Alzheimer's disease was once unique among illnesses in that a definitive diagnosis could be made only after death. Symptoms of the devastating assault on the brain could be seen from the outside, but those external signs also could be associated with other types of dementia.

Opening up the skull reveals the telltale signs: A shriveled brain with expanded, fluid-filled spaces called ventricles. On a microscopic scale, the wreckage can be observed postmortem with even more specificity.

In the past decade, diagnosis of Alzheimer’s while a patient is still living has cleared up considerably. Doctors can safely detect amyloid and tau deposits, telltale signs of the disease, in the living brain. When combined with cognitive tests, confirmation of the affliction can be made with much greater confidence than before.

Also, physicians increasingly realize that Alzheimer’s is a process that can begin decades before symptoms manifest. They’ve developed a better understanding of the changes that occur and the genetic risk factors associated with the disease.

In a new study, a group of noted researchers led by UC San Diego scientists proposes to deepen the knowledge by using known genetic risk factors to score individuals on their relative risk of developing Alzheimer’s at a given age.

This method yields what’s called a polygenic hazard score, or PHS. People in the top 1 percent of risk develop Alzheimer’s three decades earlier than those in the bottom 1 percent. Those in the top 10 percent develop Alzheimer’s more than a decade earlier than those in the bottom 10 percent.

Genetic tests costing about $50 can provide the needed genetic information. That component and the person’s age yield a personalized risk score.

Knowing the score can help doctors decide whether to give more tests to patients with cognitive problems such as memory loss, the scientific group said. The score also can help determine who are the best candidates for experimental drugs given to people who may be in the earliest stages of Alzheimer’s but haven’t exhibited any outward symptoms.

The study was posted in September on the BioRxiv website, which allows an early look at biological research before it is officially published. It’s available at j.mp/alzrisk.

Because the study isn’t published yet, it’s subject to revision. And because genuine progress against Alzheimer’s has been so slow and filled with false leads, doctors who care for Alzheimer’s patients probably will view the new method with skepticism. They will not only want to see the report’s findings verified independently, but also be shown how the individualized risk score can make a practical difference.

New viewpoint

The long-established fact that age is the greatest predictor for developing Alzheimer’s is a central assumption of the study, said Anders Dale, a UC San Diego researcher and the study’s senior author.

“We’re looking at Alzheimer’s as a condition that, it’s more a condition of when it happens than if it’s going to happen,” Dale said.

In other words, the great majority of people will eventually develop Alzheimer’s if they don’t die of something else first.

“That’s a fundamental insight that is different in this than in other studies out there — so-called polygenic risk scores, where you look at the disease as either something you’re going to get or not,” Dale said.

Genetic information from tens of thousands of people was analyzed for the study, he added.

To validate the risk findings with an independent data set, the researchers also examined the genetic details of 20,680 people that were taken from sources such as the Alzheimer’s Disease Neuroimaging Initiative and the National Institute on Aging.

Concerns

Assuming the study’s conclusions are confirmed, this new advancement raises a number of ethical issues about how to use it. Since all that’s needed to score a person are genetic information and age, it’s theoretically possible to do so for people of any age — with or without their consent.

While the ethical concerns are outside the scope of the new study, they have already been raised with other tests, such as for Huntington’s disease.

A fatal and incurable disease, Huntington’s is caused by mutations in a single gene — a prolonged repeat of the DNA letters C, A and G. The genetic test for Huntington’s is far more precise than the Alzheimer’s risk score described in the new report. Huntington’s doesn’t develop in people with 35 or fewer repeats of the targeted DNA letters. Those with 36 to 39 repeats fall into a gray area, and those with 40 or more will definitely develop the disease. In addition, the more repeats beyond 40, the earlier the patient will develop symptoms.

As with Alzheimer’s, a person who tests positive for Huntington’s cannot at this time do anything to stop or slow down the disease. Some drugs are in clinical trials, including one from Carlsbad’s Ionis Pharmaceuticals that may halt the disease at its genetic roots. (Information on that medication is available at j.mp/ionishunt.)

The question of whether to test for Alzheimer’s has proven difficult. Some people don’t want to know. Even if they want to know, the results could indicate a risk in siblings or other close relatives that alters their lives.

These considerations are even more important for the Alzheimer’s genetic risk scoring approach because that method can be used for many other diseases, Dale said. Counseling should be part of the testing, he added.

On the positive side, this information could aid doctors in diagnosing and perhaps easing the mind of older people who experience memory problems.

“Take forgetting your keys,” Dale said. “It’s a concern, and knowing your probability for your age, it turns out to be highly informative. If you are in the low-risk priority, it’s very likely that it’s a false alarm.”

Some dementias are highly treatable, Dale said. They could be caused by an underlying condition as simple as a vitamin deficiency.

Even if the news is bad, patients can still make use of the diagnosis, said Dr. Rahul Desikan of UC San Francisco, who co-authored the new study.

“We have patients who want to know, ‘Do I have to make a will? Do I want to stop driving? Do I want to have a chat with my wife about putting aside more money for my retirement?’” Desikan said.

Skepticism

Dr. Scott Riedler, a neurologist with the Sharp Rees-Stealy Medical Group, said he doesn’t see how physicians could make much use of the genetic risk score.

Riedler said the scoring test is an extension of a simpler genetic test for variants of a gene called APOE. The simpler test helps determine whether a person’s dementia is caused by Alzheimer’s or some other condition.

“The problem is, we don’t have any treatment that we would selectively give to those people at high risk” of Alzheimer’s, Riedler said. “We don’t have any disease-modifying medications at this point.”

In addition, he said the new scoring test is imprecise.

“How many people want to know that their risk of Alzheimer’s is 80 percent by age 85, if you’re not going to be able to take a drug to keep that from happening?” Riedler asked.

There are recommendations for how people might reduce their risk of Alzheimer’s, he said, but these apply to just about everyone: more physical and mental exercise, a healthy diet and so on.

Dale said the scoring test isn’t meant to be used in isolation, but as a factor along with other measures of health and mental acuity.

He also said the new report, being a purely research product at this point, needs additional work to translate the findings to clinical use. It’s possible the actual mix of patients will differ from those analyzed for the study, Dale said. So doctors need to assess the predictive power of the genetic scoring on actual patients, as opposed to a retroactive analysis.

Translation

Efforts to assess whether the genetic risk scoring can be used for actual patients is already underway at Human Longevity, a La Jolla company that’s exploring ways to extend the healthy human lifespan. Dale is on the company’s advisory board, as are a number of other UC San Diego researchers.

Founded by a group including genomics expert J. Craig Venter, Human Longevity is integrating new findings in evaluating health into a comprehensive exam at Health Nucleus, its clinical research unit. The Health Nucleus service, introduced a year ago, costs a minimum of $25,000 per person.

“The [latest] paper is important because it adds a new dimension to using multiple genetic markers for Alzheimer’s risk,” said Brad Perkins, Human Longevity’s chief medical officer.

Combining a person’s age with the genetic scoring substantially improves accuracy over just using genetic data alone, Perkins said. This makes the scoring a great candidate for use in Health Nucleus, he said.

“We’re excited about the work and are looking to apply the core concepts to other age-related diseases,” Perkins said.