This is an excellently written and compelling review, which adds to the drumbeat of support for early intervention or prevention as the design for disease modifying clinical trials in AD. The review is appropriately amyloid-centric, because amyloid has the most compelling rationale and the best defined targets for testing treatment. It is plausible that starting treatment before there is major damage to neurons or synapses may forestall a cascade from becoming self-reinforcing or independent. However, this highlights some critical gaps in our knowledge of AD: the key mechanisms involved in the cascade, and whether neurodegeneration is self-perpetuating. These questions have been difficult to address in animal models, which are best suited to model the deposition or removal of amyloid, but do not faithfully model the additional complex brain pathology of AD. If pathology is self-perpetuating, then extremely early treatment would be the most favored approach (at or before Stage 1 defined by the authors). Continuing to define mechanisms of amyloid-related toxicity and targetable...  Read more

This is an excellently written and compelling review, which adds to the drumbeat of support for early intervention or prevention as the design for disease modifying clinical trials in AD. The review is appropriately amyloid-centric, because amyloid has the most compelling rationale and the best defined targets for testing treatment. It is plausible that starting treatment before there is major damage to neurons or synapses may forestall a cascade from becoming self-reinforcing or independent. However, this highlights some critical gaps in our knowledge of AD: the key mechanisms involved in the cascade, and whether neurodegeneration is self-perpetuating. These questions have been difficult to address in animal models, which are best suited to model the deposition or removal of amyloid, but do not faithfully model the additional complex brain pathology of AD. If pathology is self-perpetuating, then extremely early treatment would be the most favored approach (at or before Stage 1 defined by the authors). Continuing to define mechanisms of amyloid-related toxicity and targetable components of the cascade could broaden treatment opportunities, and could allow combination treatments to be tested—for example, an amyloid-lowering drug plus a neuroprotective agent may provide a helpful strategy in preventing pathology from progressing from Stage 1 to Stage 2 while amyloid load is being reduced.

Another caution related to animal models is that they have not predicted toxicity found in human anti-amyloid clinical trials. For example, active immunization of transgenic mice against Aβ did not predict the development of encephalitis. Chronic treatment of transgenic mice with γ-secretase inhibitors did not anticipate problems such as skin cancer or accelerated decline that occurred in the Semagacestat clinical trial. There have been hints from mouse models that passive immunization can result in microvascular damage, but these did not predict events such as vasogenic edema that have occurred in studies of passive immunization. Trials of agents to be given to healthy people for a decade, as proposed by the authors, will require a careful approach to safety monitoring, and much can be extrapolated from trials conducted in patients with MCI or AD.

Starting the discussion about the pragmatic details of how to conduct targeted prevention studies is important. The authors appropriately emphasize the power of biomarkers to help with subject selection and to determine drug effects on amyloid burden and on aspects of neurodegeneration. Opportunities to test preventive treatment in familial AD, ApoE4 homozygotes, or in more generalizable at-risk populations are all attractive for different reasons, but, as the authors point out, there may be differences in selection of drugs or of dose ranges that are appropriate for each group.

Finally, although arguing by analogy can be misleading, the authors give the example of cholesterol lowering in the face of heart failure as a strategy that may not improve heart function. A different analogy is the treatment of osteoporosis, where a number of bisphosphonates have been shown to be effective by inhibiting osteoclast action; treatment can be started at any stage of the disease with benefits. Anti-amyloid clinical trials in MCI or in AD-dementia have generally not been conducted using methods that address whether, or by how much, amyloid burden was altered (an exception is the recent small study related to Bapineuzumab, which showed a small reduction of amyloid burden using PIB imaging; Rinne et al., 2010). So, the current wave of clinical trials using antibodies or secretase inhibitors will be important to empirically answer the questions of whether treatment lowers amyloid burden and if this is clinically helpful in a population with symptomatic AD. Even if these trials are negative, they will provide data that will be helpful to plan dosing and to select drugs for the long-term commitment of prevention trials.

