A few months ago, we got a tip from a researcher about a new product from a company called Tethys Bioscience that can quite accurately predict a person's likelihood of developing type 2 diabetes. This sounded pretty revolutionary, in large part because it has the potential to save the country billions of dollars in diabetes prevention. But does knowing your risk actually motivate people?! They've teamed up with a popular smartphone health app to find out.
Curious about this endeavor and how the pilot project turned out, we asked our columnist and correspondent Wil Dubois to take a closer look:
It may be that we haven't had a good way to really know how to direct our prevention efforts. There are many variables, but when looking at everyone with pre-diabetes, most experts agree that approximately a quarter of them will "convert" to full-blown diabetes within five years, half will stay in a pre-diabetes state, and the final quarter will "revert back" to normal blood sugar.
What if we could know, for sure, who will get the big D and who won't? Would that make a difference to our prevention efforts? If you knew, for certain, that you would get diabetes if you didn't make changes, would it motivate you to make those changes?
Well, there's now a way to test people with pre-diabetes and determine which are likely to get full-blown diabetes, and which aren't. It's medical triage in a test tube called PreDx.
Simply put, it's a blood test that predicts the risk of getting full-blown type 2 diabetes in five years once someone has pre-diabetes. Unlike traditional screening measures that identify only who's at risk, PreDx quantifies how great that risk really is.
If you administer PreDx to a group of people with pre-diabetes, the test "identifies who's moving faster towards type 2 diabetes," according to Pamela Parkes, Senior Marketing and Communications Manager for Tethys, the Emeryville, CA-based company that developed the test.
The test analyzes seven different serum biomarkers. Three are known to most people with diabetes: fasting glucose, insulin levels, and A1C. The other four require a quick explanation. They are: ferritin, a measure of the iron in the blood which has a clear association with diabetes and diabetes control; C-reactive Protein, or CRP, an inflammatory marker commonly elevated in people with type 2 diabetes; and a pair of uncommon tests called interleukin-2 receptor-É‘ (another inflammatory marker); and adiponectin, a fat cell hormone that serves as a measure of the degree of insulin resistance.
A Poker Hand?
In trying to understand how this text works, I think of it like the diabetes predictive equivalent of having a poker hand.
Who knows why we have seven cards in a poker hand, really? But when it comes to this test, there are seven items in the PreDx panel because the company's research showed that more markers didn't make the test any more accurate, according to Caitlyn Waller Kress, the company's Director of Business Development. She says the firm, which was created to develop and market the test, originally studied over 200 different markers, winnowing the list down to 58 based on the availability of commonly available reagents and equipment for testing, and then used sophisticated computer modeling of those compounds on blood samples from the patients who developed diabetes during the ongoing Inter99 study in Denmark (the largest scientific prevention study on lifestyle in Denmark ever) to narrow in on the "smoking guns" that would predict who would convert to diabetes, and how fast it would happen.
The seven lab tests are combined with the patient's gender and age, and plugged into a patented proprietary algorithm to create a risk score on a scale of 1 to 10. A score of 5 means you have a 3.4% risk of converting to full-blown type 2 diabetes in five years. The scale angles sharply upwards from there. A score of 7 means your risk is 7.5%. A score of 8 means your risk is 12.5%. At a score of 9 your risk is 25%. At a maximum score of 9.9, your risk of moving from pre-diabetes to full-blown diabetes is a whopping 78%.
A score of 10 would mean you already have full-blown diabetes, and so in poker terms... this might be something like you losing to someone with a royal flush.
About Accuracy & Cost
Juan P. Frias, Chief Medical Officer for Tethys, tells us that PreDx is "clearly better than fasting glucose or A1C," and "fairly similar to OGGT (oral glucose tolerance test) in terms of accuracy."
But he points out that the real value of PreDx is not as a screening tool. In fact, Frias is emphatic that PreDx is not a screening tool at all, but rather a way to "stratify risk" among patients who have already been diagnosed as having pre-diabetes with the less expensive fasting, fingersticks or A1C tests.
