An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Evolocumab (Repatha): Therapeutic area: Primary hyperlipidemia: Reimbursement Review [Internet]. Ottawa (ON): Canadian Agency for Drugs and Technologies in Health; 2024 Dec.
The executive summary comprises 2 tables (Table 1 and Table 2) and a conclusion.
Submitted for Review.
Summary of Economic Evaluation.
Based on the Clinical Review by Canada’s Drug Agency (CDA-AMC), results from the subgroup analyses of patients who experienced a prior MI from the FOURIER trial are indicative of a possible differential effect in favour of evolocumab (Repatha) in combination with optimized lipid-lowering therapy (LLT) over optimized LLT alone for cardiovascular event outcomes over a median follow-up period of 26 months. The sponsor also used an ad hoc subgroup analysis that assessed data from the FOURIER open-label extension (OLE) study of patients who experienced an MI before and/or during the FOURIER trial, which suggested a benefit with evolocumab. However, the CDA-AMC Clinical Review noted that the duration of follow-up in the parent trial is likely inadequate to assess the long-term relative safety of evolocumab.
As limitations with the sponsor’s model could not be adequately addressed due to the lack of alternative data and limitations with the model structure, the sponsor’s base case was used: evolocumab plus optimized LLT is associated with 0.90 incremental QALYs at an additional cost of $78,856, resulting in an incremental cost-effectiveness ratio (ICER) of $87,882 compared to optimized LLT alone. The probability of being cost-effective at a $50,000 per QALY gained threshold was 0%. Using the sponsor’s base case, a price reduction of 50% would be necessary to achieve cost-effectiveness at a $50,000 per QALY gained threshold.
Some identified limitations could not be addressed through reanalysis, and CDA-AMC noted that the clinical benefit of evolocumab remains uncertain, given that 90% of the predicted incremental QALYs for evolocumab were accrued beyond the trial period, during which there was no comparative clinical evidence.
This section is a summary of the feedback received from the patient groups, clinician groups, and drug plans that participated in the CDA-AMC review process.
No patient input was received for this review.
Clinician input was received from the following groups: Cardiac Rehabilitation and Secondary Prevention Program at Western University, the Cambridge Cardiac Rehab Program, the University of Ottawa Heart Institute, the McMaster Lipid Clinic, British Columbia Lipid Specialists, the Canadian Dyslipidemia Guideline Committee, University of British Columbia and Vancouver General Hospital and St. Paul’s Hospital Cardiac Intensive Care Unit attending physicians, and University of Toronto faculty and clinicians with expertise in atherosclerotic cardiovascular disease and/or lipid disorders. Clinician input indicated that the current standard of care for patients with atherosclerotic cardiovascular disease (ASCVD) includes lifestyle changes (i.e., diet and physical activity), statins, and ezetimibe. Clinician input received by CDA-AMC indicated that if evolocumab becomes funded for patients with ASCVD, it would be readily taken up by patients who are not achieving optimal low-density-lipoprotein cholesterol (LDL-C) control on their current treatment(s). Regarding place in therapy, clinician input indicated that evolocumab would be used as an adjunct to statins, with or without ezetimibe; however, the clinicians also noted that many patients do not tolerate statins due to adverse effects and, thus, evolocumab would also be considered for use in patients who are not receiving statin therapy. The clinicians noted that the 2021 Canadian dyslipidemia guidelines already recommend the use of a PCSK9 inhibitor (including evolocumab) for patients in whom LDL-C levels remain at or above 1.8 mmol/L while they receive maximally tolerated statins; however, the lack of funding for many patients presents a barrier to access.
Drug plan input noted that evolocumab is currently listed in many jurisdictions for patients with heterozygous familial hypercholesterolemia (HeFH). Drug plan input raised concerns about evolocumab eligibility for patients who are unable to achieve appropriate LDL-C levels due to elevated triglyceride levels. Additionally, drug plan input raised questions regarding evidence for the use of evolocumab in combination with evinacumab or inclisiran, and whether patients could move to alternative treatments if the effect of evolocumab wanes over time. Drug plan input also raised concerns regarding the budget impact, given the number of patients in Canada who have experienced recent acute coronary syndrome (ACS).
