Clinical Trials in Prodromal Alzheimer’s Disease: An Industry Perspective
Ravi Anand, M.D., Richard D. Hartman, PhD, Munaf Ali, PhD.
Currently available treatments for patients with Alzheimer’s disease (AD) have not provided substantial and/or long-lasting improvements in cognition, behavior or activities of daily living (ADL). Based on the small magnitude of the benefits accompanying these treatments in patients with established AD, their effect on important outcomes such as quality of life, increased productivity, ability to function independently, time to hospitalization, is of limited clinical, societal, or economic benefit. Experts have estimated that their impact on Quality Adjusted Life Years (QALY) is low, and consequently they are not considered cost-effective.
Many factors may explain the lack of availability of highly effective treatments for patients with AD. These include the limited ability of the five marketed drugs to produce improvement in patients with AD. It is also possible that the stage of the disease in which these agents were tested is too severe to allow for more than palliative, short-lasting, clinically significant benefits. Mild to moderate/severe AD, for which these drugs are approved, represents a stage where there is significant atrophy in various parts of the brain, extensive neuro-degeneration and synaptic loss, highly significant reduction in neurotransmitters, and a loss of high level cognitive skills and at least instrumental activities of daily living. Currently, ongoing studies with immune modulating agents such as monoclonal antibodies, whether directed against beta amyloid (Aβ), or tau protein, or inhibitors of various secretases involved in amyloid processing, all target patients with established AD that have the same brain deterioration as noted above, and are unlikely to produce substantial long-lasting benefits.
Studies in patients with Mild Cognitive Impairment (MCI) with some of these agents have also been unsuccessful in demonstrating treatment benefits. Data from autopsy studies in patients who died during the stage of MCI have indicated that a majority of these patients met pathognomonic criteria for AD, thus demonstrating that patients with MCI are at least at the stage of mild AD, although their clinical signs and symptoms are relatively mild. Based on the long duration and progressive nature of the pathological processes underlying AD, it seems logical to assume that, if treatments are to provide benefits to patients that allow them to maintain their ability to function independently, enjoy a high quality of life, and show significant increases in QALY, cost-effectiveness, and societal benefits, treatment should be started at a stage of AD, which is very early with regard to neuro-degeneration, cognitive/behavioral dysfunction, and preservation of ADLs. This stage, rather than MCI, should be considered as “Prodromal” AD.
There are many challenges in designing clinical trials to test new therapies in patients with “Prodromal” AD, including achievement of a consensus on diagnostic criteria for “Prodromal” AD, development of sensitive and specific diagnostic tools, identification of potential targets for therapeutic intervention, development of new drugs that can delay and/or modify the disease process, and determination of appropriate outcome measures and trial designs to assess the effects of drug therapies on the symptoms and signs of “Prodromal” AD.
The first major hurdle in developing therapies for “Prodromal” AD is obtaining agreement among academia, industry and regulatory agencies on the definition of the syndrome and its existence as a distinct diagnostic entity. Patients with true “Prodromal” AD represent a subset of the population of patients diagnosed with MCI, and are characterized by greater than expected deficits in cognitive function, based on age and education level, that do not interfere with functional abilities. “Prodromal” AD is difficult to diagnose and may precede the development of AD by as much as 10 years. Therefore, diagnostic criteria must be very specific, and the accurate diagnosis of “Prodromal” AD will require a specialized battery of neuropsychological tests(verbal fluency, Free and Cued Selective Reminding Test, free recall, etc. ), as well the use of neuroimaging, such as structural and functional MRI (e.g. hippocampal/ medial temporal lobe atrophy) and PET (e.g. FDG-PET, or PIB uptake) and/or other biomarkers (e.g. CSF levels of Aβ1-42 , or total tau/ Aβ1-42 ratio; neural thread protein; AD7C-NTP in CSF and urine) to ensure that patients have the disease, as well as to distinguish it from other types of MCI. Advances in genomics and new proteome-based plasma biomarkers, together with recognition of risk factors for developing AD, such as age, ApoE ε4 genotype, and PS1/2 mutations, may allow for identification of patients with incipient stages of the disease, as well as individuals at high risk for AD.
Based on the above, the following criteria are proposed for identifying “Prodromal” AD:
· Patient and/or caregiver reported deficit in cognition
· Deficit on a cognitive measure (compared to normal, based on age and education) limited to one cognitive domain
· No functional impairment, based on detailed examination
· No manifest behavioral symptoms
· MRI evidence excludes more than minimal atrophy
· Changes in CSF markers (Aβ1-42, tau/ Aβ1-42 ratio not consistent with AD).
Another key issue in designing drug trials to evaluate treatments for prodromal AD is the choice of outcome measures. The changes in cognitive function over time in these patients may be subtle, thus requiring very sophisticated and sensitive measures in order to have adequate power to distinguish differences between drug-treatment and placebo. As cognition appears to be a valid, sensitive and predictive marker of progression, change in a cognitive measure over time, rather than a paradigm such as ‘time to conversion’ to AD (which has proven to be unreliable and insensitive to change in MCI trials) should be considered as the primary efficacy measure. It is unlikely that benefits in global evaluation, behavior, and ADLs can be shown in these patients, as prodromal AD patients will not experience deficits in these parameters at this stage. Evaluation of specific biomarkers in conjunction with cognitive testing would most likely be required to assess changes over time.
A 2007 survey of the prevalence of AD in the US estimated that there were more than 5 million AD patients; this is likely to increase to7.7 million by 2030 and ~16 million by 2050. The annual cost of AD, including both direct and indirect costs, is likely to be more than $148 billion/year. The impact of effective treatments against AD will be greater in the 4th and 5th decade of life than at age 90. It is estimated that a treatment for prodromal disease that can delay the onset of AD by even 1 year will mean a gain in QALY of 0.52, and reduce the incidence of AD by 210,000 cases per year and annual costs by $19 billion. Assuming 454,000 new cases of AD will be diagnosed in 2010 and 959,000 in 2050, the cost savings, assuming $150,000 per QALY, will be $1.03 trillion. These estimates are based purely on measuring benefits of a delay in onset of disease, and not on altering disease progression. These expected cost savings should stimulate research by academia, government and industry into developing treatments and targeting patients before AD becomes established.