Reimbursement Review

Evolocumab (Repatha)

Requester: Amgen Canada Inc.

Therapeutic area: Primary hyperlipidemia

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

ACS = acute coronary syndrome; ASCVD = atherosclerotic cardiovascular disease; HeFH = heterozygous familial hypercholesterolemia; LDL-C = low-density-lipoprotein cholesterol; NOC = Notice of Compliance.

^aPrefilled syringe is not available in Canada.

Sources: Product monograph for evolocumab.¹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

Introduction

Hyperlipidemia refers to high levels of lipids in the blood, including cholesterol and triglycerides. High levels of cholesterol (also referred to as hypercholesterolemia),³ notably low-density-lipoprotein cholesterol (LDL-C), can cause atherosclerosis, defined as the buildup of fatty deposits in blood vessels leading to restriction in blood flow, which is a major cause of cardiovascular events, including heart attack, stroke, and lower-extremity and peripheral artery disease (PAD).⁴ Atherosclerotic cardiovascular disease (ASCVD), as defined in the 2021 Canadian Cardiovascular Society (CCS) Guidelines for the Management of Dyslipidemia for the Prevention of Cardiovascular Disease in Adults⁵ (hereafter referred to as the 2021 CCS dyslipidemia guidelines), comprises all clinical conditions of atherosclerotic origin, such as acute coronary syndrome (ACS), stroke, and PAD. After the first documented (index) case of ACS, a residual risk of a subsequent cardiovascular event remains.⁶ Secondary prevention refers to the treatment and management of known, clinically evident ASCVD and the prevention or delay of the onset of disease manifestations.⁵ The incidence rate for myocardial infarction (MI) was approximately 2.5 per 1,000 person-years from 2005 to 2016 in Ontario, whereas the incidence rate for unstable angina was 3.3 per 1,000 person-years in 2005 and 1.7 per 1,000 person-years in 2016.⁷ The 10-year prevalence rates for MI increased from 23.5 to 26.9 per 1,000 individuals and for unstable angina increased from 22.1 to 23.7 per 1,000 individuals between the periods of 2004 to 2013 and 2008 to 2017.⁷

ASCVD is a statin-indicated condition, according to the 2021 CCS dyslipidemia guidelines.⁵ For patients with ASCVD, the guidelines advise that proprotein convertase subtilisin kexin type 9 (PCSK9) inhibitors, with or without ezetimibe, be considered when the necessary reduction in LDL-C, apolipoprotein B (ApoB), or non–high-density-lipoprotein cholesterol (HDL-C) is substantial (i.e., LDL-C > 2.2 mmol/L or ApoB > 0.80 g/L or non-HDL-C > 2.9 mmol/L despite maximally tolerated statin dose) or for patients who have been shown to derive the largest benefit from the intensification of statin therapy with PCSK9 inhibitor therapy. This subset includes patients with recent ACS (i.e., occurring in the 52 weeks after hospitalization for the index ACS), as well as those with clinically evident ASCVD and any additional cardiovascular risk enhancers. If the necessary reduction in LDL-C, ApoB, or non-HDL-C is modest (i.e., LDL-C of 1.8 to 2.2 mmol/L or ApoB of 0.70 to 0.80 g/L or non-HDL-C 2.4 to 2.9 mmol/L despite maximally tolerated statin dose), then the guidelines advise that ezetimibe, with or without a PCSK9 inhibitor, be considered.⁵ According to the clinical experts consulted by Canada’s Drug Agency (CDA-AMC) for the purpose of this review, other lipid-lowering therapies, such as niacin, fibrates, bile acid sequestrants, mipomersen (not approved in Canada), and lomitapide (only used for homozygous familial hypercholesterolemia), are infrequently used in patients with ASCVD.

In 2016, evolocumab (Repatha) was first reviewed by the Canadian Drug Expert Committee (CDEC) for primary hyperlipidemia, including heterozygous familial hypercholesterolemia (HeFH) and clinical ASCVD. CDEC issued a recommendation that evolocumab be listed as an adjunct to diet and maximally tolerated statin therapy in adult patients with HeFH who require additional lowering of LDL-C if the prespecified clinical criteria and conditions are met. For the ASCVD component of the indication, CDEC issued a recommendation that evolocumab not be listed as an adjunct to diet and maximally tolerated statin therapy in adult patients with clinical ASCVD who require additional lowering of LDL-C.⁸ Detailed information on and reasons for the final recommendation made in 2016 by CDEC are publicly available on the CDA-AMC website.

In 2017, evolocumab was resubmitted and reviewed by CDEC for the ASCVD component of primary hyperlipidemia. CDEC issued a recommendation that evolocumab be reimbursed as an adjunct to diet and maximally tolerated statin therapy in adult patients for ASCVD who require additional lowering of LDL-C if the prespecified criterion and condition are met.⁹ However, funding is not yet in place, as negotiations concluded without an agreement in July 2019.¹⁰ Detailed information on the final recommendation made in 2017 by CDEC is publicly available on the CDA-AMC website.

The clinical experts indicated that most patients with ASCVD, who are therefore at high risk of cardiovascular events, are not meeting LDL-C (or non-HDL-C or ApoB) target levels with available treatment options. Moreover, the clinical experts indicated that nonadherence due to a real or perceived intolerance to high-intensity statins remains a challenge in clinical practice; they estimated 50% of patients discontinue their statin in the year after an ACS event. Thus, the unmet need identified by patients, clinician groups, and the clinical experts is for additional pharmacologic options that are effective in lowering LDL-C with minimal side effects in patients with primary hyperlipidemia (including ASCVD). More specifically, this unmet need is highlighted in patients who experienced recent ACS (in the previous year) and elevated LDL-C levels despite optimized statin therapy.

The objective of the present reassessment of evolocumab (ASCVD indication) is to review and critically appraise the new evidence submitted by the sponsor on the beneficial and harmful effects of evolocumab 140 mg/mL and 120 mg/mL subcutaneous injection in the treatment of primary hyperlipidemia, specifically in the requested reimbursement population of adult patients with ASCVD and recent ACS (in the previous year).

ACS comprises non-ST-segment elevation myocardial infarction (NSTEMI), ST-segment elevation myocardial infarction (STEMI), and unstable angina, with MI being the most common clinical presentation.¹¹ The clinical experts were consulted on the definition of ACS in relation to clinical practice. Since the cardiac troponin assays evolved to become highly sensitive to micromolar elevated levels of circulating troponin, unstable angina has become an exceedingly infrequent diagnosis. Thus, only MI, including STEMI and NSTEMI, was considered relevant for the purpose of this review.

Patient, Clinician, and Drug Plan Perspectives

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

Patient Input

No patient groups provided input on the present reassessment of evolocumab.

A summary of past patient input submitted by the Cardiac Health Foundation of Canada was prepared by the review team at CDA-AMC in the Clinical Review Report (Resubmission) on Evolocumab (Repatha) in December 2017, which is publicly available on the CDA-AMC website. The Cardiac Health Foundation of Canada is an organization that raises funds for and promotes programs and applied research on the rehabilitation and management of cardiovascular disease and provides education and resources on the prevention and management of cardiovascular disease in Canada. Patient input was gathered by the patient group through an online survey (N = 55) and 1 telephone interview; respondents were patients with atherosclerosis and their caregivers.

Among the survey respondents, experience with rosuvastatin, atorvastatin, ezetimibe, and bypass surgery were described with varying degrees of effectiveness. The survey respondents reported that the most common side effects associated with their current treatment were digestive-related, including gas, constipation, and upset stomach. According to the survey respondents, the most difficult-to-tolerate side effects associated with current medications were muscle pain, discomfort, and weakness.

The survey respondents identified the following unmet need: alternative treatment options to statins. More specifically, in the context of elevated cholesterol levels despite a maximally tolerated statin dose and adverse events (AEs) commonly associated with statin therapy (i.e., loss of muscle function and muscle weakness), patients expect evolocumab to lower cholesterol levels to target levels with minimal side effects. In particular, most patients indicated that a loss of muscle function is an AE they are not willing to tolerate.

Clinician Input

Input From Clinical Experts

The clinical experts indicated that most patients at high risk for cardiovascular events are not meeting LDL-C (or non–HDL-C or ApoB) target levels with available treatment options. Moreover, the clinical experts indicated that nonadherence due to a real or perceived intolerance to high-intensity statins, such as myalgias, is a challenge in clinical practice; they estimated that 50% of patients discontinue their statin in the year after an ACS event. The clinical experts further highlighted the lack of access to advanced therapies, including PCSK9 inhibitors, experienced by patients with ASCVD.

The clinical experts referenced the 2021 CCS dyslipidemia guidelines,⁵ which indicate that ezetimibe and PCSK9 inhibitors are second-line treatment options in the management of primary hyperlipidemia for secondary prevention. More specifically, the clinical experts indicated that ezetimibe and/or evolocumab would be used in addition to a maximally tolerated statin dose to meet LDL-C (or non-HDL-C or ApoB) target levels. For patients who are intolerant of or who have contraindications to statins, the clinical experts indicated that evolocumab would be an alternative therapy to statins, with or without ezetimibe.

The clinical experts referenced the 2021 CCS dyslipidemia guidelines⁵ to identify the patient population most in need of an intervention for the management of primary hyperlipidemia in secondary prevention, which is the subset of patients with ASCVD (at high cardiovascular risk) who have been shown to derive the largest benefit from the intensification of statin therapy with the addition of a PCSK9 inhibitor. This includes patients with recent ACS (occurring in the 52 weeks after hospitalization for the index ACS) and patients with additional cardiovascular risk enhancers.⁵ Additionally, the clinical experts indicated that all patients with ASCVD whose LDL-C (or non-HDL-C or ApoB) level remains above the threshold despite a maximally tolerated statin dose are suited for treatment with evolocumab.

The clinical experts indicated that although a specialist is not required to diagnose, treat, or monitor patients who receive evolocumab, administration should ideally be carried out in an outpatient clinic or hospital by a clinician who has experience with evolocumab. The clinical experts referenced the LDL-C, non-HDL-C, and ApoB thresholds in the 2021 CCS dyslipidemia guidelines⁵ as the treatment goal. According to the clinical experts, treatment response is based on the reduction in LDL-C (or non-HDL-C or ApoB) levels, assessed every 6 to 12 months in practice, depending on cardiovascular risk. When deciding to discontinue treatment with evolocumab, the clinicals experts would consider the side effects associated with treatment and competing risks from other diseases that limit life expectancy.

Clinician Group Input

A total of 9 clinician groups provided their input on the present reassessment of evolocumab: Canadian Dyslipidemia Guideline Committee; McMaster Lipid Clinic; British Columbia Lipid Specialists; Cambridge Cardiac Rehab Program; Western University, Division of Cardiology, Cardiac Rehabilitation and Secondary Prevention Program; University of British Columbia and Vancouver General Hospital and St. Paul’s Hospital Cardiac Intensive Care Unit; University of Toronto faculty and clinicians at St Michael’s Hospital; Division of Cardiology, University of Ottawa Heart Institute; and a group of primary care and specialist physicians who treat coronary artery disease and ACS across Canada.

The clinician groups identified the following limitations with currently available treatments (unmet needs) for patients with recent ACS: limited access to PCSK9 inhibitors due to cost, experience of side effects and/or intolerance to available drugs (which have an impact on adherence to treatment), and variable treatment response (e.g., treatment targets for LDL-C not met). The University of Ottawa Heart Institute highlighted that although the majority of patients with ASCVD experience a reduction in their LDL-C level to below 1.8 mmol/L with high-dose statin therapy, with or without ezetimibe, a subset of patients continues to have elevated lipid levels due to severe polygenic hypercholesterolemia and intolerance or contraindication to high-dose statin therapy. The clinician group further suggested that this subset of patients who are at high risk of recurrent cardiovascular events would benefit from additional lipid-lowering treatment in the form of a PCSK9 inhibitor.

The Canadian Dyslipidemia Guideline Committee, McMaster Lipid Clinic, and the group of primary care and specialist physicians across Canada referenced the 2021 CCS dyslipidemia guidelines⁵ to indicate that a PCSK9 inhibitor would be used as an add-on therapy after the initiation of maximally tolerated statin therapy, with or without ezetimibe, in patients with elevated LDL-C levels. More specifically, evolocumab would be used in the second line after a maximally tolerated statin dose or in the third line after statin and ezetimibe. The Canadian Dyslipidemia Guideline Committee also referenced the guidelines to identify candidates for evolocumab, comprising patients with either a recent ACS (i.e., occurring in the 52 weeks after hospitalization for the index ACS) or prior ASCVD with any of the following: diabetes or metabolic syndrome, polyvascular disease, symptomatic PAD, recurrent MI, MI in the previous 2 years, previous coronary artery bypass graft surgery, LDL-C level of 2.6 mmol/L or more, or HeFH. The clinician groups indicated that treatment response is assessed using the percent reduction in LDL-C (or non-HDL-C or ApoB) levels from pretreatment levels in practice.⁵

Drug Program Input

Input was obtained from the drug programs that participate in the CDA-AMC Reimbursement Review process. The following items were identified as key factors that could potentially impact the implementation of a recommendation for evolocumab:

• relevant comparators

• consideration for initiation of therapy

• consideration for continuation or renewal of therapy

• consideration for the prescribing of therapy

• system and economic issues.

The clinical experts consulted by CDA-AMC provided advice on the potential implementation issues raised by the drug programs (Table 6).

Clinical Evidence

Systematic Review

Description of Studies

The FOURIER trial¹² was a phase III, double-blind, placebo-controlled, randomized clinical trial (N = 27,564). The primary objective was to evaluate the effect of evolocumab, compared to placebo, on the risk of cardiovascular death, MI, stroke, hospitalization for unstable angina, or coronary revascularization, whichever occurs first, in patients with clinically evident ASCVD. The trial included patients with LDL-C of 1.8 mmol/L or more (or a non-HDL-C of 2.6 mmol/L or more) after at least 2 weeks of optimized statin therapy, with or without ezetimibe. Patients were randomized in a 1:1 ratio to receive either subcutaneous evolocumab (140 mg once every 2 weeks or 420 mg once every month, per patient preference) or matching placebo injection. Randomization was stratified by the final screening LDL-C level and by geographical region. Treatment continued until a minimum of 1,630 patients experienced an event, adjudicated by an independent, external clinical events committee (CEC) as qualifying for a key secondary end point event of cardiovascular death, MI, or stroke. The estimated study duration was 56 months from the date the first patient was randomized.

The Gencer et al.¹³ and Sabatine et al. (2018)¹⁴ studies were subgroup analyses of the FOURIER trial. The objective of the Gencer et al. study was to evaluate the risks of major adverse cardiovascular events as a function of time from the date of the qualifying MI and to evaluate the effect of evolocumab on cardiovascular outcomes in patients who experienced a recent MI (in the previous year). The objective of the Sabatine et al. (2018) study was to assess the efficacy of evolocumab in 3 subgroups in the FOURIER trial: timing from the most recent MI, number of prior MIs, and the presence of residual multivessel coronary artery disease. The subgroup of patients who experienced an MI in the previous year from the Gencer et al. study and the subgroup of patients who experienced an MI in the previous 2 years in the Sabatine et al. (2018) study were considered to be most relevant for the purpose of this review. Outcomes of clinical events (cardiovascular death, MI, stroke, hospitalization for unstable angina, or coronary revascularization) were assessed after a median follow-up of 26 months, and LDL-C (LDL-C < 1.8 mmol/L and change from baseline) was assessed at weeks 4 and 48.

In the Gencer et al. study, 2,821 patients were randomized to receive evolocumab and 2,890 patients were randomized to receive placebo in the subgroup of patients who experienced a recent MI (in the previous year). The mean age of patients was 59.7 years (standard deviation [SD] = 9.3 years) in the evolocumab group and 59.5 years (SD = 9.2 years) in the placebo group. The mean time from MI to enrolment was 5.379 months (SD = 2.965 months) in the evolocumab group and 5.355 months (SD = 2.911 months) in the placebo group. Almost all patients — 99.8% (n = 2,814) of the evolocumab group and 99.8% (n = 2,884) of the placebo group) — had at least 1 major cardiovascular risk factor or at least 2 minor cardiovascular risk factors. At baseline, the mean LDL-C was 2.453 mmol/L (SD = 0.647 mmol/L) in the evolocumab group and 2.467 mmol/L (SD = 0.647 mmol/L) in the placebo group. Almost all patients — 99.9% (n = 2,819) of the evolocumab group and 100.0% (n = 2,889) of the placebo group — were taking a statin at baseline. A total of 3.2% (n = 91) of patients in the evolocumab group and 3.3% (n = 95) of patients in the placebo group were taking ezetimibe at baseline.

In general, the baseline characteristics of patients who experienced an MI in the previous 2 years in the Sabatine et al. (2018) study were similar to the baseline characteristics of patients who experienced a recent MI (in the previous year) in the Gencer et al. study. In the Sabatine et al. (2018) study, 4,109 patients were randomized to receive evolocumab and 4,293 patients were randomized to receive placebo in the subgroup of patients who experienced an MI in the previous 2 years. The mean time from MI to enrolment was 9.191 months (SD = 6.441 months) in the evolocumab group and 9.366 months (SD = 6.544 months) in the placebo group.

Efficacy Results

Summaries of the key efficacy results from the Gencer et al. and Sabatine et al. (2018) studies are presented in Table 2 and Table 3, respectively.

Cardiovascular Death, MI, or Stroke¹⁵

Of the patients who experienced an MI in the previous year in the Gencer et al. study, the composite end point of cardiovascular death, MI, and stroke was met by 6.45% (n = 182) of patients taking evolocumab versus 8.58% (n = 248) of patients taking placebo (hazard ratio [HR] = 0.75; 95% confidence interval [CI], 0.62 to 0.91). Of the patients who were 1 year or more beyond their MI, this composite end point was met by 6.04% (n = 502) of patients taking evolocumab versus 7.04% (n = 584) of patients taking placebo (HR = 0.85; 95% CI, 0.76 to 0.96).

Of the patients who experienced an MI in the previous 2 years in the Sabatine et al. (2018) study, this composite end point was met by 6.45% (n = 265) of patients taking evolocumab versus 8.43% (n = 362) of patients taking placebo (HR = 0.76; 95% CI, 0.64 to 0.89). Of the patients who were 2 years or more beyond their MI, this composite end point was met by 5.97% (n = 419) of patients taking evolocumab versus 6.81% (n = 470) of patients taking placebo (HR = 0.87; 95% CI, 0.76 to 0.99). The absolute risk reduction was 2.9% (95% CI, 1.2% to 4.5%) in patients who experienced an MI in the previous 2 years and 1.0% (95% CI, –0.7% to 2.7%) in patients who were 2 years or more beyond their MI.¹⁴

Table 2: Summary of Key Efficacy Results From the Gencer et al. Study (Full Analysis Set)

CI = confidence interval; IQR = interquartile range; KM = Kaplan-Meier; LDL-C = low-density-lipoprotein cholesterol; MI = myocardial infarction; SD = standard deviation.

Notes: For patients in the subgroup who experienced an MI in the year before randomization, those whose most recent MI or stroke was in the 4 weeks before randomization were excluded from the FOURIER trial.

The median length of follow-up was 25.99 months (IQR, 21.72 to 30.42 months). Events occurring between the patient randomization date and the patient last confirmed survival status date, inclusive, were included. The censoring date of patients without an event was the patient last nonfatal potential end point collection date.

Multiplicity was not taken into account in subgroup analyses.

Time to hospitalization for unstable angina was not a prespecified end point in the FOURIER trial; an ad hoc analysis was performed to ensure that results were provided for each individual component of the primary end point.

^aBased on a Cox model stratified by the randomization stratification factors collected with the Interactive Voice Response System.

^bBased on a Cox model, adding subgroup and subgroup-by-treatment interaction.