Several of these concerns were addressed in the sponsor’s model, which:
included the use of evolocumab in addition to moderate-intensity or high-intensity statins, with or without ezetimibe.
CDA-AMC was unable to address the following concern raised from the patient, clinician, and drug plan input:
the use of evolocumab in patients with recent ACS who are not receiving statin therapy could not be addressed, owing to a lack of comparative clinical data specifically in the reimbursement requested population.
The current review is for evolocumab for adult patients who experienced ACS in the previous year and who have an LDL-C level of at least 1.8 mmol/L despite taking moderate-to-high intensity statin therapy, with or without ezetimibe.
The sponsor submitted a cost-utility analysis of evolocumab in combination with optimized LLT (defined as moderate-to-high intensity statin therapy, with or without ezetimibe) compared with optimized LLT alone.1 The model population comprised adult patients who experienced ACS in the previous 1 year and who have an LDL-C level of at least 1.8 mmol/L despite taking moderate-to-high intensity statin therapy, with or without ezetimibe. The modelled population was aligned with the sponsor’s reimbursement request.
The recommended dose of evolocumab is either 140 mg every 2 weeks or 420 mg once monthly; the sponsor assumed both doses to be clinically equivalent. Evolocumab is administered subcutaneously and is intended for patient self-administration. Evolocumab is provided in either a 140 mg/mL single-use, prefilled autoinjector or a 3.5 mL, 120 mg/mL single-use automated mini-doser, at a submitted price of $271.27 and $587.75, respectively. The annual per-patient cost of evolocumab was estimated to be approximately $7,053. The sponsor estimated that the annual per-patient costs of the treatments included as comparators were as follows: $66 for ezetimibe (generic), $75 for high-intensity statins, and $63 for moderate-intensity statins.
The analysis was conducted from the perspective of the Canadian public health care payer. Cost and clinical outcomes (quality-adjusted life-years [QALYs], life-years) were estimated over a lifetime horizon (52 years; 1-year cycle length). Discounting (1.5% per annum) was applied for both costs and outcomes.
The sponsor submitted a Markov model that consisted of the following health states related to the number and type of potential cardiovascular events: other ASCVD (used to capture less severe cardiovascular (CV) events such as unstable angina [UA]), myocardial infarction (MI), 2 or more MIs, ischemic stroke (IS), 2 or more ISs, post-MI, post-IS, CV death, and non-CV death.1 The acute-event health states (i.e., MI, 2 or more MIs, IS, 2 or more ISs) corresponded to a 1-year period in which the event occurred to account for a reduced quality of life and increased health care costs. The postevent health states (i.e., post-MI, post-IS, post 2 or more MIs or ISs) represent the years after the event and account for the longer-term outcomes associated with each event or combination of events. Additionally, the model included composite health states that are a combination of either 2 or 3 event health states to retain memory of previous CV events. A simplified version of the sponsor’s model structure is depicted in Figure 1.
All patients entered the model in either the other ASCVD or MI health state, depending on their recent ACS event history, based on Canadian real-world evidence (RWE).2 Patients who started in the other ASCVD health state remained in this health state until they underwent their first MI, first IS, or died from a CV or non-CV cause. At the time of a CV event, patients transitioned to the relevant acute-event health state for 1 cycle. After that cycle, patients transitioned to the appropriate postevent health state (e.g., a move to the post-MI health state from the MI health state) or could experience another CV event (e.g., a move to the IS health state from the MI health state). Patients who started in the MI health state could transition to the post-MI health state, have another CV event, or die from a CV or non-CV cause. The sponsor assumed that only 1 CV event could happen in each annual cycle.
