menu

Designing Late Phase Clinical Trials of Disease-Modifying Agents to Delay or Prevent Dementia due to Alzheimer’s Disease

Green Park Collaborative project report and recommendations · March 2018

Download the original PDF (33 pages). This page shows the full text of the PDF. The figures, some tables and the original layout are in the PDF.

Designing Late Phase Clinical Trials of Disease- modifying Agents to Delay or Prevent Dementia due to Alzheimer’s Disease

Project Report and Recommendations

The Green Park Collaborative is a major initiative of the Center for Medical Technology Policy

March 2018

to Alzheimer’s Disease

TABLE OF CONTENTS

to Alzheimer’s Disease

WRITING TEAM

DONNA A. MESSNER, PHD President, Center for Medical Technology Policy Program Director, Green Park Collaborative PETER V. RABINS, MD, PHD Professor of Psychiatry and Behavioral Sciences, Johns Hopkins School of Medicine (Retired)

The authors wish to acknowledge the substantive input and guidance of the Working Group and Steering Group throughout the project. Substantive review and feedback on this document and the associated publication manuscript (for The Journal of Prevention of Alzheimer’s Disease) was provided especially by Derek van Amerongen, MD, MS; AnnCatherine Downing, PharmD; Norman Foster, MD; Alexander Keenan; Ellen Payzant, and Lon Schneider, MD.

Many thanks go to Peter Rabins, MD for taking on the role of Working Group Chair. His deep subject matter expertise was invaluable throughout the project, from planning workgroup discussions to providing guidance and specific input on the content of meetings and reports. He has been a collegial and much welcome presence on our GPC team.

This project was conceived in discussions with sponsors and stakeholders in the second half of 2016. It launched in late 2016, with steering and workgroup meetings taking place February through September 2017. Supplemental discussions with payers took place in October through November of 2017. The final product is intended to be a fair representation of overall group positions on study design. However, CMTP/GPC retains ultimate responsibility for the contents the document and uses editorial discretion where needed to balance conflicting information or judgements.

GREEN PARK COLLABORATIVE PROJECT TEAM

Donna A. Messner, PhD, Project Lead Jennifer Al Naber, MS, MSPH, Program Manager, Green Park Collaborative Julie Simmons, CMP, Marketing and Communications Manager Janelle King, Executive Administration and Project Support

PROJECT SUPPORT

A pre-competitive, collaborative group of companies working to develop new innovative therapies for Alzheimer’s disease participated substantively in the conduct of this project, and provided financial support. GPC wishes to extend its gratitude to Eli Lilly, Genentech, Janssen, and Pfizer for their helpful and meaningful engagement on this project.

ABOUT CMTP

The Center for Medical Technology Policy (CMTP) is an independent, non-profit dedicated to developing a health care system where patients, clinicians, health care policymakers, and payers have the evidence they need to make informed health decisions. We convene and collaborate with a national and international network of thought leaders, patients, patient advocates, clinicians, policymakers, and payers. Together, we support the next generation of clinical research. We do this by providing methodological guidance, shaping health policy solutions, and transforming clinical research.

to Alzheimer’s Disease

GREEN PARK COLLABORATIVE

The Center for Medical Technology Policy hosts the Green Park Collaborative (GPC), a multi-stakeholder forum that develops condition and technology-specific study design recommendations to guide the generation of evidence needed to inform both clinical and payment decisions in the United States. GPC- USA includes a diverse mix of payers, life sciences companies, patients, clinicians, researchers, regulators, and other stakeholders.

to Alzheimer’s Disease

EXECUTIVE SUMMARY

This document reports on a project conducted by the Center for Medical Technology Policy’s Green Park Collaborative to create multi-stakeholder vetted recommendations for the design of studies in later phases of drug development to evaluate the ability of disease-modifying agents to delay or prevent the onset of dementia due to Alzheimer’s disease (AD). AD is a neurodegenerative condition characterized in its symptomatic phase by progressive loss of memory and multiple other cognitive capacities. However, characteristic neuropathology is present for many years before the development of clinically identifiable Alzheimer’s dementia. As a result, AD is increasingly seen as a continuum that begins in cognitively normal people who experience abnormal deposition and buildup of amyloid plaques and intracellular tau in the brain.

Disease-modifying therapies (DMTs) are designed to target underlying mechanisms of disease development to interfere with or halt disease progression. Currently, the main underlying mechanisms targeted are those associated with abnormal protein deposition in the brain, as evidenced in the preclinical stages by abnormal brain pathology. So far, Alzheimer’s drug development has exhibited a high failure rate and the few currently FDA-approved therapies for AD do not affect the underlying course of the disease. Several lines of evidence support the hypothesis that these trials may have failed because disease-modifying therapies for AD need to be administered years before the onset of symptoms of cognitive and functional decline. Based on this hypothesis, significant research effort is going into drug development for “disease interception,” which targets underlying pathology in cognitively normal patients having elevated risk of developing dementia due to Alzheimer’s.

This “interception” approach to Alzheimer’s disease raises many questions and challenges for evidence development because clinical trials designed to demonstrate a meaningful effect on disease development would have to run for years (possibly a decade) to establish an impact on the clinical endpoints useful to health plans and meaningful to patients. How can trials be designed to obtain meaningful evidence more quickly, and what kinds of evidence should be gathered to inform the decision-makers (payers, providers, and patients/caregivers) who will assess the value of new drugs after they are reviewed by regulators?

These are the questions our workgroup deliberated on this project. The resulting recommendations are not intended to be a comprehensive guide to designing clinical trials; they are primarily intended to inform drug developers and researchers on the evidence preferences and concerns of health plan decision-makers, patients, caregivers, providers, and other decision-makers, while dovetailing these preferences with the stated needs of regulatory authorities. A multi-stakeholder steering committee and a working group were formed to guide this project and deliberate clinical trial design features important from multiple stakeholder perspectives (group membership is provided in Appendix A).

The field is moving rapidly, and as this report shows, guidelines and disease definitions have been changing even over the course of our workgroup discussions. Hence, this is not the final word on disease interception research, but an initial contribution to what must be a longer-term exploration of meeting the needs of stakeholders.

In summary, key recommendations are:

to Alzheimer’s Disease

POPULATION

  • Criteria for inclusion in disease interception studies should include the presence of an ADassociated marker.
  • Consider enrichment strategies to enroll patients at an increased risk of progressing to Alzheimer’s-type dementia.
  • Individuals with varied educational backgrounds, ethnic status, and occupational background, should be recruited for studies of disease-modifying agents to prevent or delay progression to dementia due to AD. To the extent possible, patients having typical comorbidities for the population studied should also be included.

OUTCOMES

  • Phase 3 trials should be designed so that variables that would be included in a Phase 4 (openlabel) follow-on study will be available for analysis. Direct and indirect cost of care, measures of behavioral and neuropsychiatric symptoms, and measures of adverse effects on caregivers are especially important for studies of at-risk, cognitively normal people who progress on a biomarker or to a clinical endpoint of MCI. When patients are to be followed to a diagnosis of Alzheimer’s type dementia, measures of daily function in addition to cognitive measures should be included.
  • Consider collecting behavioral and psychological outcome data in early symptomatic disease.
  • Consider using an adjudication panel that is unaware of treatment status (active or control) to ascertain the presence or absence of conversion to dementia or MCI.

METHODS

  • Measures of Activities of Daily Life (ADLs) for early stage detection of change should be updated to reflect declines in function at work and “digital independence,” i.e., decline in ability to use common electronic devices for everyday tasks.

SUPPLEMENTAL RECOMMENDATIONS

S1. When APOE or other genetic testing for Alzheimer’s predisposition is planned to be conducted with trial participants, patients should receive genetic counseling BOTH prior to and after genetic testing, and be well-informed of their right to withdraw from the study without receiving a disclosure of genotype. S2. When amyloid status is to be conveyed to cognitively normal individuals in preclinical trials for AD, best practices for participant education, preparation, and counseling should be followed. S3. All clinical researchers of dementia should be sensitive to ethical issues underlying dementia research, including guidelines articulated in the Nuffield Council on Bioethics’ report on ethical issues in dementia. S4. Extra care should be taken in clinical trials for AD to assure that patients and their families recognize the possibility of early trial termination

to Alzheimer’s Disease

PURPOSE

The purpose of this report is to provide recommendations for the design of late phase clinical trials to evaluate the ability of disease-modifying agents to delay or prevent the onset of dementia due to Alzheimer’s disease (AD). These recommendations are not intended to be a comprehensive guide to designing clinical trials, but rather emphasize clinical evidence preferences of health plan decisionmakers in a complex neurological disease that takes decades to play itself out. Ideally the trials should reflect the concerns and decision needs of patients, caregivers, clinicians, and other decision makers. We must do this even while we concede the unique scientific challenges of this growing area of clinical drug development.

METHODS

A multi-stakeholder steering committee and working group were formed to guide this project. The working group, which deliberated specific methodological questions on study design for the purposes of this project, was composed of methods and clinical experts, payers, regulators, drug developers, and patient advocates. The project team conducted a background literature review and worked closely with the workgroup chair, Dr. Peter Rabins, to plan the agenda and discussion topics for workgroup meetings. The steering group was also composed of representatives of industry, payers, providers, patient advocates, and regulators, many of them having more of a policy orientation than methodological expertise. Appendix A provides the lists of the members of these groups.

The project began with a joint kickoff webinar with the two groups on Feb 28, 2017. A series of 6 workgroup teleconferences then took place between March and August 2017, including a joint workgroup/steering group meeting on August 29. An in-person stakeholder meeting with these two groups and additional stakeholders was held on September 28, 2017 in Baltimore, MD. Teleconferences and meetings were transcribed and key conclusions were summarized from each meeting by the project team. These summaries were provided to participants in a shared electronic folder and points of consensus from previous calls were reviewed in subsequent calls.

Due to some schedule conflicts and a change in employment, some key health plan representatives were not able to participate in the later project meetings (in the summertime teleconferences and inperson meeting). To assure robust payer input into the project recommendations, the project team reached out to an external set of payers (two commercial, with significant interest in Medicare advantage plans, and one public payer) who reviewed draft materials, provided written comments, and participated in teleconferences with the project team to discuss their comments and concerns in more detail. Two (out of three) of these payers did not wish to talk “on the record” and are not formally acknowledged in this report. We are nevertheless very grateful for their feedback, which has been incorporated into these findings.

