Real-world data, theoretical application of guidelines, cost, and access: how do we optimize non-statin therapy for LDL-C/non-HDL-C/ApoB?

This editorial refers to ‘Application of the 2019 ESC/EAS dyslipidaemia guidelines to nationwide data of patients with a recent myocardial infarction: a simulation study’†, by A. Allahyari et al., on page 3900. In this issue of the European Heart Journal, Allahyari et al.1 evaluated the implications of the 2019 ESC/EAS Guidelines for the Management of Dyslipidaemias2 for treatment of patients with recent myocardial infarction (MI) using data from the Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies (SWEDEHEART) register. Of 25 466 patients with MI in 2013–2017, follow-up visit 6–10 weeks post-MI, and LDL cholesterol (LDL-C) assessed at both time points, 86.6% were on high-intensity statin therapy and <3% on ezetimibe. Despite very high utilization of high-intensity statin therapy, 82.9% of patients would be recommended for intensification of therapy because they did not meet the goals of LDL-C <1.4 mmol/L and LDL-C reduction of ≥50% based on the 2019 ESC/EAS guidelines. Modelling to estimate the impact of sequential maximization of statin therapy, addition of ezetimibe, and addition of a proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor indicated that 50.7% would be eligible for a PCSK9 inhibitor, and addition of a PCSK9 inhibitor would enable ∼90% of all patients to achieve LDL-C <1.4 mmol/L. These findings raise important issues based on concerns about the very large number of patients who would need PCSK9 inhibitors and the cost of therapy to a healthcare system with implementation of the recommended guidelines, highlighting some of the challenges in developing guidelines and implications of different approaches. All the guidelines have in common targeting reductions in LDL-C, with the intensity of therapy dependent on the individual’s risk for developing atherosclerotic cardiovascular disease (ASCVD) events. However, dyslipidaemia guidelines developed in different parts of the world differ in emphasis and approach; some guidelines focus on providing clear and simple algorithms based on compiled evidence from clinical trials, epidemiology, and pathophysiology but excluding cost considerations,2 whereas other guidelines have a more complex approach with more risk categories, are compiled almost entirely from randomized clinical trial results, and consider both cost and cost-effectiveness.3 The study by Allahyari et al.1 highlights the potential impact on healthcare systems of these different approaches. Two major differences between the 2019 ESC/EAS guidelines and the 2018 AHA/ACC multisociety guidelines for treating patients with clinical ASCVD are risk stratification and role of LDL-C level (goal vs. threshold). In the ESC/EAS guidelines, all patients with documented ASCVD are treated alike. In contrast, the AHA/ACC guidelines recommend risk stratification for patients with clinical ASCVD to identify those at very high risk of ASCVD events, including patients with multiple major ASCVD events and those with one or more major ASCVD event and multiple high-risk conditions, as supported by clinical trials of ezetimibe or PCSK9 inhibitor therapies added to statin therapy. For example, in the Improved Reduction of Outcomes: Vytorin Efficacy International Trial (IMPROVE-IT), in patients with acute coronary syndrome (ACS), those with three or more high-risk features had much higher recurrent event rates (40% vs. 14% for 0–1 high-risk features) and greater risk reduction with addition of ezetimibe (6.3% vs. 2.2% absolute risk reduction in patients with two high-risk features).4 Similarly, in the Further Cardiovascular Outcomes Research with PCSK9 Inhibition in Subjects with Elevated Risk (FOURIER) trial of evolocumab in patients with clinical ASCVD and one or more major risk factor or two or more minor risk factors,5 patients with recent MI, ≥2 MIs, residual multivessel coronary disease,6 or pulmonary artery disease7 had higher risk of recurrent events and, therefore, much larger absolute risk reduction despite similar relative risk reduction, compared with average trial participants. A post-hoc analysis from the ODYSSEY Outcomes trial of alirocumab in patients with ACS on background statin therapy also showed greater absolute risk reduction in patients classified as very-high-risk ASCVD by the AHA/ACC guideline (63% of enrolled patients), compared with those not so classified (2.1% vs. 0.8% event rates).8 In total events analyses from this trial, ∼5 events were avoided over 4 years per 100 patients in the very-high-risk ASCVD category vs. 1.6 events in other participants. ODYSSEY Outcomes participants with a high polygenic risk score (>90th percentile) were at high risk for events and had greater absolute risk reduction (6.0% vs. 1.5%).9 Based on this clinical trial evidence, one would expect much higher absolute risk reduction with non-statin therapies in very-high-risk ASCVD patients, with lower numbers needed to treat. Although distinguishing a very-high-risk ASCVD category instead of considering all patients with clinical ASCVD as one group increases complexity, with the disadvantage of adding another step for clinicians to consider in guideline implementation, this stratification allows tailoring of therapy to patients most likely to benefit from non-statin add-on therapies, while avoiding these therapies in patients who would derive less benefit with additional pill burden (which may also affect statin adherence) and cost. Another difference in approach between the guidelines is the use of lipid levels as thresholds for initiating non-statin therapies vs. goals of therapy. Although both guidelines recommend LDL-C reduction ≥50% with high-intensity statin therapy in ASCVD patients, the AHA/ACC guideline establishes threshold levels of LDL-C (≥1.8 mmol/L, based on baseline levels in IMPROVE-IT,10 FOURIER,5 and ODYSSEY Outcomes11) or non-HDL-C (≥2.6 mmol/L) for considering addition of ezetimibe or a PCSK9 inhibitor, whereas the ESC/EAS guidelines recommend an LDL-C treatment goal of <1.4 mmol/L, with the stepwise addition of ezetimibe and then a PCSK9 inhibitor if the LDL-C