Design and rationale of the EBBINGHAUS trial: A phase 3, double‐blind, placebo‐controlled, multicenter study to assess the effect of evolocumab on cognitive function in patients with clinically evident cardiovascular disease and receiving statin background lipid‐lowering therapy—A cognitive study of patients enrolled in the FOURIER trial

Abstract

Some observational studies raised concern that statins may cause memory impairment, leading to a US Food and Drug Administration warning. Similar questions were raised regarding proprotein convertase subtilisin/kexin‐type 9 inhibitors (PCSK9i) and neurocognitive function. No prospectively designed study has evaluated the relationship between long‐term PCSK9i use and cognition changes. Patients with prior cardiovascular disease treated with maximally tolerated statin enrolled in FOURIER (the randomized, double‐blind, placebo‐controlled cardiovascular outcome study of the PCSK9i evolocumab) could participate in this prospective assessment of cognitive function (EBBINGHAUS). Key additional exclusion criteria for EBBINGHAUS were dementia, cognitive impairment, or other significant mental or neurological disorder. Cognitive testing was performed using the Cambridge Neuropsychological Test Automated Battery, a tablet‐based tool assessing executive function, working memory, memory function, and psychomotor speed at baseline, weeks 24 and 48, every 48 weeks thereafter, and study end. The primary endpoint was spatial working memory strategy index of executive function (SWMSI). The primary hypothesis was that evolocumab would be noninferior to placebo in the mean change from baseline over time in SWMSI. Fifteen hundred cognitively normal patients completing the assessments provided approximately 97% power to demonstrate that the upper 95% confidence interval for the treatment difference in mean change from baseline in SWMSI over time is <20% of the SD of the mean change in the placebo group. An exploratory analysis will compare neurocognitive function in patients with post‐baseline low‐density lipoprotein cholesterol <25 mg/dL. EBBINGHAUS will evaluate whether the addition of evolocumab to statin therapy affects cognitive function over time in patients with stable cardiovascular disease.

1. INTRODUCTION

The clinical benefits of statin treatment for lowering cholesterol in primary and secondary prevention were documented in multiple cardiovascular outcomes trials.1 Furthermore, data continue to accrue that even among patients who achieve low low‐density lipoprotein cholesterol (LDL‐C) levels (ie, <70 mg/dL), cardiovascular events may be reduced even further with additional reduction in LDL‐C.1, 2

Thus, there is an unmet need for a potent, effective nonstatin agent that will permit a significantly greater proportion of patients to achieve lower LDL‐C concentrations and further reduce cardiovascular risk.3, 4 Based on current data, evolocumab may fulfill this need and provide an important addition to the treatment of hypercholesterolemia, particularly for high‐risk patients with a history of cardiovascular disease.

1.1. Potential cognitive effects of lipid‐lowering drugs

Virtually all of the cholesterol in the brain is synthesized locally by glia and secreted in the form of lipoproteins. Given the importance of cholesterol in synapse formation and function, the potential adverse cognitive effects of cholesterol‐reducing drugs are a reasonable concern, particularly in the case of lipophilic statins that are able to cross the blood–brain barrier.5

High‐quality studies in this area, however, are scarce. Postmarketing surveillance reports, case reports, and observational studies raised the possibility of an association between statin use and impaired cognitive function. Two small (<300 patients) randomized controlled trials6, 7 of 6 months’ duration in cognitively normal patients showed small improvements in the placebo group with no similar improvement in the statin group. However, in one study,7 the results were confounded by baseline differences in test results. The authors concluded that it remained uncertain whether such minor effects would have any long‐term sequelae or occur with other cholesterol‐lowering interventions. Postmarketing reports described ill‐defined memory impairment, reversible upon statin discontinuation, and some observational studies described adverse cognitive effects that recurred with re‐challenge. Based on these observations, in 2012, the US Food and Drug Administration issued a new warning for the labeling of statin drugs regarding potential adverse effects on cognition.8 It is also unclear whether adverse cognitive effects of statins, if they do occur, can be attributed to lowering of LDL‐C or other off‐target effects unique to statins.