^cSummary statistics were based on observed data. When the calculated LDL-C was less than 40 mg/dL or triglycerides were greater than 400 mg/dL, the calculated LDL-C was replaced with ultracentrifugation LDL-C, if available.

Sources: Additional information received from the sponsor on December 21, 2023,¹⁵ and January 30, 2024.¹⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

Table 3: Summary of Key Efficacy Results From the Sabatine et al. (2018) Study (Full Analysis Set)

CI = confidence interval; IQR = interquartile range; KM = Kaplan-Meier; LDL-C = low-density lipoprotein cholesterol; MI = myocardial infarction; SD = standard deviation.

Notes: For patients in the subgroup who experienced an MI in the 2 years before randomization, those whose most recent MI or stroke was in the 4 weeks before randomization were excluded from the FOURIER trial.

The median length of follow-up was 25.99 months (IQR, 21.72 to 30.42 months). Events occurring between patient randomization date and the patient last confirmed survival status date, inclusive, were included. The censoring date of the patients without an event was the patient last nonfatal potential end point collection date.

Multiplicity was not taken into account in subgroup analyses.

^aBased on a Cox model stratified by the randomization stratification factors collected with the Interactive Voice Response System.

^bBased on a Cox model, adding subgroup and subgroup-by-treatment interaction.

^cSummary statistics were based on observed data. When the calculated LDL-C was less than 40 mg/dL or triglycerides were greater than 400 mg/dL, the calculated LDL-C was replaced with ultracentrifugation LDL-C, if available.

Sources: Additional information received from the sponsor on December 21, 2023,¹⁵ and on January 30, 2024.¹⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

Cardiovascular Death¹⁵

Of the patients who experienced an MI in the previous year, the end point of cardiovascular death was met by 1.77% (n = 50) of patients taking evolocumab versus 1.80% (n = 52) of patients taking placebo (HR = 1.00; 95% CI, 0.68 to 1.47). Of the patients who were 1 year or more beyond their MI, the end point was experienced by 1.88% (n = 156) of patients taking evolocumab versus 1.64% (n = 136) of patients taking placebo (HR = 1.15; 95% CI, 0.91 to 1.44).

This mortality end point was not assessed in the subgroup of patients who experienced an MI in the previous 2 years or in the subgroup of patients who were 2 years or more beyond their MI.

MI (Fatal or Nonfatal)¹⁵

Of the patients who experienced an MI in the previous year, the cardiovascular end point of MI (fatal or nonfatal) was met by 4.50% (n = 127) of patients taking evolocumab versus 6.61% (n = 191) of patients taking placebo (HR = 0.67; 95% CI, 0.54 to 0.84). Of the patients who were 1 year or more beyond their MI, this cardiovascular end point was met by 3.56% (n = 296) of patients taking evolocumab versus 4.57% (n = 379) of patients taking placebo (HR = 0.78; 95% CI, 0.67 to 0.91).

This cardiovascular end point was not assessed in the subgroup of patients who experienced an MI in the previous 2 years or in those who were 2 years or more beyond their MI.

Stroke (Fatal or Nonfatal)¹⁵

Of the patients who experienced an MI in the previous year, the cerebrovascular end point of stroke (fatal or nonfatal) was met by 1.06% (n = 30) of patients taking evolocumab versus 1.31% (n = 38) of patients taking placebo (HR = 0.81; 95% CI, 0.50 to 1.31). Of the patients who were 1 year or more beyond their MI, this cerebrovascular end point was met by 1.32% (n = 110) of patients taking evolocumab versus 1.65% (n = 137) of patients taking placebo (HR = 0.80; 95% CI, 0.62 to 1.03).

This cerebrovascular end point was not assessed in patients in the subgroup of patients who experienced an MI in the previous 2 years or in those who were 2 years or more beyond their MI.

Cardiovascular Death, MI, Hospitalization for Unstable Angina, Stroke, or Coronary Revascularization¹⁵

Of the patients who experienced an MI in the previous year, the composite end point of cardiovascular death, myocardial infarction, hospitalization for unstable angina, stroke, and coronary revascularization was met by 11.45% (n = 323) of patients taking evolocumab versus 14.12% (n = 408) of patients taking placebo (HR = 0.81; 95% CI, 0.70 to 0.93). Of the patients who were 1 year or more beyond their MI, this composite end point was met by 10.24% (n = 851) of patients taking evolocumab versus 11.10% (n = 921) of patients taking placebo (HR = 0.92; 95% CI, 0.84 to 1.01).

Of the patients who experienced an MI in the previous 2 years, this composite end point was met by 11.17% (n = 459) of patients taking evolocumab versus 13.72% (n = 589) of patients taking placebo (HR = 0.80; 95% CI, 0.71 to 0.91). Of the patients who were 2 years or more beyond their MI, this composite end point was met by 10.19% (n = 715) of patients taking evolocumab versus 10.73% (n = 740) of patients taking placebo (HR = 0.95; 95% CI, 0.85 to 1.05). The absolute risk reduction was 3.4% (95% CI, 1.4% to 5.3%) in patients who experienced an MI in the previous 2 years and 0.8% (95% CI, –1.1% to 2.7%) in patients who were 2 years or more beyond their MI.¹⁴

Change From Baseline in LDL-C¹⁶

Of the patients who experienced an MI in the previous year, the mean LDL-C was 2.453 mmol/L (SD = 0.647 mmol/L) in the evolocumab group and 2.467 mmol/L (SD = 0.647 mmol/L) in the placebo group at baseline. For patients who experienced an MI in the previous year, the mean percent change from baseline in LDL-C was –59.90% (SD = 30.12%) in the evolocumab group and 2.00% (SD = 27.41%) in the placebo group at week 48. Of the patients who were 1 year or more beyond their MI, the mean LDL-C was 2.563 mmol/L (SD = 0.784 mmol/L) in the evolocumab group and 2.545 mmol/L (SD = 0.711 mmol/L) in the placebo group at baseline. For patients who were 1 year or more beyond their MI, the mean percent change from baseline in LDL-C was –60.60% (SD = 30.53%) in the evolocumab group and –0.98% (SD = 25.70%) in the placebo group at week 48.

Of the patients who experienced an MI in the previous 2 years, the mean LDL-C level was 2.476 mmol/L (SD = 0.670 mmol/L) in the evolocumab group and 2.472 mmol/L (SD = 0.639 mmol/L) in the placebo group at baseline. For patients who experienced an MI in the previous 2 years, the mean percent change from baseline in LDL-C was –59.61% (SD = 31.05%) in the evolocumab group and 1.28% (SD = 26.73%) in the placebo group at week 48. For patients who were 2 years or more beyond their MI, the mean LDL-C was 2.570 mmol/L (SD = 0.796 mmol/L) in the evolocumab group and 2.557 mmol/L (SD = 0.727 mmol/L) in the placebo group at baseline. For patients who were2 years or more beyond their MI, the mean percent change from baseline in LDL-C was –60.90% (SD = 30.05%) in the evolocumab group and –1.14% (SD = 25.79%) in the placebo group at week 48.

Harms Results

Safety outcomes were not assessed by subgroups. A summary of harms results from the FOURIER trial is presented in Table 4.

Table 4: Summary of Harms Results From the FOURIER Trial (Safety Analysis Set)

AMQ = Amgen MedDRA query; HLGT = high-level group term; SAE = serious adverse event; SMQ = standard MedDRA query; TEAE = treatment-emergent adverse event.

Notes: TEAEs are presented by preferred terms and coded by Medical Dictionary for Regulatory Activities (MeDRA) Version 9.1.

Deaths by any cause were adjudicated efficacy end points in the FOURIER trial. SAEs that did not meet the criteria of adjudicated end points and were subsequently reported as AEs, but later resulted in death, are included in the SAE section.

Standardized MedDRA Queries are validated, standard sets of MedDRA terms used to support signal detection and monitoring and represent a variety of safety topics of regulatory interest. Standardized MedDRA Queries include narrow and/or broad terms; narrow terms are highly likely to represent the condition of interest.¹⁷

^aThese search strategies, SMQs and AMQs, were used to retrieve AEs potentially related to the condition under review when heterogenous medical presentations may be expected.

^bTEAEs reported in more than 0.2% of patients in any treatment group by high-level term using a narrow search strategy for potential hypersensitivity events include dermatitis and eczema; rash, eruptions, and exanthemas; nasal congestion and inflammations; urticarias; and allergic conditions not elsewhere classifiable (NEC).

^cTEAEs reported in more than 0.1% of patients in any treatment group by preferred term using a narrow search strategy for potential injection-site reaction events include injection-site pain, bruising, hematoma, erythema, and hemorrhage.

^dTEAEs using a narrow search strategy for potential muscle events include rhabdomyolysis, myopathy, and a myoglobin blood increase.

^eTEAEs by high-level group term for potential neurocognitive events include cognitive and attention disorders and disturbances, deliria (including confusion), dementia and amnestic conditions, disturbances in thinking and perception, and mental impairment disorders.

^fTEAEs reported in 1 or more patients in any treatment group by high-level term using broad and narrow search strategies for potential demyelination events and peripheral neuropathy events, respectively, include peripheral neuropathies NEC, sensory abnormalities NEC, trigeminal disorders, acute and progressive multiple sclerosis, plasma cell neoplasms NEC, acute polyneuropathies, spinal cord and nerve root disorders NEC, chronic polyneuropathies, and optic nerve disorders NEC.

^gTEAEs by preferred term using a narrow search strategy for potential hepatitis C infection include hepatitis C, chronic hepatitis, and a positive hepatitis C virus test.

^hTEAEs reported in 0.1% or more of patients in any treatment group by high-level term using a narrow search strategy for potential transaminase elevations and hepatic disorders include liver function analyses, hepatocellular damage and hepatitis NEC, hepatic and hepatobiliary disorders NEC, and coagulation and bleeding analyses.

Sources: Clinical Study Report of the FOURIER trial.¹² Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

Treatment-Emergent Adverse Events

The proportion of patients with at least 1 treatment-emergent adverse event (TEAE) or at least 1 serious adverse event (SAE) was similar in the treatment groups. A total of 10,664 patients (77.4%) in the evolocumab group and 10,644 patients (77.4%) in the placebo group reported at least 1 TEAE, with the most common TEAE being diabetes mellitus, which was reported in 1,207 patients (8.8%) and 1,130 patients (8.2%), respectively. A total of 3,410 patients (24.8%) in the evolocumab group and 3,404 patients (24.7%) in the placebo group reported at least 1 SAE, with the most common SAE being unstable angina, which was reported in 233 patients (1.7%) and 278 (2.0%), respectively.

The proportion of patients who stopped treatment due to any TEAE was similar in the treatment groups. A total of 608 patients (4.4%) in the evolocumab group and 573 patients (4.2%) in the placebo group stopped treatment due to any TEAE, with the most common TEAE being myalgia, which was reported in 37 patients (0.3%) and 46 patients (0.3%), respectively.

TEAEs of Special Interest

The proportion of patients with TEAEs of special interest, including potential hypersensitivity, injection-site reaction, muscle events, neurocognitive events, demyelination events and peripheral neuropathy, hepatitis C infection, and transaminase elevations and hepatic disorder events, was similar in the treatment groups. A total of 13 patients (< 0.1%) in the evolocumab group and 15 patients (0.1%) in the placebo group had a potential muscle-related AE (according to a narrow search strategy that included rhabdomyolysis, myopathy, and a myoglobin blood increase). A total of 1,381 patients (10.0%) in the evolocumab group and 1,344 patients (9.8%) in the placebo group had a potential muscle-related AE (according to a broader search strategy).

Critical Appraisal

Internal Validity

The Gencer et al.¹³ and Sabatine et al. (2018)¹⁴ studies were based on subgroup analyses of the FOURIER trial.¹² The subgroup analyses were based on the statistical methods from the FOURIER trial and the subgroups by timing of prior MI were prespecified; however, there was no clear hypothesis stated a priori. The P values on the test for interaction term (in general, greater than 0.05, with the exception of the primary end point in the subgroup analysis by timing of prior MI [< 2 years versus ≥ 2 years]) strongly suggest that chance cannot be excluded as a likely explanation for the differential subgroup effect. There is a lack of evidence from randomized controlled trials and large observational studies to support consistent and similar findings from the subgroup analyses. Nonetheless, the subgroup analyses results were generally consistent with the overall FOURIER trial results, with the exception of stroke, for which the HR was 0.79 (95% CI, 0.66 to 0.95);¹² the corresponding subgroup analysis results included null values.

Sample-size calculation was based on the key secondary end point of the full analysis set in the FOURIER trial, but not in the subgroup analyses. Consequently, there is an increased likelihood of producing unreliable or inaccurate results and, in particular, on cardiovascular death and stroke, components of the composite end points for which the 95% CI results included null values. Nonetheless, the sample size of subgroups was considered relatively large. Multiplicity was not accounted for in the subgroup analyses; therefore, the interpretation of the subgroup analysis results is subject to an increased likelihood of type I error.

In consideration of the aforementioned conditions, which can lower the credibility and reliability of the subgroup analysis results, the available evidence should not be viewed as conclusive; however, it may be interpreted as likely indicative of a possible subgroup effect.

External Validity

Because the sponsor’s reimbursement request focused on patients with recent ACS (in the previous year), the clinical experts were consulted on the patient population included in the subgroup analyses, which did not include patients with unstable angina and recent (in the previous 4 weeks) MI or stroke. Although evidence from these patients is lacking, overall, no key concerns were identified for the generalizability of the subgroup analysis results to the patient population in the reimbursement request.

Of note, the estimated study duration was 56 months from the date the first patient was randomized; however, the median follow-up was 26 months. In the previous review of the FOURIER trial by CDA-AMC, the length of follow-up was deemed likely too short to assess the long-term harms associated with the use of evolocumab.¹⁸

Long-Term Extension Studies

Description of Studies

Patients who completed the FOURIER trial had the option to enrol in 1 of the two 5-year extension studies (one study was conducted in North America and Eastern Europe and the other study was conducted in Western Europe) with open-label evolocumab (N = 5,305 and N = 1,600, respectively).^19,20 The primary objective of both studies was to describe the safety and tolerability of the long-term administration of evolocumab. An ad hoc subgroup analysis of the open-label extension (OLE) studies was also conducted in the subset of patients who experienced an MI before or during the parent trial. Comparisons were made between patients randomized to receive evolocumab versus placebo in the parent trial. All results reported herein are the integrated data from the 2 OLE studies.

The mean age of patients in the MI subgroup was 62.2 years (SD = 8.7 years) in the evolocumab group and 62.0 years (SD = 8.6 years) in the placebo group. Most of the participants were male in this subgroup (79.3% in the evolocumab group and 78.8% in the placebo group, respectively). At baseline, the mean LDL-C in the MI subgroup was 2.5 mmol/L (SD = 0.7 mmol/L) in both the evolocumab and placebo groups. These characteristics were similar in the overall FOURIER-OLE study population as well. Time since the most recent MI in the MI subgroup was 8.070 years (SD = 6.137 years) in the evolocumab group and 7.835 years (SD = 5.905 years) in the placebo group.

In the overall FOURIER-OLE study population, the mean time from MI to enrolment was 69.606 months (SD = 74.237 months) in the evolocumab group and 68.531 months (SD = 71.613 months) in the placebo group. Most of the participants were white (93.4% in the evolocumab group and 94.5% in the placebo group). Major and minor cardiovascular risk factors, as well as risk factor counts, were similar in the evolocumab and placebo groups in the overall OLE population. These baseline characteristics were not available for the MI subgroup population.

Efficacy Results

Change From Baseline in LDL-C

Among patients in the FOURIER-OLE studies, the median baseline reflexive LDL-C in the parent trial was 2.36 mmol/L (first quartile [Q1] and third quartile [Q3] = 2.06 mmol/L and 2.80 mmol/L, respectively); the baseline LDL-C level was similar for patients in the 2 randomized treatment groups from the parent trial.^21,22 The observed mean percent reduction from baseline in LDL-C ranged from 53.4% to 67.2% during the 260-week OLE study period.²¹

In the subset of patients (n = 5,582) who experienced an MI before and/or during the parent FOURIER trial, the mean baseline LDL-C was 2.52 mmol/L (SD = 0.695 mmol/L), which was similar in patients randomized to receive evolocumab and those randomized to receive placebo in the parent trial.²³ The mean LDL-C during the 260-week OLE study period in the MI subgroup of patients was 1.061 mmol/L (SD = 0.924 mmol/L). The mean percent reduction from baseline in LDL-C was approximately 57.7% at week 260 and was similar in patients who received evolocumab and those who received placebo in the parent trial.²³

Time to Major Cardiovascular Events

During the OLE study period, 490 patients (14.6%) originally randomized to the evolocumab group in the parent study met the FOURIER primary outcome of cardiovascular death, MI, stroke, hospitalization for unstable angina, or coronary revascularization, compared to 551 patients (16.8%) originally randomized to the placebo group (HR = 0.85; 95% CI, 0.75 to 0.96). The HR for the key secondary composite outcome of cardiovascular death, MI, and stroke was 0.80 (95% CI, 0.68 to 0.93). Of note, the HR for the individual component of cardiovascular death was 0.77 (95% CI, 0.60 to 0.99).²¹

Among patients who had an MI before and/or during the parent FOURIER trial, 406 patients (14.42%) who were randomized to receive evolocumab in the parent trial met the FOURIER primary outcome of cardiovascular death, MI, stroke, hospitalization for unstable angina, or coronary revascularization, compared with 478 patients (17.28%) who were randomized to receive placebo (HR = 0.81; 95% CI, 0.71 to 0.93). The HR for the key secondary composite outcome of cardiovascular death, MI, and stroke was 0.77 (95% CI, 0.65 to 0.90); of note, the HR for the individual component of cardiovascular death was 0.68 (95% CI, 0.51 to 0.91). Event probabilities and, consequently, the difference in event probabilities between treatment groups from the parent trial were not available for the MI subgroup analysis.

Harms Results

In the integrated OLE safety analysis set, 2,894 patients (86.3%) randomized to evolocumab in the parent study and 2,830 patients (86.4%) randomized to placebo experienced at least 1 AE during the OLE studies. The most frequently reported AE was hypertension (15% of evolocumab-treated patients and 14.6% of placebo-treated patients). Other AEs reported by at least 5% of patients in either parent study treatment group include nasopharyngitis, bronchitis, arthralgia, diabetes mellitus, atrial fibrillation, back pain, upper respiratory tract infection, angina pectoris, and pneumonia.

Approximately 43% of patients experienced at least 1 SAE during the OLE studies (43.4% in patients randomized to evolocumab in the parent study and 42.7% in patients randomized to placebo). Acute MI, angina pectoris, pneumonia, atrial fibrillation, and cardiac failure were among the SAEs reported most frequently (in 2% to 3% of patients).²⁴

Overall, approximately 8% of patients experienced an AE leading to the discontinuation of evolocumab during the OLE study (7.7% of patients who received evolocumab in the parent study and 8.0% of patients who received placebo in the parent study). The most frequently reported AEs leading to the discontinuation of evolocumab in the OLE studies were in the system organ class of neoplasms, benign, malignant and unspecified (including cysts and polyps) (2.0% to 2.1% of patients), followed by cardiac disorders (1.5% to 2.1% of patients). None of the reported AEs leading to discontinuation were reported in more than 1% of patients. The most commonly reported fatal AEs were in 3 system organ classes: cardiac disorders; neoplasms benign, malignant and unspecified (including cysts and polyps); and infections and infestations.²⁴

Notable harms reported by at least 1% of patients in any treatment group in the OLE safety analysis set included potential injection-site reaction events, potential demyelination events (peripheral neuropathy, sensory abnormalities not elsewhere classifiable [NEC], and chronic polyneuropathies), and transaminase elevations and potential hepatic disorders (liver function analyses, hepatocellular damage, and hepatitis NEC). The numbers were similar in the evolocumab and placebo groups.

