Jeffrey Cummings, MD Mary S. Easton Center for Alzheimer’s Disease Research Deane F. Johnson Center for Neurotherapeutics David Geffen School of Medicine at UCLA Los Angeles, California, USA
Dr. Cummings has provided consultation to Acadia, ADAMAS, Astellas, Bristol Meyers Squibb, Eisai, EnVivo, Forest, GlaxoSmithKline, Janssen, Lilly, Lundbeck, Medivation, Merck, Merz, Myriad, Novartis, Noven, Pfizer, Prana, reMYND, Schering‐ Plough, Signum, Sonexa, Suven, Takeda, Wyeth pharmaceutical companies. Dr. Cummings owns stock in ADAMAS, Prana, Sonexa Dr. Cummings owns the copyright of the Neuropsychiatric Inventory
Targeting Aß Why Aß may not be the best target for Alzheimer’s type dementia Targeting tau Targeting “downstream” events Some unresolved issues
Amyloid-ß protein
Amyloid protein is the central cause of AD
Autosomal dominant AD ‐ over production Sporadic AD – mostly excessive accumulation
All AD have amyloid plaques All AD mutations cause Aß overproduction AD tg mice over produce Aß Predisposing genotypes increase Aß
ApoE‐4 allele SORL1 polymorphism
The oligomer is the neurotoxic species of Aß Monomers are not neurotoxic Plaques are composed of nontoxic fibrillar amyloid Aß 42 is the neurotoxic form of Aß Aß42 oligomerizes more readily than Aß 40 0r Aß 38
BetaSecretase
GammaSecretase
Aß Degra dation
ß-amyloid
Amyloid peptide (monomer/soluble) Amyloid oligomer (soluble)
Aß Fibrillization
Plaque (insoluble)
©Jcummings, 2009
Decrease production
Beta‐secretase inhibitors Gamma‐secretase inhibitors
Increase degradation
Rosiglitazone
Decrease aggregation
PBT2, AZD 103
Increase removal
Immunotherapy (AAB‐001, IVIg)
BetaSecretase
GammaSecretase
ß-amyloid
ß or Gamma secretase inhibitors and modulators Increase degradation Aggregation inhibitors
Plaque removal; decreased deposition
©Jcummings, 2009
APP
BetaSecretase
BACE inhibitors Difficult target Few agents in trials Few anticipated off‐ target effects Example
Memapsin 2 inhibitors
Crystal structure of ß-secretase
©Jcummings, 2009
BetaSecretase
GammaSecretase
ß-amyloid
Protease inhibitors More tractable target Anticipated off‐target effects Example
LY‐450139
Components of Gamma-secretase: PS1, nicastrin, APH-1, PEN-2
©Jcummings, 2009
BetaSecretase
GammaSecretase
ß-amyloid
Modulation of Gamma‐ Secretase Cleavage site change from Aß 42 ‐> Aß 38 Less aggregation Less toxicity Fewer anticipated off‐ target effects Example Tarenflurbil Eisai compound
©Jcummings, 2009
BetaSecretase
GammaSecretase
ß-amyloid
• Insulin degrading enzyme (IDE) • Neprilysin • Inhibition of plasminogen inhibitor activator (inc plasmin) • Others - Rosiglitazone
Aß in IDE
©Jcummings, 2009
Aggregation
BetaSecretase GammaSecretase
LTP/memory effects Neurotoxicity
Inhibitors
Scyllo‐inositols
▪ AZD‐103
ß-amyloid
Metal‐protein attenuating compounds
▪ PBT2
©Jcummings, 2009
ß-amyloid
Immunotherapies
Peripheral sink with passive removal of Aß from the brain Activated microglia with active Aß removal
Vaccination Passive immunization
AAB‐001 (Bapaneuzumab) IVIg (Gammagard)
©Jcummings, 2009
Approaches to
Decrease synthesis Increase degradation Decrease aggregation Decrease entry into the CNS Increase removal
None validated; one “win” will validate the Aß hypothesis
Amyloid can exist in the brain with limited toxicity Cognitively normal aged persons may have substantial Aß in the brain 20% of normal elderly have positive PIB scans consistent with fibrillar Aß Focal Alzheimer clinical syndromes have extensive Aß in the brain (e.g., logopenic form of AD) AD‐dementia and AD‐MCI patients have equivalent PIB scans
PIB scans do not change with dementia severity (suggesting stable Aß levels) CSF Aß levels do not change after the onset of dementia Removal of Aß plaques in vaccination trials has not interrupted progressive dementia
If Aß can exist in the brain without toxicity, then Aß may not be the best anti‐AD therapy If Aß sets of a chain of events leading to dementia, then Anti‐Aß therapy might prevent AD if given early
There is no evidence that ß‐ and gamma‐secretase decrease in activity as AD advances BACE activity increases with oxidative stress Cognitive decline correlates with total Aß (soluble and insoluble) Aß is toxic in some cell culture models in the absence of tau
Tau protein
NFT correlate with dementia severity Tau aggregation can cause dementia in the absence of Aß (frontotemporal dementia) APP tg mice lacking tau to not develop behavioral deficits1 Temporal atrophy over one year correlated with CSF p‐tau2 Tau hyperphosphorylation might be an “implementation” pathway for Aß
(1Roberson ED et al, Science 2007; 316: 750-754; 2Leow AD et al, Neuroimage 2009; 45: 645-655)