References:
Rinne JO, Brooks DJ, Rossor MN, Fox NC, Bullock R, Klunk WE, Mathis CA, Blennow K, Barakos J, Okello AA, Rodriguez Martinez de Liano S, Liu E, Koller M, Gregg KM, Schenk D, Black R, Grundman M. 11C-PiB PET assessment of change in fibrillar amyloid-beta load in patients with Alzheimer's disease treated with bapineuzumab: a phase 2, double-blind, placebo-controlled, ascending-dose study. Lancet Neurol. 2010 Apr;9(4):363-72. Abstract

View all comments by Douglas Galasko

Golde et al. have done a real service to the field of Alzheimer's disease therapeutics by marshaling many cogent and well-reasoned arguments for developing primary and secondary prevention trials of potentially disease-modifying agents, especially those targeted at amyloid-β protein (Aβ). They have also laid out the scientific, medical, regulatory, and financial obstacles to executing such prevention trials. I concur with virtually all of the conclusions they reach. I would add one important point that they don’t emphasize: Given the many obstacles for conducting long and expensive prevention trials, the field also needs to focus even more creatively on designing and executing anti-Aβ trials in patients with very mild and mild Alzheimer's dementia. In other words, we have little choice in the near and intermediate term but to pursue trials that enroll mildly symptomatic patients who have clear evidence of Aβ buildup in their brains. This means especially emphasizing CSF Aβ42 levels (coupled with tau assays) as an entry criterion for ensuring that enrolled patients have an...  Read more

Golde et al. have done a real service to the field of Alzheimer's disease therapeutics by marshaling many cogent and well-reasoned arguments for developing primary and secondary prevention trials of potentially disease-modifying agents, especially those targeted at amyloid-β protein (Aβ). They have also laid out the scientific, medical, regulatory, and financial obstacles to executing such prevention trials. I concur with virtually all of the conclusions they reach. I would add one important point that they don’t emphasize: Given the many obstacles for conducting long and expensive prevention trials, the field also needs to focus even more creatively on designing and executing anti-Aβ trials in patients with very mild and mild Alzheimer's dementia. In other words, we have little choice in the near and intermediate term but to pursue trials that enroll mildly symptomatic patients who have clear evidence of Aβ buildup in their brains. This means especially emphasizing CSF Aβ42 levels (coupled with tau assays) as an entry criterion for ensuring that enrolled patients have an Aβ-driven dementia, since lumbar puncture is clinically straightforward, universally available, and much less expensive than amyloid imaging by PET scan (which could nonetheless be performed on a subset of enrollees). I envision treatment trials that enroll cognitively symptomatic but otherwise healthy patients with MMSE scores ranging from ~27 down to ~21, or an equivalent span on another cognitive screening test. Such trials would involve 18-24 months of treatment with an Aβ-lowering or Aβ-neutralizing agent.

Golde et al. correctly point out that even this mild clinical population may already have sufficient neurodegenerative changes that anti-Aβ agents would show limited or no efficacy. But experience with the very few Phase 2 AD trials to date that have shown biological signals, i.e., clear evidence of target engagement and some degree of cognitive and biomarker benefit versus placebo (see, e.g., Gilman et al., 2005; Salloway et al., 2009; Rinne et al., 2010), suggests that agents with these characteristics should be tested in Phase 3 trials confined largely to mild AD patients. While secondary prevention is a compelling way to go, even beginning now, the difficulties that Golde et al. point out for this approach should encourage us to push ahead with well-designed, well-powered Phase 3 trials of preclinically and clinically tested anti-Aβ agents in mild AD subjects.