The baseline population characteristics used to inform the model were based on a retrospective, observational, cohort study conducted in Alberta that used health administrative data.2 Specifically, the sponsor used data for patients with ASCVD who experienced a recent ACS to inform the mean age (68 years), proportion of females (32%), mean LDL-C level (2.16 mmol/L), distribution of prior CV events (81% MI, 19% other ASCVD), and distribution of optimized LLT (62% high-intensity statin, 38% moderate-intensity statin, 9% concomitant ezetimibe).
The sponsor informed the baseline CV event risk using the Alberta RWE,2 and adjusted the event rate by age and LDL-C level. The sponsor used a hazard ratio for age that was estimated using data from the Clinical Practice Research Datalink,3 and used the standardized event rate ratio from the Cholesterol Treatment Trialists’ Collaboration meta-analysis per 1 mmol/L change in LDL-C for any major vascular event.4 The sponsor further adjusted the baseline CV event rate to account for CV event history, which resulted in greater rates of recurrent CV events using data reported by Danese et al. (2021).3 Finally, the sponsor converted and disaggregated the adjusted baseline CV event rate into CV event-specific annual risks for MI, IS, and CV death. The sponsor based the disaggregation of the annual risk into CV event-specific risks using the distribution of secondary events experienced by patients in the Alberta RWE study.2
The sponsor used the LDL-C reduction observed in the FOURIER trial at 48 weeks and maintained the treatment benefit over the lifetime time horizon (i.e., there was no treatment waning over time). To estimate CV event rate ratios per 1 mmol/L LDL-C reduction, the sponsor used hazard ratios from a subgroup analysis of patients who experienced a recent MI from the FOURIER trial.5
Mortality from non-CV causes was estimated using Canadian and sex-specific general population mortality6 and the proportion of deaths from CV-related causes identified by Statistics Canada.7 The sponsor modelled mortality related to CV disease separately, and used a competing risk adjustment, in which in each cycle, non-CVD death is first taken into account and CV event-specific transition probabilities are applied conditionally upon being alive.8 The sponsor further assumed that the benefit of evolocumab on CV-related mortality would only be realized after 3 years of treatment.
Health state utilities were obtained from a publication of a Canadian cost-effectiveness analysis on left atrial appendage closure for stroke prevention in patients with atrial fibrillation.9 For health states that include more than 1 acute and/or postevent, the sponsor applied only the lowest utility value among all events within the health state.
The model included drug-acquisition costs and health care resource use costs associated with the modelled health states. The proportion of patients on the treatments included in optimized LLT was based on a Canadian RWE study.2 The sponsor categorized statin treatments and doses based on the American College of Cardiology and American Heart Association guidelines as high intensity or moderate intensity,10 and used the annual cost of each statin (from the Ontario Drug Benefit Formulary)11 and market shares (based on IQVIA drug use data)12 to estimate the weighted-average costs of statin treatment. The annual cost of evolocumab was adjusted based on estimates of treatment discontinuation from the FOURIER trial.
Health state costs were estimated from the Institute for Clinical Evaluative Sciences (ICES) claims data, based on publicly funded health services records in Ontario.13,14 Costs were estimated separately for the first year of an event and the postevent health states (calculated as the average cost from years 2 to 5 after an event). For composite health states that include a combination of 2 or 3 event states, only the highest-cost event in that state was applied. The cost for CV death was obtained from the CDA-AMC Pharmacoeconomic Review of rivaroxaban.15
The sponsor did not include the incidence, cost, or quality-of-life effects of treatment-related adverse events.
All analyses were run probabilistically (1,000 iterations for the base case and scenario analyses). The probabilistic results aligned with the deterministic results. The probabilistic findings are presented here.