RATIONALE AND SCOPE

The focus of this document is on disease-modifying agents used in people who are considered cognitively normal but at elevated risk of progressing to symptomatic Alzheimer’s disease (AD). We focus on development of disease-modifying therapies (DMTs), defined in the context of AD as “an intervention that produces an enduring change in the clinical progression of AD by interfering in the underlying pathophysiological mechanisms of the disease process that lead to cell death.”1 This stands in contrast to the few FDA-approved therapies for AD currently available, which temporarily lessen disease symptoms for patients already exhibiting overt dementia, but which do not affect the underlying course of the disease.2

to Alzheimer’s Disease

Between 2002 and 2012, candidate drugs to treat Alzheimer’s disease had a 99.6 percent failure rate.3 Several lines of evidence support the hypothesis that these trails may have failed because diseasemodifying therapies for MCI due to Alzheimer’s and dementia due to AD will need to be administered years before the onset of symptoms (rather than after onset of symptoms, the approach which has traditionally dominated Alzheimer’s therapeutics). Evidence supporting this hypothesis is that the amyloid neuropathology that defines the disease is present for 15-20 years before disease onset;4–7 that individuals with Down syndrome who all develop neuropathology by their fifth decade do not develop dementia for many years thereafter;8 and the failure of anti-amyloid and anti-tau drugs to modify disease course when administered after the onset of clinically diagnosed MCI (or prodromal disease9) or clinically identified dementia due to AD. Some evidence also suggests a need to address several features of neuropathology, not only amyloid beta.10

For these reasons, researchers are increasingly focused on early intervention, particularly intervention before a diagnosis of dementia has been made, when the individual is still considered cognitively normal or mildly symptomatic. This interventional strategy is sometimes called “disease interception.”11 This early intervention strategy hypothesizes that drugs will be more effective when administered prior to widespread neuronal loss and onset of symptoms. There are persistent questions about how early is early enough, and whether amyloid targeting will prove to be an effective strategy.10 Nevertheless, this approach is increasingly in use and raises key study design challenges that were the focus of the workgroup:

 Identifying appropriate non-symptomatic candidates for study. A disease interception study for Alzheimer disease must identify cognitively normal people more likely than an unselected population to progress to MCI or dementia due to AD and target them for inclusion in the study. Population enrichment strategies are a widely accepted approach to increasing the number of potentially observable drug response events in a clinical trial population, allowing a smaller clinical trial than would otherwise be needed. The workgroup considered the advantages and disadvantages of various enrichment strategies.

 Reliably detecting early signs of change or progression. Since following cognitively normal people to a clinical diagnosis of Alzheimer’s-type dementia will take years, studies relying on clinical endpoints requiring long follow-up are expensive and difficult to sustain, although potentially highly informative. Endpoints detectable at an earlier stage of disease development are needed for disease interception to be a viable clinical development strategy. Predictive biomarkers of progression and treatment response are lacking, and current measures of cognition and function lack sensitivity for detection of very early change. Current studies should clarify, to the extent possible in the near future, which measures (or combinations of measures and other variables – i.e., predictive models) are both sensitive and specific in this context. Such measures must be suited for repeated administration in clinical trials, be useful in persons with a wide range of educational and occupational exposure, and free of bias by ethic/racial background.

 Considering patient-important and payer-important outcomes and design features. Assuming that measures of early cognitive and functional change suitable for prediction become available as outcome measures for disease interception trials, a central question then becomes whether certain types of measures may be more meaningful to patient and payer decision-makers than others.

to Alzheimer’s Disease

BACKGROUND

Dementia due to Alzheimer’s disease is a neurodegenerative condition characterized by progressive loss of memory and multiple other cognitive capacities.12 The two pathological hallmarks of the disease are the deposition of extracellular -amyloid protein leading to the formation of -amyloid (A) plaques and the deposition of intracellular misfolded tau protein leading to the formation of neurofibrillary tangles within neurons. Loss of synaptic function and selective neuronal death are other pathological hallmarks of Alzheimer’s disease, but are shared with other neurodegenerative diseases. Although deposition of A is thought to precede the formation of tau protein, some recent evidence supports the hypothesis the presence of amyloid predisposes to the spread of neurofibrillary tangles beyond the medial temporal lobe and that both are necessary for disease progression. Alzheimer’s disease encompasses a continuum beginning with these pathologic changes, through mild cognitive impairment due to Alzheimer’s disease, progressing to dementia due to Alzheimer’s disease. In guidelines issued in 2011, “Alzheimer’s disease” refers to this full continuum, while “dementia due to Alzheimer’s” or “Alzheimer’s dementia” refers to the stage of disease in which clinically manifest dementia is present.13 (As described in more detail below, new guidelines for investigating the disease continuum, including updated disease staging and definitions, are currently in draft form.14)

AD has been understood as having two different presentations with the same apparent underlying pathology. The first, highly common presentation is “late-onset” or “sporadic” AD, which is associated with symptom onset after age 65. Annual incidence ranges from .25% at age 65 to approximately 15% at age 95, while prevalence ranges from about 3% of people aged 65-74, 17% of people aged 75-84, and 38% of people 85 or older have Alzheimer’s dementia.15 This is a complex disease for which the most robust risk factor is advancing age. In addition, a single variant in the APOE gene, called APOE-4 (hereafter APOE4) and a range of non-genetic risk factors including less education, midlife hypertension, mid-life hearing loss, type 2 diabetes and others have been identified.16

The second, relatively rare manifestation (<5% of Alzheimer’s cases), is characterized by early disease onset (before the age of 65)17 and familial inheritance patterns. This form of AD is attributable to an autosomal dominant form of the disease caused by mutations in the presenilin 1 and presenilin 2 genes (PS1 and PS2), or the amyloid protein precursor (APP) gene. The pathological similarity between autosomal dominant and sporadic forms of AD has led to the hypothesis that autosomal dominant cases are likely to benefit from therapies demonstrated to have benefit in late-onset cases.18 Assessment of this hypothesis is beyond the scope of this report; the focus of the working group was on studies for the more common, sporadic AD.

Risk of progression to sporadic AD is thus assessed through the presence of preclinical Alzheimer’s disease pathology and the presence of a risk genetic variant. When progression to symptomatic disease takes places, the most common and prominent signs of early clinically manifest AD are impairments in verbal memory, learning (especially delayed recall), and executive function. Declines in other cognitive domains (multiple aspects of executive function, language, praxis (impairment of learned motor activity without evidence of paralysis or sensory loss) and visuo-spatial function become evident as the disease progresses. The rate of cognitive decline is variable across patients and takes place on a continuum that is difficult to measure precisely, particularly in the earliest stages of decline. Prior to the development of overt dementia, subtle cognitive changes occur that are more difficult to detect and difficult to differentiate from changes related to normal aging or other (non-Alzheimer’s) pathological causes.

to Alzheimer’s Disease

Diagnostic criteria have been developed for mild cognitive impairment (MCI) due to Alzheimer’s disease and for prodromal AD.19,20 Even so, while a diagnosis of MCI, in which memory impairment is the initial symptom (amnestic MCI or “aMCI”), has been used to identify people in early stages of Alzheimer’s dementia, not all individuals with aMCI progress to AD within 5 years and some patients with dementia due to AD present with non-memory symptoms such as impaired executive function, visuo-spatial perception, or language. The presence of both amyloid and tau biomarkers together increases the likelihood that aMCI will progress to dementia due to Alzheimer’s.21 In studies of people with autosomal dominant forms of AD, A deposition has been observed to begin an estimated 20 to 30 years before the first signs of clinically manifest AD become apparent.22 However, cognitively normal people who lack the genetically dominant form of AD and exhibit A deposition do not necessarily develop symptomatic AD on follow-up. It is unclear whether this represents resistance or brain resiliency to the pathology because of the long-time course of the illness and high mortality from other illnesses in this age group. Some studies have suggested that an experience of worsening or more frequent confusion or memory loss connected with routine activities or knowledge (“subjective cognitive decline”) in otherwise cognitively normal people may be a sign of individuals at high risk of developing MCI or Alzheimer’s dementia.23–26

Since both prediction and early detection of Alzheimer’s is problematic, drug studies of preclinical disease are extremely difficult. The 2011 National Institute of Aging (NIA) guidelines noted above characterize Alzheimer’s disease mainly through its symptomatology as well as through pathology, but symptomatology may be of little use for preclinical studies. An NIA/Alzheimer’s Association (AA) workgroup has proposed a research framework characterizing preclinical dementia in terms of underlying pathologies.14 This approach groups cognitively unimpaired populations according to whether they are “normal” or “abnormal” across three biomarker-based characteristics: amyloid deposition (A), tauopathy (T), and neurodegeneration or neuronal injury (N). In this framework, “Alzheimer’s disease” refers to the presence of plaque and tangle pathophysiologic processes as evidenced by “abnormal” findings for both amyloid and tau in vivo. Combinations of these biomarker test results are mapped onto the previously used cognition-based staging terminology of “unimpaired,” “MCI,” and “dementia.” Importantly, this “ATN” framework is intended for research use, not current clinical practice, and is still (as of this writing) a draft proposal. More research is needed to illuminate the implications of biomarker changes in individuals over time.27 Even so, the framework has advantages for researchers evaluating therapies for disease interception; it enables a more precise approach to therapeutic intervention trials where specific pathways can be targeted at specific points in the disease process.

BIOMARKERS FOR DIAGNOSIS AND RESEARCH

As noted, given the difficulties of early diagnosis, and of identifying people who are good potential candidates for participation in disease interception trials, numerous biomarkers are under development for the identification and tracking of underlying disease processes. In trials of disease interception, biomarkers may be used to diagnose, assess risk, stage, evaluate response to treatment, or provide more definitive patient selection.