Subsequent studies, however, found neutral or positive effects of statins on cognition, and several systematic reviews and meta‐analyses have suggested that the preponderance of evidence favors the absence of any adverse effects of statins on cognition.9, 10, 11, 12, 13 After a review of the totality of evidence to date, in 2014 the Statin Cognitive Safety Task Force concluded that statins as a class are not associated with adverse effects on cognition.14

1.2. Evolocumab: A monoclonal antibody inhibitor of proprotein convertase subtilisin/kexin‐type 9 (PCSK9)

Evolocumab (AMG 145) is a fully human monoclonal immunoglobulin G2 that binds specifically to human proprotein convertase subtilisin/kexin‐type 9 (PCSK9) and prevents its interaction with the LDL receptor. This results in increased LDL receptor activity and subsequent decreased circulating concentrations of LDL‐C. In contrast to the mechanism of action for statins, evolocumab is not predicted to directly affect endogenous cholesterol production or cross the blood–brain barrier.15

PCSK9 activity does not appear to be associated with cognitive function. A single nucleotide polymorphism resulting in decreased PCSK9 function and reduced LDL‐C levels was not associated with cognitive function in a prospective study of pravastatin in elderly patients.16 Additionally, a family of individuals with PCSK9 loss of function mutations and lifelong exposure to reduced levels of PCSK9 and LDL‐C (including an individual with no detectable circulating PCSK9 and LDL‐C <20 mg/dL) was reported to have apparently normal cognitive function.17

Findings from phase 2 and phase 3 studies show that evolocumab lowers LDL‐C by 60% to 65% in various patient populations.18 Imbalances in adverse effects on cognitive function compared with placebo were not observed in placebo‐controlled studies; however, open‐label extension data from 2 randomized phase 3 controlled trials found an increase in reported neurocognitive events in the evolocumab group (0.9% vs 0.3%) that did not appear to be related to minimum post‐baseline level of LDL‐C.19 A similar imbalance in neurocognitive events (1.2% vs 0.5%) was observed in a 78‐week randomized placebo‐controlled study of the PCSK9 inhibitor alirocumab.20

A large cardiovascular outcomes trial of evolocumab, Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Subjects With Elevated Risk (FOURIER),21 is now underway. Given the paucity of high‐quality evidence relating LDL‐C lowering and adverse cognitive effects and the ability of PCSK9 inhibitors to reduce LDL‐C to unprecedented low levels, we prospectively designed a dedicated neurocognitive study using an established battery of tests within the FOURIER trial to specifically address this important gap in knowledge.

2. METHODS

2.1. Study design and objectives

Evaluating PCSK9 Binding Antibody Influence on Cognitive Health in High Cardiovascular Risk Subjects (EBBINGHAUS; NCT 02207634)21 is assesing cognitive function over time in a subset of patients who were enrolled in the FOURIER trial,22 in which patients were randomized in a double‐blind fashion to receive evolocumab or placebo. The FOURIER trial is currently being conducted in 27 564 high‐risk patients with clinically evident cardiovascular disease who are receiving high‐ to moderate‐intensity lipid‐lowering therapy with statin; 1974 of these patients enrolled in EBBINGHAUS.