The safety profile of evolocumab in the MI subgroup was similar to that seen in the overall study population.

Critical Appraisal

Internal Validity

An open-label study design can influence the perception of improvement and/or harms by patients and clinicians, particularly for outcomes that are subjective in measurement and interpretation. However, because all fatal or nonfatal cardiovascular events or deaths were adjudicated by an external independent CEC, the assessment of the primary and key secondary end points in the FOURIER-OLE studies were not likely to have been affected by the open-label design.

Because the descriptive analyses used in the OLE studies and the ad hoc subgroup analysis of patients who experienced a prior MI, the available evidence should only be considered suggestive of a potential treatment effect, subject to uncertainty associated with the exploratory nature of the analyses.

External Validity

The baseline characteristics of all patients enrolled in the FOURIER-OLE studies were similar to those in the randomized treatment groups from the parent FOURIER trial. Although most patients were from study sites located in Europe (> 66%), the demographics of that population were generally similar to those in the patient population in Canada. In general, the baseline characteristics of patients in the MI subgroup were similar to those in the overall OLE patient population.

Because the sponsor’s reimbursement request is focused on the patient population with recent ACS (in the previous year), it should be noted that the MI subgroup included patients who had an MI before and/or during the parent FOURIER trial. The mean time from the most recent MI to enrolment in the overall OLE patient population was 69.606 months (SD = 74.237 months) in patients randomized to evolocumab in the parent trial and 68.531 months (SD = 71.613 months) in patients randomized to placebo in the parent trial. In the subset of patients who experienced a prior MI, the mean time from the most recent MI was 8.070 years (SD = 6.137 years) in patients who were randomized to evolocumab in the parent trial and 7.835 years (SD = 5.905 years) in patients who were randomized to placebo in the parent trial.

Indirect Comparisons

No evidence on indirect treatment comparisons was submitted by the sponsor.

Study Addressing Gap in the Evidence From the Systematic Review

Description of Study

The EVOPACS study²⁵ was a phase III, double-blind, placebo-controlled, randomized trial (N = 308). The primary objective was to assess the effectiveness of evolocumab 420 mg once every month, compared to placebo, in the reduction of LDL-C at week 8 in patients receiving high-intensity statin treatment during the acute phase of ACS.

The mean age of patients was 60.5 years (SD = 12.0 years) in the evolocumab group and 61.0 years (SD = 10.7 years) in the placebo group. Most of the participants were male (83% in the evolocumab group and 80% in the placebo group). Although half the patients in both groups had history of smoking, there were more active smokers in the evolocumab group than in the placebo group (41% versus 30%). Most of the patients enrolled in this study were statin-naive (80% in the evolocumab group and 76% in the placebo group). In terms of index ACS events, 57% of patients in the evolocumab group and 70% in the placebo group had non-ST-elevation (NSTE)-ACS, and 43% in the evolocumab group and 30% in the placebo group had STEMI.

Efficacy Results

The mean change from baseline in LDL-C was –77.1% (SD = 15.8%) in the evolocumab group versus –35.4% (SD = 26.6%) in the placebo group at week 8 (least squares mean difference = –40.7%; 95% CI, –45.2% to –36.2%). The mean LDL-C level at week 8 was 0.79 mmol/L (SD = 0.46 mmol/L) in the evolocumab group and 2.06 mmol/L (SD = 0.63 mmol/L) in the placebo group. At week 8, the proportion of patients with LDL-C levels of less than 1.8 mmol/L was 95.7% in the evolocumab group compared to 37.6% in the placebo group.

Harms Results

A total of 78 of 155 patients (50.3%) in the evolocumab group and 77 of 152 patients (50.7%) in the placebo group experienced at least 1 AE during the study. Nonserious AEs, including prespecified AE categories, occurred in 73 patients (47.1%) in the evolocumab group and in 71 patients (46.7%) in the placebo group; for 2 patients (1.3%) (both in the placebo group), these AEs led to the discontinuation of the investigational product. The most common AE in the evolocumab and placebo groups was chest pain (8 [5.2%] versus 8 [5.3%]), followed by musculoskeletal pain (10 [6.5%] versus 5 [3.3%]), and nasopharyngitis (7 [4.5%] versus 4 [2.6%]).²⁶

SAEs occurred in 12 patients (7.7%) in the evolocumab group and in 11 patients (7.2%) in the placebo group; 3 patients (1.0%) (2 [1.3%] in the evolocumab group and 1 [0.7%] in the placebo group) experienced SAEs leading to the discontinuation of the investigational product. Two patients (both in the evolocumab group) died during the study; neither death was considered to be related to the investigational product by the investigator or the Data Safety and Monitoring Board, and both were adjudicated as cardiovascular deaths.²⁶

Key Take-Aways

Interpretation of the results from the EVOPACS study is limited by the small sample size and short (8-week) follow-up. The clinical experts consulted by CDA-AMC did not consider the exclusion of patients whose most recent MI or stroke was in the 4 weeks before randomization to be a major gap in the evidence. The clinical experts advised that patients with an index case of ACS are not likely to be initiated on evolocumab in the inpatient setting, as they are most likely to be statin-naive, which was the case in this study as well, where 80% and 76% of patients in the evolocumab and placebo arms, respectively, were statin-naive. As a result, these patients will first be stabilized on a statin before any add-on therapies are considered. Nonetheless, the clinical experts expect that patients with acute MI and who are stabilized will likely respond to treatment with evolocumab in a manner similar to that in patients with nonacute MI.

Although most of the baseline characteristics were similar in the 2 treatment groups, there was a slight imbalance in index ACS events (i.e., for NSTE-ACS, there were 57% and 70% of patients in the evolocumab group and placebo group, respectively; for STEMI, there were 43% and 30% of patients in the evolocumab group and placebo group, respectively). Further, because active smoking was a major risk factor for cardiovascular events in the FOURIER trial, it should be noted that there were more active smokers in the evolocumab group than in the placebo group (41% versus 30%).

Conclusion

Two subgroup analyses of patients who experienced a recent MI (< 1 year and < 2 years) in the FOURIER trial, described by Gencer et al. and Sabatine et al. (2018), informed the main body of evidence for this reassessment. New evidence from the subgroup analyses of the FOURIER trial was submitted to support the identification of a subgroup of patients who would most benefit from evolocumab, which was raised in the previous resubmission for the ASCVD component of primary hyperlipidemia. Evolocumab in addition to moderate-to-high intensity statin therapy, compared to placebo, demonstrated an absolute benefit that was likely clinically meaningful and may be amplified in the subset of patients who experienced a recent MI (i.e., in the previous year). Results from the prespecified subgroup analyses of clinical event outcomes, with the exception of stroke, were generally consistent with the overall FOURIER trial results. This possible subgroup effect on the key secondary composite end point appears to have been primarily driven by the reduction in risk of MI, but there was no difference in the risk of cardiovascular death and stroke over the median follow-up period of 26 months. A biological plausibility for the proposed subgroup effect and a greater absolute risk for cardiovascular events in patients with a recent MI than patients who were further along the trajectory of disease were noted. The ad hoc subgroup analysis of patients who experienced a prior MI in an integrated OLE analysis also informed this reassessment, which provided results on the clinical event outcomes, with the exception of coronary revascularization, that were generally consistent with the results reported in the overall population in the 5-year OLE of the FOURIER trial. Further, the ad hoc subgroup analysis of patients who experienced a prior MI may suggest a treatment benefit in patients who received evolocumab earlier than those who received delayed treatment as a result of randomization in the parent trial. Of note, this possible subgroup effect on the key secondary composite end point appears to have been driven by a reduction in the risk of cardiovascular death and MI, but there was no difference in the risk of stroke. Although no major concerns about generalizability were identified, the place in therapy of evolocumab in relation to ezetimibe that would be supported by the evidence is uncertain. The incidence of TEAEs reported in the FOURIER trial were similar in the 2 treatment groups, including muscle-related events, which are important to patients. However, it is important to note that the duration of follow-up in the parent trial is likely inadequate for assessing the long-term relative safety of evolocumab. No new safety signals were identified in the 5-year OLE studies of the FOURIER trial.

Introduction

The objective of the present reassessment of evolocumab (ASCVD indication) is to review and critically appraise the new evidence submitted by the sponsor on the beneficial and harmful effects of evolocumab 140 mg/mL and 120 mg/mL subcutaneous injection in the treatment of primary hyperlipidemia, specifically in the requested reimbursement population of adult patients with ASCVD and recent ACS (in the previous year).

Disease Background

The 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 CDA-AMC review team.

Hyperlipidemia refers to high levels of lipids in the blood, including cholesterol and triglycerides. High levels of cholesterol (also referred to as hypercholesterolemia³), notably LDL-C, can cause atherosclerosis, defined as the buildup of fatty deposits in blood vessels leading to a restriction in blood flow, which is a major cause of cardiovascular events, including heart attack, stroke, and lower-extremity and PAD. In other words, hyperlipidemia can increase an individual’s risk of developing cardiovascular disease involving the blood vessels that supply the heart (coronary artery disease), brain (cerebrovascular disease), and limbs (PAD). Other risk factors for cardiovascular disease include diabetes mellitus, high blood pressure, chronic kidney disease, cigarette smoking, and family history.⁴ ASCVD, as defined in the 2021 CCS dyslipidemia guidelines,⁵ comprises all clinical conditions of atherosclerotic origin, such as ACS, stroke, and PAD. ACS comprises NSTEMI, STEMI, and unstable angina, with MI being the most common clinical presentation.¹¹ The clinical experts were consulted on the definition of ACS used in clinical practice. Because the cardiac troponin assays have evolved to become highly sensitive to micromolar elevations in levels of circulating troponin, unstable angina has become an exceedingly infrequent diagnosis. Therefore, only MI, including STEMI and NSTEMI, was considered most relevant for the purpose of this review.

The incidence rate for MI was approximately 2.5 per 1,000 person-years over the time period from 2005 to 2016 in Ontario, whereas the incidence rate for unstable angina was 3.3 per 1,000 person-years in 2005 and 1.7 per 1,000 person-years in 2016.7 The prevalence of MI and unstable angina were 28.6 (95% CI, 28.5 to 28.7) per 1,000 individuals and 24.6 (95% CI, 24.5 to 24.7) per 1,000 individuals, respectively, in Ontario for the period from 2004 to 2017.⁷ More specifically, the 10-year prevalence rates for MI increased from 23.5 to 26.9 per 1,000 individuals and for unstable angina increased from 22.1 to 23.7 per 1,000 individuals between the periods of 2004 to 2013 and 2008 to 2017.⁷

Standards of Therapy

The 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 CDA-AMC review team.

Following an index ACS, a residual risk of subsequent cardiovascular event remains.⁶ Secondary prevention refers to the treatment and management of known, clinically evident ASCVD and the prevention or delay of the onset of disease manifestations.⁵

The 2021 CCS dyslipidemia guidelines⁵ advise on statin-indicated conditions, including ASCVD. According to the guidelines, treatment with PCSK9 inhibitors, with or without ezetimibe, should be considered for patients with ASCVD when the necessary reduction in LDL-C, ApoB, or non-HDL-C is substantial (i.e., LDL-C > 2.2 mmol/L or ApoB > 0.80 g/L or non-HDL-C > 2.9 mmol/L despite a maximally tolerated statin dose), or for patients who have been shown to derive the largest benefit from the intensification of statin therapy with the addition of a PCSK9 inhibitor. This subset includes patients with recent ACS (i.e., occurring in the 52 weeks after hospitalization for the index ACS), as well as those with clinically evident ASCVD and any additional cardiovascular risk enhancers (diabetes mellitus or metabolic syndrome, polyvascular disease, symptomatic PAD, history of MI, MI in the past 2 years, previous coronary artery bypass graft surgery, an LDL-C level of 2.6 mmol/L or more or HeFH, or a lipoprotein[a] level of 60 mg/dL or more). If the necessary reduction in LDL-C, ApoB, or non-HDL-C is modest (i.e., LDL-C of 1.8 mmol/L to 2.2 mmol/L or ApoB of 0.70 g/L to 0.80 g/L or non-HDL-C 2.4 mmol/L to 2.9 mmol/L despite a maximally tolerated statin dose), then the guidelines advise that ezetimibe, with or without a PCSK9 inhibitor, be considered. Additionally, in patients with ASCVD and a triglyceride level of 1.5 mmol/L to 5.6 mmol/L despite receiving a maximally tolerated statin dose, the guidelines advise that treatment with icosapent ethyl be considered.⁵

The key limitation of statin therapy has been the risk of myalgia, which is commonly associated with suspected underlying statin intolerance.²⁷ Other lipid-lowering therapies, such as niacin, fibrates, bile acid sequestrants, mipomersen (not approved in Canada), and lomitapide (only used for homozygous familial hypercholesterolemia), are infrequently used in patients with ASCVD, according to clinical expert input.

The clinical experts were consulted on the goal of treatment with a PCSK9 inhibitor and agreed that for patients with ASCVD who are at high cardiovascular risk, LDL-C, non-HDL-C, or ApoB should be reduced to levels below the thresholds referenced in the 2021 CCS dyslipidemia guidelines (i.e., LDL-C < 1.8 mmol/L or non-HDL-C < 2.4 mmol/L or ApoB < 0.7 g/L) for secondary prevention,⁵ thereby lowering the risk of clinical events and improving survival and quality of life.

Drug Under Review

The key characteristics of evolocumab, along with other treatments available for patients with primary hyperlipidemia (including ASCVD), are summarized in Table 5.

The indication for evolocumab under review is the reduction of elevated LDL-C in adult patients with primary hyperlipidemia (including HeFH and ASCVD) as an adjunct to diet and statin therapy, with or without other lipid-lowering therapies, in patients who require additional lowering of LDL-C and as an adjunct to diet, alone or in combination with nonstatin lipid-lowering therapies, and in patients for whom a statin is contraindicated.¹ The Notice of Compliance dates are September 10, 2015 (initial approval) and September 27, 2023 (latest revision).

In contrast to the Health Canada–approved indication, the reimbursement request for the present reassessment is for patients with recent ACS (in the previous year) who have an LDL-C level of 1.8 mmol/L or more despite taking moderate-to-high intensity statin therapy, with or without ezetimibe. This patient population is consistent with the population of patients identified in the 2021 CCS dyslipidemia guidelines⁵ for secondary prevention who were shown to derive the largest benefit from the intensification of statin therapy with the addition of a PCSK9 inhibitor, which is patients with recent ACS (i.e., occurring in the 52 weeks after hospitalization for the index ACS).

For primary hyperlipidemia in adult patients (including HeFH and ASCVD), the recommended dose of evolocumab is either 140 mg every 2 weeks or 420 mg once monthly; both doses are clinically equivalent. When switching dosage regimens, administer the first dose of the new regimen on the next scheduled date of the prior regimen. One prefilled syringe or prefilled autoinjector delivers the 140 mg every 2 week dose. One single-use automated mini-doser with a 3.5 mL prefilled cartridge delivers the 420 mg once monthly dose. Alternatively, 3 prefilled syringes or 3 prefilled autoinjectors administered consecutively within 30 minutes deliver the 420 mg once monthly dose.¹

Evolocumab is a fully human monoclonal immunoglobulin G2 that selectively binds to PCSK9 with high affinity to inhibit circulating PCSK9 from binding to the low-density-lipoprotein receptor on the surface of liver cells. This prevents the PCSK9-mediated degradation of low-density-lipoprotein receptors, thereby increasing the number of receptors available to clear low-density lipoprotein.¹

Table 5: Key Characteristics of Evolocumab, Statins, and Ezetimibe Indicated for Primary Hyperlipidemia (Including ASCVD)

ApoB = apolipoprotein B; ASCVD = atherosclerotic cardiovascular disease; DRESS = drug reaction with eosinophilic and systemic symptoms; HDL-C = high-density-lipoprotein cholesterol; HeFH = heterozygous familial hypercholesterolemia; HMG-CoA = 3-hydroxy-3-methylglutaryl-coenzyme A; IgG2 = immunoglobulin G2; LDL = low-density lipoprotein; LDL-C = low-density-lipoprotein cholesterol; LDLR = low-density lipoprotein receptor; SCAR = severe cutaneous adverse reactions; SJS = Stevens-Johnson syndrome; TEN = toxic epidermal necrolysis; TG = triglyceride; total-C = total cholesterol.

^aHealth Canada–approved indication.

Sources: Product monographs for evolocumab,¹ statins,^28-33 and ezetimibe.³⁴

Submission History

Initial Submission for Primary Hyperlipidemia

In 2016, evolocumab was first reviewed by CDEC for primary hyperlipidemia, including HeFH and clinical ASCVD. CDEC issued a recommendation that evolocumab be listed as an adjunct to diet and maximally tolerated statin therapy in adult patients with HeFH who require additional lowering of LDL-C, if the prespecified clinical criteria and condition are met. For the ASCVD component of the indication, CDEC issued a recommendation that evolocumab not be listed as an adjunct to diet and maximally tolerated statin therapy in adult patients with clinical ASCVD who require additional lowering of LDL-C.⁸ Detailed information on and the reasons for the final recommendation made by CDEC in 2016 are publicly available on the CDA-AMC website.

Resubmission for the Atherosclerotic Cardiovascular Disease Component of Primary Hyperlipidemia

In 2017, evolocumab was resubmitted and reviewed by CDEC for the ASCVD component of primary hyperlipidemia. CDEC issued a recommendation that evolocumab be reimbursed as an adjunct to diet and maximally tolerated statin therapy in adult patients with ASCVD who require additional lowering of LDL-C, if the prespecified criterion and condition are met. The criterion was that patients meet the inclusion criteria for the FOURIER trial (i.e., established cardiovascular disease and a high risk for future events, LDL-C ≥ 1.8 mmol/L or non-HDL-C ≥ 2.6 mmol/L, and receipt of a maximally tolerated dose of statins). In the double-blind, placebo-controlled, randomized clinical trial that enrolled patients with ASCVD receiving optimized statin therapy (FOURIER, n = 27,564), the composite outcome of cardiovascular death, MI, stroke, unstable angina, and revascularization was met by 9.8% of patients taking evolocumab and 11.3% of patients taking placebo over a median follow-up period of 26 months (HR = 0.85; 95% CI, 0.79 to 0.92).⁹ However, funding is not yet in place, as negotiations concluded without an agreement in July 2019.¹⁰ Detailed information on the final recommendation made by CDEC in 2017 is publicly available on the CDA-AMC website.

Basis of Present Reassessment

The 2021 CCS dyslipidemia guidelines⁵ referenced the FOURIER³⁵ and ODYSSEY³⁶ trials, which have identified subsets of patients with established cardiovascular disease (at high cardiovascular risk) who have been shown to derive the largest benefit from intensification of statin therapy with the addition of a PCSK9 inhibitor in secondary prevention. This subset includes patients with recent ACS (i.e., occurring in the 52 weeks after hospitalization for the index ACS), as well as those with clinically evident ASCVD and any additional cardiovascular risk enhancers, including diabetes mellitus or metabolic syndrome, polyvascular disease, symptomatic PAD, history of MI, MI in the past 2 years, previous coronary artery bypass graft surgery, an LDL-C level of 2.6 mmol/L or more or HeFH, and a lipoprotein(a) level of 60 mg/dL or more.⁵

Hence, the focus of the present reassessment is on the revised requested reimbursement criteria: patients with recent ACS (in the previous year) who have an LDL-C level of 1.8 mmol/L or more despite taking moderate-to-high intensity statin therapy, with or without ezetimibe.

Patient, Clinician, and Drug Plan Perspectives

Patient Group Input

No patient groups provided input on the present reassessment of evolocumab.