PreDx "adds a tremendous amount of specificity," according to Frias. The real value of the PreDx score, he says, is in allowing providers to "tailor intervention." It tells them which patients need the most aggressive treatment to prevent conversion.
The list price for the test is a whopping $585. You can't buy it yourself; it has to be ordered by your doc. If you're uninsured, there's a cash price of $300. If your uninsured and poor, you're out of luck, as the company has no patient assistance program.
Like all companies, Tethys is in the business of making money. That said, most of the 150,000 people who've had the test run over the last four years didn't pay a penny. Right now the company's test, which is CLIA Certified, is out-of-network on most insurance plans. The company is of course working hard to get on the formulary of health plans, and/or have employers subsidize the test, without billing the balance to patients.
Teaming Up with Lose It!
The folks at Tethys now have a validated working test. They're negotiating with Medicare for coverage; and in a major feather in their cap, were granted a New York state lab permit, opening them up to serving all 50 states in the U.S. In a poster presentation at the 2011 ADA Scientific sessions, a Finnish researcher was even able to show that a PreDx score could be lowered 10% following a year of intensive lifestyle intervention -- proof-positive that it's possible to change your diabetes destiny in the pre-diabetes space.
The science tied up, the marketing underway, the legal approvals behind them, the team at Tethys began to ponder broader questions. Like: Does knowing you are at severe risk of diabetes have a motivational impact on the patient? Is knowledge really power?
So they teamed up with FitNow, the folks that make the wildly popular 10-million-download weight loss app Lose It! to try to answer that question. Lose It! is a calorie and exercise tracking app that helps you spend a daily budget of calories customized to your weight loss goals. It's the Poster Child for interconnectivity, playing nice in the sand box with other fitness apps, wireless scales, pedometers, and more. It even has a social side to give you support or competition.
The App "Study"
Frias is quick to point out that the Tethys/Lose it! project was "not a clinical trial. There was no control group," or other sophisticated design. He terms it a "first look-see" at the relationship between the length of app "engagement" and knowledge of a diabetes conversation risk score.
Eighty-nine people with PreDx scores were followed, and compared to the average Lose It! user. What the researchers found was that people who had been given scientific evidence that they were at risk for developing diabetes used Lose It! 10% more than the typical user. And that result paid off. On average, over the three months, the study participants lost ten pounds each.
But did they change their health destiny? The project was designed only to look at "engagement," but five of the physicians re-tested their guinea pigs on their own at the end of the study and found that even in this very short period of time, their patients were able to lower their PreDx scores. So they lowered their risk of conversion. The star performer was a 60-year old man who lost 15 pounds over 113 days using the Lose It! app, and dropped his PreDx score from 9.7 to 5.7 — roughly cutting his five-year risk of developing diabetes from something around 70% to less than 5%.
Dropouts & Promises
Surprisingly, of the 89 people enrolled in the pilot project, 62 dropped out within two weeks, a statistic that stunned me. There was no data tracking comparing drop-outs to stay-ins, so we don't know whether higher or lower PreDx scores increased the likelihood someone would stay in the study.
Frias says he'd love to do a more sophisticated study in the future and that's absolutely being considered, but there's nothing in the works just yet. That said, according to Kress, a way to input PreDx scores and track them, by both users and their providers, will be included in the next-gen premium version of the Lose It! app — which is expected out in the near future.
As exciting as it all sounds, I know from my own clinical experience that for a few months, or even half a year, people who are diagnosed with full-blown type 2 diabetes often stay focused, only to slip back into their old habits. If that happens with a full-blown diagnosis, aren't we likely to see the same pattern with an early warning sign? I wish the Lose It! study had been longer, but at least it does show promise...
Promise that a scientific early warning, and access to easy-to-use tools, may be the first steps towards reversing the tide of the type 2 epidemic.