4 mmol/L (95% CI, −10.5 to −6.3 mmol/L; P < 0.0001).

In the SKYLINE 103 trial, the LS mean absolute change from baseline in SwCl through week 24 in the FAS was −5.1 mmol/L (SD = 0.7 mmol/L) in the VNZ–TEZ–D-IVA group (n = 294) compared with −2.3 mmol/L (SD = 0.7 mmol/L) in the ELX-TEZ-IVA group (n = 289). The MMRM analysis reported a reduction (i.e., improvement) in SwCl associated with VNZ–TEZ–D-IVA treatment compared with ELX-TEZ-IVA treatment, with an LS mean difference of −2.8 mmol/L (95% CI, −4.7 mmol/L to −0.9 mmol/L; P = 0.0034).

In the single-arm study, RIDGELINE, the within-group LS mean absolute change from baseline through week 24 (averaging weeks 16 and 24) was −8.6 mmol/L (95% CI, −11.0 mmol/L to −6.3 mmol/L) in the cohort B1 FAS (n = 77).

Fifty-Two Weeks

In the SKYLINE 102 trial, the LS mean absolute change from baseline in SwCl through week 52 in the FAS was −7.5 mmol/L (SD = 0.7 mmol/L) in the VNZ–TEZ–D-IVA group (n = 196) compared with 0.5 mmol/L (SD = 0.7 mmol/L) in the ELX-TEZ-IVA group (n = 202). The MMRM analysis reported a reduction (i.e., improvement) in SwCl associated with VNZ–TEZ–D-IVA treatment, with an LS mean difference of −8.0 mmol/L (95% CI, −9.9 mmol/L to −6.1 mmol/L).

In the SKYLINE 103 trial, the LS mean change in the VNZ–TEZ–D-IVA group (n = 284) was −5.0 mmol/L (SD = 0.6 mmol/L) and, in the ELX-TEZ-IVA group, it was −2.2 mmol/L (SD = 0.6 mmol/L). The MMRM analysis reported a reduction (i.e., improvement) in SwCl associated with VNZ–TEZ–D-IVA treatment, with an LS mean difference of −2.8 mmol/L (95% CI, −4.6 mmol/L to −1.0 mmol/L).

Proportion of Patients With an SwCl of 30 mmol/L or Lower

The 24-week pooled outcome, including data from both the SKYLINE 102 and SKYLINE 103 studies, reported an odds ratio of 2.87 (95% CI, 2.00 to 4.12; P < 0.0001), indicating patients treated with VNZ–TEZ–D-IVA were more likely to have an SwCl of 30 mmol/L or lower at week 24 than patients treated with ELX-TEZ-IVA.

In the RIDGELINE trial, 41 out of 78 patients (52.6%) had an SwCl of 30 mmol/L or lower at week 24 (95% CI, 40.9% to 64.0%).

The 52-week result in the SKYLINE 102 trial was an odds ratio of 5.77 (95% CI, 3.33 to 9.99), and in the SKYLINE 103 trial was 1.98 (95% CI, 1.36 to 2.88). While the magnitude of difference between the treatment arms appears to vary between the 2 studies, the trend was in the same direction and aligned with the 24-week outcome results reported in these studies.

Subgroup and Sensitivity Analyses

No subgroup or sensitivity analyses were reported.

Table 17: Summary of Key Efficacy Results From the 2 RCTs, SKYLINE 102 and SKYLINE 103 (Full Analysis Sets)

BMI = body mass index; CFQ-R = Cystic Fibrosis Questionnaire-Revised; CI = confidence interval; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; FAS = full analysis set; LS = least squares; PFAS = pooled full analysis set; ppFEV₁ = percent predicted forced expiratory volume in 1 second; RCT = randomized controlled trial; RD = respiratory domain; SD = standard deviation; SE = standard error; SwCl = sweat chloride; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor; vs. = versus.

Note: Analyses were based on the FAS unless noted otherwise. FAS was defined as all randomized patients who carry the intended CFTR mutation(s) and received at least 1 dose of the study drug during the treatment period. PFAS was defined as the pooled FAS from the SKYLINE 102 and SKYLINE 103 studies.

Baseline was defined as the predose day 1 value. If the predose day 1 value was missing, the most recent predose nonmissing value on or after the day −14 visit, including unscheduled visits, was used as baseline.