A summary of past patient input submitted by the Cardiac Health Foundation of Canada was prepared by the CDA-AMC review team in the Clinical Review Report (Resubmission) on Evolocumab (Repatha) in December 2017, which is publicly available on the CDA-AMC website. The Cardiac Health Foundation of Canada is an organization that raises funds for and promotes programs and applied research on the rehabilitation and management of cardiovascular disease and provides education and resources on the prevention and management of cardiovascular disease in Canada. Patient input was gathered by the patient group through an online survey (N = 55) and 1 telephone interview; respondents were patients with atherosclerosis and their caregivers.

Among the survey respondents, experience with rosuvastatin, atorvastatin, ezetimibe, and bypass surgery were described with varying degrees of effectiveness. The survey respondents reported that the most common side effects associated with their current treatment were digestive-related, including gas, constipation, and upset stomach. According to the survey respondents, the most difficult-to-tolerate side effects associated with current medications were muscle pain, discomfort, and weakness.

The survey respondents identified the following unmet need: alternative treatment options to statins. More specifically, in the context of elevated cholesterol levels despite a maximally tolerated statin dose and AEs commonly associated with statin therapy (i.e., loss of muscle function and muscle weakness), patients expect evolocumab to lower cholesterol to target levels with minimal side effects. In particular, most patients indicated that a loss of muscle function is an AE they are not willing to tolerate.

Clinician Input

Input From Clinical Experts Consulted by CDA-AMC

All CDA-AMC review teams include at least 1 clinical specialist with expertise in the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., 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 primary hyperlipidemia.

Unmet Needs

The clinical experts indicated that the association between a reduction in cholesterol and prevention of or delay in the onset of ASCVD and disease manifestation is well established. Hence, the goal of treatment with a PCSK9 inhibitor is to reduce LDL-C (or non-HDL-C or ApoB) levels to below the thresholds referenced in the 2021 CCS dyslipidemia guidelines for secondary prevention (LDL-C < 1.8 mmol/L or non-HDL-C < 2.4 mmol/L or ApoB < 0.7 g/L) in patients with ASCVD who are at high cardiovascular risk.⁵ According to the clinical experts, long-term, this translates into a reduced risk of clinical events and improved survival and quality of life.

The clinical experts indicated that most patients at high risk for cardiovascular events are not meeting LDL-C (or non-HDL-C or ApoB) target levels with available treatment options. Moreover, the clinical experts indicated that nonadherence due to real or perceived intolerance to high-intensity statins, such as intolerance due to reported myalgias, is a challenge in clinical practice; they estimated that 50% of patients discontinue their statin in the year after an ACS event.

The clinical experts highlighted the lack of access to advanced therapies, including PCSK9 inhibitors, experienced by patients with ASCVD. According to the clinical experts, lack of access is multifactorial, and includes coverage for and access to clinicians with experience using advanced therapies.

Place in Therapy

The clinical experts referenced the 2021 CCS dyslipidemia guidelines5 to indicate that ezetimibe and PCSK9 inhibitors are second-line treatment options for the management of primary hyperlipidemia for secondary prevention. More specifically, the clinical experts indicated that ezetimibe and/or evolocumab would be used in addition to a maximally tolerated statin dose to meet LDL-C (or non-HDL-C or ApoB) target levels. According to clinical expert input, 2 weeks is an appropriate trial of a statin. The clinical experts advise that the guidelines be referred to for additional context on the place in therapy of evolocumab in relation to ezetimibe (this is described in detail in the Standards of Therapy section of this report). For patients who are intolerant of or have contraindications to statins, the clinical experts indicated that evolocumab would be an alternative therapy to statins, with or without ezetimibe.

Of note, the clinical experts do not anticipate that evolocumab will become first-line therapy for the management of primary hyperlipidemia. However, they do anticipate that there will be a substantial demand for evolocumab if it becomes widely available and is reimbursed by public drug plans for the indication under review.

Patient Population

The clinical experts referenced the 2021 CCS dyslipidemia guidelines⁵ to identify the patient population most in need of an intervention for the management of primary hyperlipidemia in secondary prevention, and identified the subset of patients with ASCVD (at high cardiovascular risk) who have been shown to derive the largest benefit from the intensification of statin therapy with the addition of a PCSK9 inhibitor. This includes patients with recent ACS (occurring in the 52 weeks after hospitalization for the index ACS) and patients with additional cardiovascular risk enhancers.5 These patients are identified based on medical history, with or without a clinical exam, and standard lipid tests. The clinical experts noted that the assessment of clinical risk in secondary prevention is heterogenous and, as such, there is considerable underdiagnosis of risk that would be considered an indication for the initiation of lipid-lowering therapy.

The clinical experts indicated that all patients with ASCVD whose LDL-C (or non-HDL-C or ApoB) remains above the threshold, despite a maximally tolerated statin dose, are suited for treatment with evolocumab. According to the clinical experts, patients with end-stage renal disease, hepatic disease, New York Heart Association class IV heart failure, congestive heart failure, and an expected survival of less than 6 months are least suitable for treatment with evolocumab.

Assessing Treatment Response

The clinical experts noted that improved survival or reduced risk of cardiovascular events cannot be assessed in an individual patient in practice; they are considered established by clinical trials. Thus, the clinical experts indicated that reduction in LDL-C (or non-HDL-C or ApoB) is used to determine treatment response in practice. The clinical experts explained that requiring a specific percent reduction in these lipid parameters is arbitrary. Instead, they referenced the lipid parameter thresholds in the 2021 CCS dyslipidemia guidelines⁵ as treatment goals. Based on clinical expert input, treatment response is assessed every 6 to 12 months, depending on cardiovascular risk. For example, patients with higher cardiovascular risk are advised to have an earlier follow-up to ensure that treatment goals will be met. Additionally, the clinical experts noted that in some patients, evolocumab would allow for the down-titration of a statin that is associated with AEs and a negative impact on quality of life.

Discontinuing Treatment

When deciding whether to discontinue treatment with evolocumab, the clinicals experts said they would consider the adverse effects associated with treatment and competing risks from other diseases that limit life expectancy.

Prescribing Considerations

The clinical experts indicated that although a specialist is not required for the diagnosis, treatment, and monitoring of patients receiving evolocumab, management should ideally be carried out in an outpatient clinic or hospital setting by a clinician who has experience with evolocumab.

Clinician Group Input

This section was prepared by the CDA-AMC review team based on the input provided by clinician groups. The full clinician group submissions received are available in the consolidated clinician group input document for this review on the project website publicly accessible here.

A total of 9 clinician groups provided input on the present reassessment of evolocumab: Canadian Dyslipidemia Guideline Committee; McMaster Lipid Clinic; British Columbia Lipid Specialists; Cambridge Cardiac Rehab Program; Western University, Division of Cardiology, Cardiac Rehabilitation and Secondary Prevention Program; University of British Columbia and Vancouver General Hospital and St. Paul’s Hospital Cardiac Intensive Care Unit; University of Toronto faculty and clinicians at St Michael’s Hospital; Division of Cardiology, University of Ottawa Heart Institute; and a group of primary care and specialist physicians who treat patients with coronary artery disease and ACS in Canada.

The clinician groups identified the following limitations with currently available treatments (unmet needs) in patients with recent ACS: limited access to PCSK9 inhibitors due to cost, experience of side effects from and/or intolerance to available drugs (which have an impact on adherence to treatment), and variable treatment response (e.g., treatment targets for LDL-C not met). The University of Ottawa Heart Institute emphasized that although the majority of patients with ASCVD experience a reduction in their LDL-C level to below 1.8 mmol/L with high-dose statin therapy, with or without ezetimibe, a subset of patients continues to have elevated lipid levels due to severe polygenic hypercholesterolemia and intolerance or contraindication to high-dose statin therapy. The clinician group further suggested that this subset of patients who are at high risk of recurrent cardiovascular events would benefit from additional lipid-lowering treatment in the form of a PCSK9 inhibitor.

The Canadian Dyslipidemia Guideline Committee, McMaster Lipid Clinic, and the group of primary care and specialist physicians in Canada referenced the 2021 CCS dyslipidemia guidelines⁵ to indicate that a PCSK9 inhibitor would be used as an add-on therapy after the initiation of maximally tolerated statin therapy, with or without ezetimibe, in patients with elevated LDL-C levels. More specifically, evolocumab would be used in the second line after a maximally tolerated dose of statin or in the third line after statin and ezetimibe. The Canadian Dyslipidemia Guideline Committee also referenced the guidelines to identify candidates for evolocumab, who comprise patients with either a recent ACS (i.e., occurring in the 52 weeks after hospitalization for the index ACS) or prior ASCVD with any of the following: diabetes or metabolic syndrome, polyvascular disease, symptomatic PAD, recurrent MI, MI in the past 2 years, previous coronary artery bypass graft surgery, LDL-C level of 2.6 mmol/L or more, or HeFH. The clinician groups indicated that treatment response is assessed based on the percent reduction in LDL-C (or non-HDL-C or ApoB) levels, compared to pretreatment levels in practice.⁵

Drug Program Input

The drug programs provide input on each drug being reviewed through the CDA-AMC 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 by CDA-AMC are summarized in Table 6.

Table 6: Summary of Drug Plan Input and Clinical Expert Response

ApoB = apolipoprotein B; ASCVD = atherosclerotic cardiovascular disease; CCS = Canadian Cardiovascular Society; CDEC = Canadian Drug Expert Committee; HDL-C = high-density lipoprotein cholesterol; HoFH = homozygous familial hyperlipidemia; LDL-C = low-density lipoprotein cholesterol; SC = subcutaneous.

^aSubstantial refers to an LDL-C level greater than 2.2 mmol/L, an ApoB level greater than 0.80 g/L, or a non-HDL-C level greater than 2.9 mmol/L despite a maximally tolerated statin dose.

^bRecent ACS is defined in the guidelines as occurring in the 52 weeks after hospitalization for the index ACS.

^cCardiovascular risk enhancers, according to the guidelines, include diabetes mellitus or metabolic syndrome, polyvascular disease, symptomatic PAD, history of MI, MI in the past 2 years, previous coronary artery bypass graft surgery, an LDL-C level of 2.6 mmol/L or more or HeFH, and a lipoprotein(a) level of 60 mg/dL or more.

^dModest refers to an LDL-C level of 1.8 to 2.2 mmol/L, an ApoB level of 0.70 to 0.80 g/L, or a non-HDL-C level of 2.4 mmol/L to 2.9 mmol/L despite a maximally tolerated statin dose.

Clinical Evidence

The objective of the Clinical Review Report in the present reassessment is to review and critically appraise the new clinical evidence submitted by the sponsor on the beneficial and harmful effects of evolocumab 140 mg/mL and 120 mg/mL subcutaneous injections in the treatment of primary hyperlipidemia in the subgroup of adult patients with ASCVD who have recent ACS (in the previous year). The focus will be placed on comparing evolocumab to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the present reassessment of evolocumab is presented in 3 sections, with our critical appraisal of the evidence included at the end of each section. The first section, the Systematic Review, includes a pivotal study that was selected in accordance with the sponsor’s systematic review protocol. The second section includes sponsor-submitted long-term extension studies. The third section includes an additional study that was considered by the sponsor to address an important gap in the systematic review evidence.

Included Studies

The pivotal trial, the FOURIER trial, was previously reviewed and critically appraised by CDA-AMC in the December 2017 Clinical Review Report (Resubmission) on Evolocumab (Repatha), which is publicly available on the CDA-AMC website. In addition to the FOURIER trial, new clinical evidence from the following is included in the CADTH reassessment and appraised in this document:

• 2 subgroup analyses of the pivotal FOURIER study identified in the systematic review

• 2 long-term extension studies of the FOURIER trial

• 1 additional study addressing a gap in the systematic review evidence.

Systematic Review

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

Description of Studies

Characteristics of the included studies are summarized in Table 7.

The FOURIER Trial¹²

Detailed information on the FOURIER trial can be found in the December 2017 Clinical Review Report (Resubmission) on Evolocumab (Repatha), which is publicly available on the CDA-AMC website. The study design of the FOURIER trial is presented in Figure 1.

The FOURIER trial was a phase III, double-blind, placebo-controlled, randomized clinical trial (N = 27,564). The primary objective was to evaluate the effect of evolocumab, compared with placebo, on the risk of cardiovascular death, MI, stroke, hospitalization for unstable angina, and coronary revascularization, whichever occurs first, in patients with clinically evident ASCVD. Patients were enrolled between February 8, 2013, and June 5, 2015, at 1,242 clinical centres, 46 of which were located in Canada.² Patients who were taking stable (i.e., for at least 4 weeks), optimized lipid-lowering therapy at screening with no planned or anticipated changes for the duration of the trial proceeded directly to the final screening visit. For patients who required a change in lipid-lowering therapy, their dose was titrated until their lipid levels were considered optimally managed; they then proceeded to the final screening visit. However, these patients could only be randomized after at least 4 weeks (i.e., stable) of optimized lipid-lowering therapy. All procedures up to randomization were designed to be completed no more than 15 weeks after patients provided informed consent.

Patients who met all of the inclusion criteria and none of the exclusion criteria were randomized in a 1:1 ratio to receive either subcutaneous evolocumab or matching placebo injection (with a dosing schedule reflecting patient preference) in addition to their stable, optimized, background lipid-lowering therapy (i.e., at least a statin, with or without ezetimibe). Treatment allocation was based on a computer-generated randomization schedule prepared by the sponsor before the start of the study, using the Interactive Voice or Web Response System. Randomization was stratified by the final screening LDL-C level (< 2.2 mmol/L versus ≥ 2.2 mmol/L) and by geographic region (Europe, North America, Latin America, and the Asia-Pacific).

Laboratory assessments were performed during screening, on day 1, and at weeks 4, 12, and 24, and every 24 weeks thereafter. Patients were to complete all planned visits, regardless of adherence to study drug. To maintain blinding of the treatment assignment, central laboratory results were not reported to the investigator until unblinding of the clinical database. A patient’s treatment assignment was only unblinded when required for further management of the patient. As this was an event-driven study, treatment continued until a minimum of 1,630 patients experienced an event adjudicated by an independent external committee as qualifying for a key secondary end point event of cardiovascular death, MI, or stroke. The estimated study duration was 56 months from the date the first patient was randomized. Patients who completed the trial had the option to enrol in an OLE study^19,20 of the FOURIER trial.

Gencer et al. Study¹³

The aim of this study was to evaluate the risks of major adverse cardiovascular events as a function of time from the date of the qualifying MI and to evaluate the effect of evolocumab on cardiovascular outcomes in patients who experienced an MI in the previous year. Note that only the latter objective is considered relevant for the purpose of this review.

Sabatine et al. (2018) Study¹⁴

The aim of this study was to assess the efficacy of evolocumab in 3 subgroups in the FOURIER trial: timing from the most recent MI, number of prior MIs, and the presence of residual multivessel coronary artery disease. Note that only the subgroup by timing of prior MI is considered relevant for the purpose of this review.

Populations

Key inclusion and exclusion criteria used in the included studies are summarized in Table 7.

FOURIER Trial¹²

The trial included patients aged 40 to 85 years, inclusive, with a history of ASCVD, evidenced by a history of MI or nonhemorrhagic stroke, or symptomatic PAD, and additional characteristics that placed them at high cardiovascular risk. Other key inclusion criteria were a most recent fasting LDL-C level of 1.8 mmol/L or more (or a non-HDL-C level of 2.6 mmol/L or more) after at least 2 weeks of stable, optimized lipid-lowering therapy. Key exclusion criteria were the most recent MI or stroke occurring in the 4 weeks before randomization; NYHA class III or IV heart failure or a last known left ventricular ejection fraction of less than 30%; and a planned or expected cardiac surgery or revascularization in the 3 months after randomization.

Table 7: Details of the FOURIER Trial and the Gencer et al. and Sabatine et al. (2018) Studies

ALT = alanine aminotransferase; ApoA1 = apolipoprotein A1; ApoB = apolipoprotein B; AST = aspartate aminotransferase; ECG = electrocardiogram; eGFR = estimated glomerular filtration rate; HDL-C = high-density lipoprotein cholesterol; hsCRP = high-sensitivity C-reactive protein; KM = Kaplan-Meier; LDL-C = low-density-lipoprotein cholesterol; LLN = lower limit of normal; Lp(a) = lipoprotein (a); MI = myocardial infarction; NA = not applicable; NYHA = New York Heart Association; OLE = open-label extension; PAD = peripheral artery disease; RCT = randomized controlled trial; SC = subcutaneous; TEAE = treatment-emergent adverse event; TIA = transient ischemic attack; TSH = thyroid stimulating hormone; ULN = upper limit of normal; VLDL-C = very-low-density-lipoprotein cholesterol.

^aMajor risk factors included type 1 or type 2 diabetes, age of 65 years or older, a qualifying MI or stroke in the 6 months before screening, current daily cigarette smoking, an additional prior MI or nonhemorrhagic stroke (excluding the qualifying diagnosis), or symptomatic PAD if enrolled with a history of MI or nonhemorrhagic stroke.

^bMinor risk factors included a history of non-MI-related coronary revascularization, residual coronary artery disease with 40% or greater stenosis in 2 or more large vessels, HDL-C levels of less than 1.0 mmol/L for males and greater than 1.3 mmol/L for females, hsCRP greater than 2.0 mg/L, LDL-C level of 3.4 mmol/L or more or non-HDL-C level of 4.1 mmol/L or more, and metabolic syndrome as defined in the study protocol.

^cTime to hospitalization for unstable angina was not a prespecified end point in the FOURIER trial; an ad hoc analysis was performed to ensure that results were provided for each individual component of the primary end point.

Sources: Clinical Study Report of the FOURIER trial,¹² additional information received from the sponsor on January 30, 2024,¹⁶ Sabatine et al. (2018),¹⁴ and Gencer et al.¹³ Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

Figure 1: Study Design of the FOURIER Trial

EOS = end of study; EP = end point; HDL-C = high-density-lipoprotein cholesterol; IP = investigational product; LDL-C = low-density lipoprotein cholesterol; PO = oral administration; Q12W = every 12 weeks; Q2W = every 2 weeks; Q24W = every 24 weeks; QM = once every month; SC = subcutaneous.

Source: Clinical Study Report of the FOURIER trial.¹²

Gencer et al. Study¹³

This study included patients from the FOURIER trial with a known date of prior MI.

Sabatine et al. (2018) Study¹⁴

This study included patients in the FOURIER trial who experienced a prior MI.

Interventions

FOURIER Trial¹²

Intervention and Comparator

Eligible patients were randomized in a 1:1 ratio to receive either subcutaneous evolocumab (140 mg once every 2 weeks administered as 1 prefilled autoinjector or pen or 420 mg once every month administered as 3 prefilled autoinjectors or pens) or matching placebo injection. The dosing schedule was selected in accordance with patient preference. On day 1, week 2 (for the every-2-week schedule only), and week 4, patients received instruction on and supervision in the use of the prefilled autoinjector or pen and remained for at least 30 minutes after injection for observation. Thereafter, patients administered the assigned intervention at a location external to the study centre. Dose adjustments of interventions were not permitted during the trial, with the exception of a switch in dose schedule. Every 3 months after the first 24 weeks of treatment, patients had the opportunity to switch between the 2 dosing schedules in accordance with their preference and subject to study drug availability.

Concomitant Lipid-Lowering Therapies

At randomization, all patients received optimized, background, lipid-lowering therapy, per study protocol (Table 8). Optimized, background, lipid-lowering therapy was defined as an effective statin dose (i.e., high-to-moderate intensity; at least atorvastatin 20 mg daily or equivalent), or, where locally approved, as a highly effective statin dose (i.e., at least atorvastatin 40 mg daily or equivalent, or any permitted dose of background statin therapy in combination with ezetimibe). Definitions of statin intensity were based on joint American College of Cardiology and the American Heart Association Guidelines (Table 9). In addition to background statin therapy, ezetimibe and other commercially available lipid-lowering therapy at dosages approved by local regulatory authorities were permitted.