The primary objective of the EBBINGHAUS study is to evaluate the change over time in executive function, as assessed by the Cambridge Neuropsychological Test Automated Battery23 (CANTAB, http://www.cambridgecognition.com). CANTAB is a computerized cognitive‐assessment tool that uses sensitive touch‐screen neuropsychological tests of cognition specifically designed to assess central nervous system (CNS) disorders and cognitive function across a range of domains, including episodic and working memory, executive function, psychomotor speed, and attention. These assessments were designed to be language‐ and culture‐independent and are therefore suitable for large multinational clinical studies. CANTAB has been utilized in >160 clinical trials over the past 3 decades in all phases of drug development, and it is sensitive to both positive and negative effects of drugs on cognition. These trials span a broad range of therapeutics, including monoclonal antibodies, vaccines, and partial neurotransmitter antagonists, and include studies in both normal subjects and in populations with neuropsychiatric disorders, such as schizophrenia and Alzheimer disease.23, 24, 25

The primary objective of this study was to compare the Spatial Working Memory (SWM) strategy index of executive function, as assessed by CANTAB, between patients receiving evolocumab and patients receiving placebo. Secondary objectives include comparisons between randomized treatment groups in:

• Working memory, as assessed by the CANTAB SWM test between‐errors score

• Memory function, as assessed by the CANTAB Paired Associates Learning (PAL) test

• Psychomotor speed, as assessed by the CANTAB Reaction Time (RTI) test

These tests were selected to comprise a broad range of cognitive domains, because it is unclear in which domains a hypothetical effect on cognition might manifest. An assessment of executive function was selected as the primary endpoint, as this domain integrates several cognitive processes.

Exploratory analyses will evaluate the change over time in the CANTAB global composite score of cognitive function and will summarize the primary and secondary endpoints in patients with vs without ≥1 post‐baseline LDL‐C <25 mg/dL.

The primary hypothesis is that in patients receiving statin therapy in combination with evolocumab, the mean change from baseline over time in executive function will be noninferior to mean change from baseline over time in patients receiving statin therapy in combination with placebo.

2.2. Study population

Eligibility criteria for the FOURIER trial have been published previously.21 In brief, patients with prior myocardial infarction, prior ischemic stroke, or symptomatic peripheral arterial disease with an LDL‐C ≥70 mg/dL (or non–high‐density lipoprotein cholesterol ≥100 mg/dL) on an optimized statin regimen at least as potent as atorvastatin 20 mg daily, were randomized 1:1 to receive evolocumab (either 140 mg subcutaneously every 2 weeks or 420 mg subcutaneously every month, according to patient preference) or matching placebo injections. Major exclusion criteria were myocardial infarction or stroke within 4 weeks, New York Heart Association class III or IV heart failure or left ventricular ejection fraction <30%, prior hemorrhagic stroke, planned revascularization in the next 3 months, or severe renal dysfunction.

Patients were eligible to enroll in EBBINGHAUS provided they had been randomized in FOURIER and had not passed the week 12 study visit. Since the natural courses of temporal changes in cognitive function are not the same in patients with, vs those without, a history of mental disability, patients with a diagnosis of dementia, cognitive impairment, or other significant mental or neurological disorders that could confound the study results or interfere with the patient's participation were excluded. Additional eligibility criteria and participating countries are listed in Table 1.

Table 1.

Eligibility criteria for EBBINGHAUS

Inclusion criteria
Signed written informed consent for the EBBINGHAUS study
Consented and randomized into the FOURIER trial
Exclusion criteria
Current or known past diagnosis of dementia or mild cognitive impairment
Any condition or situation, including other significant mental or neurological disorders that, in the investigator's opinion, may confound the study results or may interfere significantly with the patient's participation in either EBBINGHAUS or FOURIER
Participating countries
Australia	Germany	Malaysia	South Africa
Belgium	Greece	Netherlands	Spain
Canada	Hong Kong	New Zealand	Sweden
Czech Republic	Hungary	Norway	Turkey
Denmark	Italy	Poland	United Kingdom
Estonia	Japan	Portugal	United States
Finland	Latvia	Russia
France	Lithuania	Slovakia

Abbreviations: EBBINGHAUS, Evaluating PCSK9 Binding Antibody Influence on Cognitive Health in High Cardiovascular Risk Subjects; FOURIER, Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Subjects With Elevated Risk.

