Now, after years of work detailing this devastating finale, scientists are increasingly tracking the path of Alzheimer's back to its roots.
For a society facing an epidemic of Alzheimer's - an estimated 360,000 new cases each year - getting at these early stages is critical. But it's also significant because years and possibly even decades before symptoms surface, subtle changes are already under way in the brain.
"It's like an iceberg building," said Dr. Dennis Selkoe, a professor of neurologic diseases at Harvard Medical School. "It takes a while until it breaks the water."
Doctors want to find a way to intervene. But Alzheimer's raises some of the most daunting questions in medicine, because the condition takes over the human body's most complex organ, one that is still largely a mystery: the brain.
Thirty years ago, physicians thought Alzheimer's was an inevitable consequence of aging. But they came to see that only certain people suffered from what was, in fact, a disease. Its hallmarks are sticky protein deposits, or plaques, and abnormal tangles in nerve cells.
By killing brain cells, Alzheimer's causes patients - who have included Ronald Reagan, Iris Murdoch, E.B. White, Rita Hayworth and Willem de Kooning - to deteriorate to an infant-like state. Their thinking and memory erode, their personality changes, and they eventually forget to feed or bathe themselves.
"We can tell you in horrifying detail all the later stages of the disease," said Dr. Marcelle Morrison-Bogorad, associate director of neuroscience and neuropsychology at the National Institute on Aging. "Now, we're beginning to get a handle on the earliest stages."
This month, Selkoe's study of rats in the journal Nature showed which form of the protein involved in Alzheimer's - beta amyloid - builds up in the brain. The work proved that the protein does its damage by interfering with memory circuits.
Many scientists are trying to figure out ways to pick up these and other brain changes. Researchers have already identified several genes that can cause Alzheimer's. They've also discovered that some neuropsychological tests can reveal deficits, such as visual memory errors, that are prevalent in people who go on to develop the disease.
Some physicians are enhancing imaging techniques so they can see the microscopic deposits and tangles in the brains of living patients. Right now, there is no way to view this destruction until autopsy.
"Somehow, you have to look in there and see if the treatments are working," said Dr. William Klunk, an associate professor of psychiatry at the University of Pittsburgh School of Medicine. He and Dr. Chester Mathis are testing a dye that can penetrate the blood-brain barrier and highlight the plaques during a PET scan.
In another approach, a recent study suggests a blood test might be able to detect Alzheimer's pathology in the brain, even before a person shows symptoms.
Dr. David Holtzman, an associate professor of neurology at Washington University School of Medicine, along with Dr. Steven Paul at the pharmaceutical company Eli Lilly, reported that intravenously injecting mice with an antibody to the beta amyloid protein decreases the level of the harmful protein. Apparently, the antibody flushes it from the brain to the bloodstream. There, the amyloid doesn't appear to cause any problems - and can be measured.
Other scientists are working their way back even earlier in the progression of Alzheimer's, by trying to prevent it.
About 5 percent to 10 percent of people who develop the condition have the inherited, or familial, form, which means they'll get Alzheimer's early, between the ages of 40 and 60.
But the bulk of patients with the disease develop it after age 65. One in 10 adults over 65, and nearly half of those over age 85, have Alzheimer's. These people's risk is probably influenced by factors that can be changed, such as environment or diet.
Some findings in mice offer hope that even simple steps, such as low-calorie diets or taking folic acid, might help nerve cells endure aging and resist the disease, according to Dr. Mark P. Mattson, chief of the neurosciences lab at the National Institute on Aging's Gerontology Research Center.
Nationwide, seven large-scale clinical trials are looking at ways to prevent or delay the disease in healthy people. Physicians are testing agents such as estrogen, anti-inflammatory drugs and gingko.
One of the more intriguing avenues scientists are exploring is the complex relationship between Alzheimer's and heart disease. According to several studies, lowered blood pressure in mid-life seems to correlate with a reduced risk of Alzheimer's later. At the same time, other research has shown that the most common type of cholesterol-lowering drugs, statins, might also decrease one's risk of Alzheimer's.
The findings hold out promise that treatments proved safe and effective for one major disease could be tapped for another. But experts aren't expecting easy answers.
"We can describe things, but we can't say how they work. Quite often, our attempts to explain how they work is just sort of skating over the surface of our ignorance - not because we're stupid, but because there's so much to know," said Morrison-Bogorad of the National Institute on Aging. "There are huge questions."
Scientists need to figure out why Alzheimer's starts, where it starts, how it progresses and what can stop it. They have to decipher why some neurons are vulnerable to the disease, and conversely, what protective factors enable other nerve cells to survive.
But the brain is a tough region to access for study. And with a disease that lasts years and years, testing treatments takes a long time. Making matters harder, no animal gets Alzheimer's, and scientists had to create an experimental model - a genetically modified mouse.
Researchers were reminded of these difficulties recently when a promising clinical trial in Alzheimer's patients had to be halted. The therapy, a vaccine, had reversed the formation of plaques in mice. But in the human trial, a small percentage of participants suffered a serious side effect, brain inflammation.
In the meantime, to help the rest of us ward off Alzheimer's, a stream of studies suggest all sorts of steps, from doing crossword puzzles to taking vitamins and even engaging in leisure activities such as watching movies.
But the truth is, no one knows what can be done to stop this slow killer. At least not yet.