Table 8: Permitted Background Statin Therapy in the FOURIER Trial

Note: Only the statins listed in Table 5 were used during study participation. Other lipid-lowering therapy was not a requirement of the FOURIER trial.

^aFor patients with LDL-C greater than 2.6 mmol/L and not receiving highly effective statin therapy, the investigator was required to confirm that the selected dose of statin was optimized and appropriate for the patient for the duration of the study (i.e., a higher dose of statin therapy was not appropriate for the patient because of patient refusal, dose not tolerated, dose not available in the country, or other significant concern).

^bThis dose was considered a highly effective therapy in the trial. Additionally, any of the statin doses listed in the table that were administered in combination with ezetimibe were considered to be highly effective therapy.

^cSimvastatin 80 mg was not available in all participating countries in the trial. Approval by the local regulatory authority was required for patients using the simvastatin 80 mg dose in the trial.

Source: Clinical Study Report of the FOURIER trial.¹²

Table 9: Background Statin Intensity Defined in the FOURIER Trial

ACC = American College of Cardiology; AHA = American Heart Association; NA = not applicable.

Source: Clinical Study Report of the FOURIER trial.¹²

Prohibited Treatments

The following treatments were prohibited during the trial:

• all lipid therapies not being taken at the final screening visit, with the exception of ezetimibe, described later

• mipomersen, lomitapide, and fibrates and derivatives, with the exception of fenofibrate at a stable (≥ 6 weeks before the final screening visit), optimized dose, per study protocol

• any PCSK9 inhibitor other than evolocumab that was provided in the trial.

In general, no changes were made to patients’ background lipid-lowering therapy from baseline (i.e., end of screening) to the end of the study. However, because of the results of the IMPROVE-IT study,³⁸ the addition of ezetimibe to a patient’s lipid-lowering regimen may be considered in the setting of an on-study ACS. The IMPROVE-IT study included adult patients who had been hospitalized in the preceding 10 days for an ACS, defined as an acute MI, with or without ST-segment elevation on electrocardiography, or high-risk unstable angina.³⁸ Additionally, if a concomitant drug with the potential to impact the metabolism of a patient’s particular statin was required during the study, it may be necessary to withdraw or change the background statin therapy.

Outcomes

A list of efficacy end points assessed in this Clinical Review Report is provided in Table 10, followed by descriptions of the outcome measures. Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence,² as well as any outcomes identified as important to this review by the clinical experts consulted by CDA-AMC and input from patient and clinician groups and public drug plans. Using the same considerations, the CDA-AMC review team selected end points that were considered to be most relevant to expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. Select notable harms outcomes considered to be important to expert committee deliberations were also assessed.

Table 10: Outcomes Summarized From the Gencer et al. and Sabatine et al. (2018) Studies

LDL-C = low-density-lipoprotein cholesterol; MI = myocardial infarction.

^aThe estimated study duration was 56 months from the date the first patient was randomized.

^bMultiplicity was not taken into account for subgroup analyses in the Gencer et al.¹³ and Sabatine et al. (2018)¹⁴ studies.

Sources: Clinical Study Report of the FOURIER trial,¹² Sabatine et al. (2018),¹⁴ and Gencer et al.¹³ Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

FOURIER Trial¹²

Clinical Outcomes

The primary end point was time to cardiovascular death, MI, hospitalization for unstable angina, stroke, or coronary revascularization, whichever occurs first (including ischemic and hemorrhagic stroke). The key secondary end point was time to cardiovascular death, MI, or stroke, whichever occurs first. These end points were selected because they have the potential to be modified by a reduction in LDL-C. Other secondary end points included components of the composite end point, death by any cause, first hospitalization for worsening heart failure, and transient ischemic attack.

All fatal and nonfatal events, as well as events of new-onset diabetes, that occurred from randomization up to each patient’s end-of-study visit were collected by the investigator as potential end points for adjudication by an independent, external CEC, in the form of the Thrombolysis in Myocardial Infarction Study Group in Boston, Massachusetts. Additionally, the sponsor study team conducted regular reviews of the electronic case report forms used to identify any SAEs, lab parameters, and electrocardiogram findings that could indicate a potential end point. Such cases were brought to the attention of the study sites so that they could report the event as a potential end point for adjudication.

Adjudication of events was based on standardized definitions in the 2012 Standardized Definitions for Cardiovascular and Stroke End point Events in Clinical Trials and the 2012 Third Universal Definition of Myocardial Infarction.³⁹ The CEC was blinded to treatment allocation and reviewed events according to prespecified criteria. Each potential end point was independently reviewed by 2 assigned adjudicators on the committee. If the adjudication results were concordant, then the adjudication of that potential end point was complete. If the adjudication results were discordant, then the potential end point was discussed at a committee meeting until a consensus was reached. If the committee was unable to reach consensus, then the case was submitted to the chairperson for final adjudication. If a reported potential end point was adjudicated as negative (i.e., did not meet the definition of an end point), the event was reclassified as an AE or SAE and was reported to regulatory agencies, as required.

Lipid Parameter Outcomes

Lipid parameter outcomes related to LDL-C were exploratory end points. For analysis of complete lipid profiles, fasting (at least 9 hours) blood samples were transported to 1 of the 2 generally used central laboratories, where standard laboratory procedures were used for lipid assessments. For all analyses related to LDL-C, unless otherwise specified, a reflexive approach based on the Friedewald equation was used to calculate LDL-C. If the calculated LDL-C was less than 1.0 mmol/L or if triglycerides were greater than 4.5 mmol/L, preparative ultracentrifugation LDL-C was determined.

Throughout the study, the central laboratory compared LDL-C concentrations with the patient’s prior assessed LDL-C level, without unblinding the study team, investigator, or site staff. If the measured LDL-C exceeded a preset threshold, the study centre was notified by an automated system to instruct the patient on adherence to the study drug, statin, ezetimibe, and any other concomitant lipid-lowering therapy, if applicable, and diet. To maintain the blind, the same reminder was provided to additional patients in each treatment group based on an algorithm that balanced the frequency of alerts for both treatment groups. Investigators were also informed if triglycerides were greater than 11.3 mmol/L to facilitate appropriate follow-up with the patient.

The Lipid Monitoring Committee monitored the LDL-C separation between treatment groups over the study period and was involved in the follow-up of patients meeting lipid-alert thresholds used by the central laboratory. This was a separate external, independent committee that reviewed unblinded lipid results by treatment group to ensure that the design parameter was being met, as described in the study protocol. This committee was advisory in nature and did not have access to clinical outcomes data; its primary function was to review treatment adherence and other relevant nonoutcomes data.

Patients were also monitored for very low LDL-C levels (< 0.6 mmol/L); these data were provided to the Data Monitoring Committee. This was a separate external, independent committee that reviewed the accumulating data from all ongoing studies of evolocumab for avoidable increased risk of harm to patients.

Notable Harms

The incidence of TEAEs was a safety end point in the FOURIER trial. All AEs were coded, using the Medical Dictionary for Regulatory Activities (MedDRA) Version 9.1, to a system organ class and preferred term. TEAEs were defined as any untoward medical occurrence in a patient, including worsening of a preexisting medical condition, that occurred between the first dose of the study drug and the end of the study.

When comparing evolocumab with relevant comparators for the indication under review, the following harms outcomes were considered important to expert committee deliberations: hypersensitivity, injection-site reactions, muscle events, neurocognitive events, demyelination events and peripheral neuropathy, hepatitis C infection, transaminase elevations, and potential hepatic disorders.

Statistical Analysis

FOURIER Trial¹²

The sample-size calculation was based on the key secondary end point (composite of cardiovascular death, MI, and stroke).¹² The following assumptions were made: a 26-month enrolment period, a placebo event rate of 2% per year,^40-48 a 3% loss-to-follow-up rate over the study duration of 56 months, attenuation of the treatment effect would occur due to a 3-month treatment lag, and noncompliance with study drug of 10% per year over the study. The HR for the triple composite end point was assumed to be 0.80, based on the 2010 Cholesterol Treatment Trialists' (CTT) Collaboration meta-analysis, which assessed the relation between LDL-C reduction and cardiovascular events.⁴⁹ The overall type I error was controlled at a 0.05 significance level. Based on these assumptions, the attenuated HR was assumed to be 0.85, and based on a 2-sided log-rank test of demonstrating the superiority of evolocumab over placebo, a total sample size of 27,500 patients, with 1,630 patients experiencing a key secondary end point event, was required to ensure 90% power.⁵⁰

The sponsor indicated that the subgroup analyses conducted in the Gencer et al. and Sabatine et al. (2018) studies were based on the efficacy analysis methods described in the FOURIER trial.¹⁵ The statistical analysis of efficacy end points in the Gencer et al. and Sabatine et al. (2018) studies is summarized in Table 11. Multiplicity was not taken into account for subgroup analyses.^2,12

The primary analysis of all time-to-event end points, including the primary and secondary end points, was derived from a 2-sided log-rank test stratified by the randomization stratification factors to compare the survival functions of each treatment group. Kaplan-Meier curves were estimated and graphically displayed by treatment, and Kaplan-Meier estimates and corresponding 95% CIs were calculated. HRs and corresponding 95% CIs were estimated from a Cox model stratified by the randomization stratification factors. For all time-to-event end points, there were no imputations of data, with the exception of incomplete dates of events.

Events from the patient randomization date to the patient end-of-study date were included in the primary analyses of the primary and secondary efficacy end points. For all time-to-event calculations, the starting point was the patient’s date of randomization, and the event onset date was the onset date adjudicated by the CEC. The censoring date for patients without an event was the patient’s last nonfatal potential end point collection date. For patients who discontinued the study early (due to withdrawal of consent or loss to follow-up), vital status data were collected during the end-of-study visit period, before the overall study end date, as permitted by local law. All adjudicated cases of death collected up to the study end date, including cardiovascular death, noncardiovascular death, and undetermined death, were included in the analysis based on the adjudicated results.

Gencer et al. Study¹³

For the subgroup analyses, patients with a known date of prior MI were stratified according to recent MI (in the year before randomization) or remote MI (more than 1 year before randomization). Of note, patients who experienced an MI in the 4 weeks before randomization were excluded.

Sabatine et al. (2018) Study¹⁴

For the subgroup analyses, patients were stratified according to the number of prior MIs; the timing of prior MI (in relation to randomization); and the extent of coronary disease, defined as stenosis of 40% or more in 2 or more large vessels. According to the investigator, this was part of a prespecified analysis; the subgroup by timing of prior MI (< 2 years versus ≥ 2 years) was derived from previous studies of patients with a history of MI that identified time since MI to be a feature of patients who are at high cardiovascular risk and who would derive a greater relative and/or absolute risk reduction from therapies. For the purpose of this review, only the subgroup stratified by timing of prior MI is included in this report.

Analysis Populations

FOURIER Trial¹²

The sponsor indicated that the subgroup analyses conducted in the Gencer et al. and Sabatine et al. (2018) studies were based on the efficacy analysis methods described in the FOURIER trial.15 Therefore, the full analysis set (Table 12) was used for the primary analysis of the primary and secondary efficacy end points; all patients were analyzed according to their randomized treatment assignment.

Table 11: Statistical Analysis in the Gencer et al. and Sabatine et al. (2018) Studies

CI = confidence interval; HR = hazard ratio; KM = Kaplan-Meier; LDL-C = low-density-lipoprotein cholesterol.

Sources: Clinical Study Report of the FOURIER trial¹² and additional information received from the sponsor on December 21, 2023.¹⁵ Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

Table 12: Analysis Populations in the FOURIER Trial

Note: For the purpose of this review, the pharmacokinetic and electrocardiogram analysis sets are not included in Table 9.

Sources: Clinical Study Report of the FOURIER trial¹² and additional information received from the sponsor on December 21, 2023.¹⁵ Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

Results

FOURIER Trial¹²

Detailed information — including patient disposition, baseline characteristics, exposure to the study drug and concomitant lipid-lowering therapies, and efficacy and harms — on results from the FOURIER trial, which compared treatment groups, can be found in the December 2017 Clinical Review Report (Resubmission) on Evolocumab (Repatha), which is publicly available on the CDA-AMC website.

Patient Disposition

Gencer et al. Study¹³

A summary of patient disposition from the Gencer et al. study is presented in Table 13. The proportion of patients who discontinued the study was similar in the 2 treatment groups by subgroup, and were relatively low (i.e., ≤ 1.0% of patients in each treatment group).¹⁶

Table 13: Summary of Patient Disposition From the Gencer et al. Study (All Randomized Patients)

FAS = full analysis set; MI = myocardial infarction; NA = not available; PP = per protocol.

Source: Additional information received from the sponsor on January 30, 2024.¹⁶

Sabatine et al. (2018) Study¹⁴

A summary of patient disposition from the Sabatine et al. (2018) study is presented in Table 14. The proportion of patients who discontinued the study was similar in the 2 treatment groups by subgroup, and was relatively low (i.e., < 1.0% of patients in each treatment group).¹⁶

Table 14: Summary of Patient Disposition From the Sabatine et al. (2018) Study (All Randomized Patients)

FAS = full analysis set; MI = myocardial infarction; NA = not available; PP = per-protocol.

Source: Additional information received from the sponsor on January 30, 2024.¹⁶

Baseline Characteristics

Gencer et al. Study¹³

Summaries of baseline characteristics and concomitant lipid-lowering therapy use at baseline from the Gencer et al. study are presented in Table 15 and Table 16, respectively. The proportion of patients according to baseline demographics, relevant medical history, lipid parameter, and concomitant lipid-lowering therapy use was similar in the 2 treatment groups by subgroup.

Subgroup of Patients Who Experienced a Prior MI in the Previous Year¹⁶

A total of 2,821 patients were randomized to receive evolocumab and 2,890 patients were randomized to receive placebo. The proportion of patients who were male was 78.6% (n = 2,217) in the evolocumab group and 77.0% (n = 2,225) in the placebo group. The proportion of patients who were female was 21.4% (n = 604) in the evolocumab group and 23.0% (n = 665) in the placebo group. Most patients were white (85.5% [n = 2,413] of patients in the evolocumab group and 85.0% [n = 2,457] of patients in the placebo group). The remaining patients were Asian (9.9% [n = 278] of patients in the evolocumab group and 9.6% [n = 276] of patients in the placebo group), Black or African American (1.7% [n = 47] and 1.9% [n = 56] of patients, respectively), American Indian or Alaska Native [wording from original source] (0.4% [n = 12] and 0.7% [n = 20] of patients, respectively), and Native Hawaiian or other Pacific Islander (< 0.1% [n = 1] and 0.1% [n = 4] of patients, respectively). The mean age of patients was 59.7 years (SD = 9.3 years) in the evolocumab group and 59.5 years (SD = 9.2 years) in the placebo group.

Table 15: Summary of Baseline Characteristics From the Gencer et al. Study (All Randomized Patients)

HDL-C = high-density-lipoprotein cholesterol; hsCRP = high-sensitivity C-reactive protein; LDL-C = low-density-lipoprotein cholesterol; MI = myocardial infarction; NA = not available; PAD = peripheral artery disease; SD = standard deviation.

^aMetabolic syndrome was defined as at least 3 of the following: waist circumference of greater than 40 inches for males and greater than 35 inches for females, triglyceride level of 1.7 mmol/L or greater, HDL-C level of less than 1.0 mmol/L for males and less than 1.3 mmol/L for females, systolic blood pressure of 130 mm Hg or greater or diastolic blood pressure of 85 mm Hg or greater or hypertension treated with medication, and fasting glucose level of 5.6 mmol/L or greater identified by the central laboratory at final screening.

Source: Additional information received from the sponsor on January 30, 2024.¹⁶

The mean time from MI to enrolment was 5.379 months (SD = 2.965 months) in the evolocumab group and 5.355 months (SD = 2.911 months) in the placebo group. Almost all patients had at least 1 major cardiovascular risk factor or at least 2 minor cardiovascular risk factors (99.8% [n = 2,814] of patients in the evolocumab group and 99.8% [n = 2,884] of patients in the placebo group). The remaining patients had no major risk factors and fewer than 2 minor risk factors (0.2% [n = 7] of patients in the evolocumab group and 0.2% [n = 6] of patients in the placebo group). At baseline, the mean LDL-C level was 2.453 mmol/L (SD = 0.647 mmol/L) in the evolocumab group and 2.467 mmol/L (SD = 0.647 mmol/L) in the placebo group.

Almost all patients were taking a statin at baseline (99.9% [n = 2,819] of patients in the evolocumab group and 100.0% [n = 2,889] of patients in the placebo group). The most commonly used statin — atorvastatin — was used by 84.6% (n = 2,387) of patients in the evolocumab group and 84.4% (n = 2,438) of patients in the placebo group. Most patients were taking high-intensity statins (77.7% [n = 2,191] of patients in the evolocumab group and 77.0% [n = 2,224] of patients in the placebo group) and moderate-intensity statins (22.2% [n = 626] of patients in the evolocumab group and 22.7% [n = 657] of patients in the placebo group). A total of 3.2% (n = 91) of patients in the evolocumab group and 3.3% (n = 95) of patients in the placebo group were taking ezetimibe at baseline. Overall, the proportion of patients taking a nonstatin lipid-lowering therapy — including fibrates (1.1% [n = 30] and 1.2% [n = 35], respectively), nicotinic acid and derivatives (0.2% [n = 5] and 0.1% [n = 3], respectively), bile acid sequestrants (< 0.1% [n = 1] in both treatment groups), and other lipid-lowering therapies (5.0% [n = 140] and 5.2% [n = 150], respectively) — was similar in the evolocumab and placebo groups.

Subgroup of Patients Who Were More Than 1 Year Beyond Their MI¹⁶

In general, the baseline characteristics of patients who were 1 year or more beyond their MI were similar to those of patients who experienced an MI in the previous year. A total of 8,308 patients were randomized to receive evolocumab and 8,301 patients were randomized to receive placebo. The mean time from MI to enrolment was 85.739 months (SD = 75.162 months) in the evolocumab group and 85.746 months (SD = 74.518 months) in the placebo group. Note that there was limited information available on cardiovascular risk factors for this subgroup.

Table 16: Summary of Baseline Concomitant Lipid-Lowering Therapy From the Gencer et al. Study (All Randomized Patients)

ACC = American College of Cardiology; AHA = American Heart Association; HMG-CoA = 3-hydroxy-3-methylglutaryl-coenzyme A; MI = myocardial infarction; NR = not reported.

Note: Coded using WHODrug Version June 1, 2016.

^aThe most commonly used drug (identified as a preferred term) by medication category listed is based on a frequency of greater than 100 patients in any treatment group at baseline.

Source: Additional information received from the sponsor on February 7, 2024.⁵¹

Sabatine et al. (2018) Study¹⁴

Summaries of baseline characteristics and concomitant lipid-lowering therapy use at baseline from the Sabatine et al. (2018) study are presented in Table 17 and Table 18, respectively. The proportion of patients according to baseline demographics, relevant medical history, lipid parameter, and concomitant lipid-lowering therapy use were similar in the 2 treatment groups by subgroup.

Subgroup of Patients Who Experienced an MI in the Previous 2 Years¹⁶

In general, the baseline characteristics of patients who experienced an MI in the previous 2 years were similar to those of patients who experienced an MI in the previous year. A total of 4,109 patients were randomized to receive evolocumab and 4,293 patients were randomized to receive placebo. The mean time from MI to enrolment was 9.191 months (SD = 6.441 months) in the evolocumab group and 9.366 months (SD = 6.544 months) in the placebo group.

Subgroup of Patients Who Were 2 Years or More Beyond Their MI¹⁶

In general, the baseline characteristics of patients who were 2 years or more beyond their MI were similar to the those of patients who were 1 year or more beyond their MI. A total of 7,020 patients were randomized to receive evolocumab and 6,898 patients were randomized to receive placebo. The mean time from MI to enrolment was 98.252 months (SD = 75.324 months) in the evolocumab group and 99.600 months (SD = 74.460 months) in the placebo group. Note that there was limited information available on cardiovascular risk factors for this subgroup.