In the sponsor’s base-case analysis, treatment with evolocumab plus optimized LLT was associated with incremental costs of $78,856 and a gain of 0.90 QALYs, compared with optimized LLT, over the lifetime time horizon, resulting in an ICER of $87,882 per QALY gained (Table 3). The probability of evolocumab plus optimized LLT being cost-effective at a $50,000 per QALY gained threshold compared to optimized LLT alone was 0%. Approximately 90% of the incremental QALYs in the sponsor’s base case were accrued after 7 years, which is the combined median duration of follow-up of the FOURIER and FOURIER-OLE trials. The submitted analysis is based on the publicly available list prices of all treatments, aside from evolocumab.
Summary of the Sponsor’s Economic Evaluation Results.
The sponsor conducted several scenario analyses, including using baseline CV event rates from the FOURIER trial, alternative approaches to applying utilities (i.e., using the lowest values or multiplicative values for composite health states), and assuming that the delay in mortality benefit with evolocumab was 1 year. The conclusions of the sponsor’s base case were most sensitive to the approach for utility values applied in the model, with a 46% to 47% increase in the ICER due to a decline in incremental QALYs.
The sponsor conducted a scenario analysis from a societal perspective that included additional costs associated with short-term absenteeism, presenteeism, and caregiver time. In this analysis, relative to optimized LLT, the ICER was $87,882 per QALY gained. This result was the same as the sponsor’s base-case analysis using a health care payer perspective.
CDA-AMC identified several key limitations to the sponsor’s analysis that have notable implications for the economic analysis:
The relationship between treatment with evolocumab and CV events is uncertain. The sponsor used hazard ratios derived from the Gencer et al. subgroup analysis of the FOURIER trial to inform the rate ratios per 1 mmol/L LDL-C reduction for nonfatal MI and nonfatal IS in the submitted model, and used the FOURIER-OLE prior MI subgroup data to inform the risk of CV death after 3 years.5,16 The CDA-AMC Clinical Review reported that the results for the recent MI subgroup are indicative of a possible benefit when evolocumab is added to optimized LLT, based on the primary and key secondary composite outcomes over 26 months. The association between treatment with evolocumab and CV-related mortality (after the 3-year delay in benefit) was informed by results of the FOURIER-OLE study for the subgroup of patients with a history of MI. The results of the sponsor’s model are particularly sensitive to assumptions about the CV-related mortality benefit, for which there is uncertainty in the clinical evidence. The CDA-AMC Clinical Review noted that the results of the FOURIER-OLE study suggest that patients with a median of 24 months of additional exposure to evolocumab may have contributed to a reduction in CV death in the subgroup of patients who experienced an MI before or during the FOURIER trial. However, the CDA-AMC Clinical Review also noted limitations associated with the subgroup analysis, including the fact that the sample-size calculation was based on the full analysis set in the FOURIER trial but not in the subgroup analyses, and that multiplicity was not accounted for in the subgroup analyses. These limitations lead to some uncertainty in the analysis.
In light of the uncertainty regarding CV mortality and its impact on the results, CDA-AMC conducted 2 scenario analyses in which the lower and upper limits of the credible interval for the CV-related mortality risk ratio were applied to assess the impact on the cost-effectiveness of evolocumab compared to optimized LLT.
Long-term treatment discontinuation was not modelled. In their submitted model, the sponsor assumed that by year 3, 15% of patients would have discontinued treatment with evolocumab and that no further discontinuation would take place. The assumption that patients would be fully compliant with their treatment over their lifetime is likely inappropriate, according to the clinical experts consulted by CDA-AMC. In general, it has been established that adherence to long-term therapies can be low, especially for chronic diseases in which medications are prescribed as a preventive measure.17,18 For LLTs specifically, several studies have demonstrated the extent and impact of nonadherence to LLTs on CV health.19-23 Barriers to treatment adherence have also been identified specifically for LLTs, including patient, health care system, and treatment-related factors.24 Although research on LLT adherence has largely been focused on statin therapies, the long-term adherence to newer treatments like evolocumab remains unknown. However, the safety profile and reduced treatment frequency of evolocumab may promote better adherence.24
Because of its structure and programming, the sponsor’s model did not allow for the consideration of treatment discontinuation. The modelled population retained the full costs and benefits of evolocumab over the lifetime horizon. The impact of treatment discontinuation and subsequent treatment-effect waning on the cost-effectiveness of evolocumab over a lifetime time horizon is unknown.