Sources: Clinical Study Reports.^9,10

Table 18: Summary of Key Efficacy Results From the Single-Arm Study, RIDGELINE 103

BMI = body mass index; CFQ-R = Cystic Fibrosis Questionnaire-Revised; CI = confidence interval; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; FAS = full analysis set; LS = least squares; PFAS = pooled full analysis set; ppFEV₁ = percent predicted forced expiratory volume in 1 second; RD = respiratory domain; SD = standard deviation; SE = standard error; SwCl = sweat chloride; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

Notes: Baseline was defined as the predose day 1 value. For patients who were on stable ELX-TEZ-IVA, if the predose day 1 value was missing, the most recent available predose value, including the screening assessment, was used as baseline.

Source: Clinical Study Report.¹¹

Harms

Harms data are presented for the SKYLINE 102 and SKYLINE 103 trials in Table 19 and in Table 20 for the single-arm RIDGELINE trial.

To contextualize the safety data from both the SKYLINE 102 and SKYLINE 103 trials, it is important to note that the majority of patients had received commercial ELX-TEZ-IVA before study enrolment, with a median exposure of approximately 2 years, and all patients received ELX-TEZ-IVA for 4 weeks during the run-in period. Additionally, patients who had a prior history of intolerance to ELX-TEZ-IVA were not eligible to enrol, and patients who developed intolerance to ELX-TEZ-IVA during the run-in period were discontinued from the study. Therefore, the safety data from the ELX-TEZ-IVA group are reflective of the experience in patients who have tolerated ELX-TEZ-IVA and continued on an existing treatment regimen, and the safety data from the VNZ–TEZ–D-IVA group are reflective of patients who were switched to a new treatment regimen. Similarly, in the RIDGELINE trial, all patients either had prior experience on ELX-TEZ-IVA or participated in a run-in period in which they received ELX-TEZ-IVA and then switched to VNZ–TEZ–D-IVA during the study period.

Adverse Events

In the SKYLINE 102 and SKYLINE 103 trials, 94.4% and 96.5% of patients, respectively, who were treated with VNZ–TEZ–D-IVA experienced an AE, while 97% and 94.5% of patients, respectively, who were treated with ELX-TEZ-IVA experienced an AE. In both studies, the most common AEs were infective pulmonary exacerbations related to CF (range, 26.8% to 34.7% across the treatment groups in both studies), COVID-19 (range, 20.4% to 26.7%), cough (range, 20.3% to 23.0%), and nasopharyngitis (range, 17.3% to 23.0%).

In the RIDGELINE trial, the proportion of patients with at least 1 AE was 70.6% and 96.2% in cohorts A1 and B1, respectively.

Serious Adverse Events

In the SKYLINE 102 and SKYLINE 103 trials, among patients treated with VNZ–TEZ–D-IVA, 14.3% and 14.1%, respectively, experienced an SAE, while 20.3% and 13.8% of patients, respectively, who were treated with ELX-TEZ-IVA experienced an SAE. The proportion of patients with any SAE appears elevated in the ELX-TEZ-IVA group of the SKYLINE 102 trial relative to the VNZ–TEZ–D-IVA group, but this was not the case in the SKYLINE 103 trial, where the values were similar across the 2 treatment groups. One life-threatening AE occurred, which was an infective pulmonary exacerbation of CF in a patient in the ELX-TEZ-IVA treatment group of the SKYLINE 103 trial.

In the 2 RCTs, the most common SAE was an infective pulmonary exacerbation of CF. In the SKYLINE 102 trial, this occurred in 5.6% of patients treated with VNZ–TEZ–D-IVA and 11.4% treated with ELX-TEZ-IVA. In the SKYLINE 103 trial, this occurred in 6.3% and 4.2% of patients, respectively. Other SAEs were individually uncommon (from 0 to approximately 2% of patients for each SAE) but included influenza, hemoptysis, pneumonia, suicidal ideation, syncope, COVID-19, ALT increased, ALT or AST increased, constipation, distal intestinal obstruction syndrome, gamma-glutamyl transferase increased, cholelithiasis, and nephrolithiasis.

In the RIDGELINE trial, no patients in cohort A1 had an SAE, and 6 patients (7.7%) in cohort B1 had an SAE. In cohort B1, the SAEs included infective pulmonary exacerbations (n = 2); among these 2 patients, 1 also had an SAE of failure to thrive. Other SAEs that each occurred in 1 patient (1.3%) were adenovirus infection, constipation, pulmonary function test decreased, and cough. No life-threatening AEs occurred.