Table 17: Summary of Baseline Characteristics From the Sabatine et al. (2018) Study (All Randomized Patients)

HDL-C = high-density-lipoprotein cholesterol; hsCRP = high-sensitivity C-reactive protein; LDL-C = low-density-lipoprotein cholesterol; MI = myocardial infarction; NA = not available; PAD = peripheral artery disease; SD = standard deviation.

^aMetabolic syndrome was defined as at least 3 of the following: waist circumference of greater than 40 inches for males and greater than 35 inches for females, triglyceride level of 1.7 mmol/L or greater, HDL-C level of less than 1.0 mmol/L for males and less than 1.3 mmol/L for females, systolic blood pressure of 140 mm Hg or greater or diastolic blood pressure of 85 mm Hg or greater or hypertension treated with medication, and fasting glucose level of 5.6 mmol/L or greater identified by the central laboratory at final screening.

Source: Additional information received from the sponsor on January 30, 2024.¹⁶

Table 18: Summary of Baseline Concomitant Lipid-Lowering Therapy From the Sabatine et al. (2018) Study (All Randomized Patients)

ACC = American College of Cardiology; AHA = American Heart Association; HMG-CoA = 3-hydroxy-3-methylglutaryl-coenzyme A; MI = myocardial infarction; NR = not reported.

Note: Coded using WHODrug Version June 1, 2016.

^aMost commonly used drug (identified as a preferred term) by medication category, listed based on a frequency of greater than 100 patients in any treatment group at baseline.

Source: Additional information received from the sponsor on February 7, 2024.⁵¹

Exposure to Study Treatments

Gencer et al. Study¹³

Study Drug¹⁶

A summary of patient exposure from the Gencer et al. study is presented in Table 19.

Subgroup of patients who experienced an MI in the previous year. Total study drug exposure was 5,899 patient-years in the evolocumab group and 6,063 patient-years in the placebo group. The median duration of study drug exposure was 25.791 months (interquartile range [IQR], 20.534 to 31.540 months) in the evolocumab group and 25.561 months (IQR, 20.271 to 31.934 months) in the placebo group. Total study exposure was 6,359 patient-years in the evolocumab group and 6,535 patient-years in the placebo group. The median duration of study exposure was 27.203 months (IQR, 22.111 to 32.493 months) in the evolocumab group and 27.269 months (IQR, 22.111 to 32.624 months) in the placebo group. Information on adherence was not available.

Subgroup of patients who were 1 year beyond their MI. The total patient-years of study drug exposure was 16,718 in the evolocumab group and 16,644 in the placebo group. The total patient-years of study exposure was 17,985 in the evolocumab group and 17,968 in the placebo group. The median durations of study drug exposure and study exposure were similar to the corresponding durations in the subgroup of patients who experienced an MI in the previous year. Information on adherence was not available.

Table 19: Summary of Patient Exposure From the Gencer et al. Study (Safety Analysis Set)

IQR = interquartile range; MI = myocardial infarction; NA = not available; SD = standard deviation.

Source: Additional information received from the sponsor on January 30, 2024.¹⁶

Sabatine et al. (2018) Study¹⁴

Study Drug¹⁶

A summary of patient exposure from the Sabatine et al. (2018) study is presented in Table 20.

Subgroup of patients who experienced an MI in the previous 2 years. The total patient-years of study drug exposure was 8,598 in the evolocumab group and 8,980 in the placebo group. The total patient-years of study exposure was 9,266 in the evolocumab group and 9,691 in the placebo group. The median durations of study drug exposure and study exposure were similar to the corresponding durations in the subgroup of patients who experienced an MI in the previous year. Information on adherence was not available.

Subgroup of patients who were 2 years or more beyond their MI. The total patient-years of study drug exposure was 14,018 in the evolocumab group and 13,727 in the placebo group. The total patient-years of study exposure was 15,078 in the evolocumab group and 14,812 in the placebo group. The median durations of study drug exposure and study exposure were similar to the corresponding durations in the subgroup of patients who were 1 year or more beyond their MI. Information on adherence was not available.

Table 20: Summary of Patient Exposure From the Sabatine et al. (2018) Study (Safety Analysis Set)

IQR = interquartile range; MI = myocardial infarction; NA = not available; SD = standard deviation.

Source: Additional information received from the sponsor on January 30, 2024.¹⁶

Efficacy

FOURIER Trial¹²

Detailed information on the key efficacy outcomes in the FOURIER trial can be found in the December 2017 Clinical Review Report (Resubmission) on Evolocumab (Repatha), which is publicly available on the CDA-AMC website.

The median length of follow-up was 25.99 months (IQR, 21.72 to 30.42 months) for all patients in the full analysis set of the FOURIER trial.

Clinical Outcomes From the Gencer et al.¹³ and Sabatine et al. (2018)¹⁴ Studies

The key efficacy results from the subgroup analyses in the Gencer et al. and Sabatine et al. (2018) studies are summarized in Table 22 and Table 21, respectively.

Cardiovascular Death, MI, or Stroke¹⁵

Of the patients who experienced an MI in the previous year, the composite end point of cardiovascular death, MI, and stroke was met by 6.45% (n = 182) of patients taking evolocumab and by 8.58% (n = 248) of patients taking placebo (HR = 0.75; 95% CI, 0.62 to 0.91). Of the patients who were 1 year or more beyond their MI, this composite end point was met by 6.04% (n = 502) of patients taking evolocumab and by 7.04% (n = 584) of patients taking placebo (HR = 0.85; 95% CI, 0.76 to 0.96).

Of the patients who experienced an MI in the previous 2 years, the composite end point of cardiovascular death, MI, and stroke was met by 6.45% (n = 265) of patients taking evolocumab and by 8.43% (n = 362) of patients taking placebo (HR = 0.76; 95% CI, 0.64 to 0.89). Of the patients who were 2 years or more beyond their MI, this composite end point was met by 5.97% (n = 419) of patients taking evolocumab and by 6.81% (n = 470) of patients taking placebo (HR = 0.87; 95% CI, 0.76 to 0.99). The absolute risk reduction was 2.9% (95% CI, 1.2% to 4.5%) in patients who experienced an MI in the previous 2 years and 1.0% (95% CI, –0.7% to 2.7%) in patients who were 2 years or more beyond their MI.¹⁴

Cardiovascular Death¹⁵

Of the patients who experienced an MI in the previous year, the end point of cardiovascular death was met by 1.77% (n = 50) of patients taking evolocumab and by 1.80% (n = 52) of patients taking placebo (HR = 1.00; 95% CI, 0.68 to 1.47). Of the patients who were 1 year or more beyond their MI, this end point was met by 1.88% (n = 156) of patients taking evolocumab and by 1.64% (n = 136) of patients taking placebo (HR = 1.15; 95% CI, 0.91 to 1.44).

This mortality end point was not assessed in the subgroups of patients who experienced an MI in the previous 2 years or who were 2 years or more beyond their MI.

MI (Fatal or Nonfatal)¹⁵

Of the patients who experienced an MI in the previous year, the end point of fatal or nonfatal MI was met by 4.50% (n = 127) of patients taking evolocumab and by 6.61% (n = 191) of patients taking placebo (HR = 0.67; 95% CI, 0.54 to 0.84). Of the patients who were 1 year or more beyond their MI, this cardiovascular end point was met by 3.56% (n = 296) of patients taking evolocumab and by 4.57% (n = 379) of patients taking placebo (HR = 0.78; 95% CI, 0.67 to 0.91).

This cardiovascular end point was not assessed in the subgroup of patients who experienced an MI in the previous 2 years or who were 2 years or more beyond their MI.

Stroke (Fatal or Nonfatal)¹⁵

Of the patients who experienced an MI in the previous year, the end point of fatal or nonfatal stroke was met by 1.06% (n = 30) of patients taking evolocumab and by 1.31% (n = 38) of patients taking placebo (HR = 0.81; 95% CI, 0.50 to 1.31). For the patients who were 1 year or more beyond their MI, this cerebrovascular end point was met by 1.32% (n = 110) of patients taking evolocumab and by 1.65% (n = 137) of patients taking placebo (HR = 0.80; 95% CI, 0.62 to 1.03).

This cerebrovascular end point was not assessed in the subgroup of patients who experienced an MI in the 2 previous years or who were 2 years or more beyond their MI.

Cardiovascular Death, MI, Hospitalization for Unstable Angina, Stroke, or Coronary Revascularization¹⁵

Of the patients who experienced an MI in the previous year, the composite end point of cardiovascular death, MI, hospitalization for unstable angina, stroke, and coronary revascularization was met by 11.45% (n = 323) of patients taking evolocumab and by 14.12% (n = 408) of patients taking placebo (HR = 0.81; 95% CI, 0.70 to 0.93). Of the patients who were 1 year or more beyond their MI, this composite end point was met by 10.24% (n = 851) of patients taking evolocumab and by 11.10% (n = 921) of patients taking placebo (HR = 0.92; 95% CI, 0.84 to 1.01).

Of the patients who experienced an MI in the previous 2 years, the composite end point of cardiovascular death, MI, hospitalization for unstable angina, stroke, and coronary revascularization was met by 11.17% (n = 459) of patients taking evolocumab and by 13.72% (n = 589) of patients taking placebo (HR = 0.80; 95% CI, 0.71 to 0.91). Of the patients who were 2 years or more beyond their MI, this composite end point was met by 10.19% (n = 715) of patients taking evolocumab and by 10.73% (n = 740) of patients taking placebo (HR = 0.95; 95% CI, 0.85 to 1.05). The absolute risk reduction was 3.4% (95% CI, 1.4% to 5.3%) in patients who experienced an MI in the previous 2 years and 0.8% (95% CI, –1.1% to 2.7%) in patients who were 2 years or more beyond their MI.¹⁴

Coronary Revascularization¹⁵

Of the patients who experienced an MI in the previous year, the end point of coronary revascularization was met by 7.30% (n = 206) of patients taking evolocumab and by 9.79% (n = 283) of patients taking placebo (HR = 0.74; 95% CI, 0.62 to 0.89). Of the patients who were 1 year or more beyond their MI, this cardiovascular end point was met by 5.89% (n = 489) of patients taking evolocumab and by 6.95% (n = 577) of patients taking placebo (HR = 0.84; 95% CI, 0.75 to 0.95).

This cardiovascular end point was not assessed in the subgroup of patients who experienced an MI in the previous 2 years or who were 2 years or more beyond their MI.

Other Clinical Outcomes of Importance

Transient ischemic attack and death by any cause were identified as relevant to expert committee deliberations, but these outcomes were not assessed in the subgroup analyses of the FOURIER trial.

Lipid Parameter Outcomes From the Gencer et al.¹³ and Sabatine et al. (2018)¹⁴ Studies

Treatment Response Based on LDL-C¹⁶

Of the patients who experienced an MI in the previous year, 244 of 2,821 patients (8.6%) in the evolocumab group and 198 of 2,889 patients (6.9%) in the placebo group had an LDL-C level of less than 1.8 mmol/L at baseline. Of the patients who experienced an MI in the previous year, 2,468 of 2,692 patients (91.7%) in the evolocumab group and 502 of 2,764 patients (18.2%) in the placebo group achieved a treatment response of an LDL-C level of less than 1.8 mmol/L at week 4. Of the patients who were 1 year or more beyond their MI, 517 of 8,308 patients (6.2%) in the evolocumab group and 560 of 8,301 patients (6.7%) in the placebo group had an LDL-C level of less than 1.8 mmol/L at baseline. Of the patients who were 1 year or more beyond their MI, 7,283 of 8,047 patients (90.5%) in the evolocumab group and 1,365 of 7,994 patients (17.1%) in the placebo group achieved a treatment response of an LDL-C level of less than 1.8 mmol/L at week 4.

Of the patients who experienced an MI in the previous 2 years, 331 of 4,109 patients (8.1%) in the evolocumab group and 284 of 4,292 patients (6.6%) in the placebo group had an LDL-C level of less than 1.8 mmol/L at baseline. Of the patients who experienced an MI in the previous 2 years, 3,608 of 3,941 patients (91.6%) in the evolocumab group and 753 of 4,094 patients (18.4%) in the placebo group achieved a treatment response of an LDL-C level of less than 1.8 mmol/L at week 4. Of the patients who were 2 years or more beyond their MI, 430 of 7,020 patients (6.1%) in the evolocumab group and 474 of 6,898 patients (6.9%) in the placebo group had an LDL-C level of less than 1.8 mmol/L at baseline. Of the patients who were 2 years or more beyond their MI, 6,143 of 6,798 patients (90.4%) in the evolocumab group and 1,114 of 6,664 patients (16.7%) in the placebo group achieved a treatment response of an LDL-C level of less than 1.8 mmol/L at week 4.

Change From Baseline in LDL-C¹⁶

Of the patients who experienced an MI in the previous year, the mean LDL-C level was 2.453 mmol/L (SD = 0.647 mmol/L) in the evolocumab group and 2.467 mmol/L (SD = 0.647 mmol/L) in the placebo group at baseline. Of the patients who experienced an MI in the previous year, the mean percent change from baseline to week 48 in LDL-C was –59.90% (SD = 30.12%) in the evolocumab group and 2.00% (SD = 27.41%) in the placebo group. Of the patients who were 1 year or more beyond their MI, the mean LDL-C level was 2.563 mmol/L (SD = 0.784 mmol/L) in the evolocumab group and 2.545 mmol/L (SD = 0.711 mmol/L) in the placebo group at baseline. Of the patients who were 1 year or more beyond their MI, the mean percent change from baseline to week 48 in LDL-C level was –60.60% (SD = 30.53%) in the evolocumab group and –0.98% (SD = 25.70%) in the placebo group.

Table 21: Summary of Key Efficacy Results From the Gencer et al. Study (Full Analysis Set)

CI = confidence interval; IQR = interquartile range; KM = Kaplan-Meier; LDL-C = low-density-lipoprotein cholesterol; MI = myocardial infarction; SD = standard deviation.

Notes: In the subgroup of patients who experienced an MI in the previous year, those whose most recent MI or stroke was in the 4 weeks before randomization were excluded from the FOURIER trial.

The median length of follow-up was 25.99 months (IQR, 21.72 to 30.42 months). Patients with events that occurred between patient randomization date and the patient last confirmed survival status date, inclusive, were included. The censoring date of patients without an event was the patient last nonfatal potential end point collection date.

Multiplicity was not taken into account for subgroup analyses.

Time to hospitalization for unstable angina was not a prespecified end point in the FOURIER trial; an ad hoc analysis was performed to ensure that the results were provided for each individual component of the primary end point.

^aBased on a Cox model stratified by the randomization stratification factors collected with the Interactive Voice Response System.

^bBased on a Cox model, adding subgroup and subgroup-by-treatment interaction.

^cNumber of patients observed at each visit; the analysis set was all randomized patients.

^dSummary statistics were based on observed data. When the calculated LDL-C was less than 40 mg/dL or triglycerides were greater than 400 mg/dL, the calculated LDL-C was replaced with ultracentrifugation LDL-C, if available.

Sources: Additional information received from the sponsor on December 21, 2023,¹⁵ and January 30, 2024.¹⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

Of the patients who experienced an MI in the previous 2 years, the mean LDL-C level was 2.476 mmol/L (SD = 0.670 mmol/L) in the evolocumab group and 2.472 mmol/L (SD = 0.639 mmol/L) in the placebo group at baseline. Of the patients who experienced an MI in the previous 2 years, the mean percent change from baseline to week 48 in LDL-C level was –59.61% (SD = 31.05%) in the evolocumab group and 1.28% (SD = 26.73%) in the placebo group. Of the patients who were 2 years or more beyond their MI, the mean LDL-C level was 2.570 mmol/L (SD = 0.796 mmol/L) in the evolocumab group and 2.557 mmol/L (SD = 0.727 mmol/L) in the placebo group at baseline. Of the patients who were 2 years or more beyond their MI, the mean percent change from baseline to week 48 in LDL-C level was –60.90% (SD = 30.05%) in the evolocumab group and –1.14% (SD = 25.79%) in the placebo group.

Harms

FOURIER Trial¹²

Detailed information on harms outcomes in the FOURIER trial can be found in the December 2017 Clinical Review Report (Resubmission) on Evolocumab (Repatha), which is publicly available on the CDA-AMC website.

Treatment-Emergent Adverse Events

The proportion of patients with at least 1 TEAE or with at least 1 SAE was similar in the 2 treatment groups. A total of 10,664 patients (77.4%) in the evolocumab group and 10,644 patients (77.4%) in the placebo group reported at least 1 TEAE, with the most common TEAE being diabetes mellitus, which was reported in 1,207 patients (8.8%) and 1,130 patients (8.2%), respectively. A total of 3,410 patients (24.8%) in the evolocumab group and 3,404 patients (24.7%) in the placebo group reported at least 1 SAE, with the most common SAE being unstable angina, which was reported in 233 patients (1.7%) and 278 patients (2.0%), respectively.

The proportion of patients who stopped treatment due to any TEAE was also similar in the 2 treatment groups. A total of 608 patients (4.4%) in the evolocumab group and 573 patients (4.2%) in the placebo group stopped treatment due to any TEAE, with the most common TEAE being myalgia, which was reported in 37 patients (0.3%) and 46 patients (0.3%), respectively.

Table 22: Summary of Key Efficacy Results From the Sabatine et al. (2018) Study (Full Analysis Set)

CI = confidence interval; IQR = interquartile range; LDL-C = low-density-lipoprotein cholesterol; MI = myocardial infarction; SD = standard deviation.

Notes: For patients in the subgroup who experienced an MI in the previous 2 years, those whose most recent MI or stroke was in the 4 weeks before randomization were excluded from the FOURIER trial.

The median length of follow-up was 25.99 months (IQR, 21.72 to 30.42 months). Events occurring between the patient randomization date and the patient last confirmed survival status date, inclusive, were included. The censoring date of the patients without an event was the patient last nonfatal potential end point collection date.

Multiplicity was not taken into account for subgroup analyses.

^aBased on a Cox model stratified by the randomization stratification factors collected with the Interactive Voice Response System.

^bBased on a Cox model, adding subgroup and subgroup-by-treatment interaction.

^cNumber of patients observed at each visit; the analysis set was all randomized patients.

^dSummary statistics were based on observed data. When the calculated LDL-C was less than 40 mg/dL or triglycerides were greater than 400 mg/dL, the calculated LDL-C was replaced with ultracentrifugation LDL-C, if available.

Sources: Additional information received from the sponsor on December 21, 2023¹⁵ and on January 30, 2024.¹⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.²

TEAEs of Special Interest

The proportion of patients with TEAEs of special interest — including potential hypersensitivity, injection-site reaction, muscle events, neurocognitive events, demyelination events and peripheral neuropathy, hepatitis C infection, and transaminase elevations and hepatic disorder events — was similar in the 2 treatment groups. A total of 13 patients (< 0.1%) in the evolocumab group and 15 patients (0.1%) in the placebo group had a potential muscle-related AE (according to a narrow search strategy that included rhabdomyolysis, myopathy, and a myoglobin blood increase). A total of 1,381 patients (10.0%) in the evolocumab group and 1,344 patients (9.8%) in the placebo group had a potential muscle-related AE (according to a broader search strategy). A summary of harms results from the FOURIER trial is presented in Table 23.

Table 23: Summary of Harms Results From the FOURIER Trial (Safety Analysis Set)

AMQ = Amgen MedDRA query; HLGT = high-level group term; SAE = serious adverse event; SMQ = standard MedDRA query; TEAE = treatment-emergent adverse event.

Notes: TEAEs are presented by preferred terms and coded by Medical Dictionary for Regulatory Activities (MedDRA) Version 9.1.