CDA-AMC was unable to address this limitation.
The long-term efficacy of evolocumab on LDL-C lowering has not been established. In the submitted model, patients are assumed to receive the full benefit of the LDL-C reduction observed at 48 weeks in the FOURIER trial for up to 52 years (the model time horizon). The submitted model did not explore the impact of potential treatment waning over time. Although there is a lack of long-term data for the modelled time horizon (and thus uncertainty) to inform the persistence of LDL-C reduction over time, the clinical experts advised that the maintained benefit for patients who continue treatment may be a reasonable assumption. However, 90% of the sponsor’s predicted incremental health benefits were accrued beyond the time for which there are data, so the cost-effectiveness results may be influenced by changes in assumptions around the long-term effects of evolocumab.
CDA-AMC was unable to address this limitation in reanalysis.
The clinical evidence did not align with the requested population. The sponsor modelled patients with recent ACS (MI or UA). However, the evidence used to inform the clinical efficacy data in the model was predominantly from patients with history of MI only (i.e., Gencer et al., FOURIER-OLE MI subgroup analysis).5,16 As such, the cost-effectiveness of evolocumab in patients with UA is uncertain. However, clinical expert feedback indicated that, in practice, the preference would be to mostly treat MI, and that UA would make up a small proportion of cases.
CDA-AMC was unable to address this limitation in reanalysis. However, the impact on the cost-effectiveness results may be small, given the small proportion of UA cases.
The submitted model lacked transparency. The submitted model file relied on calculations that incorporated data held in multiple worksheets, some of which contained considerable numbers of blank columns between cells that needed to be checked. For example, on the worksheets related to transition probabilities, there are more than 900 columns, with only a small portion of them containing data. Formulas on these sheets refer to blank (zeroed out) columns, complicating the validation process. The model implies functionality that is not present (e.g., providing trace rows for an average starting age of 18 but preventing the user from inputting these values). These limitations make thorough auditing of the sponsor’s model impractical.
CDA-AMC was unable to address this limitation.
Additionally, the key assumptions detailed in Table 4 were made by the sponsor and have been appraised by CDA-AMC.
Key Assumptions of the Submitted Economic Evaluation (Not Noted as Limitations to the Submission).
Key limitations of the sponsor’s model could not be adequately addressed due to the lack of alternative data and limitations with the model structure (i.e., treatment waning and treatment discontinuation). As such, the sponsor’s base case was maintained.
Given that the CDA-AMC Clinical Review noted some uncertainty surrounding the relationship between treatment with evolocumab and CV-related mortality, CDA-AMC conducted 2 scenario analyses involving different values for CV-related mortality: the first used the lower credible interval of the hazard ratio for CV mortality from the FOURIER-OLE trial, and the second used the upper credible interval. These changes are described in Table 10.
In scenario analysis 1, evolocumab plus optimized LLT was associated with 1.35 incremental QALYs at an additional cost of $92,755, compared with optimized LLT alone, resulting in an ICER of $68,809 per QALY gained. In scenario analysis 2, evolocumab plus optimized LLT was associated with 0.39 incremental QALYs at an additional cost of $63,829, compared with optimized LLT alone, resulting in an ICER of $164,205 per QALY gained. A summary of the CDA-AMC scenario analysis results can be found in Table 5. The probability of cost-effectiveness at a $50,000 per QALY gained threshold was 0% in both scenario analysis 1 and 2.
CDA-AMC conducted price reduction analyses based on the sponsor’s base case (Table 6). These analyses demonstrated that using the sponsor’s base case, a price reduction of 50% would be necessary to achieve cost-effectiveness at a $50,000 per QALY gained threshold.
Summary of CDA-AMC Scenario Analysis Results.
CDA-AMC Price Reduction Analyses.
