Alzheimer’s Disease: ethics of early diagnosis and practicalities of prevention

Drugs for Alzheimer’s have not been conspicuously successful. Do we need to think more about lifestyle interventions? Or should we treat pharmacologically, but earlier in the disease? The latter approach involves ethical and practical dilemmas in diagnosing prodromal AD.

Amyloid deposition seems to start a decade or so before the appearance of mild cognitive impairment (MCI), which in turn precedes dementia by many years. But the progression from MCI to dementia seems not to be inevitable: perhaps 30% of people with MCI do not develop severe cognitive dysfunction, according to Giovanni Frisoni, University Hospital of Geneva, Switzerland.

We have the concept of state -- the presence of biomarkers of amyloidopathy or tau, or both – and stage, which can be defined clinically as asymptomatic, prodromal, and advanced. In theory, we have a window of opportunity of perhaps twenty years in which we could intervene, if we had the means to make a difference, and if we could be sufficiently sure about the diagnosis of prodromal Alzheimer’s Disease (AD).


“Normal” pathology


One reason for caution is the devastating consequences of diagnosing “pre-AD” based on biomarkers. This is especially so given that AD-like lesions are found in cognitively normal people. In our mid forties, half of us have neurofibrillary tangles. By the age of 75, we almost all have them. This is perhaps “primary age-related tauopathy”.

Plaque prevalence and dementia prevalence both rise with age, but the dementia curve lags 20 years behind 

It has recently been argued that early tau pathology in the entorhinal cortex and hippocampus is always associated with the development of AD, but this is contested. One study suggests that only 26% of people with tau lesions progressed to dementia over five years.


Do drugs make a difference?


There is widespread scepticism about cholinesterase inhibitors in AD. But this is not really justified, Professor Frisoni argued. Cholinesterase inhibitors do work. The problem is that the mean change is relatively small and may be difficult to discern in individual patients with a fluctuating course of disease, and – after six to twelve months – progression resumes.

In contrast to the position with statins and blood lipids, for example, there is no biomarker of efficacy. In we could monitor a measure of cholinergic activity, we would have greater faith in treatment.

Is there any reason to suppose cholinesterase inhibitors would be more impressive if used in the prodromal phase? The trial would need to be conducted in patients with MCI who had proven positive on an AD biomarker – and not in all-comers.

Tau pathology precedes amyloid: at 75 years, almost all of us have neurofibrillary tangles but only half of us have amyloid deposits


The right to know and the consequences of knowing


Many people present to memory clinics with mild cognitive problems and request a specialist opinion. There is much discussion about whether it is ethical to diagnose AD in such people when we lack more effective means of intervention. But patients are uncomfortable with uncertainty and ask for hope, Professor Frisoni said. Biomarkers probably can distinguish a state of early AD that differs from normal aging; and cholinesterase inhibitors even now represent a reasonable option.

That said, very careful consideration must be given to assessing the patient’s true wishes in relation to disclosure of any biomarker-based diagnosis of prodromal AD, the uncertainties surrounding such a diagnosis, the potential involvement of family members and close friends, and the provision of counselling.

We have an obligation to do no harm; and, quite apart from the implications of prodromal AD for the individual and family, there is a risk the diagnosis may restrict access to health insurance.

In favour of giving patients as much information as possible, with appropriate caveats, is the principle that the patient has the right to know. Knowledge also brings the ability to plan for the future, and perhaps take all possible preventive measures. 


Prospects for primary prevention


Recent data from the population-based Rotterdam study show a lower age-adjusted incidence of dementia in the cohort recruited in the 2000s than among people recruited in the 1990s (4.92 vs 6.56 cases per thousand person years). This decline in incidence is consistent with that seen in other studies.

The factors thought responsible include a reduction in cardiovascular risk factors, supported by MRI evidence of less vascular change, and higher levels of education. The brain is healthier and brain volumes are larger.

Turning to AD specifically, Philip Scheltens (VU Medical Centre, Amsterdam, The Netherlands) reported that around a third of the condition worldwide is attributable to modifiable risk factors such as midlife hypertension and obesity, diabetes, physical inactivity and depression.

Randomized controlled trials of primary prevention in people at increased risk of dementia have reported some encouraging results. The Finnish FINGER study involved 1260 people aged sixty and above and used a two-year intervention including advice on nutrition, exercise, cognitive training and social activity. The intervention group showed a significant enhancement of cognitive performance and lower risk of cognitive decline, but longer follow-up is needed to assess incidence of dementia in general or AD in particular.

Data from the placebo-controlled, French Multidomain Alzheimer Preventive Trial are still to be fully published, but there seems evidence of a significant beneficial effect on cognition through combining multimodality lifestyle counselling with an omega-3 supplement.

So, it may be possible to prevent, or at least delay the onset of AD in individual patients. But, Professor Scheltens warned, a 10-20% reduction in risk factors would have to be achieved every ten years to reduce the worldwide prevalence of AD in 2050 by only 8-15%.

Our correspondent’s highlights from the symposium are meant as a fair representation of the scientific content presented. The views and opinions expressed on this page do not necessarily reflect those of Lundbeck.