Death from any cause was an adjudicated efficacy end point in the FOURIER trial. SAEs that did not meet the criteria for adjudicated end points and were subsequently reported as AEs, but later resulted in death, are included in the SAE section.

Standardized MedDRA Queries are validated, standard sets of MedDRA terms used to support signal detection and monitoring and represent a variety of safety topics of regulatory interest. Standardized MedDRA Queries include narrow and/or broad terms; narrow terms are highly likely to represent the condition of interest.¹⁷

^aTEAEs reported by 5% or more of patients in any treatment group are listed in descending order of frequency in the evolocumab group.

^bTreatment-emergent SAEs reported by 1% or more of patients in any treatment group are listed in descending order of frequency in the evolocumab group.

^cTEAEs leading to discontinuation of the study drug reported by more than 10 patients in any treatment group are listed in descending order of frequency in the evolocumab group.

^dThe SMQ and AMQ search strategies were used to retrieve AEs potentially related to the condition under review when heterogenous medical presentations may be expected.

^eTEAEs reported in more than 0.2% of patients in any treatment group by high-level term using a narrow search strategy for potential hypersensitivity events included dermatitis and eczema; rash, eruptions, and exanthemas; nasal congestion and inflammations; urticarias; and allergic conditions NEC.

^fTEAEs reported in more than 0.1% of patients in any treatment group by preferred term using a narrow search strategy for potential injection-site reaction events included injection-site pain, bruising, hematoma, erythema, and hemorrhage.

^gTEAEs using narrow search strategy for potential muscle events included rhabdomyolysis, myopathy, and a myoglobin blood increase.

^hTEAEs by high-level group term for potential neurocognitive events included cognitive and attention disorders and disturbances, deliria (including confusion), dementia and amnestic conditions, disturbances in thinking and perception, and mental impairment disorders.

ⁱTEAEs reported in 1 or more patients in any treatment group by high-level term using broad and narrow search strategies for potential demyelination events and peripheral neuropathy, respectively, included peripheral neuropathies NEC, sensory abnormalities NEC, trigeminal disorders, acute and progressive multiple sclerosis, plasma cell neoplasms NEC, acute polyneuropathies, spinal cord and nerve root disorders NEC, chronic polyneuropathies, and optic nerve disorders NEC.

^jTEAEs by preferred term using a narrow search strategy for potential hepatitis C infection included hepatitis C, chronic hepatitis, and a positive hepatitis C virus test.

^kTEAEs reported in 0.1% or more of patients in any treatment group by high-level term using a narrow search strategy for potential transaminase elevations and hepatic disorders included liver function analyses, hepatocellular damage and hepatitis NEC, hepatic and hepatobiliary disorders NEC, and coagulation and bleeding analyses.

Sources: Clinical Study Report of the FOURIER trial.¹² Details included in the table are from the sponsor’s Summary of Clinical Evidence.2

Gencer et al.¹³ and Sabatine et al. (2018)¹⁴ Studies

Safety outcomes were not assessed by subgroup in either the Gencer et al. study or the Sabatine et al. (2018) study.

Critical Appraisal

Internal Validity

FOURIER Trial

The FOURIER trial¹² was a phase III, double-blind, placebo-controlled, randomized clinical trial. Based on the previous review of the FOURIER trial in the December 2017 Clinical Review Report (Resubmission) on Evolocumab (Repatha), the key appraisal points are as follows.

Allocation concealment was facilitated throughout the randomization process using an Interactive Voice or Web Response System and a computer-generated randomization schedule that was prepared before the start of the trial. Blinding was facilitated with the matched placebo injection. Local injection-site and hypersensitivity reactions are associated with PCSK9 inhibitors.⁵² The proportion of patients with these AEs were similar in the 2 treatment groups and low overall (< 8.0% of patients in each group), so unblinding of the study drug was not likely. The adjudication of clinical events according to standardized definitions was performed by an independent external committee that was blinded to treatment allocation. Of note, information on the number of events identified by the investigator that were negatively adjudicated by the committee and the corresponding reasons to support these negative adjudications were not available. Although the study protocol included a suggestion that the study drug be self-administered under the supervision of site staff at each study visit, information was not available to support adherence (or possible poor adherence) to the study drug when administered was external to the study site. However, instructions on adherence to the study drug and lipid-lowering regimen were provided to patients if their LDL-C levels exceeded a preset threshold; these reminders were carried out in a manner that maintained the double-blind study design.

Gencer et al. and Sabatine et al. (2018) Studies

The Gencer et al.¹³ and Sabatine et al. (2018)¹⁴ studies were based on subgroup analyses of the FOURIER trial.¹² The subgroup analyses were based on the statistical methods used in the FOURIER trial, and the subgroups by timing of prior MI were prespecified; however, there was no clear hypothesis stated a priori. The P values on test for interaction term (in general, greater than 0.05, with the exception of the primary end point in the subgroup analysis by timing of prior MI [< 2 years versus ≥ 2 years]) strongly suggest that chance cannot be excluded as a likely explanation for the differential subgroup effect. There is a lack of evidence from randomized controlled trials and large observational studies to support consistent and similar findings from the subgroup analyses. Nonetheless, the subgroup analyses results were generally consistent with the overall FOURIER trial results, with the exception of stroke, for which the HR was 0.79 (95% CI, 0.66 to 0.95),¹² and the corresponding subgroup analysis results included null values.

After consultation with the clinical experts, additional treatment-effect modifiers in patients with primary hyperlipidemia were identified, including family history of primary hyperlipidemia, HeFH, early age of MI, elevated lipoprotein (a), and high on-treatment LDL-C level. Randomization was not stratified by, and the statistical model was not adjusted for, these effect modifiers. Moreover, information on these factors at baseline was not available to support their balance (or possible imbalance) in treatment groups. However, randomization was likely preserved in the 2 treatment groups by subgroup, as baseline characteristics were similar in the treatment groups by subgroup. Furthermore, the proportion of patients who discontinued the study was similar in the 2 treatment groups by subgroup, and was low overall (< 1.0% of patients in each group).

The sample-size calculation was based on the key secondary end point for the full analysis set in the FOURIER trial, but not for the subgroup analyses. Consequently, there is an increased likelihood of unreliable or inaccurate results, particularly for cardiovascular death and stroke, which are components of the composite end points for which the 95% CI results included null values. Nonetheless, the sample size of subgroups was considered relatively large. Multiplicity was not accounted for in the subgroup analyses; therefore, the interpretation of the subgroup analysis results is subject to an increased likelihood of type I error.

Because the aforementioned conditions can lower the credibility and reliability of the subgroup analysis results, the available evidence should not be viewed as conclusive; however, it may be interpreted as likely indicative of a possible subgroup effect.

External Validity

Gencer et al. and Sabatine et al. (2018) Studies

Because the sponsor’s reimbursement request focused on patients with recent ACS (in the previous year), the clinical experts were consulted on the patient population included in the subgroup analyses, which did not include patients with unstable angina or with recent (in the previous 4 weeks) MI or stroke. Although evidence in these patients is lacking, the experts did not identify any major concerns related to generalizing the subgroup analysis results to these patients.

Patients in the evolocumab group received subcutaneous evolocumab 140 mg once every 2 weeks (self-administered as 1 prefilled autoinjector or pen) or evolocumab 420 mg once every month (self-administered as 3 prefilled autoinjectors or pens), per patient preference. Dose adjustments were not permitted, with the exception of a switch between dose schedules. Thus, the use of evolocumab in the trial aligns with the recommended dose for the indication under review, per the product monograph.¹

At randomization, all patients were on an optimized, background lipid-lowering regimen, defined as an effective statin dose (i.e., high-to-moderate intensity of at least atorvastatin 20 mg daily or equivalent). Ezetimibe and other commercially available lipid-lowering therapy at dosages approved by local regulatory authorities were also permitted in the trial. Based on input from key interest-holders and in consultation with members of the expert committee, these were considered to be relevant comparators to evolocumab in the Canadian setting. Moreover, the clinical experts agreed that the permitted statins and their intensity, defined by American College of Cardiology and American Heart Association guidelines, were generally reflective of clinical practice. Regarding other PCSK9 inhibitors, funding is not yet in place for alirocumab, as negotiations concluded without an agreement in October 2019 for the indication of ASCVD.¹⁰

Outcomes on clinical events and LDL-C levels were assessed in this Clinical Review Report. The 2021 CCS dyslipidemia guidelines⁵ reference a meta-analysis by Silverman et al.⁵³ on the association between reduction in LDL-C and reduction in risk of major vascular events (a composite of cardiovascular death, acute MI or other ACS, coronary revascularization, and stroke) across lipid-lowering drugs. Of note, 2 trials of PCSK9 inhibitors were included; however, dedicated cardiovascular outcome trials for PCSK9 inhibitors had not yet been completed at the time of the meta-analysis. Based on the 18 secondary prevention trials with statin therapy, reduction in LDL-C by 1 mmol/L was associated with a relative risk of 0.79 (95% CI, 0.73 to 0.86) for major vascular events.⁵³ This finding supports the approximate linear relation between an absolute reduction in LDL-C with statin therapy and a proportional reduction in the incidence of major vascular events reported in past meta-analyses by the CTT Collaboration.^49,54 More specifically, the CTT Collaboration reported that a reduction of 1 mmol/L in LDL-C that is sustained for 5 years with statin therapy is associated with a proportional reduction, by approximately one-fifth, in major coronary events, coronary revascularization, and stroke, largely irrespective of the initial lipid profile or other presenting characteristics.^49,54 The updated CTT meta-analysis also demonstrated that additional reductions in LDL-C (to approximately 1 mmol/L to 2 mmol/L) with intensive statin therapy was associated with further reductions in the incidence of these clinical events.⁴⁹ The absolute benefit is dependent on the absolute baseline risk of such events and the absolute reduction in LDL-C.^5,54

Overall, no key concerns were identified related to the generalizability of the subgroup analysis results to the patient population in the reimbursement request. Of note, the estimated study duration was 56 months from the date the first patient was randomized; however, the median follow-up was 26 months. In the previous review of the FOURIER trial, the length of follow-up was deemed to be likely too short to assess the long-term harms associated with the use of evolocumab.¹⁸

Long-Term Extension Studies

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

This section summarizes the integrated results from 2 5-year OLE studies of the FOURIER trial (FOURIER-OLE). Patients who completed the parent trial had the option of enrolling in 1 of the 2 OLE studies. Additionally, this section summarizes an ad hoc subgroup analysis, based on the OLE studies, of patients who experienced an MI before and/or during the parent trial. All results reported herein are the integrated data from the 2 OLE studies.

Description of Studies

The FOURIER-OLE studies comprised 2 phase IIIb, multicentre, single-arm, 5-year OLE studies assessing the safety, tolerability, and clinical effects of long-term evolocumab administration in patients who completed the FOURIER trial (parent trial). The first OLE study (the first patient was enrolled on September 2, 2016) included patients who completed the parent trial at selected study sites located in North America or Eastern Europe (N = 5,035). The second OLE study (the first patient was enrolled on March 13, 2017) included patients who completed the parent trial at selected study sites located in Western Europe (N = 1,600).^21,55,56

Populations

After completing the parent trial in which they were randomized to receive either evolocumab or placebo, patients were eligible to enrol in 1 of the 2 OLE studies. The key inclusion and exclusion criteria used in the FOURIER trial are described in Table 7.

Interventions

Upon enrolment in the OLE studies, all patients received open-label evolocumab (140 mg administered subcutaneously once every 2 weeks [1 prefilled autoinjector or pen] or 420 mg administered subcutaneously once every month [3 prefilled autoinjectors or pens], according to patient preference). Patients were permitted to switch between the 2 dosing schedules based on preference. Patients were advised to continue the same background lipid-lowering therapy, including a statin, which was consistent with the protocol for the parent trial.^21,55,56

Outcomes

The primary end point of the OLE studies was the incidence of AEs. The secondary end points included percent change from baseline in LDL-C and the proportion of patients with a LDL-C level of less than 1.03 mmol/L at each scheduled visit (i.e., week 12, week 24, and every 24 weeks thereafter for 260 weeks [approximately 5 years]). The incidence of major cardiovascular events (i.e., cardiovascular death, MI, stroke or hospitalization for unstable angina, or coronary revascularization) were prespecified exploratory end points.^21,55,56

Statistical Analysis

In the (descriptive) analysis of the primary and secondary end points, the integrated OLE safety analysis set, which includes all patients who received at least 1 dose of open-label evolocumab in the OLE studies, was used. Of note, the primary analysis of the primary and secondary end points only used data from the OLE studies (i.e., data from the parent FOURIER trial were not included). No formal hypothesis was tested; additionally, no statistical inference or missing value imputation was planned. For all end points, results were summarized according to the randomized treatment group in the parent trial and for the overall OLE study population.⁵⁶

In the analysis of safety outcomes, patients were censored 30 days after permanent drug discontinuation or at the end of the study, whichever occurred first. For clinical event outcomes, the HR and associated 95% CI were estimated from stratified Cox models with the prespecified stratification factors at randomization. If the proportional hazards assumption was violated, modified Poisson regression was conducted as a sensitivity analysis. Landmark analyses were annually performed on the primary and key secondary composite end points.²¹

Ad Hoc Subgroup Analysis of Patients Who Experienced a Prior MI

An ad hoc subgroup analysis of the FOURIER-OLE studies was conducted in patients who experienced an MI before and/or during the parent FOURIER trial. A Cox model stratified by the FOURIER-OLE study was used to estimate the HR and associated 95% CI in the randomized treatment groups in the parent trial (evolocumab and placebo). A 2-sided log-rank test stratified by the OLE study was used to compare survival functions in the randomized treatment groups in the parent trial.²³

Efficacy end points included:²³

• time to cardiovascular death, MI, hospitalization for unstable angina, stroke, or coronary revascularization, whichever occurred first

• time to cardiovascular death, MI, or stroke, whichever occurred first

• time to cardiovascular death

• time to death by any cause

• time to first MI

• time to first stroke

• time to first coronary revascularization

• time to first hospitalization for unstable angina

• time to coronary heart disease or MI

• time to coronary heart disease death

• LDL-C.

For patients who experienced an MI before and/or during the parent trial, the following outcomes were also summarized:²³

• enrolment by region

• time from most recent MI to enrolment in the FOURIER-OLE study

• duration of exposure during the OLE study only and during the parent FOURIER trial plus the FOURIER-OLE study

• TEAEs

• LDL-C using data from both the parent trial and the FOURIER-OLE study.

Results

Patient Disposition

A total of 6,635 patients were enrolled in the 2 OLE studies and included in the integrated analysis. The integrated OLE safety analysis set included 6,630 patients (99.9%) who received at least 1 dose of evolocumab in the OLE studies, 3,353 of whom received evolocumab in the parent trial and 3,277 of whom received placebo. A total of 5,209 patients (78.5%) completed approximately 5 years of treatment with evolocumab in the OLE studies and 1,421 patients (21.4%) discontinued treatment early. The most common reasons for treatment discontinuation were death (433 patients [6.5%]), patient request (425 patients [6.4%]), and AEs (281 patients [4.2%]).⁵⁶

A total of 5,655 patients (85.2%) in the integrated OLE safety analysis set completed the OLE studies and 980 patients (14.8%) discontinued the study early. The most common reason for study discontinuation was death (646 patients [9.7%]). Final vital status was not available for 48 patients (0.7%), 34 who withdrew consent and 14 who were lost to follow-up. Of these, 7 patients were missing their final vital status date.⁵⁶

Note that 1 of the 2 OLE studies continued for 260 weeks (approximately 5 years), whereas the other study was terminated early, at a median duration of exposure of approximately 240 weeks. The early termination of the study was not due to any safety findings.

Baseline Characteristics

A summary of the baseline characteristics of all patients in the FOURIER-OLE studies and included in the integrated analysis are presented in Table 24.

Ad Hoc Subgroup Analysis of Patients Who Experienced a Prior MI

A summary of the baseline characteristics of the subgroup of patients who had an MI before and/or during the parent FOURIER trial are presented in Table 25.

Table 24: Summary of Baseline Characteristics in the FOURIER-OLE Studies (Integrated OLE Safety Analysis Set)

HDL-C = high-density-lipoprotein cholesterol; hsCRP = high-sensitivity C-reactive protein; LDL-C = low-density-lipoprotein cholesterol; MI = myocardial infarction; OLE = open-label extension; PAD = peripheral artery disease; SD = standard deviation.

Source: Amgen FOURIER-OLE studies, Common Technical Document Module 2.7.3, Summary of Clinical Efficacy.²²

Table 25: Summary of Baseline Characteristics in the MI Subgroup of the FOURIER-OLE Studies (Integrated Results)

LDL-C = low-density-lipoprotein cholesterol; MI = myocardial infarction; N = number of patients who received at least 1 dose of open-label evolocumab in the OLE studies; OLE = open-label extension; SD = standard deviation.

Source: Amgen FOURIER-OLE studies. Amgen data on file.²³

Exposure to Study Treatments

Patients randomized to receive evolocumab in the parent trial had a median of 24.3 months of additional exposure to evolocumab (median, 84.2 months; Q1 and Q3 = 78.1 months and 89.8 months, respectively) compared to those who were randomized to receive placebo (median, 59.8 months; Q1 and Q3 = 52.8 months and 60.3 months, respectively). Overall, the FOURIER-OLE studies provided an additional 29,064 patient-years of exposure to evolocumab (beyond the parent study).⁵⁶

Concomitant Lipid-Lowering Therapies

All patients in the parent trial (n = 6,630) reported the use of at least 1 lipid-lowering medication at baseline, as did 98.0% (n = 6,495) of patients at the start of the FOURIER-OLE studies, with almost all patients taking a statin. Atorvastatin was the most common background statin therapy used at baseline in the parent trial (78.8%) and at the start of the FOURIER-OLE studies (75.6%), followed by rosuvastatin (14.6% and 14.7%, respectively) and simvastatin (5.9% and 6.2%, respectively). Nearly all patients were taking a moderate-intensity (21.2%) or high-intensity (78.6%) statin at baseline in the parent trial; the corresponding proportions were similar at the start of the FOURIER-OLE studies (22.2% and 74.8%, respectively).⁵⁶ Other commonly used lipid-lowering concomitant medications included fish oil, fibrates, and ezetimibe (approximately 3% to 6% each).

Ad Hoc Subgroup Analysis of Patients Who Experienced a Previous MI

In the MI subgroup, the mean duration of exposure to evolocumab was 81.42 months (SD = 14.3 months) in patients who were randomized to receive evolocumab in the parent trial and 52.70 months (SD = 14.4 months) in patients who received placebo in the parent trial.²³

Concomitant Lipid-Lowering Therapies

No data were available on the concomitant lipid-lowering therapies used in the MI subgroup.

Efficacy

Change From Baseline in LDL-C

Among patients in the FOURIER-OLE studies, the median baseline reflexive LDL-C level in the parent trial was 2.36 mmol/L (Q1 and Q3 = 2.06 mmol/L and 2.80 mmol/L, respectively); the baseline LDL-C level was similar in the 2 randomized treatment groups of the parent trial.^21,22

The percent change from baseline (in the parent study) in reflexive LDL-C at each scheduled visit of the FOURIER-OLE studies is presented in Figure 2. The observed mean percent reduction from baseline in LDL-C ranged from 53.4% to 67.2% during the 260-week OLE study period.²¹

Ad Hoc Subgroup Analysis of Patients Who Experienced a Previous MI

In the subset of patients (n = 5,582) who experienced an MI before and/or during the parent FOURIER trial, the mean baseline LDL-C level in the parent trial was 2.52 mmol/L (SD = 0.695 mmol/L), which was similar in patients randomized to receive evolocumab and those randomized to receive placebo in the parent trial.²³ The mean LDL-C level during the 260-week OLE study period for the MI subgroup of patients was 1.061 mmol/L (SD = 0.924 mmol/L). The mean percent reduction from baseline to week 260 in LDL-C was approximately 57.7%, and was similar in patients who received evolocumab and those who received placebo in the parent trial.²³ A summary on the LDL-C parameter is presented in Table 26.