Relevant biomarkers span three primary categories of techniques: brain imaging, cerebrospinal fluid (CSF) analysis, and blood-based analysis. In terms of the information they provide relative to AD pathophysiology, these techniques are perhaps more usefully characterized as:20

to Alzheimer’s Disease

  • Biomarkers of -amyloid deposition
  • CSF (cerebrospinal fluid) A42 and A42 /tau ratio
  • PET (positron emission tomography) amyloid imaging
  • Biomarkers of neurofibrillary tangle tau accumulation
  • CSF phosphorylated tau
  • PET tau imaging
  • Biomarkers combining -amyloid and tau measures
  • A42 /tau ratio
  • Biomarkers of neuronal injury
  • CSF tau
  • Hippocampal volume or medial temporal atrophy by volumetric measures or visual rating
  • Rate of brain atrophy
  • FDG-PET imaging (flurodeoxyglucose PET)
  • SPECT perfusion imaging (single photo emission tomography)
  • Markers of biochemical change
  • Inflammatory biomarkers (cytokines)
  • Oxidative stress (isoprostanes)
  • Other markers of synaptic damage and neurodegeneration, such as cell death

Less well validated biomarkers of neuronal injury include functional MRI (fMRI) activation studies, resting BOLD (blood oxygen level-dependent) functional connectivity, MRI perfusion, MR spectroscopy, diffusion tensor imaging, and voxel-based and multivariate measures.20

Positron emission tomography (PET) imaging with amyloid-beta tracers (amyloid PET) has been shown to image the deposition of amyloid proteins consistent with autopsy measures of amyloid deposition.28 Since amyloid buildup has been shown to begin well before clinical signs of dementia become apparent,29 the amyloid PET has become a focus of attention as a diagnostic tool. Amyloid PET is FDAapproved for imaging of -amyloid neuritic plaque density in brains of adults with cognitive impairment being evaluated for AD and other causes of cognitive decline. It is not indicated for predicting development of AD or monitoring responses to therapies, although for research purposes it is currently being studied for identifying preclinical AD and evaluating pharmacodynamic response to amyloidtargeted therapies. Proposed appropriate use criteria focus on patients with objective cognitive impairment and uncertain diagnosis after comprehensive evaluation by a dementia specialist, where knowledge of amyloid status is expected to alter clinical management.30 A negative scan can rule out the presence of amyloid and therefore rule out a diagnosis of AD. A positive scan is consistent with moderate to frequent amyloid plaques in brain, but it cannot definitively establish whether AD is the cause of symptoms (because amyloid can be a co-pathology with other causes of cognitive impairment and because it has reduced value for predicting AD as a cause of symptoms beyond the age of 70).22,31

Amyloid PET is not in routine diagnostic use. This is due in part to lack of a clear reimbursement pathway. The Centers for Medicare and Medicaid Services (CMS) is currently allowing for Medicare coverage of scanning only in the contexts of clinical trials to evaluate the clinical usefulness of amyloid PET given the absence of a disease-modifying AD treatment, the high cost of scanning, and the variable availability of the technique across communities in the U.S.32

to Alzheimer’s Disease

PET tracers that demonstrate the presence of tau protein in the brain have also been more recently developed as a marker of neurofibrillary tangle burden. Tau PET correlates with clinical symptoms, disease stage, and AD clinical progression.14

Cerebrospinal fluid (CSF) markers for A42 and tau proteins have been shown to correlate with the presence and progression of dementia due to AD. (A42, an isoform of the amyloid protein containing 42 amino acids, is the dominant form of amyloid found in plaque depositions. CSF A42 levels are reduced in AD, presumably due to sequestration in amyloid plaques. Tau and p-tau levels are elevated in AD.) Comparisons between these measures and imaging measures are underway. However, some observers have expressed concern that the invasive nature of CSF testing may be a barrier to patient recruitment in clinical studies.33 Blood-based markers would be a preferable alternative for many patients, but these currently do not perform as well as imaging or CSF markers for characterization of patients with emerging disease pathology or symptoms.

PET scanning with fluorodeoxyglucose tracers (FDG-PET) provides a view of glucose metabolism and hence cellular function in the brain. Specific patterns of hypometabolism are associated with different forms of dementia. For example, evidence from the Alzheimer Disease Neuroimaging Initiative (ADNI) Study demonstrated that occipital hypometabolism accurately predicts concurrent Lewy body pathology.34 FDG-PET hypometabolism also was a more sensitive measure of change than the ADAS-Cog, a standard cognitive measure used in clinical trials.35 The best specificity was found with a combination of the Trail Making Test-B and FDG PET.36 For more clinically apparent AD, the UK’s National Institute for Health and Care Excellence (NICE) recently published draft guidelines for comment suggesting that FDG- PET is one biomarker (among several alternatives) that may be used with cognitive testing to help establish a diagnosis of Alzheimer’s-type dementia.37 FDG-PET may also be useful as a biomarker for disease progression and useful in demonstrating slowing or arresting of dementia due to AD in clinical trials.

Functional MRI also assesses alterations in regional cerebral blood flow and might have advantages over FDG-PET in monitoring synaptic and cellular function as markers of slowing or stopping of disease progression. Rates of whole brain and hippocampal atrophy can be calculated from volumetric MRI scans performed serially (and these repeated scans can be done without the radiation exposure for patients associated with PET scanning).38,39 While hippocampal atrophy can also be associated with frontotemporal dementia and other disorders, these biomarkers can be useful with other information to help characterize preclinical AD, and might also be used as evidence of the slowing of neurodegeneration. In carriers of early onset AD genes, functional and structural MRI abnormalities were evident more than two decades in advance of cognitive decline, and well in advance of amyloid deposition as measured by PET or CSF biomarkers.5,40 MRI may also be more available than amyloid PET scanning in non-research centers.

When well developed, these markers could not only provide insight into underlying pathophysiology of the disease, but also function as inclusion and exclusion criteria for clinical disease interception trials and as surrogate outcomes predictive of clinical benefit in trials. Currently, none of the available biomarkers are used in routine clinical practice in the U.S., in part because of the complexity of the procedures, variable availability, and because data on positive predictive value and negative predictive value of many markers are still being developed. This information is likely to be available within several years. Because these tests (particularly the scans) are associated with significant expense, payers have concerns regarding how they might ultimately be used. With appropriate validation, they would prefer to see such techniques employed not for population screening, but rather for more targeted use in

to Alzheimer’s Disease

diagnosis or treatment. In the context of Alzheimer’s prevention clinical trials, however, these biomarkers are essential tools.

IDENTIFYING PRE-CLINICAL CANDIDATES FOR TRIAL PARTICIPATION

No formal criteria have been developed for identifying pre-clinical individuals. For research purposes, therefore, a model is needed of how individuals progress from a cognitively non-impaired state to the earliest stages of detectable Alzheimer’s dementia.

Current researchers in this field have informally adopted the idea that candidates for assessment of preclinical AD can be staged as shown in Box 1.41

In this scheme, prospective research subjects are staged after establishing normal cognition through clinical evaluation i.e., an MMSE (mini-mental state examination) score of 25 – 30, Logical Memory II (Weschler Memory Scale) of 6 – 18, and CDR (Clinical Dementia Rating - Sum of Boxes) of 0. If the Stage 1 biomarkers (amyloid PET or amyloid- CSF test) are positive, then Stage 2 biomarkers are assessed (FDG-PET, tau, MRI). By Stage 3, subtle cognitive changes are evident as compared to baseline, but the individual still does not meet the diagnostic criteria for MCI.

As noted earlier, a new biomarker-based draft framework has been proposed for clinical research in Alzheimer’s disease.14 In that framework, pre-clinical AD is defined, at a minimum, by intact cognitive function and presence of amyloid pathology in the brain. The hypothetical disease framework posits that progression is associated with increasing neurofibrillary tangle formation, neurodegeneration, and clinical symptoms. The overall mapping of this biomarker scheme onto traditional cognitive staging of Alzheimer’s disease is shown in Table 1. Box 1. Preclinical Staging for Alzheimer’s (Research Only)

STAGE 1 – Asymptomatic amyloidosis

 High PET amyloid tracer retention  Low CSF A1-42

STAGE 2 – Amyloidosis + Neurodegeneration

 Neuronal dysfunction on FDG-PET/fMRI  High CSF-tau/p-tau  Cortical thinning/hippocampal atrophy on sMRI

STAGE 3 – Amyloidosis + Neurodegeneration + Subtle Cognitive Decline

 Evidence of subtle change from baseline level of cognition  Poor performance on more challenging cognitive tests  Does not yet meet criteria for MCI

to Alzheimer’s Disease

Table 1. Framework for Biomarker Mapping with Clinical Staging of Alzheimer's Disease14

[This table or figure is not reproduced here. See page 14 of the PDF.]

A- denotes normal amyloid as determined by CSF Ab 42 or 42/40 ratio, or amyloid PET T+ denotes an abnormal tau finding: aggregated tau (neurofibrillary tangles) or associated pathologic state by CSF (phosphorylated tau) or tau PET T- denotes normal tau as determined by CSF (phosphorylated tau) or tau PET N+ denotes an abnormal neuronal finding: evidence of neurodegeneration or neuronal injury as determined by anatomic MRI, FDG PET, or CSF total tau N- denotes a normal neuronal finding: no evidence of neurodegeneration or neuronal injury as determined by anatomic MRI, FDG PET, or CSF total tau

While the identification of early pathology can serve as a useful screening tool to enrich clinical studies with individuals more likely to progress to MCI and Alzheimer’s-type dementia, use of these markers as surrogate outcomes (as in, for example, reductions in depositions of amyloid-beta or stabilization of hippocampal volume changes) has not yet been established (i.e., the markers cannot yet be used to reliably anticipate clinical disease development). Therefore, incorporation of candidate predictive biomarkers in longitudinal studies and successful therapeutic trials are critical in building the evidence base for these biomarkers.

REGULATORY CONTEXT

While the focus of this document is decision needs of “post-regulatory” decision-makers (payers, providers, patients, caregivers), these considerations must be harmonized as appropriate with regulatory requirements. We will here summarize relevant guidance provided by two of the most salient regulatory authorities, the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA).

to Alzheimer’s Disease

Just as the current project and report was being finalized, the FDA issued revised draft Guidance for Industry on drug development for early Alzheimer’s disease.42 The new draft guidance creates four categories to characterize individuals who may participate in AD clinical trials:

 Stage 1: Patients with characteristic pathophysiologic changes of AD but no evidence of clinical impact  Stage 2: Patients with characteristic pathophysiologic changes of AD and subtle detectable abnormalities on sensitive neuropsychological measures, but no functional impairment  Stage 3: Patients with characteristic pathophysiologic changes of AD, subtle or more apparent detectable abnormalities on sensitive neuropsychological measures, and mild but detectable functional impairment  Stage 4: Patients with overt dementia

According to the FDA, the ideal outcome measure to demonstrate therapeutic effect in Stage 3 patients would provide an assessment of meaningful cognitive function. An integrated scale that meaningfully assesses both daily function and cognitive effects in these patients would be acceptable as a single primary efficacy endpoint.

For trials of Stage 2 patients, it may be difficult to establish a clinically meaningful effect on the basis of subtle cognitive deficits in a trial of “reasonable” duration. However, if the trial is designed to be of sufficient duration for patients to transition from Stage 2 to Stage 3, then Stage 3 endpoints would be acceptable. Alternatively, the FDA will consider “strongly justified arguments” of a persuasive effect on neuropsychological performance (preferably demonstrated across multiple individual tests). (See the Guidance for additional detail on contextual factors that help to make a convincing argument.) In Stage 1 patients, a clinically meaningful benefit cannot be measured (in a trial of reasonable duration) because there is no clinical impairment to assess. In this case, an effect on characteristic pathophysiologic changes of AD, as demonstrated by an effect on a biomarker, may be the basis for an accelerated approval with post-approval study requirements to confirm predicted clinical benefit. A pattern of treatment effects seen across multiple individual biomarker measures would strengthen the persuasiveness of the result. At present, insufficient evidence exists that an effect on any biomarker is reasonably likely to predict clinical benefit, but the FDA urges the research community to continue work towards this goal.

The FDA additionally noted that a time-to-event survival analysis would be an acceptable primary efficacy measure, and that a randomized start or randomized withdrawal trial design would be the most convincing approach to demonstrate a persistent effect on disease course.