Evolocumab has undergone review by CDA-AMC twice, receiving a Reimburse With Conditions recommendation both times; the first was for HeFH25 and the second was for HeFH and ASCVD.26 Although evolocumab received a recommendation to reimburse with conditions from the Canadian Drug Expert Committee, negotiations with the pan-Canadian Pharmaceutical Alliance concluded without agreement for ASCVD.27
At the time of this writing, inclisiran (Leqvio) is under review for the treatment of HeFH and ASCVD as an adjunct to lifestyle changes in adults who are on maximally tolerated dose of a statin, with or without other LLTs. Given that this indication significantly overlaps with that being reviewed for evolocumab, inclisiran may be a relevant comparator that could not be included in the present analysis. The cost-effectiveness of evolocumab compared to inclisiran is unknown. Additionally, it is uncertain how the introduction of inclisiran would affect market-share expectations and, subsequently, the budget impact of evolocumab.
Clinician input noted that the 2021 Canadian Cardiovascular Society Guidelines for the Management of Dyslipidemia for the Prevention of Cardiovascular Disease in Adults recommends the use of PCSK9 inhibitors, such as evolocumab, as second-line or third-line therapy (after statins and ezetimibe) in post-ACS patients with LDL-C levels greater than 1.8 mmol/L.28 However, clinicians noted that they are unable to follow the guidelines in a considerable proportion of patients because of the funding status of PCSK9 inhibitors, which are not accessible without private insurance.
Based on the CDA-AMC Clinical Review, results from the subgroup analyses of patients who experienced an MI in the previous year or who were 1 year or more beyond their MI from the FOURIER trial are indicative of a possible subgroup effect in favour of evolocumab in combination with optimized LLT over optimized LLT alone for CV event outcomes over a median follow-up period of 26 months. The sponsor also used an ad hoc subgroup analysis that assessed data from the FOURIER-OLE study of patients who experienced an MI before and/or during the FOURIER trial, which suggested a benefit. However, the Clinical Review noted that the duration of follow-up in the parent trial is likely inadequate to assess the long-term relative safety of evolocumab.
Using the sponsor’s base case, evolocumab plus optimized LLT is associated with 0.90 incremental QALYs at an additional cost of $78,856, resulting in an ICER of $87,882 compared to optimized LLT alone. The probability of being cost-effective at a $50,000 per QALY gained threshold was 0%, and a 50% price reduction would be required to achieve cost-effectiveness at this threshold. There is uncertainty in the predicted benefit of evolocumab due to limitations in comparative clinical efficacy. Given the uncertainty surrounding the relationship between treatment with evolocumab and CV-related mortality, CDA-AMC conducted 2 scenario analyses to assess the impact of alternative assumptions for this parameter. CDA-AMC’s scenario analyses applied the lower and upper credible intervals of the hazard ratio for CV-related mortality derived from the FOURIER-OLE subgroup analysis of patients with a history of MI. These analyses represent the greatest and smallest CV-related mortality benefits associated with treatment with evolocumab.
Results from scenario analysis 1 and scenario analysis 2 were generally aligned: evolocumab is not cost-effective at a willingness-to-pay threshold of $50,000 per QALY gained. In scenario analysis 1 (assuming the greatest mortality benefit associated with treatment with evolocumab), evolocumab was associated with an ICER of $68,809 per QALY gained compared to optimized background LLT alone. In scenario analysis 2 (assuming the smallest mortality benefit associated with treatment with evolocumab), evolocumab was associated with an ICER of $164,205 per QALY gained compared to optimized LLT alone. Using CDA-AMC’s scenario analyses, a price reduction of 35% to 69% would be necessary to achieve cost-effectiveness at a $50,000 per QALY gained threshold.