Withdrawals Due to Adverse Events

In the SKYLINE 102 trial, 2.0% of patients treated with VNZ–TEZ–D-IVA and 4.5% of patients treated with ELX-TEZ-IVA had an AE leading to treatment discontinuation. In the SKYLINE 103 trial, this occurred in 4.9% and 3.1% of patients, respectively. The AEs leading to treatment discontinuation were each individually uncommon (0% to approximately 2% of patients for each AE) and included a variety of types of AEs.

In the RIDGELINE trial, 1 patient (1.3%) had treatment interrupted due to a seizure, and 1 patient stopped treatment due to AEs of fatigue and coughing. Both patients belonged to cohort B1.

Mortality

No deaths occurred in the SKYLINE 102, SKYLINE 103, or RIDGELINE trials.

Notable Harms

Identified AEs of special interest included elevated aminotransferase levels, rash, elevated creatine kinase levels, cataracts, hypoglycemia, and neuropsychiatric events.

In the VNZ–TEZ–D-IVA and ELX-TEZ-IVA groups of the SKYLINE 102 trial, the proportion of patients with AEs of elevated aminotransferase levels was 8.2% and 6.4%, respectively. Rash occurred in 8.7% and 7.4%, elevated creatinine occurred in 9.2% and 11.9%, hypoglycemia occurred in 2.6% and 3.5%, and neuropsychiatric AEs occurred in 8.7% and 13.9%, respectively. No patients in either treatment group had AEs of █████████. No patients treated with VNZ–TEZ–D-IVA had serious ████████ ████████████, rash, creatine kinase elevation, ████████, or hypoglycemia events. Serious neuropsychiatric events occurred in 3 patients (1.5%) treated with VNZ–TEZ–D-IVA and in 2 patients (1.0%) treated with ELX-TEZ-IVA.

In the VNZ–TEZ–D-IVA and ELX-TEZ-IVA groups of the SKYLINE 103 trial, the proportion of patients with AEs of elevated aminotransferase levels was 9.5% and 7.6%, respectively. Rash occurred in 12.7% and 8.0%, elevated creatine kinase levels occurred in 8.8% and 5.9%, cataracts occurred in ████ ███ ████, hypoglycemia occurred in 1.1% and 3.8%, and neuropsychiatric AEs occurred in 13.4% and 10.7%, respectively. Serious neuropsychiatric events occurred in 1 patient (0.4%) treated with VNZ–TEZ–D-IVA and 2 patients (0.7%) treated with ELX-TEZ-IVA. Serious elevated aminotransferase–related events occurred in ███ ██████ patients treated with VNZ–TEZ–D-IVA and ████ treated with ELX-TEZ-IVA. Events of serious elevated creatine kinase occurred in 1 patient (0.4%) in the VNZ–TEZ–D-IVA group and in no patients in the ELX-TEZ-IVA group. No patients had serious rash, ████████, or hypoglycemia events.

In cohort A1 of the RIDGELINE trial, there were no events of ALT increased, AST increased, elevated creatine kinase levels, █████████, hypoglycemia, or neuropsychiatric AEs, while 3 patients (17.6%) had rash. In cohort B1, ALT increase occurred in 5.1% of patients, AST increase occurred in 2.6%, rash occurred in 5.1%, elevated creatine kinase levels occurred in 2.6%, cataracts occurred in ████, and neuropsychiatric events occurred in 5.1%; no patients in cohort B1 had hypoglycemia. No patients in either cohort had serious ████████ ████████████, rash, creatine kinase, ████████, hypoglycemia, or neuropsychiatric events.

Table 19: Summary of Harms Results in the SKYLINE 102 and SKYLINE 103 Trials (Safety Analysis Set)

AE = adverse event; ALT = alanine aminotransferase; AST = aspartate aminotransferase; CF = cystic fibrosis; DIOS = Distal Intestinal Obstruction Syndrome; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; GGT = gamma-glutamyl transferase PT = preferred term; SAE = serious adverse event; TEAE = treatment-emergent adverse event; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

Sources: Clinical Study Reports.^9,10

Table 20: Summary of Harms Results in the RIDGELINE Trial (Safety Analysis Set)

AE = adverse event; ALT = alanine aminotransferase; AST = aspartate aminotransferase; CF = cystic fibrosis; COVID-19 = coronavirus disease 2019; NR = not reported; PEx = pulmonary exacerbation; PT = preferred term SAE = serious adverse event; TEAE = treatment-emergent adverse event; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

^aThe AEs displayed here are those experienced by at least 5% of the patients in cohort B1. Due to the small sample size of cohort A1, a single patient represents greater than 5% of the cohort’s population.

