LOU-ELLEN BARKAN (LEB): Richard, you’ve been engaged in Alzheimer’s research for many years, how did you become interested in this work?

RICHARD MAYEUX (RM): When I was an intern in the ’70s, patients with Alzheimer’s disease were left alone on open wards at Boston City Hospital. No one seemed to be interested in working with them. When I moved to New York from Boston to begin as an assistant professor I became interested in Alzheimer’s disease, but I did not know how to begin. Dr. Robert Katzman at Albert Einstein, had written a short editorial in the Archives of Neurology, about a pending epidemic of Alzheimer’s disease. And I said, “You know, he’s right, he’s absolutely right.”

Katzman was a professor and head of the department. I contacted him to say I was really interested in his work and that I had some background studying memory and other neuro-degenerative diseases.With that conversation started a long friendship, which enabled me to start my career. While I never really trained with him, I had the benefit of his advice. I saw how he approached the problem and modeled a lot of what we were doing on his work. Eventually, I had enough experience to begin projects here at Columbia, but that was how I became interested in the field.

LEB: Your interest was stimulated by academic work and the work of your peers.

RM: Katzman and Terry were ahead of the curve, recognizing that Alzheimer’s disease was a degenerative brain disease and not just part of normal aging. Together they opened up an area that no one had talked or thought much about. At that time, there were few serious publications on the topic and very few new ideas on how to approach the problem.

LEB: It sounds like you were out on your own in a lonely field.

RM: Not entirely, because I was able to talk with people like Katzman. But I was the only one at Columbia, so in the beginning, it was tough. I hired a terrific post-doc, Yaakov Stern, who’s now a professor here. We put together a couple of research grants to study Alzheimer’s. We then recruited Mary Sano, who’s now the Alzheimer’s Center Director at Mt. Sinai. Eventually, we grew from a small staff to our current operation.

LEB: It must be exciting to be the first.

RM: It was and we had some funny experiences. Once, we thought one of our grants was running out. We had just enough money to keep going for a couple of months, when we met a nice man whose father had died of Alzheimer’s disease. He came to see us because he couldn’t decide where to give his donation, Katzman at Albert Einstein or Price at Johns Hopkins, and he came to ask my opinion. I asked him, “Well, how much money are we talking about?” He told me “$25,000.” I said, “That’s not going to mean much to them, but it’ll mean a lot to me.” He has donated to our research for twenty years. Bill Robertson gave us our break, allowing us to get that first major grant.

Jed Levine (JL): That first seed money is so important. We know that your major areas of interest are genetics and epidemiology. Can you tell us more about your NIH funded genetics initiative?

RM: We’ve had a grant from NIH to study Latino families mostly from the Dominican Republic. We chose to study those families because of a project that had started in northern Manhattan. We had reported that the rates of Alzheimer’s disease were a little higher than we expected and we were concerned. The strongest predictor of disease in that population was having a family history of the disease. When we examined the pedigrees of the families, it was clear that the average 70-80 year old Dominican person with Alzheimer’s disease had several living brothers and sisters. This was ideal for study, so we launched the project.

Once we started ascertaining families, I realized that we weren’t going to find anything until we had several hundred families, so our first job was to get the families together. Early on we discovered that there was a single presenilin mutation for early onset familial Alzheimer’s disease, which was highly unusual. Because there had not been a mass migration to the island, this was an ideal place to do genetic studies. People there are so wonderful and wanted us to help them look at this problem.

In 2002, NIH invited several of us who had grants to a meeting—even though our group had no findings except for this novel presenilin mutation. What was clear was that since the discovery of APOE in 1993, there had been no major genetic breakthroughs. NIH asked us what could they do to facilitate research and, as we talked, we saw there were overlapping studies. People were studying the same families but not sharing information or communicating. I remember talking with Gerry Schellenberg (from the University of Washington) and we agreed that what was needed was access to a group of families. We all want to do our studies, analyze and publish our data first, but if the NIA wants to expedite this project they need a national sample that any qualified person can access.

The reason for this is the cost to find these families, characterize them, and have a big enough family size to make it available to study. So if every grant has to start off with a major financial investment, there’s no way this is going to happen. We said, “What if we did this collaboratively, but right from the beginning and these samples are going to be available to everybody, including the clinical data?” The other scientists said, “We already have a repository in Indiana, but it doesn’t have enough families.” People were already using that resource, but not depositing the data anywhere.

