According to the American Diabetes Association, seven million people have undiagnosed type 2 diabetes, and a staggering 79 million more have pre-diabetes. Early detection of type 2 diabetes can save lives, which is what drives the ADA's recent Diabetes Alert Day. But Dr. Michael Snyder, a geneticist at Stanford University, has taken early detection to a whole new level.
Dr. Snyder and his team have spent the last three years of their lives years working on analyzing genomes (a person's hereditary DNA info) as a way to predict an individual's future health. The changes in a genome can sometimes predict the early onset of disease, such as type 2 diabetes, which has several genes linked to it. Called an integrative Personal "Omics" Profile, or iPOP, the analysis examines all sorts of nitty-gritty details about a person's body.
What's fascinating is that Dr. Snyder unwittingly became part of his own experiment! Over the course of about a year, he and his team watched as his body slowly developed type 2 diabetes. He was diagnosed last year at age 55, and because he caught his diabetes so early, he's been able to manage it so far with just diet and exercise. This is the first documented instance of anyone watching their own body develop diabetes at such a detailed level, so we were especially intrigued to talk with Dr. Snyder — to find out what impact this research could have on both diabetes care and on the future of healthcare in general.
DM) What exactly were you and your team working on when you discovered your own case of type 2 diabetes?
American Diabetes Association Names New CEO
Non-profit leader Kevin L. Hagan named as new chief exec of national diabetes org after six-month search.
FDA Approves New Basal Insulin
Sanofi's Troujeo has 'flatter profile' of action that helps to avoid lows.
Daytona Win for Racecar Driver with Diabetes!
Type 1 driver Ryan Reed wins first NASCAR series race at Daytona on Feb. 21.
MS) The nature of our research is in characterizing genomes and doing large-scale analysis of genes and proteins. We were interested in learning how much your genetic information can be integrated into your healthcare. Your health is a product of two things: your genome/ genetics, and also a product of your environmental exposure, like pathogens, chemicals, things that you've been exposed to over the years. That's really what determines your health.
Although people talk about genetics in health in general terms, no one has really done much in how it exactly works. Now it's possible to get your whole genetic code sequence from all 6 billion bases. There are still issues with accuracy, but it's possible to do it. So the question is, can you do that and get useful information and incorporate it somehow into your healthcare?
So genome testing would become commonplace — something people do all the time?
When you think about it, when you go to a doctor, you'll get a physical exam, get your blood pressure done, and you'll also get a blood test. That test measures 15 or so different things in your blood. But there are technologies out there that can measure tens of thousands of things. The idea is that instead of measuring so few things, we should measure as many things as possible.
Our project was designed to sequence genome and see if that process is useful in healthcare, and can we follow the genome and see if it would be useful for disease diagnosis. So that's what launched the study.
How were you involved as a subject in the study?
We've now been profiling me for two years, and we've done it extensively when I'm sick. Because I have two kids, I've been sick four times. When I'm healthy, we do a genomic scan every two months, and when I'm sick we do it more often, between seven and nine times. We see what kind of things we can learn.
So what kinds of things did you learn?
There were several surprises, including that I was found to be at risk for type 2 diabetes based on my genetic sequence. I had several known genes that put me at risk. A lot of people know they have a family history of a disease, but this was a surprise because I didn't have a family history. We started tracking my glucose levels, which were normal and then started going up. I signed up for a very fancy glucose metabolism test. They found that my blood sugar was high, but they wanted to do it again. It was still high. Then they did an A1c test and sure enough I was diagnosed with type 2 diabetes.
The genome can in fact predict disease from your genome sequence. Because of that, I can actually follow the relevant things that are happening, and I could catch it early. Once I could see my blood sugar levels spike up, I essentially completely changed my diet and exercise routine and started taking baby aspirin. Gradually my blood sugars came back down to normal levels. My genome turned me on to something I should be aware of before anything (negative) happened.