Source: Clinical Study Report.¹¹

Critical Appraisal

Internal Validity

SKYLINE 102 and SKYLINE 103 Trials

The SKYLINE 102 and SKYLINE 103 trials were double-blind, parallel-group, randomized active-controlled trials. The treatments were administered in a double-blind manner, including the use of identical placebo formulations to prevent unblinding due to the twice-daily dosing schedule of ELX-TEZ-IVA versus the once-daily dosing schedule of VNZ–TEZ–D-IVA. The similarity in mechanism of action, expected efficacy, and expected AEs between the 2 therapies further reduced the likelihood of unintentional unblinding. Moreover, all patients received ELX-TEZ-IVA during the run-in period, and the majority had prior experience with ELX-TEZ-IVA or other CFTRm therapy, minimizing the potential for treatment inference based on subjective experience.

Randomization and allocation concealment were conducted in an appropriate manner, and randomization was stratified by relevant factors. Patient disposition was also generally well balanced between the arms. A few patients were randomized but not dosed: in the SKYLINE 102 trial, 4 in the VNZ–TEZ–D-IVA group and 3 in the ELX-TEZ-IVA group; in the SKYLINE 103 trial, 1 in the VNZ–TEZ–D-IVA group and in no patients in the ELX-TEZ-IVA group, indicating a low risk of bias due to postrandomization exclusions.

Discontinuation of treatment or study was under 10% in all groups across the 2 RCTs and the reasons were generally balanced. However, in the SKYLINE 102 trial, more patients discontinued treatment due to AEs in the ELX-TEZ-IVA group than the VNZ–TEZ–D-IVA group. This may just be attributable to random variation rather than a systematic issue, especially as it was not observed in the SKYLINE 103 trial.

Baseline demographic and disease-related characteristics were generally well balanced across treatment groups in both the SKYLINE 102 and SKYLINE 103 trials. Important protocol deviations were observed in 5% or fewer patients in each study and were not differentially distributed between the treatment groups or apparently driven by any systemic bias. The efficacy and safety results were generally consistent between the SKYLINE 102 and SKYLINE 103 trials, with minor differences that may be due to chance, such as the previously mentioned higher number of AE-related discontinuations in the ELX-TEZ-IVA group in the SKYLINE 102 trial.

Analyses of the efficacy outcomes of interest (which includes lung function [ppFEV₁], body weight [BMI and BMI z score], HRQoL [CFQ-R respiratory domain], and SwCl at weeks 24 and 52) were conducted using MMRM using restricted maximum likelihood, and adjustment factors included fixed categorical effects for treatment, visit, age at screening (except for in the assessment of BMI z score which included only patients aged 20 years and younger), and treatment-by-visit interaction, with baseline ppFEV₁ and baseline SwCl as continuous covariates. Notably, the sample size for BMI z score was very small (n = 32 to 53 in each treatment group) owing to the inclusion of only patients aged 20 years or younger, and this age category does not reflect the randomization stratification factors in the study design (which was divided by age 18 rather than 20 years), which suggests it may break randomization.

The primary end point of ppFEV₁ at 24 weeks, and the key secondary end points of absolute change from baseline in SwCl through week 24, the proportion of patients with an SwCl level of 60 mmol/L or less at week 24, and the proportion of patients with an SwCl level of 30 mmol/L or less at week 24, were controlled for type I error using a hierarchical testing procedure. However, this does not appear to apply to the 52-week outcomes for the change in ppFEV₁ or change in SwCl, which were the related end points that CDA-AMC assessed using GRADE, nor does it apply to any of the other outcomes assessed in this report. Nonetheless, the consistency between the 24-week and 52-week end point for the change in ppFEV₁ bolsters confidence in the results.

