The 2009 Advanced Technologies and Treatments for Diabetes Meeting (ATTD)
The first major diabetes technology conference for this year was the Advanced Technologies and Treatments for Diabetes conference (or ATTD), held in Athens, Greece and bringing diabetes experts from all over the world. The biggest updates from this conference concerned continuous glucose monitoring (CGM) and progress towards a closed loop system using a pump connected to a CGM. We attended some presentations regarding a cure for type 1 diabetes, but this area of research continues to move very slowly.
continuous glucose monitoring
CGM systems continue to be a hot topic everywhere we go, and it seems like there’s more news for every issue! More and more researchers are studying CGM, and they are definitely improving in terms of accuracy and ease of use. The accuracy improvements represent great news – the more evidence there is that CGM can help improve management, the more big insurance companies will continue to climb onboard and reimburse for CGM systems, making them more accessible to patients who can’t afford them on their own. We know, of course, that there are still some major limitations because of all the people without insurance, with high deductibles, and in countries outside the US where CGM isn’t yet approved or reimbursed. We are glad the products are improving but sorry that insurance remains a barrier for some patients.
Diabetes technology expert Dr. John Mastrototaro, Vice President of Global, Medical, Scientific and Health Affairs at Medtronic, spoke at the conference and filled the audience in on some CGM improvements that the company has in the works. These include: 1) increasing the life of each sensor from three days to six days, 2) decreasing the startup time from two hours to 30 minutes, 3) reducing (or even eliminating) the need for traditional fingerstick calibration, 4) reducing sensor size and combining the sensor and insulin pumping site, and 5) improving the accuracy of the sensor. Of these improvements, we think that longer sensor life is closest to being ready, but Medtronic didn’t give any definite timelines - we imagine these are from near-term to long-term. We’ll be following this closely.
diaTribe Advisory Board member Dr. Lois Jovanovic of the Sansum Diabetes Research Institute (in Santa Barbara, CA), well-known as the world expert on pregnancy and diabetes, talked about her work using CGM systems during pregnancy. It’s well accepted that hypoglycemia and hyperglycemia during pregnancy increases the risk of birth defects, and it’s often very hard for pregnant women with diabetes to maintain the stellar glycemic control recommended by Dr. Jovanovic. But, one silver lining on glycemic control and pregnancy of late – Dr. Jovanovic has found that using CGM can dramatically help patients to detect hyperglycemia and hypoglycemia that they might not otherwise have noticed, allowing them to dramatically improve their glucose-management therapy (with their doctor’s help, of course) and protect their baby’s health. This applies to people who have type 1, type 2, and gestational diabetes (meaning diabetes that starts during pregnancy and then resolves). Talk to your doctor about CGM if you fit into any of these categories and are pregnant or thinking about becoming pregnant. Planning is key - internal organs begin development in humans within the 3rd to 8th weeks in utero (this is called organogenesis) and so ideally, blood glucose control ought to be very good before pregnancy begins. Dr. Jovanovic recommends patients work with their doctors and diabetes educators to get an A1c of “well below 6.0” (she prefers around 5.0 – 5.5) prior to pregnancy for those with established type 1 or type 2 diabetes. Although this is a major challenge for virtually anyone with diabetes, fertility, as she says, is a wonderful motivator – and CGM can be an incredibly helpful tool on this path.
the artificial pancreas—revisited
This meeting was filled with compelling talks about the artificial pancreas—some projects are further along than others, but overall there is clearly meaningful forward progress. We were particularly impressed by two presentations, one by Dr. John Mastrototaro and the second by Dr. Eyal Dassau also of the Sansum Diabetes Research Institute. These talks were focused on efforts to refine the computer program (algorithm) that will be used to give automatic insulin dosing from an insulin pump based on glucose readings from a CGM.