There were several limitations that add uncertainty to these results, including uncertainty about the clinical evidence that informed the relationship between treatment with evolocumab and CV events and structural assumptions around treatment discontinuation and waning. These limitations could not be addressed through reanalysis, and the predicted clinical benefit of evolocumab remains uncertain, given that 90% of the predicted incremental QALYs for evolocumab were accrued beyond the trial period, for which there is no clinical evidence.
acute coronary syndrome
atherosclerotic cardiovascular disease
budget impact analysis
Canada’s Drug Agency
cardiovascular
heterozygous familial hypercholesterolemia
incremental cost-effectiveness ratio
ischemic stroke
low-density lipoprotein cholesterol
lipid-lowering therapy
myocardial infarction
open-label extension
quality-adjusted life-year
real-world evidence
unstable angina
CDA-AMC Cost Comparison for the Treatment of Acute Coronary Syndrome.
Note that this appendix has not been copy-edited.
Submission Quality.
Note that this appendix has not been copy-edited.
Model Structure.
Disaggregated Summary of the Sponsor’s Economic Evaluation Results.
Note that this appendix has not been copy-edited.
CDA-AMC Revisions to the Submitted Economic Evaluation for Scenario Analyses.
Disaggregated Summary of CDA-AMC Scenario Analysis 1.
Disaggregated Summary of CDA-AMC Scenario Analysis 2.
CDA-AMC Price Reduction Analyses, Scenario Analyses.
Note that this appendix has not been copy-edited.
Summary of Key Take-Aways.
The sponsor submitted a budget impact analysis (BIA) estimating the incremental budget impact of reimbursing evolocumab as an adjunct to optimized LLT for the treatment of recent ACS within the past 1 year who have LDL-C ≥ 1.8 mmol/L despite taking moderate-to-high intensity statin therapy, with or without ezetimibe.31 The sponsor used an epidemiologic approach using a participating public drug plan perspective, over a 3-year time horizon (2025 to 2027). The reference scenario includes optimized LLT as the comparator. Beginning with the population in Canada aged 18 years and older, excluding Quebec, the sponsor narrowed the population using estimates of HeFH prevalence,32,33 the proportion with ASCVD,34,35 the proportion of patients with recent ACS,2 the proportion of those receiving optimal LLT,13 the proportion of patients with LDL-C ≥ 1.8 mmol/L,36 and estimates of public drug coverage.37 Key inputs to the BIA are documented in Table 15.
The sponsor’s BIA included the following key assumptions:
The sponsor excluded the cost of optimized LLT from the BIA because it is not expected to change with the introduction of evolocumab for the requested population.
The sponsor assumed that no patients would discontinue evolocumab over the BIA time horizon.
Summary of Key Model Parameters.
The sponsor estimated the net budget impact of funding evolocumab for the requested population will be $93,570,873 in year 1, $126,733,581 in year 2, and $160,817,043 in year 3. The total incremental expenditure is estimated to be $381,121,498 over the first 3 years of listing evolocumab.
CDA-AMC identified several key limitations to the sponsor’s analysis that have notable implications on the results of the BIA:
The sponsor’s derivation of the eligible population was inappropriate. The sponsor used a prevalence-based approach to estimate the eligible population beginning with the Canadian population prevalence of ASCVD.34,35 However, the sponsor then narrowed the ASCVD population with an estimate of the proportion of patients with ASCVD found to have had an ACS event during the follow-up period (mean duration of follow-up of 40.8 months) from the Alberta RWE study (an incidence-based approach).2 CDA-AMC identified limitations with this approach. First, the proportion of patients with ASCVD who had an ACS event was identified over 3 years and thus overestimated the annual incidence. Second, the sponsor’s mixed use of prevalence- and incidence-based sources leads to uncertainty. As a result, the sponsor overestimated the number of patients with ACS that happen per year in participating jurisdictions, with an estimate of approximately 183,000 people. Given the reimbursement request (i.e., patients with recent ACS within the past 1 year) a fully incidence-based approach is more appropriate.