2.3. Treatment protocol

All participating centers obtained approval for the EBBINGHAUS and FOURIER protocols and each patient provided separate written informed consent for both protocols. Investigators were encouraged to enroll patients in EBBINGHAUS prior to the first dose of study drug in the FOURIER trial, although enrollment in EBBINGHAUS was permitted up to the FOURIER week 12 study visit. Enrollment in EBBINGHAUS was registered via an interactive voice response system/interactive web response system.

CANTAB assessments were performed at the study site during the regularly scheduled visits, and by study personnel trained in administering the tests. The initial CANTAB assessment was conducted during a screening visit for training purposes (to mitigate training effects that could influence test performance) and whenever possible, prior to the first dose of blinded study drug in FOURIER, and ≥1 week prior to the baseline CANTAB assessment. For patients enrolled in EBBINGHAUS after randomization in the FOURIER trial, the CANTAB screening assessment was performed within 4 weeks of the FOURIER week 12 study visit. Subsequent CANTAB testing occurred at weeks 24 (±6 weeks) and 48 (±6 weeks), then every 48 weeks (±6 weeks), and finally at the end‐of‐study visit for the FOURIER trial.

Details of the CANTAB tests and validation studies are provided (see Supporting Information, Appendices 1 and 2, in the online version of this article) and are also available at http://www.cambridgecognition.com/clinicaltrials/cantabsolutions/tests. The following 3 tests, which correlate with traditional neuropsychological testing in clinical practice,26 were conducted:

• Spatial Working Memory (SWM)

  Patients search for colored tokens hidden inside boxes on the screen by touching them. The critical instruction is that once a token has been found inside a box, there will never be a token hidden inside that box again, so patients must not return to a box where a token has been found.

• Paired Associates Learning (PAL)

  Boxes on the screen open up one at a time to reveal patterns; patients must remember where the patterns are located. Once all the patterns have been shown, the patients must touch the box where they think each pattern was hidden.

• Reaction Time (RTI)

  Patients touch and hold down a button on the screen until they see a yellow spot flash inside one of 5 target circles. Once they see the spot, they must let go of the button and touch the circle where the spot appeared as quickly as they can.

2.4. Study endpoints

The primary endpoint is the SWM strategy index of executive function. The strategy index score represents the number of times a patient begins a search with a different box in a SWM test having ≥6 boxes. A high score represents an inefficient use of strategy and planning, whereas a low score represents an efficient use of strategy (range, 4–28).

The 3 secondary endpoints are SWM between‐errors score, PAL total errors adjusted, and RTI median 5‐choice reaction time. The SWM between‐errors score is the number of times that a patient revisits a box in which a token has previously been found in an SWM test having ≥4 boxes. The PAL total errors adjusted comprises the number of errors committed by a patient in a PAL test plus an adjustment for the estimated number of errors the patient would have made on any stages that were not reached. The RTI median 5‐choice reaction time is the median duration between the onset of the stimulus and the release of the button in an RTI test.

The CANTAB global composite score of cognitive function is a prespecified exploratory endpoint and calculated by averaging the combined Z‐scores of each of the 4 measures above, where the Z‐score for any one measure is the difference between the patient score and the mean of the baseline score for all patients divided by the SD of the baseline score for all patients.

2.5. Statistical design and analysis

The primary analysis set in this study includes all dosed patients enrolled in EBBINGHAUS who completed a baseline cognitive functional assessment prior to, or on the first date of, administartion of study of administration of study drug and who have ≥1 post‐baseline cognitive measure. Patients are analyzed according to the treatment arm assigned at randomization. Additional analyses will use the full analysis set of all enrolled patients, adding patients who completed the baseline cognitive functional assessment after the day of the first dose of study drug. Data in patients who experience a stroke after enrollment will be included in the primary analysis up until the day of the stroke. Unless specified otherwise, the Z‐score will be used in efficacy analyses.