The primary end point of change in ppFEV₁ at 24 weeks (and secondary at 52 weeks) was a noninferiority assessment tested at a 1-sided alpha level of 0.025, which was met according to the prespecified noninferiority margin of −3%. The selection of a 3% noninferiority margin, based on the approach used, is consistent with FDA and European Medicines Agency guidance and has been used in previous noninferiority studies of treatments for CF.^7,8 As mentioned previously, a 5% improvement in ppFEV₁ has been recognized by CDA-AMC and other health technology assessment bodies and Cystic Fibrosis Canada as an MID, indicating that the 3% margin remains within the bounds of clinical relevance and supports the appropriateness of the noninferiority margin. The clinical experts consulted by CDA-AMC agreed that the noninferiority margin used in the trials was appropriate. The statistical plan for the primary end point was structured in the estimand framework, which was appropriate. The treatment policy strategy was used to handle the use of a nonstudy drug CFTRm for greater than 3 days during the run-in or treatment periods before week 24, as well as treatment discontinuation before week 24, which means that observed ppFEV₁ values were used if available after the event. This approach was considered appropriate because it better reflects the effectiveness in a real-world setting where treatment discontinuation and modifying therapies are a part of practice. A sensitivity analysis was conducted for the primary end point in which intercurrent events were instead addressed using the hypothetical strategy, meaning that values were set to ‘missing’ after the intercurrent event, and the results were aligned. Only 2 intercurrent events were defined; ergo, it is hypothetically possible there could be overlooked sources of potential bias; however, the intercurrent events described were expected by CDA-AMC to be the most common and impactful events because deaths did not occur during these studies and changes in therapies other than CFTRm treatments are expected to be less impactful on the efficacy results due to the nature of the disease and the typical roster of concomitant medications used in CF. As such, this did not raise major concerns regarding increased risk of bias.

Other outcomes were assessed using 1-sided alpha levels of 0.05. The end point of pulmonary exacerbation rate did not appear to be calculated using a similar MMRM as described previously, which may suggest it was not adjusted for similar factors and may raise the risk of bias for that end point.

All missing data were assumed to be missing at random and were not imputed, and there was a generally low number of patients (less than 10%) with missing data, and this was well balanced between the treatment groups for each end point as of the 52-week outcomes in both studies. Given the expected similarity in the treatment efficacy and safety profiles, this aligned with expectations, albeit there was a slight imbalance in discontinuations due to AEs in 1 study, as previously mentioned.

The 52-week end points assessed using GRADE were not included in the hierarchical testing procedure and were not adjusted for multiplicity. This may raise concerns for an increased risk of type I error for these end points, but it was not considered to be particularly significant in the context of these noninferiority studies.

There were no concerns regarding elevated risk of bias with regard to the measurement of outcomes or selection of reported results for any of the study end points. The ppFEV₁ is standardized and repeatable, and all outcomes assessed herein, other than the CFQ-R respiratory domain, are objective. The CFQ-R is a self-reported (or caregiver-reported) metric that is relatively well validated in assessing the HRQoL of patients with CF. Although these data are generated by either a child or caregiver, the double-blind randomized nature of the study mitigates the elevated risk of bias in this type of end point when comparing between treatment groups, except in the event of accidental unblinding, which was not considered to be likely in general.

Overall, the risk of bias for the SKYLINE 102 and SKYLINE 103 trials was considered low to moderate.

RIDGELINE Trial

The RIDGELINE study was a single-arm, noncomparative, nonrandomized, open-label study that assessed outcomes similar to the SKYLINE 102 and SKYLINE 103 studies in terms of within-group change from baseline after 24 weeks of therapy with VNZ–TEZ–D-IVA. There was no formal primary end point, no adjustment for multiplicity or hierarchical testing procedure, and no sensitivity or subgroup analyses were conducted. The MMRM approach was similar to that previously described, but with fixed categorical effects for genotype group and visit, and baseline SwCl as a continuous covariate. All patients either had prior treatment experience with ELX-TEZ-IVA or participated in a run-in period with it before the treatment period, during which they then received VNZ–TEZ–D-IVA. There was a very low rate of missing data (which were assumed to be missing at random and were not imputed) and a very low rate of discontinuations.

Because this is a noncomparative, nonrandomized study, it is impossible to conclude whether the observed effects can be attributable to the effects of VNZ–TEZ–D-IVA. As such, the risk of bias is high and the degree of certainty in the results in comparison with ELX-TEZ-IVA is very low. This is especially the case for the self-reported (or caregiver-reported) HRQoL metric of the CFQ-R respiratory domain.