Given that nearly all ACS events happen in people with ASCVD (whether diagnosed or not), an incidence-based approach to estimating the eligible population may be a more appropriate method. According to the Canadian Chronic Disease Surveillance System, the annual incidence of MI in adults aged 20 years and older is 214 per 100,000 people.38 Applying this incidence to the population in Canada (excluding Quebec) without HeFH results in an estimate of approximately 55 thousand people with recent MI. This estimate aligns with the Canadian Institute for Health Information’s estimate that approximately 51 thousand patients with acute MI are admitted to Canadian hospitals (excluding Quebec) annually.39 Clinical experts consulted by CDA-AMC indicated that to account for the incidence of UA, increasing the MI incidence by 10% would capture additional patients with UA. This results in an estimated incidence of 235 per 100,000 population. While conditions other than ASCVD could result in MI, the clinical experts indicated that those are rare, and that assuming all ACS is a result of ASCVD is reasonable.
CDA-AMC used the annual incidence of MI in Canada in 2020 to 2021 (214 per 100,000)38 with an increase of 10% to capture the incidence of UA to derive the eligible population. CDA-AMC maintained the sponsor’s narrowing of the population by those on optimal LLT (83.1%) and those whose LDL-C remains above the 1.8 mmol/L threshold (42.3%).
The clinical experts indicated that UA is a disappearing diagnosis due to the high sensitivity cardiac troponin assays. Thus, the clinical experts suggested that consideration of MI only, as opposed to the definition of ACS that includes UA, may be more relevant for the purpose of this review. As such, CDA-AMC performed a scenario analysis using only the annual incidence of MI (214 per 100,000) to estimate the patient population.
The market uptake of evolocumab is uncertain. The sponsor estimated market shares of evolocumab using internal forecasts. The clinical experts consulted by CDA-AMC indicated that the sponsor’s projections may be greater than what will happen in Canadian clinical practice. The clinical experts suggested that by year 3 of reimbursement the market uptake of evolocumab may reach 30%. However, the trajectory of that uptake including the year 1 estimate is uncertain and may be influenced by patient choice and physician education programs.
CDA-AMC maintained the sponsor’s estimates of market uptake in the base-case analysis.
CDA-AMC conducted a scenario analysis informed by clinical experts, assuming that the market uptake of evolocumab will be 12.5%, 20%, and 30% in years 1, 2, and 3, respectively.
CDA-AMC revised the sponsor’s submitted analysis by using an incidence-based approach to estimating the eligible population. The changes applied to derive the CDA-AMC base case are described in Table 16.
CDA-AMC Revisions to the Submitted BIA.
The results of the CDA-AMC step-wise reanalysis are presented in summary format in Table 17 and a more detailed breakdown is presented in Table 18.
The CDA-AMC base case suggests that reimbursing evolocumab in the requested population would be associated with an incremental cost of $31,417,178 in year 1, $42,551,826 in year 2, and $53,995,624 in year 3, for a 3-year budgetary impact of $127,964,628. The CDA-AMC incidence-based approach resulted in a smaller eligible population than the sponsor’s estimate, which resulted in a 66% decline in the predicted 3-year budget impact.
Summary of CDA-AMC Reanalyses of the BIA.
CDA-AMC conducted the following scenario analyses to address remaining uncertainty, using the CDA-AMC base case (results are provided in Table 18):
assuming that only patients with recent MI make up the eligible population (i.e., excluding patients with recent UA)
assuming that the market shares in years 1, 2, and 3 are 12.5%, 20%, and 30%, respectively
assuming that the price of evolocumab is reduced by 50% (CDA-AMC’s estimated price reduction from the sponsor’s base-case analysis).
Detailed Breakdown of CDA-AMC Reanalyses of the BIA.
Copyright © 2024 - Canadian Agency for Drugs and Technologies in Health. Except where otherwise noted, this work is distributed under the terms of a Creative Commons Attribution-NonCommercial- NoDerivatives 4.0 International licence (CC BY-NC-ND).
Your browsing activity is empty.
Activity recording is turned off.
See more...