Moving from the recent FDA guidance to draft guidance from the EMA issued in 2016,43 for diseasemodifying agents, the EMA guidance states that:

  • A placebo control is needed.
  • Any effect on preclinical/prodromal AD cannot be assumed to generalize to other disease stages. Separate trials would be needed, or longer-term follow up to establish efficacy in later stages of AD.
  • A statistical comparison of rates of change in clinical symptoms over time (slope analysis) between treatment and control is a generally acceptable approach, but (similar to FDA) cautions that the linearity of each curve cannot be assumed. The EMA suggests that to overcome this

to Alzheimer’s Disease

possible deficiency, measurements should be taken at regular intervals in the parallel group design. Delayed-start or -withdrawal design would enhance this approach. The guidance also notes that the length of follow-up is a critical variable to see if or when the response curves come together.

  • Alternatively, in earlier disease especially, EMA suggests a time-to-event analysis on a clinically important measure (e.g., time to diagnosis of Alzheimer’s dementia) may be used.
  • Evidence of delay in rate of decline should be accompanied by evidence of delay in progression of brain neurodegeneration through biomarkers.

For pharmacologic AD prevention studies, the EMA also notes that:

  • Placebo controlled trials should be done in enriched populations.
  • A preclinical disease model still needs to be validated (time course of disease and adaptations to brain changes vary individually).
  • From the EMA perspective, the main goal of preclinical studies is prevention of cognitive impairment, since no biomarker is yet considered a valid surrogate.
  • Prevention trials require large samples and long follow-up, at least 5 years typically. No firm recommendations are made on duration needed, as this will vary with the trial.

RECOMMENDATIONS

The following recommendations are based on the discussions of the working group, and steering group, and multi-stakeholder deliberation in the in-person meeting. As noted, these are not intended to be a generalized guide to doing clinical trials for disease-modifying agents in AD, but rather to home in on design issues of importance to key stakeholders. The recommendations seek to do this with sensitivity to the developing science and regulatory requirements, and in dialog with other sets of related recommendations (as noted in both the Recommendations and Discussions sections below).

POPULATION

  • Criteria for inclusion in disease interception studies should include the presence of an ADassociated marker.

Rationale: Based on the current understanding of the disease, the presence of amyloid plaques in the brain is considered to be a defining feature of preclinical AD. Hence, at a minimum, a positive amyloid- scan should be required for enrollment in disease intervention studies. Elevated CSF phosphorylated tau or increased signal on tau PET may provide further information on AD specific pathological staging.

Implementation: Many current disease intervention studies require a positive amyloid- scan or low CSF A42. If other biomarkers are developed that can serve as surrogate measures of amyloid positivity, these criteria may be modified. (Note that this recommendation is a strategy for selecting study populations, not a recommendation for the use of any particular therapeutic strategy or outcome measure.)

  • Consider enrichment strategies to enroll patients at an increased risk of progressing to Alzheimer’s-type dementia.

Rationale: Enrichment (recruiting people more likely than the general population to progress to dementia) is widely used in preclinical studies of Alzheimer’s-type dementia and MCI because the

to Alzheimer’s Disease

phenomenon researchers are attempting to observerelatively younger, cognitively normal people developing early signs of dementia within a few years of study initiationis rare. Thus, designing and successfully implementing efficacy trials that provide enough cases for conversion from at-risk to disorder (e.g., from aMCI to AD dementia) using an unenriched population would be very difficult to conduct from cost, recruitment, and retention perspectives.

In addition, it would be ethically questionable to expose healthy individuals who are never likely to develop aMCI or AD dementia to treatments with potential adverse effects.

Implementation: We have recommended (above) that early abnormal accumulation of amyloid (as detected by amyloid PET scanning or CSF testing) should be used to select candidates for clinical trial participation among cognitively normal individuals. While selection of an “A+” population (a cognitively normal population with abnormal amyloid accumulation; see Table 1) assures that all patients who later exhibit signs of neurological, cognitive, or functional deterioration fit an Alzheimer’s-type pattern of progression, not all A+ individuals will progress to MCI or dementia. If prevention or delay of progression to dementia is the ultimate goal of the investigational treatments, then until well validated predictive tools are available, additional strategies must be used to assure the clinical trial population is composed of people having an elevated risk of progressing.

The prevalence of amyloid positivity increases with age and presence of APOE e4 alleles.22 To reduce the number of asymptomatic people who need to be screened for amyloid, prescreening criteria based on age, family history, and APOE can be established. For example, a preclinical trial could recruit people age 65 and older for amyloid screening, but also extend to people aged 60-65 with the presence of an APOE e4 allele. Alternatively, among people having amyloid deposition, further enrichment can be considered, e.g., evidence of increased tau or neurodegeneration. The potential benefits and limitations with regards to screen failure rate, generalizability, and current knowledge of progression rates in subgroups need to be considered.

While enrichment strategies are clearly necessary for studying very early development of this condition, some strategies have the potential disadvantage of limiting the generalizability of a positive study. For example, the TOMORROW study (which was recently discontinued, but nevertheless illustrates the point) used genotyping of the APOE and the TOMM40 genes to assess the efficacy of low doses of pioglitazone (a PPAR gamma agonist and an anti-diabetes drug). By considering age together with these two genotypes, one can identify a group of people not only highly likely to develop Alzheimer's pathology and dementia, but a group that is particularly likely to respond to pioglitazone because of its mechanism of action. Hence, the results of this study will only be relevant to people having that genotype.

More generically, the use of APOE-e4 status as a biomarker for inclusion has a number of advantages. Its presence heightens the likelihood of earlier conversion to AD dementia. The neuropathology of progression appears the same regardless of genotype, leading to the hypothesis that drugs effective in people having the APOE-e4 genotype will also be effective in people with other genotypes. Moreover, a recent National Academies of Medicine report favors enrichment by APOE-4 genotyping because of a current secular trend in the U.S. towards a decline in dementia incidence hypothesized as related to a parallel decrease in cardiovascular morbidity and mortality due to better prevention and treatment.50 The rationale is that an APOE-4 selected group would be less subject to this trend. Even so, there are some potential disadvantages. Only 20% of the population is heterozygous for APOE-e4 and <5% is homozygous. This is a relatively small proportion of the overall population that might be treated with a

to Alzheimer’s Disease

successful therapy, and while a reasoned pathophysiological rationale exists to say that differences in APOE genotype will not translate into differences in drug effectiveness, conclusive evidence is not yet available. Note also that APOE testing is seen by some observers as problematic (see Recommendation

  • and may create an impediment to clinical trial recruiting in some countries.

In any event, for studies of certain disease-modifying agents, it may be more important to enrich on the basis of -amyloid pathology or other factors. As noted above, APOE may be a useful tool for screening prior to -amyloid testing, to reduce the number of participants who must be scanned or lumbar punctured.

Finally, a recent report from the Lancet identifies a series of modifiable risk factors for AD each of which is linked to a stage of life in which intervention for the risk factor may be most effective;16 i.e., early life education; mid-life hearing loss, obesity, and hypertension; and late-life smoking, depression, physical inactivity, social isolation, and diabetes. Studies using these types of behavioral risk factors as part of entry criteria or baseline data collection should be attentive to both the factor and the period of life relevant to that factor.

  • Individuals with varied educational backgrounds, ethnic status, and occupational background, should be recruited for studies of disease-modifying agents to prevent or delay progression to dementia due to AD. To the extent possible, patients having typical comorbidities for the population studied should also be included.

Rationale: Older African Americans are approximately twice as likely to have Alzheimer’s (or other dementias) as older whites in the U.S., and Hispanics are approximately 1.5 times as likely to have dementia as older whites.12 Higher risk for Alzheimer’s is also associated with fewer years of formal education, which in turn could implicate occupations that are less mentally challenging and lower socioeconomic status. These and other associations point to complex and mutually implicating risk factors for Alzheimer’s, some of which, such as midlife hypertension, are modifiable. However, racial minorities and individuals with less education are generally underrepresented in clinical trials. These trends tend to bias clinical trial results and subsequent treatment guidelines towards interventions that benefit white, educated subjects (and healthier ones who suffer fewer comorbidities), with less information available on any differential benefit of these subgroups. Better diversification of trial populations would therefore enhance generalizability of study results to affected populations. In addition, in the case of Alzheimer’s clinical trials, the white, educated individuals more likely to volunteer for trials are a lower-risk group for progression to Alzheimer’s type dementia. Prevention trials in this lower-risk group must increase the numbers of subjects screened and enrolled for trials, and must rely more heavily on biomarker-based enrichment strategies, to increase opportunities to observe prevention or delay of progression to Alzheimer’s-type dementia.

Implementation: A National Academies report on Preventing Cognitive Decline and Dementia noted that community-based participatory approaches to recruitment have had some success in enrolling minority and underrepresented populations.44 However, examples offered by this report and associated citations are for non-drug intervention trials, such as the Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) trial45 or the Mental Activity and eXercise (MAX) trial,46 or for observational studies. Researchers implementing interventional drug trials may experience greater barriers, particular when attempting to recruit asymptomatic subjects.47 These challenges of underrepresentation of minorities in clinical trials are experienced across all clinical research as well as in studies of dementia.48 Some success has been achieved in building registries of healthy volunteers from diverse communities

to Alzheimer’s Disease

willing to participate in trials, however sustained engagement is required for meaningful levels of recruitment to pharmacologic prevention trials.49

OUTCOMES

  • Phase 3 trials should be designed so that variables that would be included in a Phase 4 (openlabel) follow-on study will be available for analysis. Direct and indirect cost of care, measures of behavioral and neuropsychiatric symptoms, and measures of adverse effects on caregivers are especially important for studies of at-risk, cognitively normal people who progress on a biomarker or to a clinical endpoint of MCI.

Rationale: Variables such as the incidence and severity of behavioral and neuropsychiatric symptoms, caregiver distress, participant quality of life, health care utilization and patient medical costs might be used to demonstrate benefit or impacts of therapy in phase 4 trials in addition to delay of cognitive and functional decline. Evidence suggests that both direct medical costs and indirect costs (e.g., need for more supervision by family members and others) are higher for people at the MCI stage of diagnosis, and some costs are measurably elevated even for people who are still testing as cognitively normal a year before a diagnosis of MCI occurs (see Discussion below).51–53 This increase is also correlated with a reduction in paid and volunteer work for these individuals.54

Implementation: Collecting baseline information on a limited set of variables that would be used in Phase 4 extension studies would be efficient and offer the possibility of demonstrating benefits of active treatment in addition to delay, modulation, or prevention of cognitive and functional decline.

  • Consider collecting behavioral and psychological outcome data in early symptomatic disease.

There is a dearth of behavioral and psychological outcome data in early symptomatic disease, even though behavioral disturbances, increased anxiety and irritability are prevalent in Alzheimer's-type dementia.55 Recent data suggests that these behaviors are seen in earlier stages of disease development, and may represent an important domain for capturing earlier changes.56 Subtle behavioral changes may be an especially useful early indicator of emergent disease, since individuals can consciously practice some cognitive and functional test skills for better evaluation results; these behavioral and psychological test measures are less modifiable through practice.