External Validity

The SKYLINE 102 and SKYLINE 103 trials reflect the patient population aged 12 years and older with CF, while patients aged 6 to 11 years (inclusive) were recruited for cohorts A1 and B1 of the RIDGELINE study. The eligibility criteria and baseline characteristics were considered appropriate and reflective of the patient population with CF in Canada within those age groups and those expected to be eligible for treatment with VNZ–TEZ–D-IVA, based on clinical expert input. The intervention, comparator, and treatment setting were relevant to the study question and to real-world clinical practice because the active comparator, ELX-TEZ-IVA, is the standard of care for patients with CF who have eligible CF-causing genotypes and is received by the majority of patients with CF in Canada, in the setting of specialty CF clinics. Concomitant medications were evaluated by the clinical experts to be reflective of their experience in treating patients with CF in Canada, and the evaluated end points captured clinically relevant data, albeit some end points are not routinely assessed in clinical practice (e.g., CFQ-R respiratory domain and testing for SwCl on an ongoing basis after diagnosis), although they are still informative and useful in the context of a clinical trial setting for regulatory and reimbursement decision-making. The end point of SwCl is typically assessed in clinical studies of CF, but is not regularly assessed in clinical practice after diagnosis, and the importance of further reductions in SwCl with VNZ–TEZ–D-IVA relative to that seen with ELX-TEZ-IVA has not been established; moreover, the magnitude of difference in SwCl observed in these studies was not considered to be clinically important by the CF experts consulted by CDA-AMC, based on the data available to date.

Because all of the included patients had treatment experience with ELX-TEZ-IVA and/or participated in run-in periods with ELX-TEZ-IVA treatment, and patients intolerant to or ineligible for ELX-TEZ-IVA were not included, the efficacy and safety data during the treatment period of the studies reflect the experience of patients switching from 1 CFTRm to another (if assigned to VNZ–TEZ–D-IVA), or reflects the experience of patients continuing on a CFTRm therapy to which they have demonstrated tolerance (if assigned to ELX-TEZ-IVA in the 2 RCTs). This represents a gap in the generalizability of these studies because no data were captured on patients naive to CFTRm therapy who were beginning treatment with VNZ–TEZ–D-IVA. However, this is expected to be a minority of patients, as per clinical expert input. Some proportion of these patients in the real world would be those whose genotypes are rare and not captured in any phase III clinical trial of CF, which is a generalizability issue that was also present in the clinical trial program of ELX-TEZ-IVA. Data for rarer genotypes will be discussed in the next section, wherein in vitro evidence was considered in lieu of clinical data.

For RIDGELINE, the 24-week single-arm study conducted in patients aged 6 to 11 years, the generalizability of results such as ppFEV₁ and weight-related outcomes may be potentially confounded by the shorter study duration due to the propensity for pediatric patients to catch seasonal illnesses.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered most relevant to inform expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group.^71,72

• High certainty: We are very confident that the true effect lies close to that of the estimate of the effect.

• Moderate certainty: We are moderately confident in the effect estimate. The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. We use the word “likely” for evidence of moderate certainty (e.g., “X intervention likely results in Y outcome”).

• Low certainty: Our confidence in the effect estimate is limited. The true effect may be substantially different from the estimate of the effect. We use the word “may” for evidence of low certainty (e.g., “X intervention likely results in Y outcome”).

• Very low certainty: We have very little confidence in the effect estimate. The true effect is likely to be substantially different from the estimate of effect. We describe evidence of very low certainty as “very uncertain.”

For RCTs: Following the GRADE approach, evidence from RCTs started as high-certainty evidence and could be rated down for concerns related to study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias.

For single-arm trials: Although GRADE guidance is not available for noncomparative studies, the CDA-AMC review team assessed pivotal single-arm trials for study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias to present these important considerations. Because the lack of a comparator arm does not allow for a conclusion to be drawn on the effect of the intervention versus any comparator, the certainty of evidence for single-arm trials started at very low certainty, with no opportunity for rating up.

When possible, certainty is typically rated in the context of the presence of an important (nontrivial) treatment effect; if this is not possible, certainty is rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. In the context of the current assessment, the objective of the studies was to establish noninferiority, so unless signalled otherwise, the certainty was generally rated in the context of suggesting there may be little to no difference between treatments for a given outcome, where applicable. For most outcomes, the target-of-certainty assessment was therefore null. As described in Table 10, an MID of 4 points was identified for the CFQ-R respiratory domain.