View all comments by Dennis Selkoe

In their recent perspective in Neuron, Golde, Schneider, and Koo provide a cogent synthesis of the arguments for shifting clinical trials for Alzheimer’s disease from treatment of symptomatic individuals to disease prevention. We strongly agree with these authors that preclinical studies in APP transgenic mice should be considered in the context of prevention rather than treatment of symptomatic disease, and that studies in mice should be designed to parallel more closely subsequent human studies (Zahs and Ashe, 2010). In addition to testing interventions in mice with already established amyloid deposits, as suggested by Golde et al., we have proposed that biomarkers similar to those indicated for use in human prevention trials be used as outcome measures in animal studies; that agents be tested in multiple lines of mice, in multiple background strains, and preferably confirmed in multiple laboratories; and that care be taken to ensure that studies are adequately powered. Implementation of these suggestions will increase the cost of preclinical testing, but we believe that...  Read more

In their recent perspective in Neuron, Golde, Schneider, and Koo provide a cogent synthesis of the arguments for shifting clinical trials for Alzheimer’s disease from treatment of symptomatic individuals to disease prevention. We strongly agree with these authors that preclinical studies in APP transgenic mice should be considered in the context of prevention rather than treatment of symptomatic disease, and that studies in mice should be designed to parallel more closely subsequent human studies (Zahs and Ashe, 2010). In addition to testing interventions in mice with already established amyloid deposits, as suggested by Golde et al., we have proposed that biomarkers similar to those indicated for use in human prevention trials be used as outcome measures in animal studies; that agents be tested in multiple lines of mice, in multiple background strains, and preferably confirmed in multiple laboratories; and that care be taken to ensure that studies are adequately powered. Implementation of these suggestions will increase the cost of preclinical testing, but we believe that adoption of these recommendations is necessary to increase the probability that a successful intervention in mice will also succeed in humans.

In addition to scientific issues, the Neuron perspective also addresses the regulatory and financial challenges to conducting prevention trials for AD. As others before them (Alzheimer's Study Group, 2009; Reiman et al., 2010), Golde, Schneider, and Koo have highlighted the need for increasing incentives to pharma in order to encourage industry participation in long and costly prevention trials. We believe that another approach—one that focuses on the potential therapeutic benefits of affordable generic drugs and over-the-counter supplements—is also warranted. Several such compounds have been reported to benefit APP transgenic mice (Zahs and Ashe, 2010), although more rigorous preclinical testing is required to identify the strongest candidates for testing in humans. As pharmaceutical companies would have little or no financial incentive to support clinical trials of these compounds, a different type of public-private partnership will be needed, which might include philanthropic and healthcare organizations.

References:
Alzheimer's Study Group (2009) A National Alzheimer’s Strategic Plan: The Report of the Alzheimer’s Study Group.

Reiman EM, Langbaum JB, Tariot PN (2010) Alzheimer's prevention initiative: a proposal to evaluate presymptomatic treatments as quickly as possible. Biomark Med 4:3-14. Abstract

Zahs KR, Ashe KH (2010) 'Too much good news' - are Alzheimer mouse models trying to tell us how to prevent, not cure, Alzheimer's disease? Trends Neurosci 33:381-389. Abstract

View all comments by Karen Hsiao Ashe
View all comments by Kathleen Zahs

Re: "It is generally accepted that about every fifth person diagnosed as having AD turns out after death not to have had amyloid pathology. If AD is defined by its signature pathology, Feldman said, then these 20 percent of cases should not be called Alzheimer’s."

As someone with a strong interest but not a professional in this field, I have to ask, How can what is merely a hypothetical signature pathology be used to exclude cases that do not exhibit that pathology?

All that would prove is A = A!

View all comments by Helen Cunningham

If we accept that the pathological accumulation of amyloid-β in the brain precedes dementia for over a decade, prevention trials in Alzheimer’s disease with disease-modifying drugs, including anti-amyloid therapy, should be conducted in subjects in their early sixties and not in their early seventies, when the disease is probably too advanced, unless aggressive brain repair/regenerative treatments are administered, also.

View all comments by Elena Galea