Table 26: Summary of LDL-C Parameters in the MI Subgroup of the FOURIER-OLE Studies (Integrated Results)

LDL-C = low-density-lipoprotein cholesterol; OLE = open-label extension; Q1 = first quartile; Q3 = third quartile; SD = standard deviation.

Source: Amgen FOURIER-OLE studies.²³

Time to Major Cardiovascular Events

A summary of major adverse cardiovascular events in the FOURIER-OLE studies, according to original randomization assignment in the parent study (N = 3,355 in the evolocumab group in the parent study and N = 3,280 in the placebo group in the parent study), is presented here.

Figure 2: Mean Percent Change From Baseline in LDL-C by Scheduled Visit and Treatment Group Using Data From the Parent Study and the OLE Studies (Integrated OLE Safety Analysis Set)

EvoMab = evolocumab; LDL-C = low-density-lipoprotein cholesterol; N = number of patients who received at least 1 dose of open-label evolocumab in the FOURIER-OLE studies; OLE = open-label extension.

Note: The vertical line represents the standard error around the mean.

Plot is based on observed data; no imputation was used for missing values.

Baseline was defined as the parent study baseline.

Source: Amgen FOURIER-OLE studies, Common Technical Document Module 2.7.3, Summary of Clinical Efficacy.22

Cardiovascular Death, MI, Stroke, Coronary Revascularization, or Hospitalization for Unstable Angina²¹

During the OLE study period, 490 patients (14.6%) originally randomized to the evolocumab group in the parent study experienced the FOURIER primary outcome of cardiovascular death, MI, stroke, hospitalization for unstable angina, or coronary revascularization, compared to 551 patients (16.8%) originally randomized to the placebo group (HR = 0.85; 95% CI, 0.75 to 0.96). The KM estimate for the primary outcome at 60 months was 15.38% (95% CI, 14.11% to 16.63%) for patients randomized to receive evolocumab in the parent trial and 17.50% (95% CI, 16.15% to 18.83%) for patients randomized to placebo, with a difference in event probability of 2.12% (95% CI, 0.28% to 3.96%).

Cardiovascular Death, MI, or Stroke²¹

The HR for the key secondary composite outcome of cardiovascular death, MI, and stroke was 0.80 (95% CI, 0.68 to 0.93), with 309 patients (9.2%) randomized to the evolocumab group and 374 (11.4%) randomized to the placebo group in the parent study experiencing these outcomes. The KM estimate for this key secondary composite outcome at 60 months was 9.75% (95% CI, 8.70% to 10.78%) in patients randomized to the evolocumab group in the parent trial and 11.90% (95% CI, 10.75% to 13.03%) in patients randomized to the placebo group, with a difference in event probability of 2.15% (95% CI, 0.61% to 3.70%).

Cardiovascular Death²¹

For cardiovascular death, 107 (3.19%) and 138 patients (4.21%) originally randomized to the evolocumab group and the placebo group in the parent study experienced this outcome, respectively. The HR for the individual component of cardiovascular death was 0.77 (95% CI, 0.60 to 0.99). The KM estimate for cardiovascular death at 60 months was 3.33% (95% CI, 2.70% to 3.95%) in patients randomized to the evolocumab group in the parent trial and 4.42% (95% CI, 3.69% to 5.15%) in patients randomized to the placebo group, with a difference in event probability of 1.10% (95% CI, 0.14% to 2.06%).

Death From Any Cause²¹

For death from any cause, 338 patients (10.07%) and 344 patients (10.49%) originally randomized to the evolocumab group and the placebo group in the parent study experienced this outcome, respectively. The HR for the individual component of death from any cause was 0.97 (95% CI, 0.83 to 1.13). The KM estimate for death from any cause at 60 months was 10.34% (95% CI, 9.28% to 11.39%) in patients randomized to the evolocumab group in the parent trial and 10.71% (95% CI, 9.62% to 11.79%) in patients randomized to the placebo group, with a difference in event probability of 0.37% (95% CI, –1.15% to 1.89%).

Myocardial Infarction²¹

For MI, 151 patients (4.5%) and 194 patients (5.91%) originally randomized to the evolocumab group and the placebo group in the parent study experienced this outcome, respectively. The HR for the individual component of MI was 0.75 (95% CI, 0.60 to 0.92). The KM estimate for MI at 60 months was 4.83% (95% CI, 4.08% to 5.59%) in patients randomized to the evolocumab group in the parent trial and 6.20% (95% CI, 5.35% to 7.05%) in patients randomized to the placebo group, with a difference in event probability of 1.37% (95% CI, 0.23% to 2.51%).

Stroke²¹

For stroke, 102 patients (3.04%) and 94 patients (2.87%) originally randomized to the evolocumab group and the placebo group in the parent study experienced this outcome, respectively. The HR for the individual component of stroke was 1.05 (95% CI, 0.80 to 1.39). The KM estimate for stroke at 60 months was 3.26% (95% CI, 2.63% to 3.88%) in patients randomized to the evolocumab group in the parent trial and 3.09% (95% CI, 2.47% to 3.71%) in patients randomized to the placebo group, with a difference in event probability of –0.17% (95% CI, –1.05% to 0.71%).

Coronary Revascularization²¹

For coronary revascularization, 280 patients (8.35%) and 313 patients (9.54%) originally randomized to the evolocumab group and the placebo group in the parent study experienced this outcome, respectively. The HR for the individual component of coronary revascularization was 0.85 (95% CI, 0.73 to 1.00). The KM estimate for coronary revascularization at 60 months was 8.91% (95% CI, 7.90% to 9.91%) in patients randomized to the evolocumab group in the parent trial and 10.13% (95% CI, 9.05% to 11.19%) in patients randomized to the placebo group, with a difference in event probability of 1.22% (95% CI, –0.25% to 2.69%).

Ad Hoc Subgroup Analysis of Patients Who Experienced a Previous MI

A summary of the major adverse cardiovascular events in the MI subgroup of the FOURIER-OLE studies is presented in Table 27. Among patients who experienced an MI before and/or during the parent FOURIER trial, 406 patients (14.42%) who were randomized to receive evolocumab in the parent trial experienced the FOURIER primary outcome of cardiovascular death, MI, stroke, hospitalization for unstable angina, or coronary revascularization, compared with 478 patients (17.28%) who were randomized to receive placebo in the parent trial (HR = 0.81; 95% CI, 0.71 to 0.93). The HR for the key secondary composite outcome of cardiovascular death, MI, and stroke was 0.77 (95% CI, 0.65 to 0.90); of note, the HR for the individual component of cardiovascular death was 0.68 (95% CI, 0.51 to 0.91). Event probabilities and, consequently, the difference in event probabilities between treatment groups from the parent trial were not available for the MI subgroup analysis.

Table 27: Major Adverse Cardiovascular Events in the MI Subgroup of the FOURIER-OLE Studies (Integrated Results)

CI = confidence interval; HR = hazard ratio; MI = myocardial infarction; OLE = open-label extension.

Note: N refers to the number of patients who experienced an MI before and/or during the parent FOURIER trial.

Source: Amgen FOURIER-OLE studies.²³

Harms

In the integrated OLE safety analysis set, 2,894 patients (86.3%) randomized to evolocumab in the parent study and 2,830 patients (86.4%) randomized to placebo experienced at least 1 AE during the OLE studies (Table 28). The most frequently reported AE was hypertension (15% of patients). Other AEs reported by at least 5% of patients in either treatment group in the parent study include nasopharyngitis, bronchitis, arthralgia, diabetes mellitus, atrial fibrillation, back pain, upper respiratory tract infection, angina pectoris, and pneumonia (Table 28).

Approximately 43% of patients experienced at least 1 SAE during the OLE studies (43.4% of patients randomized to the evolocumab group in the parent study and 42.7% of patients randomized to placebo). SAEs that occurred in at least 1% of patients in either group in the parent study are summarized in Table 28. Acute MI, angina pectoris, pneumonia, atrial fibrillation, and cardiac failure were among the SAEs reported most frequently (in 2% to 3% of patients).

The incidence rates of TEAEs leading to the discontinuation of evolocumab in at least 5 patients in the OLE safety analysis set were low and are summarized in Table 28. Overall, approximately 8% of patients experienced an AE leading to discontinuation of evolocumab during the OLE study (7.7% of patients who received evolocumab in the parent study and 8.0% of patients who received placebo in the parent study). The most frequently reported AEs leading to the discontinuation of evolocumab in the OLE studies were in the system organ class of neoplasms, benign, malignant and unspecified (including cysts and polyps) (2.0% to 2.1% of patients), followed by cardiac disorders (1.5% to 2.1% of patients). None of the reported AEs leading to discontinuation were reported in more than 1% of patients.

The overall incidence of fatal AEs in the OLE safety analysis set was similar in patients randomized to evolocumab in the parent study (6.7% [223 patients]) and those randomized to placebo (6.5% [213 patients]). The most commonly reported fatal AEs were in 3 system organ classes: cardiac disorders; neoplasms benign, malignant and unspecified (including cysts and polyps); and infections and infestations. The incidence of fatal AEs in the cardiac disorders system organ class was 1.5% (50 patients) and 2.3% (74 patients) for those randomized to evolocumab and placebo, respectively, in the parent study. For the system organ class of neoplasms benign, malignant and unspecified (including cysts and polyps), the incidence of fatal AEs was 1.9% (64 patients) and 1.6% (51 patients) for those randomized to evolocumab and placebo, respectively, in the parent study.²⁴

Notable harms identified for this review include potential hypersensitivity, injection-site reaction, muscle events, neurocognitive events, demyelination events and peripheral neuropathy, hepatitis C infection, and transaminase elevations and hepatic disorder events. Notable harms reported by at least 1% of patients in any treatment group in the OLE safety analysis set included potential injection-site reaction events, potential demyelination events (peripheral neuropathy, sensory abnormalities NEC, and chronic polyneuropathies), and transaminase elevations and potential hepatic disorders (liver function analyses and hepatocellular damage and hepatitis NEC) (Table 28). The numbers were similar in the evolocumab and placebo groups.

Ad Hoc Subgroup Analysis of Patients Who Experienced a Previous MI

The evolocumab safety profile in the MI subgroup was similar to that seen in the overall study population. Details, including TEAEs, SAEs, AEs leading to discontinuation, and device-related AEs, are presented in Table 29.

Table 28: Summary of Harms in the FOURIER-OLE Studies (Integrated OLE Safety Analysis Set)

AE = adverse event; NEC = not elsewhere classifiable; OLE = open-label extension; SAE = serious adverse event; TEAE = treatment-emergent adverse event.

^aTEAEs reported by 5% or more of patients in any treatment group are listed in descending order of frequency in the evolocumab group.

^bTreatment-emergent SAEs reported by 1% or more of patients in any treatment group are listed in descending order of frequency in the evolocumab group.

^cTEAEs that led to the discontinuation of the study drug reported by more than 5 patients in any treatment group are listed in descending order of frequency in the evolocumab group.

^dIncluding cysts and polyps.

^eExcluding islet cell carcinoma and carcinoid tumours.

^fAEs of special interest reported by 1% or more of patients in any treatment group are listed in descending order of frequency in the evolocumab group.

^gBoth parent study and open-label study data are used.

^hOpen-label study data only.

Source: Clinical Summary Report Module 2.7.4, Summary of Clinical Safety.²⁴ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 29: Summary of Harms in the MI Subgroup of the FOURIER-OLE Studies (Integrated Results)

AE = adverse event; SAE = serious adverse event; TEAE = treatment-emergent adverse event.

Source: Amgen FOURIER-OLE studies, Amgen data on file.²³

Critical Appraisal

Internal Validity

This section summarizes the integrated results of 2 phase IIIb, multicentre, single-arm, 5-year OLE studies that included patients who completed the parent FOURIER trial. An open-label study design can influence the perception of improvement and/or harms by patients and clinicians, particularly in outcomes that are subjective in measurement and interpretation. However, because all fatal and nonfatal cardiovascular events and deaths were adjudicated by an external, independent CEC, the assessment of the primary and key secondary end points in the FOURIER-OLE studies were not likely to have been affected by the open-label design.

Because the descriptive analyses used in the OLE studies and the ad hoc subgroup analysis of patients who experienced a previous MI, the available evidence should only be considered suggestive of a potential treatment effect, subject to the uncertainty associated with the exploratory nature of the analyses.

External Validity

The baseline characteristics of all patients enrolled in the FOURIER-OLE studies were similar in the randomized treatment groups from the parent FOURIER trial. Although most patients were from the study sites located in Europe (> 66%), their demographics were generally similar to the patient population in Canada. In general, the baseline characteristics of patients in the MI subgroup were similar to the overall OLE patient population.

Because the sponsor’s reimbursement request is focused on the patient population with recent ACS (in the previous year), it should be noted that the MI subgroup included patients who experienced an MI before and/or during the parent FOURIER trial. The mean time from the most recent MI to enrolment in the overall OLE patient population was 69.606 months (SD = 74.237 months) in patients who were randomized to evolocumab in the parent trial and 68.531 months (SD = 71.613 months) in patients who were randomized to placebo. In the subset of patients who experienced a previous MI, the mean time from the most recent MI was 8.070 years (SD = 6.137 years) in patients who were randomized to evolocumab in the parent trial and 7.835 years (SD = 5.905 years) in patients who were randomized to placebo.

Study Addressing Gap in the Systematic Review Evidence

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

This section summarizes 1 study that was submitted by the sponsor to address a gap in the systematic review evidence. A summary of the evidence gap identified by the sponsor and a brief description of the study that addresses this gap are presented in Table 30.

Table 30: Summary of Gaps in the Systematic Review Evidence

ACS = acute coronary syndrome; LDL-C = low-density-lipoprotein cholesterol; MI = myocardial infarction.

Source: Details included in the table are from the sponsor’s Summary of Clinical Evidence.2

Description of Study

The EVOPACS study²⁶ is a randomized, double-blind, placebo-controlled, multicentre study that assessed the efficacy and safety of evolocumab 420 mg administered subcutaneously every month, compared to matching placebo, in the reduction of LDL-C when administered immediately after an ACS event (N = 308). The primary objective was to evaluate the effectiveness of evolocumab, compared to placebo, administered in the acute phase of ACS in the reduction of LDL-C at week 8 in patients receiving guideline-recommended high-intensity statin treatment (i.e., atorvastatin 40 mg once daily). The secondary objective was to assess the safety and tolerability of early administration of evolocumab in the acute phase of ACS.²⁶

Screening was performed on patients who were medically stabilized as soon as possible after hospital admission. The first patient was enrolled on January 23, 2018; patients were enrolled at 7 clinical centres in Switzerland and were randomized in a 1:1 ratio to receive either evolocumab or placebo. The analyses presented in this report are based on a database snapshot date of June 6, 2019.²⁶

Populations

Patients eligible for inclusion were adults (aged ≥ 18 years) who were hospitalized for a recent ACS (NSTEMI or unstable angina in < 72 hours or STEMI in < 24 hours before screening) and who had LDL-C levels as follows:²⁶

• LDL-C of at least 1.8 mmol/L or non-HDL-C of at least 2.6 mmol/L in patients receiving stable treatment with a high-intensity statin for at least 4 weeks before enrolment (i.e., continuous treatment that was unchanged with regard to statin intensity over the previous 4 weeks)

• LDL-C of at least 2.3 mmol/L or non-HDL-C of at least 3.1 mmol/L in patients receiving stable treatment with a low-intensity or moderate-intensity statin for at least 4 weeks before enrolment (i.e., continuous treatment that was unchanged with regard to statin intensity over the previous 4 weeks)

• LDL-C of at least 3.2 mmol/L or non-HDL-C of at least 4.0 mmol/L in patients who were statin-naive or who had not been on a stable (unchanged) statin regimen for at least 4 weeks before enrolment.

Patients were excluded if their clinical status was unstable, or they experienced uncontrolled cardiac arrhythmia, severe renal dysfunction, active liver disease or hepatic dysfunction, intolerance to atorvastatin or statins, and previous use of evolocumab or another PCSK9 inhibitor.²⁶

Interventions

Patients received 2 doses of evolocumab during the EVOPACS study; the first dose was administered by subcutaneous injection on day 1 (at the study site during hospitalization for the index ACS), and the second dose was administered at week 4 during a scheduled study-site visit. Evolocumab 420 mg and placebo were administered subcutaneously once monthly.

All patients received effective, high-intensity statin background therapy consisting of atorvastatin 40 mg once daily for the duration of the study, starting on day 1 and continuing through to week 8. Other medical therapies initiated during hospitalization for the index ACS event were administered in accordance with current guidelines for the management of patients with ACS and the clinical judgment of treating physicians.²⁶

Outcomes

The primary end point was the percent change from baseline to week 8 in the calculated LDL-C level. The secondary end points included AEs, SAEs, product complaints, and laboratory abnormalities from baseline to week 8.²⁶

Statistical Analysis

This study was a superiority trial powered to assess the primary end point: percent change from baseline to week 8 in LDL-C. It was assumed an average percent change of –30% in the placebo arm (atorvastatin 40 mg once daily) and –44% in the active treatment arm (evolocumab 420 mg administered subcutaneously once on day 1 plus atorvastatin 40 mg once daily) would yield a difference in LDL-C percent reduction of 14%. A common SD of 36% was adopted. To achieve 90% power at a significance level of 5% for a 2-sided t test, and assuming a dropout rate of 10% at week 8, a total of 308 patients (154 in each treatment group) were to be recruited.²⁶

Efficacy and safety analyses were performed on the full analysis set, which included all randomized patients who received at least 1 dose of evolocumab. The superiority of evolocumab over placebo was assessed for all efficacy end points, without applying multiplicity adjustments. In all efficacy and safety analyses, patients were grouped according to their randomized treatment group assignment.²⁶

The primary analysis — percent change in LDL-C from day 1 to 8 weeks — was conducted with a linear mixed-effects model, adjusted for the stratification factors (with the presence of stable statin treatment for at least 4 weeks before enrolment as a fixed effect and study centre as a random effect). Analysis was based on the intention-to-treat principle and patients contributed to the group to which they were randomized. For analyses of safety end points, missing data were not imputed.²⁶

Results

Patient Disposition

The patient disposition for the EVOPACS trial is illustrated in Table 31.

Baseline Characteristics

A summary of baseline characteristics in the EVOPACS study is presented in Table 32.

Exposure to Study Treatments

A summary of treatment exposure in the EVOPACS study is presented in Table 33. Measures of patient compliance were not required for this study.

Co-interventions and Concomitant Medications

Patients were advised to follow a stable, therapeutic lifestyle and diet, as outlined in adult treatment panel III of the National Cholesterol Education Program, or an equivalent diet for the duration of the study. Lifestyle and dietary habits, as well as the level of physical exercise, were to be maintained throughout the study duration.²⁶

Most patients (235 [76%]) were not taking a statin at baseline, 37 (12%) were on low-intensity to moderate-intensity statins, and 36 (12%) were on high-intensity statins. Twenty-four percent were taking Aspirin at baseline. By week 4, 281 patients (97%) were receiving high-intensity statins and 278 (96%) were on an Aspirin regimen. The use of these medications remained consistent throughout the rest of the study. In addition, the use of ticagrelor (an anticoagulant) by 4 patients (1%) at baseline increased to 198 patients (68%) by the time of discharge and remained at approximately those levels through week 8.²⁶

Table 31: Summary of Patient Disposition in the EVOPACS Study

^aPatient was in very bad health and did not have any energy to spare for the study, based on the information from hospital letters.

Source: Amgen Clinical Study Report, EVOPACS.²⁶

Table 32: Summary of Baseline Characteristics in the EVOPACS Study