Findings from the SKYLINE 102 and SKYLINE 103 trials, the RCTs that recruited patients aged 12 years or older, were considered together and summarized narratively in Table 2 because these studies were similar in populations, interventions, design, and outcome measures. The RIDGELINE trial was considered separately in Table 3, given the different patient population (aged 6 to 11 years), different end points (24 weeks rather than 52 weeks), and single-arm design.

Long-Term Extension Studies

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

No long-term extension studies with results were submitted, although there are some presently ongoing for which results data are not yet available.

Indirect Evidence

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

The sponsor submitted indirect comparisons between VNZ–TEZ–D-IVA and best supportive care. Given the presence of direct evidence against ELX-TEZ-IVA, which is the standard of care, as previously discussed, this indirect comparison with supportive care was considered by CDA-AMC to have limited relevance; ergo, it was not summarized or critically appraised for the purpose of this review.

Studies Addressing Gaps in the Systematic Review Evidence

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

Description of Studies

The sponsor submitted 4 nonclinical studies evaluating the effect of VNZ–TEZ–D-IVA on cells in an in vitro setting. For CFTR mutations with low prevalence, the submission suggested that a conventional clinical trial approach to demonstrate efficacy is not always feasible.^2,12 To evaluate the potential clinical benefit of VNZ–TEZ–D-IVA in these circumstances, additional mutations were identified using 3 methods: the in vitro FRT system, the in vitro human bronchial epithelial primary cell culture model, and mechanistic data.¹³

The FRT assay was used to identify additional CFTR mutations for which it is predicted that VNZ–TEZ–D-IVA may offer clinical benefit. The FRT assay was developed to systematically characterize the responsiveness of individual CFTR mutations to CFTRm therapies.⁷³ This in vitro FRT cell model has been used previously in the approval of ELX-TEZ-IVA in Canada and in other countries.

Using the FRT assay, the function of the CFTR at the cell surface was assessed in Ussing chambers, which quantify the CFTR-mediated transport in FRT cells expressing each mutant CFTR form as a fraction of the chloride transport in FRT cells expressing normal CFTR (percentage of normal). In vitro responsiveness in the FRT cell model is based on an increase of 10% or more over baseline in chloride transport as a percentage of normal CFTR chloride transport. This threshold was selected because analyses of CFTR activity and disease phenotype showed that a 10% increase in CFTR activity is typically associated with improved clinical status,⁵⁶ and the clinical experts consulted by CDA-AMC concurred that this is an appropriate threshold; it was also the threshold applied in previous reviews of ELX-TEZ-IVA. The sponsor stated that this is a conservative threshold, and some mutations that did not meet this threshold were further studied.⁷⁴

The N1303K CFTR mutation previously did not meet the 10% threshold for responsiveness in in vitro assessments of ELX-TEZ-IVA submitted in previous reviews, but real-world observational evidence suggested some potential for responsiveness in clinical outcomes for patients with the N1303K genotype when treated with ELX-TEZ-IVA. To further validate predicted efficacy for the N1303K CFTR mutation in the context of treatment with VNZ–TEZ–D-IVA, it was tested using in vitro human bronchial epithelial cells from 3 CF donors, 1 with a homozygous N1303K mutation, and 2 with a heterozygous N1303K and a minimal function mutation, in addition to clinical benefit demonstrated in published observational studies.

Results

Based on all data available, there were 303 CFTR mutations that were considered by the sponsor to be responsive to VNZ–TEZ–D-IVA; of these, 26 were confirmed as responsive by the clinical data from the phase III clinical studies. An overview of the study designs and key results is presented in Table 21, and a listing of all mutations considered by the sponsor to be responsive to VNZ–TEZ–D-IVA based on the in vitro assessments is presented in Table 22.