Because we're could see the "signature" of these processes, I could see biochemical pathways were changing in away that no one had ever seen before. There were some very interesting changes. It was basically following a type 2 diabetes onset in a way that it wasn't ever seen before.
What was your initial reaction when you found out that you had type 2 diabetes?
I was surprised. I thought maybe it was just a transient glucose spike that would go away, but it happened over several months. That's why I took action by switching my diet. That was something I had to deal with. I tend to be pretty pragmatic about things. I saw something and I dealt with it.
What kind of impact do you think this could have on the diabetes community at large?
My own view is that genomics and genetics should be integrated into healthcare and that we can be carefully following things. I think some type of information is actionable, but some isn't, so you have to be careful what you return to patients.
For example, when I went in for this glucose test, the woman said, 'There's no way that you have type 2.' I agreed, but said that my genome says I'm at risk. I didn't know a lot about type 2 diabetes. They repeated the test, and I was elevated to the point where she said, 'You have type 2 diabetes.' The data spoke for itself!
I don't believe that type 2 diabetes is a heterogeneous disease. Even amongst type 2s, there are many many different types of this disease. there are many reasons why someone is at risk. I'm not overweight, I'm not the stereotype. I'm hopeful this will tease out the different types of diabetes. Some respond to metformin, some don't. Some respond to anti-inflammatory medicine, some don't. Diabetes is really hundreds of diabetes, and they just have one common characteristic which is a high level of glucose. We need to be able to classify that better. I'm hoping this improved classification can help us in treatment.
We hope to extend the technology to test all the different variants from the same drop of blood. You should be able to get 5,000 tests done with one drop so we can get a better idea of what's going on in your body. I think people who are at risk for certain diseases could do a simple home test. You could probably monitor yourself every month so you can catch diseases early. Most diseases, if you catch them early, can be prevented or treated. But if you catch them late, it's a real problem. By then there's a breakdown in the body, and things can be quite irreversible. The key is to catch it earlier.
With the help of genomics, medicine of the future should be completely transformed. It will be different than today.
It's nice to give the patient the opportunity to be an important part of their health care. The information is more than one physician can handle anyway. People should take responsibility for their healthcare, so I think it's important to give them that option.
Of course, some people are worriers and getting all their genomic information wouldn't be a good idea for them. They need to perhaps work with a genetic interpreter to help them decide what kind of information they can return to you. The patient should have control over that decision, as in, 'I don't want to learn this or I do want to learn that.'
There could be something like increased risk for breast cancer, and my own view is that people would want to know that. It's probably a good thing to learn because then you would know to monitor yourself closely. But perhaps not Huntington's disease, because there's nothing you can do about it if you're at risk.
Diabetes is definitely one that you should know about because what you eat and do can affect it. It definitely should be a dialogue between the patient and the doctor and the geneticist. I think the more you can empower people with their own health, the better that is.
You mentioned in an article that your life insurance premiums went up because you now have diabetes. Patients aware of health risks through genetic testing might also end up paying higher insurance premiums, no? So do you think documenting those risks is really worth it?
Right now, the testing itself is expensive, but in some areas, I think you can make a strong case. For example, for people who are at risk for cardiovascular disease, or something with an immediate health risk that runs in their family. Having a stroke is incredibly expensive, so knowing you're at risk and making some changes to avoid it is worth it.
Ultimately, the scans would be sufficiently cheap. It would be as just as cheap to test the whole genome, rather than doing these individual tests.
People need to be careful if they want this stuff shared. They may not want to know or may not want to share this. These are all relevant issues. I do think it can be argued economically, at least in some areas, saving peoples lives is something you can't put in economic terms.
There is still lots more to learn. I think my case is a very first, but hopefully the first of many case studies where your genome is part of your health. Hopefully this will help not just the diabetes community, but all health communities.
Thank you, Dr. Snyder, for volunteering yourself to science and helping us understand a little bit more about how genome sequencing may transform healthcare in the near future!