So the NIA issued supplemental grants to the Alzheimer’s Disease Centers—just a small amount in order to get the centers to collaborate. Then I got a call asking,“Would you be willing to take the lead on this and coordinate the whole thing?” And I said, “Under certain conditions. That you’re really going to follow through; that everybody can have access to the data; and that I can organize the study in a way that I think would be reasonable.” They said, “Go for it.”

We quickly worked with statisticians to determine how many families we needed and we decided that we needed to recruit 1,000 families. We needed a partner to get the work out. That is how Bill Thies (Vice President, Medical & Scientific Affairs of the National Alzheimer’s Association) got involved. We did a couple of public announcements and set up an 800 number and 18 centers across the country. We wanted “perfect” families. The minimal requirement was two people over 60. One had to have probable or definite Alzheimer’s disease and the other could have possible, along with any other relative in the same age range.

The project took off and about midway, I said, “We’re nearing 300 families. What if we do one more thing, submit the samples to the Center for Inherited Disease Research at NIH and have a first line genotype done, so there could be some linkage data?” The idea was that now a scientist who’s starting from scratch can say, “I’m interested in Chromosome 20. My gene’s there.” The scientist logs in and says, “Oh, there’s a linkage peak there, right around where my gene is.” He can now request access to the samples, access to the clinical data, all without spending a penny. And his grant proposal can focus on getting a smaller grant just to do the genotyping in this area. Now, many more scientists are using these samples to supplement their own studies. The bottom line is that all these samples are available to every scientist.

JL: How many families are in the studies?

RM: We have 900 families in our sites and are still recruiting. We have one sample from about 700 families and about 450-500 samples that are qualifying. That means we have enough data to do a linkage study. No family is ever finished, because we keep adding families. We go back and see them and adjust the diagnosis. If there is an autopsy, we correct it if there is an interesting outcome or something that we didn’t expect. We are still looking for families because not all those 900 families will work out. It takes us about two years from the time we identify a person to get the whole family evaluated. This is a rigorous evaluation. If we don’t see them in person, we have to review the medical records in detail, so it’s high quality data.

LEB: I am curious about one thing. You said that you need one person with Alzheimer’s disease and one relative. I once heard Dr. Mary Sano ask how we can confirm family records for individuals who are now in their 80s or 90s. After all, many years ago, the process of diagnosing Alzheimer’s was very difficult and uncertain?

RM: You mean when the death certificate says Alzheimer’s or dementia and the individual did not have an autopsy?

LEB: For example, many people diagnosed in 1970s had a diagnosis based on anecdotes rather than hard scientific data.

RM: Our rule is, tell me what you have and I’ll post the data. But I will note if the diagnosis comes from a General Practitioner, the family or an autopsy. Some doctors may not want to use certain samples, because they worry about the quality of diagnosis. They don’t have to use them. But others may say that certain samples are all they need for a first pass linkage study. What is important is that the information is there.

JL: I know a few months ago there was a lot of press about the SORL1 gene. What does this mean for families who are now coping with Alzheimer’s disease? Is this related to the genetic initiative?

RM: Genetic discoveries, including the early onset Alzheimer’s disease genes APOE and now, SORL1, all point to one specific set of pathways that are key to understanding the causes of Alzheimer’s. The three mutations result in an overproduction of amyloid beta and result in changing the type of amyloid that’s produced. The amyloid folds on itself, and when folded, unchaperoned proteins aggregate. What makes the amyloid toxic is not the aggregate itself, but it is the process of changing forms that may actually kill nerve cells.

All of the biological therapies that are on the horizon—the gamma secretase inhibitors, the beta secretase inhibitors, the anti-aggregates, the fluorizine treatments—could not have been developed unless we knew the pathway. But we don’t know everything there is to know about the pathway. We can’t explain the tangles. We don’t really know the complete story of how amyloid is processed within the cell.

The SORL1 story is a complicated story. And that’s because Peter St. George-Hyslop (from University of Toronto) and I were working together looking at a set of genes and focused on a region on Chromosome 10 where there were three genes in a gene family called vacuolar protein sorting genes. These are important for moving proteins within the cell organelles and then out to the cell surface. It turns out that one of the proteins that it moves around is an amyloid precursor protein (APP). Around the same time, Scott Small in our group was using microarray in hippocampi from Alzheimer’s patients and normal controls. He found that one of these protein-sorting proteins decreased in an Alzheimer’s hippocampus relative to controls. Jim Lah and Alan Levey (Emory University) had just published a paper describing another protein in this family. This time it was SORL1.