The primary analysis is a noninferiority comparison of evolocumab vs placebo of the mean change from baseline over time in the CANTAB SWM strategy index of executive function. A repeated measures mixed‐effect linear model will be used to estimate treatment difference and associated 95% confidence interval. The model will include stratification factor, age, education level, baseline SWM strategy index (Z‐score), treatment group, visit, and treatment by visit interaction. We estimate that approximately 1500 patients will provide ~97% power to show the upper limit of a 95% confidence interval for the treatment difference in mean change from baseline over time (placebo‐evolocumab) is <20% of the observed SD of the mean change (ie, Cohen's d <0.20 in the placebo group). Cohen's d or effect size27 is a generally accepted and commonly used metric in neuropsychological literature.28 Although qualitative interpretation of effect sizes must be done cautiously, d < 0.20 is typically considered small.27, 29 Given the demographics of the EBBINGHAUS population and the expected rates of cognitive change over time, the noninferiority margin is expected to be approximately equivalent to the effect of healthy aging over a 5‐ to 6‐year period.30

If noninferiority is demonstrated, superiority testing will be performed using the rate of change in SWM strategy index. For superiority testing, assuming an effect size of 0.15 of the rate of change, 1500 patients will provide ~83% power to demonstrate a difference at an α level of 0.05.

Additional exploratory analyses will include, but are not limited to, the following: (1) additional repeated measures mixed‐effect models analyses with additional covariates (eg, baseline LDL‐C, region, sex, prior history of stroke); (2) subgroup analyses of the primary endpoint; (3) by‐visit analyses; (4) analyses on secondary efficacy endpoints; (5) analyses by achievement of ≥1 post‐baseline LDL‐C <25 mg/dL; and (6) analyses in patients who received study drug ≥1 day after the baseline cognitive assessment in combination with the primary analysis set.

2.6. Study organization

The EBBINGHAUS protocol was a collaborative effort of the Steering Committee of the EBBINGHAUS trial (see Supporting Information, Appendix 3, in the online version of this article), an appointed member of the FOURIER trial Executive Committee (RPG), Cambridge Cognition (KZ), and the sponsor (Amgen), who will monitor the progress of the trial. The Thrombolysis In Myocardial Infarction (TIMI) Study Group, an academic research organization within Brigham and Women's Hospital (Boston, Massachusetts) and Cambridge Cognition (Cambridge, United Kingdom) will have full access to the complete EBBINGHAUS database after the study completion and will independently generate analyses. The EBBINGHAUS Steering Committee will be responsible for submitting the results of the study for publication to a peer‐reviewed medical journal. The authors are solely responsible for drafting and editing both the current and the results manuscripts and for their final content.

3. RESULTS

Patient enrollment began on September 10, 2014, and was completed on August 6, 2015. A total of 2442 patients were screened and 1974 were enrolled. There were 1204 patients who completed a neurocognitive assessment prior to (or on the same day as) the first dose of evolocumab and had ≥1 post‐baseline cognitive measure and who represent the primary analysis set, resulting in 93% power for the assessment of noninferiority. Baseline characteristics in EBBINGHAUS and FOURIER were generally similar (Table 2). Primary results of both the EBBINGHAUS and FOURIER trials are anticipated in early 2017.

Table 2.

