Experts on Alzheimer's disease are proposing new criteria for diagnosing the dementia which would pick it up at an earlier stage and should get more patients onto treatment or into trials of new drugs.
An international expert group said the new guidelines would revise the definition of Alzheimer's to take into account recent scientific developments - including the use of so-called biomarkers, or biological signals, which can show if a person is at risk of the disease before they have any symptoms.
This pre-clinical stage, which can be about 10 years before dementia sets in, is widely seen as the best time to intervene in Alzheimer's. Recent studies have shown that brain scans, spinal fluid analyses and other tests can help predict who will develop Alzheimer's and they are becoming crucial to researchers and drug firms trying to develop new treatments.
"It's very important for us to move from the old way of seeing Alzheimer's disease to a new one that incorporates the importance of biomarkers," said Bruno Dubois from France's Salpetriere Hospital.
"There is no longer a reason to wait until patients have developed full-blown dementia," said Dubois, who leads the International Working Group for New Research Criteria for the Diagnosis of Alzheimer's Disease.
Alzheimer's, the most common form of dementia, is a fatal brain disease in which people gradually lose their memories and their abilities to reason and care for themselves. It affects more than 26 million people globally and there is no cure.
A report last month said the worldwide costs of coping with dementia will reach $604 billion in 2010, more than one percent of global GDP output, and those costs will soar further as the number of sufferers triples by 2050.
Currently only a post-mortem examination can absolutely confirm that a person has Alzheimer's. But in a paper in The Lancet Medical journal on Monday, the Dubois team said Alzheimer's should be defined as a clinical-biological syndrome to allow a diagnosis to be made on the basis of biomarkers in living patients and at an early stage of disease.
To meet these new criteria for a diagnosis of Alzheimer's, patients would not necessarily have clinical symptoms of dementia, but would have episodic memory impairment together with at least one positive biomarker shown on a brain scan or in test called a cerebrospinal fluid (CSF) analysis.
"The value of these definitions is their potential application in clinical trials of disease-modifying drugs," Dubois's team wrote.
Several large clinical trials of potential Alzheimer's drugs from firms such as Pfizer, Medivation and Eli Lilly have failed in recent years - partly, experts say, because they were tested in people whose brains had already been too damaged by the disease.
Alzheimer's advocacy groups are concerned that these costly failures may dissuade drugs companies and government-funded institutions from investing further in Alzheimer's research.
Speaking at the launch last week of a new advocacy group, USAgainstAlzheimer's, Republican Senator Susan Collins said the balance of spending on care versus research was badly skewed.
"For every dollar that the federal government spends today on the costs of Alzheimer's care, it invests less than a penny in research to find a cure. That simply does not make sense," she told reporters on a conference call.