When Peter and I decided to look at this set of genes, we decided to look at all of the genes in this gene family, because they looked like they worked together and could be part of a complex.

We screened them genetically and SORL1 stood out from the rest. We already knew SORL1 was decreased in patients with Alzheimer’s disease from the work at Emory. Peter and I took populations that were as genetically homogenous as possible and powerful enough to see in a single association.

I had my family study in the Latinos,Peter had his family study from Northern Europe and we compared those two. Then Lindsay Farrer had African-Americans, whites, and the Israeli Arabs in his study from Boston. One surprise was that Israeli Arabs and Dominicans looked genetically the same with regard to SORL1.

Then we included collaborators from Mayo Clinic. We ended up with 6,000 people, one of the largest studies ever. We have also just finished a study of SORL1 in three different ethnic groups in Washington Heights, which was published this month in Archives of Neurology.

What we don’t have is a specific mutation or polymorphism in the gene itself. We do have regions where we think the defects in the gene are located. The importance of this is that it looks as if the genetic variant produces less of the SORL1 protein. Making less SORL1 allows APP to be processed through the beta secretase pathway, so it makes more of the abnormal protein, amyloid beta. It’s a subtle change.

JL: I was interested when I read the first report that you have very different ethnic groups that shared this one factor. What explains that?

RM: The point I’m going to try to make at the Academy of Neurology meeting is that we have to put away barriers to collaboration, because we’re not solving anything working on our own. Look at APOE. One group found it, but hundreds of people have received grants and written papers based on the initial observation. Yes, it was a key observation. Yes, one person does get a lot of credit for having identified it. But, there were multiple people at the institution that led to the discovery. There’s going to be plenty for all of us to do; identifying the genes and mapping the pathways which will become targets for therapeutic investigations.

JL: Is it realistic for us to talk about preventing the disease, delaying onset, even a cure in our lifetime?

RM: I’m an optimist. I think we have to look at Alzheimer’s disease the same way we look at cardiovascular disease, as a very complex disease with multiple factors. If you lose weight, don’t have diabetes, don’t smoke, and exercise regularly, you might get your risk down to almost nothing. But what if you have a bad gene that raises your cholesterol or gives you diabetes? There are now drugs like Lipitor for high cholesterol or aspirin which reduces your risk because it affects the platelet aggregation. In the worse case scenario, if you have diabetes and all the other risk factors, we now have stents and other invasive procedures.

That’s what’s going to happen with Alzheimer’s disease. Once we map out these genes, there will be some very aggressive,more invasive therapies that people can consider. But there will also be easier therapies which I’ll give you a very simple example. Let’s pretend that a vaccine is the best therapy we will ever have, and that 6% to 10% of the people who have the vaccine develop encephalitis—a very serious and sometimes fatal complication. However, in our example, if the vaccine works, it cures the disease. Let’s say you are a person with an early onset mutation and your chances of getting the disease by age 40 are 100% percent. If you have the vaccine, you know you have a 10% chance of a serious, if not fatal complication. Now you have an option and can make an informed decision.

This is an extreme and somewhat artificial example, but I think this is going to be our approach in five or ten years. Some therapies will be very aggressive, very risky, and individuals will be guided by their risk profiles.

However, what is very important is for people to understand that the risk factor is not being overweight when you’re 70, it’s being overweight when you’re 40—the midlife cardiovascular risk factors that are among the most important and probably very modifiable.

Look for Part II of this in-depth interview with Dr. Richard Mayeux in our Fall issue.

Dr. Mayeux is the Gertrude H. Sergievsky Professor of Neurology, Psychiatry and Epidemiology at Columbia University. He is also the director of the Sergievsky Center, and the co-director of the Taub Institute for Research on Alzheimer’s Disease and the Aging Brain, both at Columbia University Medical Center.

Dr. Mayeux graduated with distinction from the University of Oklahoma School of Medicine and completed training in Internal Medicine and Neurology in Boston and New York before joining the faculty of Columbia University in 1979. In 1991, he completed a Masters of Science degree in Epidemiology.

He has authored over 300 papers, chapters and books on the epidemiology and genetic epidemiology of Alzheimer’s disease and other degenerative diseases of the aging brain. Dr. Mayeux received the Leadership and Excellence in Alzheimer’s Disease Award from the National Institute of Aging in 1992 and the Potamkin Prize in 2007. He was elected to the Association of American Physicians and the American Epidemiological Society. He was appointed as a Fellow of the New York Academy of Science, and elected to the Institute of Medicine of the National Academies of Science.

Previous | Next