goal is not achieved with statin, reflecting achieved LDL-C levels in non-statin therapy trials and data showing that patients with lower achieved LDL-C have lower recurrent ASCVD event rates. These different approaches to lipid levels lead to important considerations raised by this study. First, the benefits of LDL-C–lowering therapy are related to the absolute reduction in LDL-C level, with each 1 mmol/L reduction associated with ∼22% reduction in major adverse cardiovascular events.12 Thus, a 50% reduction in LDL-C from a baseline level of 1.5 mmol/L provides an absolute 0.75 mmol/L reduction but a 1.0 mmol/L reduction from a baseline of 2.0 mmol/L. Therefore, the lower the LDL-C level at which non-statin therapy is added, the less the absolute LDL-C reduction, and the lower the relative risk reduction for ASCVD events. The greatest challenges with recommending an LDL-C goal of <1.4 mmol/L are the implication that it should be attained in all patients with ASCVD and, as pointed out clearly in this study, the large number of patients who would require ezetimibe and PCSK9 inhibitor therapy. Identifying patients with the highest risk for recurrent ASCVD events among those with clinical ASCVD and recommending consideration of additional non-statin therapies for LDL-C ≥1.8 mmol/L, as in the AHA/ACC guidelines, reduces the numbers needed to treat. Application of the AHA/ACC guidelines to a cohort of >1 million ASCVD patients in the U.S. Veterans Administration healthcare system showed that 43% were in the very-high-risk category and, in models with optimization of statin therapy and sequential use of ezetimibe followed by a PCSK9 inhibitor, ∼10% of all ASCVD patients (24% of very-high-risk patients) would need a PCSK9 inhibitor.13 As noted by Allahyari et al., coverage decisions in most healthcare systems and nations consider both cost-effectiveness and affordability (budget impact). In addition to ASCVD risk and absolute LDL-C [or non-HDL-C or apolipoprotein B (apoB)] reduction, the benefit of lipid therapy is related to duration of and adherence to therapy. Although both FOURIER5 and ODYSSEY Outcomes11 included patients with ASCVD plus additional risk factors and with elevated baseline LDL-C (≥1.8 mmol/L) or non-HDL-C (≥2.6 mmol/L), in both trials the short duration of therapy was insufficient to assess fully cardiovascular mortality and total mortality, the latter a critical endpoint in determining cost-effectiveness based on years of life saved. As we move into the biotech world of precision medicine, careful patient selection and clinical trial duration are critical to provide evidence for guideline development and cost-effectiveness evaluation. Future guidelines should provide easy-to-use tools for quantitative risk assessment in both primary and secondary prevention, utilizing clinical data including imaging, biomarkers, and genetics, as well as potential benefit of therapy, based on absolute ASCVD risk, by calculating expected LDL-C (or non-HDL-C or apoB) reduction with a specific therapy. While the study by Allahyari et al. focused on the theoretical cost of adding a PCSK9 inhibitor to 50% of post-MI patients, their data demonstrate that despite outstanding use of high-intensity statin (∼87%), the actual use of combination lipid therapy was extremely low: 3% treated with ezetimibe and only 0.03% (13/44 890 SWEDEHEART patients; excluded from this analysis) with a PCSK9 inhibitor. In contrast to treatment of hypertension and diabetes, combination therapy is rarely used to treat hypercholesterolaemia as recommended by all current guidelines. Although PCSK9 inhibitors are extremely effective in lowering LDL-C, have outstanding intermediate-term safety data, and have been shown to improve outcomes, they are rarely used even in very-high-risk ASCVD patients with severely elevated LDL-C14 because of restrictive policies by many payers that deny access even to appropriate patients. Professional societies, guideline committees, and payers must work together to help healthcare providers identify and treat high-risk patients with elevated levels of atherogenic lipoproteins, balancing concerns about theoretical future cost of overutilization of additional non-statin therapy while addressing the current undertreatment of high-risk ASCVD patients who would benefit from additional non-statin therapy (Table 1). Approaches to optimize benefits of LDL-C lowering with non-statin therapies post–acute coronary syndrome: SWEDEHEART 13/44 980 prospective patients; excluded from the analysis because there were so few. Future approaches to optimize clinical benefit and cost-effectiveness: •Quantitative risk assessment in both primary and secondary prevention. •Calculation of predicted absolute reduction in LDL-C (or non-HDL-C or apoB) with addition of non-statin therapy. Approaches to optimize benefits of LDL-C lowering with non-statin therapies post–acute coronary syndrome: SWEDEHEART 13/44 980 prospective patients; excluded from the analysis because there were so few. Future approaches to optimize clinical benefit and cost-effectiveness: •Quantitative risk assessment in both primary and secondary prevention. •Calculation of predicted absolute reduction in LDL-C (or non-HDL-C or apoB) with addition of non-statin therapy. Conflict of interest: C.M.B. has received grant/research support from Abbott Diagnostic, Akcea, Amgen, Esperion, Novartis, Regeneron, and Roche Diagnostic; and is a consultant fort Abbott Diagnostics, Akcea, Amarin, Amgen, Arrowhead, Astra Zeneca, Boehringer Ingelheim, Corvidia, Denka Seiken, Esperion, Intercept, Janssen, Matinas BioPharma Inc., Merck, Novartis, Novo Nordisk, Regeneron, Roche Diagnostic, and Sanofi-Synthelabo. S.S.V. has received grant support from the Department of Veterans Affairs, World Heart Federation, and the Tahir and Jooma Family; an honorarium for the American College of Cardiology (Associate Editor for Innovations, ACC.org); and is on the steering committeeof the Patient and Provider Assessment of Lipid Management (PALM) registry at the Duke Clinical Research Institute (no financial remuneration). The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology or of the U.S. Department of Veterans Affairs.