Implementation: Collection of behavioral and psychological data on clinical trials may represent added patient burden on trials, since trial participants may also be taking cognitive and functional tests. Yet, as noted, for trials requiring multiple testing sessions over time, as noted, behavioral testing has the advantage of being less modifiable through practice.

  • Consider using an adjudication panel that is unaware of participant treatment status (active or control) to ascertain the presence or absence of conversion to dementia or MCI.

Rationale: The validity of the earliest evidence of symptomatic dementia across individual patients is unclear. Use of a panel of experts having information on the participant’s baseline condition can allow individualization in the ascertainment of functional decline (for an adjudication of dementia) or memory, executive function or other cognitive domain impairments as an indicator of MCI or AD. Study partners (spouses or family members) are often helpful in detecting early change and providing baseline information. Earliest signs of change may be noted (but not be acknowledged) by the subject

to Alzheimer’s Disease

themselves or coworkers. As noted below, some currently used functional measures may benefit from updates.

Implementation: While an effective method to detect early change in participants indicative of disease progression, adjudication of progression by study panel may be infeasible in very large trials.

METHODS

  • Measures of Activities of Daily Life (ADLs) for early stage detection of change should be updated to reflect declines in function at work and “digital independence,” i.e., decline in ability to use common electronic devices for everyday tasks

Rationale: Impairment of the ability to perform high level activities of daily living is characteristic of the onset of dementia. These are key indicators of a decline in the ability to be employed and independent, of an increased need for caregiver assistance, and, ultimately, of placement in long term care. Assessment of these capacities, called Instrumental Activities of Daily Living (IADL) is therefore used in part to establish a diagnosis of dementia, and can also be used as outcomes in pharmacologic trials. ADLs categorized as instrumental (IADL), entail higher order skills such as managing finances, and basic (ADL) skills for self-care such as dressing, feeding oneself, and bathing. Because a decline in basic ADL function occurs later in dementia, basic ADLs are less relevant to studies that seek to delay or prevent disease onset. Therefore, activities that become impaired earlier in the disease process are more sensitive indicators of disease onset and impairment.

While most of the basic skills questions posed to caregivers or a study partner on these assessments continue to be relevant and valid (“ate without physical help and used a knife”), most IADL questions were developed prior to the rapid expansion of smartphones, tablets, email, and the Internet. Yet for detection of early changes in function, these devices may provide opportunities for more fine-grained assessment. For example, an individual may be able to use a smartphone for calling people already listed as a contact, but may have lost the ability to add new contacts. Additionally, many contemporary IADLs, such as shopping, money management, and leisure activities, are commonly done with electronic devices. Hence, these heretofore separately assessed activities require electronic literacy and skills not currently taken into account. Measurement of ability to use electronic devices in these contexts may allow for earlier detection and possibly greater sensitivity to early changes. In addition, these types of functional deficits may be perceived by affected individuals and family members as more relevant than, e.g., the ability to write a check.

Another set of complex functional changes that may be detected early in some patients is the ability to work effectively. Depending on job function, some patients may demonstrate subtle losses of competence that are noticeable to long-time coworkers.

Implementation: Research is needed to develop sensitive and specific measures of functional change associated with electronic devices and of changes in function at work and to incorporate them into existing validated tools that assess changes in IADL function.

SUPPLEMENTAL RECOMMENDATIONS

to Alzheimer’s Disease

The following recommendations, while not specifically methodological, were strongly voiced by members of the group, and thus are included here as “supplemental recommendations” related to ethical and sensitive conduct of research in this area.

S1. When APOE or other genetic testing for Alzheimer’s predisposition is planned to be conducted with trial participants, patients should receive genetic counseling BOTH prior to and after genetic testing, and be well-informed of their right to withdraw from the study without receiving a disclosure of genotype.

Rationale: Research has shown that well-counseled patients generally do well with APOE disclosures when they opt to receive them.57 However, such studies counsel patients closely before testing, have been carried out with white, relatively affluent populations, and see a significant level of drop-out before the disclosure step (see the recruitment flow chart provided in ref.57 ). Less work has been done on patient reactions to APOE disclosures when adequate advance counseling is not provided, but such disclosures can be emotionally harmful.58 Few if any studies have been published on the effects of APOE disclosure in less advantaged populations.

Implementation: Informed consent materials should include the requirement to be counseled on the significance and interpretation of genetic information for late onset Alzheimer’s disease and make clear that the patient may elect to withdraw from the study after receiving genetic counseling and anytime up to receiving a disclosure. Patient rights under the Genetic Information Nondiscrimination Act (GINA) should be reviewed, with clarity on ways the Act does, and does not, protect the patient. Additional genetic counseling should take place following disclosure to focus more fully on the interpretation and implications of the patient’s specific genotype. Such counseling should be offered even if patients have low-risk genotypes, since individuals may misconceive the meaning of “low-risk.”

S2. When amyloid status is to be conveyed to cognitively normal individuals in preclinical trials for AD, best practices for participant education, preparation, and counseling should be followed.

Rationale: Just as for genetic disclosures discussed in Recommendation S1, amyloid disclosures to cognitively normal individuals may result in anxiety or depression, and may be associated with significant misunderstandings regarding the interpretation and significance of the information received. Researchers must therefore assure that participants are well informed before consenting to the scan, are appropriately counseled after the scan, and that follow-up and additional counseling resources are available, if needed.

Implementation: Researchers developed an approach to amyloid disclosures for the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease (A4) Study that can be adapted to other preclinical AD studies.59 The approach includes an educational process built into informed consent in which participants learn what is known and unknown about amyloid imaging and how to interpret results; screening for anxiety and depression to assess the participant’s readiness for amyloid imaging results; checks for comprehension of the information provided and for signs of distress; after testing, in-person disclosure with assessment of mood and willingness to receive results; and procedures for follow-up. Care should be taken to adapt these procedures for amyloid testing by PET scanning or lumbar puncture. Researchers should consider whether other biomarker scans require similar procedures.

to Alzheimer’s Disease

S3. All clinical researchers of dementia should be sensitive to ethical issues underlying dementia research, including guidelines articulated in the Nuffield Council on Bioethics’ report on ethical issues in dementia.

Rationale: While all clinical trials should be conducted in a respectful and sensitive manner, the emotional impact of cognitive impairment is great for both study participants and care partners. Retaining participants is especially challenging when trials are expected to last for many months, even years. Added to this is the burden of care partners who must accompany and monitor the participant. The Nuffield Council on Bioethics report provides guidance on providing for emotional support.60,61 Study investigators should also coordinate the clinical care of trial participants with health care providers.

Implementation: Study investigators take practical steps study protocols consistent with the guidelines in the Nuffield report to provide patient and family support. Investigators should identify and provide evidence of continuing communications with the health care providers of each participant.

S4. Extra care should be taken in clinical trials for AD to assure that patients and their families recognize the possibility of early trial termination.

Rationale: While these disclosures are required for all clinical trials, patient and caregiver advocates participating in this project felt that greater awareness was needed of the hopeful expectations that often accompany interventional clinical trials for Alzheimer’s disease, despite conscious knowledge that the trial may not be positive. When trials are terminated, families can be caught by surprise and devastated by the news. This stark disappointment is often compounded by having the news delivered through an impersonal letter arriving “out of the blue.”

Implementation: Clinical trial planners should use extra sensitivity in considering how trial information will be conveyed to patients and families. The informed consent process will include clear information on the role of the data safety monitoring board and the possibility that the trial could be discontinued before it is complete. This information should be reinforced through discussion at the time of enrollment. The participant should also be told what kinds of communications to expect should the trial be terminated early. While it may be difficult for large trials, if possible, participants should be notified of trial termination through direct conversations (e.g., phone calls or other direct interactions through the personal physician’s staff). As for all clinical trials, patients should be provided with a contact number for more information and opportunities to ask questions.

DISCUSSION

DRUG DEVELOPMENT STRATEGIES

Candidate predictive biomarkers should be (and usually are) included in clinical trials of diseasemodifying agents for disease interception. Developing a greater understanding of the predictive utility of biomarkers and their relationship to clinical endpoints is critical to reducing clinical trial size, making coverage decisions and individual treatment recommendations. It should be recognized, however, that while the focus of this document has been primarily on phase 3 drug trials, outcome and epidemiologic researchers in settings outside of commercial drug development also have a role to play. Clinical drug trials alone are necessary but not sufficient to solve the riddle of preventing Alzheimer’s disease.

to Alzheimer’s Disease

CONSISTENCY IN OUTCOMES USED IN PREVENTION TRIALS

Recent reviews of early symptomatic and prevention trials of AD have shown broad inconsistency in the clinical outcome measures used to measure cognitive and functional declines associated with AD. These inconsistencies complicate comparisons between drug trials. As trials multiply and new therapies become available, this will make it more difficult for patients, caregivers, clinicians, and payers to make informed decisions on covered benefits and appropriate treatment options. For this reason, the National Academies of Medicine called for greater consistency in the selection of outcomes in AD studies and a 2012 Institute for Clinical and Economic Review (ICER) report called for a common core set of outcomes for development of diagnostic tests for Alzheimer’s.64 Similarly, the 2016 Alzheimer’s Disease-Related Dementias Summit recommended harmonizing test batteries (but avoided mandating a specific test).65 Also of note: a recent AHRQ systematic review suggested that the use of the Alzheimer’s disease Assessment Scale – Cognitive Subscale (ADAS-Cog) has improved consistency of methodology across studies.66

As these reports imply, comparative effectiveness across trials would be promoted by the development of a core outcome set. According to the COMET (Core Outcome Measures in Effectiveness Trials) Initiative, a core set is a minimum set of outcomes recommended to be collected in all clinical trials for a given condition.67 Core sets do not limit the number or type of (additional) outcomes that could be collected in a trial, but are intended to establish a baseline of data that should be collected consistently across trials.