Kinases phosphorylate tau GSK3‐ß inhibitors (valproate, lithium) CDK5 inhibitors Phosphatases dephosphorylate tau PP2A modulators Other approaches Aggregation inhibitors
Aß peptide
Oligomer Plaque
Neurofibrillary Tangles
Oxidation
Ecitotoxicity
Inflammation
Mitochondrial Effects
Apoptosis
Synaptoxicity, Cell Dysfunction, and Cell Death
©Jcummings, 2009
Agent Dimebon AL 208 Methylene blue (Rember) Anavex 141 Bryostatin HDAC inhibitors Resveratrol
Process/Pathology Mitochondrial permeability transition pore inhibition Decrease tau hyperphosphorylation Dec tau aggregation Anti-apoptotic effects Increases synaptogenesis Restore transcriptional balance; neuroprotection Anti-oxidant; neuroprotection
Intracellular Aß
• Aß injury to mitochondria • Calcium enters cells (through mPTP) • Generation of reactive oxygen species • Caspase activation • Cell death • Therapeutic target - Dimebon
Cell Death
©Jcummings, 2009
Defining early AD Deciding when to go to Phase III Defining disease‐modification Determining a minimally clinically important change Some unintended consequences of disease‐ modification
Current definitions define Alzheimer’s dementia
DSM – dementia of the Alzheimer type NINCDS‐ADRDA – probable AD, requires the presence of dementia
Disease‐modifying therapy should be implemented early to minimize cognitive loss Definition of pre‐dementia AD is required
Episodic memory impairment Documented Progressive Isolated or part of a larger syndrome Biomarker evidence of AD Medial temporal atrophy on MRI Biparietal hypometabolism on FDG PET Amyloid signal on amyloid PET CSF: decreased Aß, increased tau/p‐tau First degree relative with APP or PS mutation
(Dubois B, et al. Lancet Neurol 2007; 6: 734-746)
Includes both
MCI of the Alzheimer type Dementia of the Alzheimer type
MCI would no longer be a separate condition Assist in constructing trial populations and drug development Facilitate regulatory discussions Likely to be highly specific with some sacrifice of sensitivity; requires study Challenging to apply in clinical practice
Increasing Alzheimer’s Pathology Mild Cognitive Symptoms
No Symptoms
Time
Dementia
• Increasingly severe phenotype • Biomarkers assist in identifying the underlying pathology • Biomarker changes may precede clinically detectable changes
©Jcummings, 2009
Traditional Model
P1: Safety P2: Clinical POC; Dose P3: Efficacy (Pivotal) FDA; Approval
Disease‐Modifying Drug Model
P1: Safety P2 P3: Efficacy (Pivotal) FDA; Approval
POC – proof of concept
Adequately powered, 18 mo P2 study IS P3
Proceed to P3 with Pharmacologic, not Clinical POC
©Jcummings, 2009
Clinical evidence – requires larger sample (≈ 400‐600 per arm) and long exposure (12‐18 months) Biomarkers
Limited experience in trials ADNI data are helpful There is no established predictive relationship between any treatment‐related biomarker change and the clinical response to therapy
MRI – no precedent; standardization issues CSF Aß and tau – magnitude and time course are uncertain PIB – plaques are themselves probably not the main target; time course of reduction (if any) is uncertain Aß production – promising; relationship to clinical outcomes is uncertain
• •
No consensus definition of D‐M Two elements required
• Effect the underlying disease process leading to cell death • Delay the progress of Alzheimer’s disease
Slowing of decline Delay of significant disease milestones Reduced slope of decline Increasing tx‐pbo difference over time
(Cummings JL. Alz Dem 2006; 2: 263-271)
D‐M: Increasing Drug‐Placebo Different Over Time
20
10 1 2 3 Years
©Jcummings, 2009
MMSE
4
5
25% slowing is considered a reasonable therapeutic target
Assuming 2 MMSE points loss per year 25% slowing ‐> 1 MMSE point difference at 2 years ‐> 2 MMSE points at 4 years
25% drug‐placebo difference would equate to 6 month delay after 2 years of treatment, 1 year delay after 4 years of treatment
Would mean 25% longer in each stage of disease Long term persistence required for benefit Benefit limited Minimal acceptable difference
How much? Who defines? What is the role of FDA in this question? How will payers regard this?
The amyloid oligmer hypothesis informs much of the current therapeutic quest in AD Aß therapeutics may involve production, degradation, oligomerization, or removal of Aß Aß therapies might be optimal for AD prevention Tau abnormalities correlate with the emergence of cognitive symptoms
Non‐Aß “Downstream” targets may offer therapeutic opportunities There are many uncertainties in AD therapeutics
When to treat How to know the target has been engaged Relationship of biomarkers to clinical outcomes Minimal acceptable clinical benefits