Potential cardiovascular risk reduction with evolocumab in the real world: a simulation in patients with a history of myocardial infarction from the HEYMANS register
Abstract Background/Introduction FOURIER included 22,351 patients with a history of myocardial infarction (MI) and a median low-density lipoprotein cholesterol (LDL-C) of 2.4 mmol/L. Reducing LDL-C with evolocumab reduced the risk of major cardiovascular (CV) events by 1.3%, in absolute terms, over 2.2 years. Whether similar benefits might be observed in real-world evidence from evolocumab use is unknown. Purpose Simulate CV risk and assess the potential CV risk reduction among a large European cohort of evolocumab users with a history of MI. Methods We used interim data from HEYMANS, a register of patients initiating evolocumab in routine clinical practice across 12 European countries, from August 2015 with follow-up through July 2020. Demographic and clinical characteristics, lipid-lowering therapy (LLT), and lipid values were collected from routine medical records (6 months prior to evolocumab initiation through 30 months post initiation). Patients with a history of MI were considered and two sub-cohorts were created: recent MI (MI ≤1 year before evolocumab initiation) and remote MI (MI >1 year before evolocumab initiation). For each patient, we 1) simulated their CV risk using three different sources, correcting for age and LDL-C: i) the REACH equation, ii) FOURIER, iii) an observational study including FOURIER-like patients; 2) calculated their absolute LDL-C reduction on evolocumab; 3) simulated their relative risk reduction (RRR) by randomly sampling from the inverse probability distribution of the rate ratio per 1 mmol/L from the key secondary endpoint in the FOURIER landmark analysis; 4) calculated their absolute risk reduction (ARR) and number needed to treat (NNT) over 2 years (recent MI) or 10 years (remote MI). Results Our analysis included 90 recent MI and 489 remote MI patients initiating evolocumab in clinical practice per local reimbursement criteria, with up to 24 months follow-up. Median (inter-quartile range) age was 59 (53–67) and 61 (53–68) years in recent MI and remote MI patients, respectively. LDL-C before evolocumab was 3.8 (3.2–4.6) and 3.6 (3.0–4.5) mmol/L. Absolute LDL-C reduction on evolocumab was 2.2 (1.4–2.8) and 2.2 (1.6–2.8) mmol/L, meaning relative LDL-C reduction of 60% (44%-73%) and 62% (47%-72%), respectively. Predicted ARR with evolocumab was substantial, whether over 2 years (recent MI) or over 10 years (remote MI). See Table 1. Conclusions This cohort of evolocumab users in clinical practice had a higher baseline LDL-C and CV risk than patients enrolled in FOURIER. LDL-C reduction and RRR were very similar in recent MI and remote MI patients. However, patients with a recent MI had a higher short-term CV risk and therefore showed a larger ARR on evolocumab. FUNDunding Acknowledgement Type of funding sources: Private company. Main funding source(s): Amgen