One group has recently published recommendations for a core outcome set for assessment of interventions in mild to moderate dementia due to Alzheimer’s.68 That group recommended that the mini-mental state examination (MMSE) and the ADAS-Cog be used to measure outcomes in all diseasemodifying trials. It also recommended that serial structural MRIs be used for assessment of disease modification (but kept this measure as optional for patients). Some of our workgroup members noted that neither the MMSE nor ADAS-Cog is typically used in clinical practice. The MMSE requires payment to the copyright owner and is less sensitive, especially for MCI than the freely available Montreal Cognitive Assessment.69 Likewise the ADAS-Cog is almost never used outside clinical trial setting by either physicians or neuropsychologists. This makes it difficult for clinicians to judge the clinical relevance of treatment outcomes and may impair clinical acceptance of new treatments. Notably, while the core outcome development process described in the publication was relatively robust in its inclusion academic experts and, to a lesser extent, patient/caregivers, it was not a multi-stakeholder exercise: payers, health technology assessors, regulators, and life sciences industry representatives, among others, were not included in the deliberations.70 This could explain why the core set largely reflects the priorities of researchers. The group’s full report indicates that consideration of additional stakeholders’ views may take place in the future.70

In any event, that published core set applies only to mild and moderate dementia, not disease interception. The recommended outcome measures are not yet sufficiently sensitive for detecting the earliest changes in cognition. (Cognitive instruments designed for investigation of preclinical AD do exist, such as the Preclinical Alzheimer Cognitive Composite,71 which is now in clinical trials.72) Hence, the need still exists for articulation of a core set for use in DMT/interception studies. Nevertheless, as noted, measurement tools and biomarker-based models describing the development and onset of dementia are still a matter of investigation. The expert working group therefore felt more time was needed before recommendations can be made for core outcome measures to demonstrate efficacy or disease modification in the preclinical or pre-dementia stages of clinical study.

to Alzheimer’s Disease

However, the project team’s consultation with external payers pointed to the need for a better understanding of patient and caregiver priorities and concerns for research, care, and resource use across the disease spectrum. In particular, they called for a patient-centered, patient-informed framework for defining drug effects (the character and magnitude of effects) that should be considered in the context of value assessment. This would be an important preparatory step for then defining core outcomes of importance to patients and caregivers across the disease spectrum. If informed also by the needs of regulators, payers, and other stakeholders, the resulting core outcomes could form the heart of a value assessment framework for therapies intended to treat this condition.

COVERAGE, REIMBURSEMENT AND “MEANINGFUL” THERAPIES

Our health plan informants on this project offered personal opinions that a new agent demonstrated to delay progression to dementia and approved by the FDA would have to be a covered benefit in a health plan. However, a concern is that the first successful trials of disease-modifying agents will be initially approved in the U.S. via accelerated approval, which (given the new draft FDA guidance described above) may provide health plans with little information regarding clinical benefits. The concern is that early successful DMTs will have evident effects on underlying pathology but modest effects on the progression to dementia (as some of our informants believe). For these observers, it is critical to establish the nature and magnitude of clinical effect as early as possible to understand the potential value of the new therapy.

Accordingly, a 2015 report from the Tapestry Group called for agreement of a set of “reimbursement endpoints,” presumably meaning measures needed by payers for decision-making.73 The group noted the challenge of modeling long-term outcomes on the basis of short-term data; they affirmed that stakeholders need to agree on endpoints that, taken together, could provide a composite picture of long-term effectiveness; they stressed that study designs must reflect varying priorities of stakeholders (clinically and economically relevant to payers, realistic to developers, and important to patients and caregivers); and noted that longer term follow-up for endpoint validation must be supported through broad-scale registries, analysis of electronic medical records, and other databases. The group indicated that in their forum, “HTAs and payers demonstrated a clear preference for objective end points such as functional measures or resource consumption measures, whereas patient advocates placed greater emphasis on quality-of-life metrics.”

The Tapestry Group recommended the following endpoints for disease-modifying agents:

Clinical effectiveness endpoints, 2 years after launch:  Prevention/lessening of cognitive decline  Better ADL functioning scores in treated group  Patient autonomy/maintenance of independence  Patient and caregiver consumption of resources  QoL or satisfaction (PROs or caregiver report)  Biomarkers linked to slowing disease progression  Safety and reduction of adverse events

Clinical effectiveness endpoints, 2 – 5 years after launch:  Duration of response  Time to next stage of disease progression  Delay to nursing home/reduced rate of institutionalization

to Alzheimer’s Disease

 nursing home readiness score  Mortality/longer life  Compliance with treatment

Note that minimally clinically important differences for these endpoints were not specified. In addition, for cognitively normal patients having an elevated risk of progression (the population that would be targeted for prevention trials), some of these measures may not produce measurable change in the time horizons suggested. For example, in asymptomatic populations or those with only mild cognitive impairment, there may be no discernable change in quality of life, independence or autonomy, or frequency of adverse events during a two-year span. Similarly, for these populations, significant differences in mortality may not be seen within the 2 to 5-year timeframe. Hence, overall, the Tapestry recommendations and timeline are more applicable to clinical studies focused on symptomatic disease stages more than prevention studies of DMTs.

Nevertheless, many of these measures and the reported stakeholder priorities for effectiveness accord well in general with the feedback we received on this project from external payers. Our external informants indicated that, for them, “meaningful” outcomes would demonstrate long-term differences that can be perceived by patients, families, and providers. Chief among these would be maintaining the ability for individuals with dementia to remain at home in a safe, reasonably well-functioning manner, rather than being institutionalized. While, as noted in the Tapestry report, payers are concerned about resource use, our informants underscored that use of medical resources can serve as an effective proxy for patient quality of life. Our external payer informants recognized that these are long-term outcomes rather far in the future from the disease interception studies at issue in this report. Nevertheless, they continued to emphasize “real” change, regardless of the stage of disease. Changes in biomarkers or even subtle changes in early cognitive testing should be able to be linked to early functional measures, such as an ability to keep working, and other delays in progression that patients, providers, and families can perceive.

While our project was underway, another view of the problem took place in a Duke Margolis Center workshop on “Exploring Outcomes and Value across the Spectrum of Alzheimer’s Disease,” held in June 2017 in Washington DC.74 Meeting participants reviewed and discussed current available therapies, the importance of identifying early signs of AD, the challenges of effective screening and diagnosis, challenges of designing clinical trials for earlier stages of disease development, and the need for new approaches to paying for Alzheimer’s care and future treatment. The workshop concluded with enumeration of a series of high priority next steps to make progress on these issues. They included more rapid standardization and validation of biomarkers; improved predictive modeling to identify candidates for early treatment and clinical trials; improvement of data infrastructure and systems for real-world evidence capture; and development of value-based payment models to support efficient, patientcentered AD care delivery. From the point of view of many payers, however, the top priorities in this list are those directed at establishing the science of effective therapeutics and the evidence base to support them. Until an example of a clinically meaningful therapy exists, reevaluating approaches to paying for treatment is seen by many as a purely academic exercise.75

CONCLUDING REMARKS

While any new disease-modifying therapies entering the market will require a longer-term development strategy, we have argued that several measures of direct and indirect medical costs, early subtle cognitive change (possibly very early functional changes with respect to use of electronic devices), and

to Alzheimer’s Disease

lower rates of behavioral and neuropsychiatric symptoms would be useful data to collect throughout phase 3 and 4 studies. (Note that these potentially important psychosocial endpoints are missing from the Tapestry Networks list above. From the point of view of some of our workgroup members, this is a serious omission.) These may be very early signs of changes predictive of progression and may thus serve as early data points for long-term models. In addition, as noted, current FDA guidance suggests that accelerated approval could be granted “if an effect could be shown on a valid and reliable cognitive assessment used as a single primary efficacy measure.” Hence, these measures of subtle early cognitive change may also provide an important strategy for regulatory review, as well as providing a possible anchor for the types of meaningful clinical outcomes payers hope to see.

Further, if these early clinical measures are carefully correlated with physiological (biomarker) changes and later progression, may also reduce the number of scans, lumbar punctures, and other techniques required for assessment. Given the often-stated concerns over the cost, availability, and consistency of some of the biomarker tests, these measures could represent important advances towards cost savings in future clinical diagnostics. Expectations about the necessity of using biomarkers in these innovative clinical treatments should be discussed with payers and other stakeholders. Biomarkers are likely to be a significant part of the discussion of the clinical utility of disease interception strategies. Providing information to payers useful for coverage and payment decision-making is of paramount interest for industry stakeholders, and (given the need to pay for any new therapies) of urgent importance to patients, their families, and their physicians. As noted earlier, the question arises not only what are meaningful therapeutic effects, but also what are meaningful effect magnitudes or durations. Robust longitudinal datasets to understand the clinical course and outcomes of populations exhibiting characteristics of preclinical and very early stage AD are lacking, obscuring the larger context in which randomized trials for preclinical AD should be interpreted. In addition, at this stage of drug development, it is difficult to anticipate the types and magnitude of effects that will be seen in a new generation of therapies.

Some observers perceive the payer emphasis on “real” change as representing expectations that may be difficult to meet in the near term, leading possibly to lack of coverage for promising newly approved agents. On the other hand, some payer informants expressed concern that health plans would be compelled to pay for new drugs having effect on a biomarker and but unknown value in terms of lived patient experience. Both of these concerns point to the continuing need for multi-stakeholder deliberation and agreement on the minimal clinical differences and early measures of change that will be meaningful.

The GPC project team has found that our discussions with project participants, industry representatives, colleagues, and payers all point to the need for multi-stakeholder deliberation with robust patient and caregiver input to establish meaningful elements of value in this context. Such a framework would be meaningless without translation into a set of preferred trial outcomes. Our discussions with external payer informants further confirmed the importance of core outcomes for new product review and value assessment. The GPC team believes this work is critical to pave the way for new therapies and, as the field continues to evolve, will continue to engage with stakeholders and others working in this area to define the priorities that should drive outcome selection for evaluation of benefit and value in Alzheimer’s therapeutics. Only in this manner can we assure that future clinical trials for Alzheimer’s therapies consistently produce information that is patient-centered, relevant to key decision-makers, and comparable across studies.