Baseline characteristics

	EBBINGHAUS	FOURIER16
Number of patients enrolled	1974	27 564
Mean age, y (SD)	62.8 (8.7)	62.5 (9.0)
Female sex	543 (27.5)	6769 (24.6)
Race
White	1818 (92.1)	23,458 (85.1)
Black or African American	60 (3.0)	669 (2.4)
Asian or other	96 (4.9)	3437 (12.5)
Region
North America	479 (24.3)	4571 (16.6)
Europe	1359 (68.8)	17 335 (62.9)
Latin America	01	1823 (6.6)
Asia Pacific or South Africa	136 (6.9)	3835 (13.9)
Cardiac risk factors
HTN	1657 (83.9)	22,084 (80.1)
DM	691 (35.0)	10,081 (36.6)
Current cigarette use	661 (33.5)	7777 (28.2)
History of vascular disease
MI	1481 (75.0)	22,351 (81.1)
Nonhemorrhagic stroke	388 (19.7)	5337 (19.4)
PAD	370 (18.7)	3642 (13.2)
Statin use
High intensity	1408 (71.3)	19,103 (69.3)
Moderate intensity	564 (28.6)	8392 (30.4)
Low intensity or unknown	2 (0.1)	69 (0.3)
Ezetimibe	101 (5.1)	1440 (5.2)
Median lipid values at baseline, mg/dL (Q1–Q3)
LDL‐C2	92 (80‐108)	92 (80‐109)
Total cholesterol	168 (151‐189)	168 (151‐189)
HDL‐C	45 (38‐54)	44 (37‐53)
TG	131 (99‐179)	133 (100‐182)

Abbreviations: DM, diabetes mellitus; EBBINGHAUS, Evaluating PCSK9 Binding Antibody Influence on Cognitive Health in High Cardiovascular Risk Subjects; FOURIER, Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Subjects With Elevated Risk; HDL‐C, high‐density lipoprotein cholesterol; HTN, hypertension; LDL‐C, low‐density lipoprotein cholesterol; MI, myocardial infarction; PAD, peripheral arterial disease; Q, quartile; SD, standard deviation; TG, triglycerides.

Data are presented as n (%) unless otherwise indicated.

¹Countries in Latin America did not participate in EBBINGHAUS.

²LDL‐C was calculated based on the Friedewald equation, unless the calculated value was <40 mg/dL or the measured TG were >400 mg/dL, in which case preparative ultracentrifugation was performed with LDL‐C measured by β quantification.

4. DISCUSSION

Several studies have confirmed that cholesterol in the brain is generated and regulated by mechanisms largely independent from serum cholesterol, which does not cross the blood–brain barrier. Cholesterol levels in the brain, therefore, are not influenced primarily by dietary uptake or hepatic synthesis.31, 32 On the other hand, debate around the impact of serum cholesterol regulation on neurocognitive function remains. For example, serum cholesterol metabolites, such as oxysterols, cross the blood–brain barrier independently of serum cholesterol levels and can have a negative impact on brain function.31 Oxysterols are also known to be major regulators of cholesterol homeostasis in the CNS.31 Furthermore, excess serum cholesterol can induce signaling to the brain via cholesterol metabolites, pro‐inflammatory mediators and antioxidant processes, thereby possibly influencing neurocognitive function.31

A significant body of evidence indicates that elevated serum LDL‐C may be detrimental to human learning and memory and also may be a risk factor for mild cognitive impairment33, 34 and dementia.35 Other studies, however, show that increased cholesterol may be positively associated with learning and memory.36, 37

Taken together, these data may suggest that the relationship between cholesterol levels and adult learning and memory changes as a function of age. High serum cholesterol may be detrimental in middle age, whereas it may be beneficial or protective in the very old. Interestingly, a recent study found that there was no association between cognitive measures and serum cholesterol concentrations among the young.38

4.1. Statins and ezetimibe

A limited body of literature consisting of case reports and 2 small clinical trials has suggested that statins might have a negative impact on cognition.6, 7 However, the Prospective Study of Pravastatin in the Elderly at Risk (PROSPER),13 the largest prospective randomized placebo‐controlled study of a statin on cognitive function, evaluated 5804 patients with 4 neuropsychological performance tests at 6 different time points over 42 months mean follow‐up and found no significant effect on cognitive function or disability in patients treated with pravastatin compared with placebo. In addition, several meta‐analyses,39, 40, 41 including a recent comprehensive evaluation of randomized controlled trials with statins,42 suggest that statin use is not associated with adverse effects on cognition and that there may even be a modest association of statin use with reduced risk of dementia. Moreover, recent data obtained on ezetimibe from Improved Reduction of Outcomes: Vytorin Efficacy International Trial (IMPROVE‐IT)2 also are reassuring regarding very low LDL‐C levels induced by ezetimibe on neurocognitive function, as well as for the risk for hemorrhagic stroke.43