Table 21: Summary of Nonclinical Studies Submitted With Key Results

CF = cystic fibrosis; CI = chloride; DIVA = deuterated ivacaftor (deutivacaftor); EC = effective concentration; ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; F/F = homozygous for F508del; F/G = heterozygous for F508del and a gating mutation; F/MF = heterozygous for F508del and a minimal function mutation; F/other = heterozygous for F508del and a non-F508del mutation that does not fit into 1 of the other specified mutation categories (i.e., gating, residual function, minimal function, triple combination–responsive); FRT = Fischer rat thyroid; HBE = human bronchial epithelial; IVA = ivacaftor; LUM-IVA = lumacaftor-ivacaftor; SwCl = sweat chloride; TEZ = tezacaftor; TEZ-IVA = tezacaftor-ivacaftor; VNZ = vanzacaftor; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

^aFRT cells were each engineered to express 1 of the rare mutations, and the following control FRT cell lines were included as comparators for the western blot and Ussing chamber studies: untransfected (parental) FRT cells (used to establish a baseline and noise threshold for the system); wild-type (normal) CFTR (used to normalize CI transport to normal [% normal]), and F508del CFTR mutation (the most common CF-causing mutation and the subject of FDA approvals for LUM-IVA, TEZ-IVA, and ELX-TEZ-IVA therapies). The positive controls were 2 CFTR mutations (G551D and R117H) that were responsive in previous FRT studies and have been demonstrated in clinical trials to be IVA-responsive (studies 770-102, 770-103, and 770-110), and the negative controls were 3 CFTR mutations (G1061R, R1066C, and N1303K) that were not responsive to IVA or TEZ-IVA in previous in vitro studies.

Sources: Details included in the table are from the sponsor’s Summary of Clinical Evidence and Nonclinical Study Reports.^75-78

Table 22: Summary of Mutations Suggested by the Sponsor as Being Responsive to VNZ–TEZ–D-IVA Based on the Sponsor’s Interpretation of the Submitted Nonclinical Studies

ELX-TEZ-IVA = elexacaftor-tezacaftor-ivacaftor and ivacaftor; FRT = Fischer rat thyroid; IVA = ivacaftor; TEZ-IVA = tezacaftor-ivacaftor; VNZ–TEZ–D-IVA = vanzacaftor-tezacaftor-deutivacaftor.

^aComplex/compound mutations in which a single allele of the CFTR gene has multiple mutations; these exist independent of the presence of mutations on the other allele.

Source: Sponsor-submitted Summary of Clinical Evidence.⁷⁹

Perspective of the Clinical Experts

The clinical experts consulted for this review noted that responsiveness to VNZ–TEZ–D-IVA using the in vitro model applied by the sponsor in the clinical development program for patients with rare CFTR mutations is sufficient evidence to support initial prescribing to these patients. In addition to regulatory approval based on this information (e.g., FDA approvals), the clinical experts cited the following considerations that would support the use of VNZ–TEZ–D-IVA in clinical practice:

• There is a lack of alternative treatment options for some of these patients, particularly those who are not eligible for or cannot tolerate ELX-TEZ-IVA but may potentially experience a response to VNZ–TEZ–D-IVA based on their CF-causing genotype.

• The nature of CF is severe, progressive, and life-limiting.

• In vitro data demonstrating activity on the CFTR channel for patients with these rare mutations support extrapolation of the clinical benefit demonstrated in studies involving more common CF-causing mutations in the CFTR gene.

• There are concerns about equity for those living with CFTR mutations for which the incidence is sufficiently low to preclude the generation of robust clinical evidence. The clinical experts noted that CF with a non-F508del mutation is more likely to be diagnosed in racialized persons with CF who may already be encountering systemic disadvantages within the health care system. Based on the rarity of these mutations, it would be impossible to enrol patients with each mutation into clinical trials. The current focus on clinical trial data alone for drug approval is resulting in ethnic and racial inequity in medication access. It is well known that these patient groups are underrepresented in clinical trials for a multitude of complex reasons.

• Based on prior experience with ELX-TEZ-IVA for the treatment of CF in Canada and the rest of the world, there is a mounting body of evidence from case reports, case series, and clinical experience that ELX-TEZ-IVA can have a clinically meaningful impact on patients with rare CFTR mutations shown to be responsive based on in vitro data, which supports the use of this in vitro data in this manner in the context of VNZ–TEZ–D-IVA. The clinical experts consulted by CDA-AMC noted that, in this small minority of patients, it may be appropriate to conduct repeat sweat testing to build evidence for responsiveness to CFTRm therapy, although that is not expected to be standard clinical practice for the majority of patients.

Health Canada Conclusions

The 266 CFTR-responsive mutations listed in the product monograph for Alyftrek (VNZ–TEZ–D-IVA) are provided in Table 23.

Table 23: CFTR Mutations Included in the Health Canada Indication and Listed in the Current Product Monograph as Responsive to VNZ–TEZ–D-IVA^a