to Alzheimer’s Disease

REFERENCES

  • Cummings J, Fox N. Defining Disease-modifying Therapy for Alzheimer’s Disease. J Prev Alzheimers Dis. 2017;4(2):109-115.
  • Feldman HH, Haas M, Gandy S, et al. Alzheimer’s disease research and development: a call for a new research roadmap: Alzheimer’s disease research and development. Ann N Y Acad Sci. 2014;1313(1):1-16. doi:10.1111/nyas.12424
  • Cummings JL, Morstorf T, Zhong K. Alzheimer’s disease drug-development pipeline: few candidates, frequent failures. Alzheimers Res Ther. 2014;6(4):37. doi:10.1186/alzrt269
  • Villemagne VL, Burnham S, Bourgeat P, et al. Amyloid β deposition, neurodegeneration, and cognitive decline in sporadic Alzheimer’s disease: a prospective cohort study. Lancet Neurol. 2013;12(4):357-367. doi:10.1016/S1474-4422(13)70044-9
  • Reiman EM, Quiroz YT, Fleisher AS, et al. Brain imaging and fluid biomarker analysis in young adults at genetic risk for autosomal dominant Alzheimer’s disease in the presenilin 1 E280A kindred: a case-control study. Lancet Neurol. 2012;11(12):1048-1056. doi:10.1016/S1474- 4422(12)70228-4
  • Jack CR, Lowe VJ, Weigand SD, et al. Serial PIB and MRI in normal, mild cognitive impairment and Alzheimer’s disease: implications for sequence of pathological events in Alzheimer’s disease. Brain. 2009;132(5):1355-1365. doi:10.1093/brain/awp062
  • Bateman RJ, Xiong C, Benzinger TLS, et al. Clinical and Biomarker Changes in Dominantly Inherited Alzheimer’s Disease. N Engl J Med. 2012;367(9):795-804. doi:10.1056/NEJMoa1202753
  • Head E, Powell D, Gold BT, Schmitt FA. Alzheimer’s Disease in Down Syndrome. Eur J Neurodegener Dis. 2012;1(3):353-364.
  • Cummings JL, Dubois B, Molinuevo JL, Scheltens P. International Work Group Criteria for the Diagnosis of Alzheimer Disease. Med Clin North Am. 2013;97(3):363-368. doi:10.1016/j.mcna.2013.01.001
  • Murphy MP. Amyloid-Beta Solubility in the Treatment of Alzheimer’s Disease. N Engl J Med. 2018;378(4):391-392. doi:10.1056/NEJMe1714638
  • Walsh C, Drinkenburg WHIM, Ahnaou A. Neurophysiological assessment of neural network plasticity and connectivity: Progress towards early functional biomarkers for disease interception therapies in Alzheimer’s disease. Neurosci Biobehav Rev. 2017;73:340-358. doi:10.1016/j.neubiorev.2016.12.020
  • Alzheimer’s Association. 2017 Alzheimer’s disease facts and figures. Alzheimers Dement. 2017;13(4):325-373. doi:10.1016/j.jalz.2017.02.001
  • McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):263- 269. doi:10.1016/j.jalz.2011.03.005
  • Jack CR, Bennett DA, Blennow K, et al. 2018 NIA-AA Research Framework to Investigate the Alzheiemr’s Disease Continuum (DRAFT). September 2017. https://alz.org/aaic/_downloads/draftnia-aa-7-18-17.pdf. Accessed February 6, 2018.
  • Hebert LE, Weuve J, Scherr PA, Evans DA. Alzheimer disease in the United States (2010-2050) estimated using the 2010 census. Neurology. 2013;80(19):1778-1783. doi:10.1212/WNL.0b013e31828726f5
  • Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, and care. The Lancet. July 2017. doi:10.1016/S0140-6736(17)31363-6

to Alzheimer’s Disease

  • Lanoiselée H-M, Nicolas G, Wallon D, et al. APP, PSEN1, and PSEN2 mutations in early-onset Alzheimer disease: A genetic screening study of familial and sporadic cases. Miller BL, ed. PLOS Med. 2017;14(3):e1002270. doi:10.1371/journal.pmed.1002270
  • Progress and Challenges in Alzheimer’s Disease Drug Development: An Update on Scientific, Clinical, and Regulatory Considerations from an Industry Perspective | Global CEO Initiative on Alzheimer’s Disease. http://www.ceoalzheimersinitiative.org/node/90. Accessed February 2, 2018.
  • Petersen RC, Lopez O, Armstrong MJ, et al. Practice guideline update summary: Mild cognitive impairment: Report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology. Neurology. 2018;90(3):126-135. doi:10.1212/WNL.0000000000004826
  • Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):270-279. doi:10.1016/j.jalz.2011.03.008
  • Schneider JA, Arvanitakis Z, Leurgans SE, Bennett DA. The neuropathology of probable Alzheimer disease and mild cognitive impairment. Ann Neurol. 2009;66(2):200-208. doi:10.1002/ana.21706
  • Jansen WJ, Ossenkoppele R, Knol DL, et al. Prevalence of Cerebral Amyloid Pathology in Persons Without Dementia: A Meta-analysis. JAMA. 2015;313(19):1924. doi:10.1001/jama.2015.4668
  • Gifford KA, Liu D, Lu Z, et al. The source of cognitive complaints predicts diagnostic conversion differentially among nondemented older adults. Alzheimers Dement. 2014;10(3):319-327. doi:10.1016/j.jalz.2013.02.007
  • Reisberg B, Shulman MB, Torossian C, Leng L, Zhu W. Outcome over seven years of healthy adults with and without subjective cognitive impairment. Alzheimers Dement. 2010;6(1):11-24. doi:10.1016/j.jalz.2009.10.002
  • Jessen F, Wolfsgruber S, Wiese B, et al. AD dementia risk in late MCI, in early MCI, and in subjective memory impairment. Alzheimers Dement. 2014;10(1):76-83. doi:10.1016/j.jalz.2012.09.017
  • Buckley RF, Maruff P, Ames D, et al. Subjective memory decline predicts greater rates of clinical progression in preclinical Alzheimer’s disease. Alzheimers Dement. 2016;12(7):796-804. doi:10.1016/j.jalz.2015.12.013
  • Jack CR, Wiste HJ, Weigand SD, et al. Age-specific and sex-specific prevalence of cerebral βamyloidosis, tauopathy, and neurodegeneration in cognitively unimpaired individuals aged 50–95 years: a cross-sectional study. Lancet Neurol. 2017;16(6):435-444. doi:10.1016/S1474- 4422(17)30077-7
  • Clark CM. Use of Florbetapir-PET for Imaging β-Amyloid Pathology. JAMA. 2011;305(3):275. doi:10.1001/jama.2010.2008
  • Ritchie K, Ritchie CW, Yaffe K, Skoog I, Scarmeas N. Is late-onset Alzheimer’s disease really a disease of midlife? Alzheimers Dement Transl Res Clin Interv. 2015;1(2):122-130. doi:10.1016/j.trci.2015.06.004
  • Johnson KA, Minoshima S, Bohnen NI, et al. Appropriate Use Criteria for Amyloid PET: A Report of the Amyloid Imaging Task Force, the Society of Nuclear Medicine and Molecular Imaging, and the Alzheimer’s Association. J Nucl Med. 2013;54(3):476-490. doi:10.2967/jnumed.113.120618
  • Ossenkoppele R, Jansen WJ, Rabinovici GD, et al. Prevalence of Amyloid PET Positivity in Dementia Syndromes: A Meta-analysis. JAMA. 2015;313(19):1939. doi:10.1001/jama.2015.4669
  • CMS. Amyloid PET. January 2018. https://www.cms.gov/Medicare/Coverage/Coverage-with- Evidence-Development/Amyloid-PET.html.
  • Watson JL, Ryan L, Silverberg N, Cahan V, Bernard MA. Obstacles And Opportunities In Alzheimer’s Clinical Trial Recruitment. Health Aff (Millwood). 2014;33(4):574-579. doi:10.1377/hlthaff.2013.1314

to Alzheimer’s Disease

  • Toledo JB, Cairns NJ, Da X, et al. Clinical and multimodal biomarker correlates of ADNI neuropathological findings. Acta Neuropathol Commun. 2013;1(1):65. doi:10.1186/2051-5960-1- 65
  • Landau SM, Harvey D, Madison CM, et al. Associations between cognitive, functional, and FDG-PET measures of decline in AD and MCI. Neurobiol Aging. 2011;32(7):1207-1218. doi:10.1016/j.neurobiolaging.2009.07.002
  • Ewers M, Brendel M, Rizk-Jackson A, et al. Reduced FDG-PET brain metabolism and executive function predict clinical progression in elderly healthy subjects. NeuroImage Clin. 2014;4:45-52. doi:10.1016/j.nicl.2013.10.018
  • National Instutute for Health and Care Excellence (NICE). Dementia: Assessment, management and support for people living with dementia and their carers. Clinical Guideline. Draft for Consultation. January 2018. https://www.nice.org.uk/guidance/GID- CGWAVE0792/documents/full-guideline-updated. Accessed January 3, 2018.
  • Frisoni GB, Fox NC, Jack CR, Scheltens P, Thompson PM. The clinical use of structural MRI in Alzheimer disease. Nat Rev Neurol. 2010;6(2):67-77. doi:10.1038/nrneurol.2009.215
  • Ahmed RM, Paterson RW, Warren JD, et al. Biomarkers in dementia: clinical utility and new directions. J Neurol Neurosurg Psychiatry. 2014;85(12):1426-1434. doi:10.1136/jnnp-2014-307662
  • Hane FT, Robinson M, Lee BY, Bai O, Leonenko Z, Albert MS. Recent Progress in Alzheimer’s Disease Research, Part 3: Diagnosis and Treatment. J Alzheimers Dis. 2017;57(3):645-665. doi:10.3233/JAD-160907
  • Sperling RA, Aisen PS, Beckett LA, et al. Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):280- 292. doi:10.1016/j.jalz.2011.03.003
  • US Food and Drug Administration. Early Alzheimer’s Disease: Developing Drugs for Treatment, Guidance for Industry (Draft Guidance). February 2018. https://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/U CM596728.pdf. Accessed February 23, 2018.
  • Committee for Medicinal Products for Human Use, European Medicines Agency. Draft guideline on the clinical investigation of medicines for the treatment of Alzheimer’s disease and other dementias. January 2016. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2016/02/WC500 200830.pdf.
  • Preventing Cognitive Decline and Dementia: A Way Forward. Washington, DC: National Academies Press; 2017. https://www.ncbi.nlm.nih.gov/books/NBK436397/.
  • Rebok GW, Ball K, Guey LT, et al. Ten-Year Effects of the Advanced Cognitive Training for Independent and Vital Elderly Cognitive Training Trial on Cognition and Everyday Functioning in Older Adults. J Am Geriatr Soc. 2014;62(1):16-24. doi:10.1111/jgs.12607
  • Barnes DE, Santos-Modesitt W, Poelke G, et al. The Mental Activity and eXercise (MAX) Trial: A Randomized Controlled Trial to Enhance Cognitive Function in Older Adults. JAMA Intern Med. 2013;173(9):797. doi:10.1001/jamainternmed.2013.189
  • Khachaturian ZS, Barnes D, Einstein R, et al. Developing a national strategy to prevent dementia: Leon Thal Symposium 2009. Alzheimers Dement. 2010;6(2):89-97. doi:10.1016/j.jalz.2010.01.008
  • Shin J, Doraiswamy PM. Underrepresentation of African-Americans in Alzheimer’s Trials: A Call for Affirmative Action. Front Aging Neurosci. 2016;8. doi:10.3389/fnagi.2016.00123
  • Romero HR, Welsh-Bohmer KA, Gwyther LP, et al. Community Engagement in Diverse Populations for Alzheimer Disease Prevention Trials: Alzheimer Dis Assoc Disord. 2014;28(3):269-274. doi:10.1097/WAD.0000000000000029