4.2. Proprotein convertase subtilisin/kexin‐type 9 (PCSK9)

In the process of human development, PCSK9 is expressed in specific brain regions where active neurogenesis takes place, and the question has therefore been raised whether impacting PCSK9 levels could affect neurocognitive function. On one hand, very low levels of circulating LDL‐C caused by genetic mutations leading to PCSK9 loss of function have not shown to affect CNS function. Also, genome‐wide association studies found no association between PCSK9 and Alzheimer disease44 or the risk of dementia.45 Although scarce, these genetic data are reassuring regarding natural PCSK9 inhibition and brain disorders. On the other hand, the impact of pharmacological compounds regulating PCSK9 levels on neurocognitive function remains to be clarified.

In a recent meta‐analysis reporting all clinical data available from phase 2 and phase 3 anti‐PCSK9 monoclonal antibody (mAb) studies,46 neurocognitive adverse events were low in absolute numbers (55/9581) but with significantly higher rates in the anti‐PCSK9 mAb‐treated groups compared with placebo. However, it is of note that no association was shown between neurocognitive events and very low LDL‐C levels obtained with either evolocumab19 or alirocumab.20 Furthermore, the recent Global Assessment of Plaque Regression With a PCSK9 Antibody as Measured by Intravascular Ultrasound (GLAGOV) trial47 showed no difference between PCSK9 add‐on treatment vs placebo regarding neurocognitive function, even in patients achieving very low LDL‐C (mean, 36.6 mg/d). However, these results should be considered with care, as the study population was small (N = 968) and follow‐up time was only 18 months.

5. CONCLUSION

If the marked lipid‐lowering effects observed with anti‐PCSK9 mAb48 do actually result in significant reductions in cardiovascular events in the ongoing outcome studies, their long‐term use in high‐risk patients on statin who require additional lowering of LDL‐C is likely to increase. It is therefore essential that the effect of PCSK9 mAb on neurocognitive function be carefully investigated in sufficiently large patient groups with appropriate follow‐up times. This is the rationale behind the EBBINGHAUS study, which will be the first to prospectively explore the relationship between extremely low LDL‐C levels and neurocognitive function. However, further studies evaluating patients with impaired neurocognition will be necessary to establish whether the inhibition of PCSK9 could affect patients with abnormal baseline function. Dedicated neurocognitive studies on intensive lipid‐lowering effects, such as EBBINGHAUS, should provide much‐needed data for this field.

Conflicts of interest

Dr. Giugliano has received honoraria for CME lectures and/or consulting from Amarin, the American College of Cardiology, Amgen, CVS Caremark, and Merck for his work in the lipid field.

Supporting information

Appendix 1 ‐ Details of the CANTAB Tests

Appendix 2 ‐ Key Outcome Measures and Validation Studies for CANTAB Tests

Appendix 3 ‐ Members of the EBBINGHAUS Steering Committee

Giugliano RP, Mach F, Zavitz K, Kurtz C, Schneider J, Wang H, Keech A, Pedersen TR, Sabatine MS, Sever PS, Honarpour N, Wasserman SM, Ott BR and on behalf of the EBBINGHAUS Investigators . Design and rationale of the EBBINGHAUS trial: A phase 3, double‐blind, placebo‐controlled, multicenter study to assess the effect of evolocumab on cognitive function in patients with clinically evident cardiovascular disease and receiving statin background lipid‐lowering therapy—A cognitive study of patients enrolled in the FOURIER trial. Clin Cardiol, 2017;40(2):59–65.