to Alzheimer’s Disease

  • National Academies of Sciences E. Biomarker Tests for Molecularly Targeted Therapies: Key to Unlocking Precision Medicine.; 2016. https://www.nap.edu/catalog/21860/biomarker-tests-formolecularly-targeted-therapies-key-to-unlocking-precision. Accessed December 9, 2016.
  • Zhu CW, Sano M, Ferris SH, Whitehouse PJ, Patterson MB, Aisen PS. Health-Related Resource Use and Costs in Elderly Adults with and without Mild Cognitive Impairment. J Am Geriatr Soc. 2013;61(3):396-402. doi:10.1111/jgs.12132
  • Leibson CL, Long KH, Ransom JE, et al. Direct medical costs and source of cost differences across the spectrum of cognitive decline: A population-based study. Alzheimers Dement. 2015;11(8):917- 932. doi:10.1016/j.jalz.2015.01.007
  • Lin P-J, Zhong Y, Fillit HM, Chen E, Neumann PJ. Medicare Expenditures of Individuals with Alzheimer’s Disease and Related Dementias or Mild Cognitive Impairment Before and After Diagnosis. J Am Geriatr Soc. 2016;64(8):1549-1557. doi:10.1111/jgs.14227
  • Sano M, Zhu CW, Whitehouse PJ, et al. ADCS Prevention Instrument Project: pharmacoeconomics: assessing health-related resource use among healthy elderly. Alzheimer Dis Assoc Disord. 2006;20(Supplement 3):S191-S202.
  • Steinberg M, Shao H, Zandi P, et al. Point and 5-year period prevalence of neuropsychiatric symptoms in dementia: the Cache County Study. Int J Geriatr Psychiatry. 2008;23(2):170-177. doi:10.1002/gps.1858
  • Geda YE, Roberts RO, Mielke MM, et al. Baseline Neuropsychiatric Symptoms and the Risk of Incident Mild Cognitive Impairment: A Population-Based Study. Am J Psychiatry. 2014;171(5):572- 581. doi:10.1176/appi.ajp.2014.13060821
  • Green RC, Roberts JS, Cupples LA, et al. Disclosure of APOE Genotype for Risk of Alzheimer’s Disease. N Engl J Med. 2009;361(3):245-254. doi:10.1056/NEJMoa0809578
  • Messner DA. Informed choice in direct-to-consumer genetic testing for Alzheimer and other diseases: lessons from two cases. New Genet Soc. 2011;30(1):59-72. doi:10.1080/14636778.2011.552300
  • Harkins K, Sankar P, Sperling R, et al. Development of a process to disclose amyloid imaging results to cognitively normal older adult research participants. Alzheimers Res Ther. 2015;7(1). doi:10.1186/s13195-015-0112-7
  • Nuffield Council on Bioethics. Dementia: Ethical Issues. London: Nuffield Council on Bioethics; 2009.
  • Black BS, Taylor HA, Rabins PV, Karlawish J. Study partners perform essential tasks in dementia research and can experience burdens and benefits in this role. Dementia. May 2016:147130121664879. doi:10.1177/1471301216648796
  • Burback D, Molnar FJ, St. John P, Man-Son-Hing M. Key Methodological Features of Randomized Controlled Trials of Alzheimer’s Disease Therapy. Dement Geriatr Cogn Disord. 1999;10(6):534- 540. doi:10.1159/000017201
  • Caselli RJ, Woodruff BK. Clinical Impact of Amyloid Positron Emission Tomography—Is It Worth the Cost? JAMA Neurol. 2016;73(12):1396. doi:10.1001/jamaneurol.2016.3792
  • Pearson SD, Ollendorf DA, Colby JA. Diagnostic Tests for Alzheimer ’s Disease : Generating and Evaluating Evidence to Inform Insurance Coverage Policy. December 2012. https://pdfs.semanticscholar.org/2750/577ecc59b14aaaaa8d0bffb02aebba8c8498.pdf. Accessed January 5, 2018.
  • Holtzman DM, Corriveau RA. 2016 Alzheimer’s Disease-Related Dementias Summit: Report to the National Advisory Neurological Disorders and Stroke Council. https://www.ninds.nih.gov/sites/default/files/ADRD2016_summit_council_rpt_508Cpdf.pdf. Published September 15, 2016. Accessed November 16, 2017.

to Alzheimer’s Disease

  • Agency for Healthcare Research and Quality. Interventions To Prevent Age-Related Cognitive Decline, Mild Cognitive Impairment, and Clinical Alzheimer’s-Type Dementia. https://ahrq-ehcapplication.s3.amazonaws.com/media/pdf/cognitive-decline_research-2017.pdf. Published Maarch 2017. Accessed November 16, 2017.
  • Home :: Core Outcome Measures in Effectiveness Trials Initiative (COMET). http://www.cometinitiative.org/. Accessed November 16, 2017.
  • Webster L, Groskreutz D, Grinbergs-Saull A, et al. Core outcome measures for interventions to prevent or slow the progress of dementia for people living with mild to moderate dementia: Systematic review and consensus recommendations. PlosOne. June 2017. doi:doi.org/10.1371/journal.pone.0179521
  • for the Alzheimer’s Disease Neuroimaging Initiative, Trzepacz PT, Hochstetler H, Wang S, Walker B, Saykin AJ. Relationship between the Montreal Cognitive Assessment and Mini-mental State Examination for assessment of mild cognitive impairment in older adults. BMC Geriatr. 2015;15(1). doi:10.1186/s12877-015-0103-3
  • Webster L, Groskreutz D, Grinbergs-Saull A, et al. Development of a core outcome set for disease modification trials in mild to moderate dementia: a systematic review, patient and public consultation and consensus recommendations. Health Technol Assess. 2017;21(26):1-192. doi:10.3310/hta21260
  • Donohue MC, Sperling RA, Salmon DP, et al. The Preclinical Alzheimer Cognitive Composite: Measuring Amyloid-Related Decline. JAMA Neurol. 2014;71(8):961. doi:10.1001/jamaneurol.2014.803
  • An Efficacy and Safety Study of JNJ-54861911 in Participants Who Are Asymptomatic at Risk for Developing Alzheimer’s Dementia - Full Text View - ClinicalTrials.gov. https://clinicaltrials.gov/ct2/show/NCT02569398. Accessed February 23, 2018.
  • Bradley P, Akehurst R, Ballard C, et al. Taking stock: A multistakeholder perspective on improving the delivery of care and the development of treatments for Alzheimer’s disease. Alzheimers Dement. 2015;11(4):455-461. doi:10.1016/j.jalz.2014.01.007
  • Exploring Outcomes and Value across the Spectrum of Alzheimer’s Disease | Margolis Center for Health Policy. https://healthpolicy.duke.edu/events/exploring-outcomes-and-value-acrossspectrum-alzheimer%E2%80%99s-disease. Accessed February 23, 2018.
  • Karlsberg Schaffer S, Messner D, Mestre-Ferrandiz J, Tambor E, Towse A. Paying for Cures: Perspectives on Solutions to the ‘Affordability Issue.’ Value in Health 2018; 21: 276-279.

to Alzheimer’s Disease

APPENDIX A

ALZHEIMER’S DISEASE INITIATIVE PARTICIPANTS

Name Position Project Role

Christin Bexelius, PhD International Payer Strategy Leader (IPSL) Alzheimer's

Disease, Roche Steering Committee

Joel Bobula Senior Director, Outcomes & Evidence, Pfizer Steering Committee

Omar Dabbous, MD, MPH Head of Quantitative Science Payer Evidence Group, Halozyme (Independent private consultant on project) Steering Committee and Methods Workgroup AnnCatherine Downing, PharmD Principal Clinical Research Scientist Lilly Biomedicines Methods Workgroup

HowardFillit, MD Founding Executive Director and Chief Science Officer

The Alzheimer's Drug Discovery Foundation Steering Committee

Norman Foster, MD Director, Center for Alzheimer's Care, Imaging and

Research Chief, Division of Cognitive Neurology Professor, Department of Neurology Senior Investigator, The Brain Institute University of Utah Steering Committee and Methods Workgroup

Sadie Gabler, BS, CCRC Director of Research Participant Advocacy

Office of Research Integrity and Compliance University of Utah Methods workgroup

Jens Grueger, PhD Vice President, Head of Global Pricing and Market

Access, Roche Steering Committee

Devika Gupta Director, Market Access, Pfizer Methods

Workgroup

Allison Hoffman, PhD Office of Science, U.S. Food and Drug Administration Steering Committee

and Methods Workgroup

Joseph Hutter, MD LCDR, US Public Health Service, Centers for Medicare

and Medicaid Services (CMS) Methods Workgroup

Michael Irizarry, MD Senior Medical Fellow, Neurosciences,Eli Lilly Methods

Workgroup

Alexander Keenan Director Global Strategic Analytics,J&J Methods

Workgroup

Randy Krakauer, MD RetiredasVice President and National Medical Director

for Medical Strategy, Aetna Steering Committee and Methods Workgroup

Lindsay Lair, MD Global Medical Affairs Leader, GSCO Neuroscience,J&JMethods

Workgroup

Doris Lotz, MD, MPH Chief Medical Officer, New Hampshire Department of

Health and Human Services Steering Committee and Methods Workgroup

to Alzheimer’s Disease

Name Position Project Role

Michelle Mujoomdar, PhD Canadian Agency for Drugs and Technologies in Health (CADTH); European Network for Health Technology Assessment (EUnetHTA) Steering Committee and Methods Workgroup

Ellen Payzant Caregiver/Advocate Steering Committee

and Methods Workgroup

Creighton Phelps, PhD Freelance Neuroscientist - advisor for Neurosciences,

Alzheimer's disease and related disorders (Former Deputy Director of the Division of Neuroscience, NIA and the Director for the Alzheimer’s Disease Centers Program.) Methods Workgroup

Peter Rabins, MD, MPH Retired as Richman Family Professor of Alzheimer's and

Related Diseases, Department of Psychiatry and Behavioral Sciences, Johns Hopkins University Workgroup Chair

James Rollins, MD Director, Division of Items and Devices, Coverage and

Analysis Group, Centers for Medicare and Medicaid Services (CMS) Methods Workgroup

Mary Sano, PhD Director, Alzheimer Disease Research Center

Professor, Department of Psychiatry Associate Dean, Clinical Research Mount Sinai School of Medicine Director of Research Methods Workgroup

Lon S. Schneider, MD Professor of Psychiatry, Neurology, and Gerontology,

Keck School of Medicine of the University of Southern California Professor of Gerontology, USC Davis School of Gerontology Director, USC California Alzheimer's Disease Center Clinical Core Director, USC NIH Alzheimer's Disease Research Center Methods Workgroup

Eric Sokol Senior Vice President of Public Policy, Alzheimer's

Foundation of America Steering Committee

Greg White Senior Director, Global Market Access Policy,J&J Steering Committee

Jessica Wind, MEng Senior Product Manager, Managed Market Launch

Strategy, Genentech Methods Workgroup

Kristin Wrobleski, PhD Director, Global Patient Outcomes & Real-World

Evidence, Eli Lilly Steering Committee

Jerry Yang, PhD, MBA Senior Director, Medical Affairs, Pfizer Methods

Workgroup

Download the original PDF