Dr. Mastrototaro gave a very interesting presentation about the work Medtronic is doing to close the loop. Medtronic has a working algorithm that has been tested in three different trials. Closed-loop (i.e. automatic glycemic control without human intervention) studies using this algorithm have shown good nighttime glucose control, which is really exciting – basically, this is easier than daytime control because there aren’t outside factors like food and stress that the system has to address. During the day, the system still has challenges coping with carbohydrates from meals. To help solve this problem, the company has added features that let you tell the system when you’re about to eat a meal and give manual meal boluses (this is called a hybrid closed-loop system). In a study done at Yale University by diaTribe Advisory Board member Dr. Bill Tamborlane and his team, a person who started at an A1c of 7.1% (certainly good control) was able to increase his ‘time in the zone’ from 58% to 82% using a trial closed-loop system. We want this! Exercise is, of course, another variable that can affect blood glucose near to the time of exercise as well as hours later. Researchers are still designing systems that can easily handle the effects of exercise.
Dr. Dassau presented an alternative idea for an algorithm that may be better able to control blood glucose automatically. Essentially, Dr. Dassau’s idea is to have a specific type of algorithm (called MPC, for model predictive control) that is in direct control of administering insulin, and then have that algorithm watched by a second algorithm called the ILC (iterative learning control) that would ‘learn’ the ups and downs and general patterns of a person’s blood sugar over the course of the day and fine-tune the performance of the MPC algorithm. Although it’s more complicated, this type of approach has performed exceptionally well in computer models. Even when researchers simulated variations in meals without giving any notice to the algorithm, the system was able to avoid (simulated) hypo- or hyperglycemia 96% of the time! We think that this technology shows great promise for use in future closed-loop systems. However, the important question of patient safety is still an issue, and it seems that safety will be harder to prove with more complicated algorithms than with simpler ones.
slow progress toward a cure
Cures for type 1 diabetes were not a primary focus at this meeting, but were addressed during several talks. We’re happy to see a number of different research groups continue to work on the problem and a number of different approaches being tried. We’d like to fill you in on a new approach that you may not have heard about. We note there isn’t really news here yet in terms of establishing successful treatment but for anyone who wants to read about what’s in early research on this front, read on ...
The new treatment idea is to make a diabetes ‘vaccine’. As you may be aware, type 1 diabetes is an auto-immune disease, which means that a person’s immune system inappropriately recognizes his insulin-producing beta cells as foreign and destroys the cells. Dr. Tihamer Orban, a researcher at the world-renowned Joslin Diabetes Center in Boston, has devised an injectable drug that is designed to desensitize a newly diagnosed patient’s immune system to insulin, stopping the attack on their beta cells. The drug doesn’t work like a typical vaccine (which introduces a foreign substance into your body), but more like the shots available to protect people from hay fever or pet allergies. He’s already completed a two-year phase 1 trial in newly diagnosed type 1 patients. Phase 1 trials are designed to show whether the treatment is safe – phase 2 trials should show the efficacy, or whether it works. We’ll be watching to see what happens next. Read more about Dr. Orban’s work.
Dr. Yong Zhao of the University of Illinois is working on the same problem but taking a very different tack. His research group has discovered that the T cells that are responsible for immune attack on the pancreas in diabetes, can be suppressed using stem cells from a person’s umbilical cord. For the tests he’s done in mice so far, he removes T cells from the animals and exposes them to the stem cells. The T cells are then reinjected into the mice. The exposure to stem cells seems to diminish the T cells’ response to insulin and is believed to reduce T cell attacks on the pancreas. His group has shown that this treatment can completely normalize the blood sugar levels of diabetic mice, and he thinks it would be safe and even low-cost when applied to humans. It’s impossible at this point to say where his team’s work might lead, but it’s likely that the diabetes community will start to see more stem-cell-based trials in the next few years with the progressive steps taken by President Obama thus far.
In our view, it’s too early to say which, if any, of the various approaches will succeed or when such treatments are likely to become available. But, we wanted to let diaTribe readers know what was happening in early research. Two promising companies that we know of are currently working on diabetes vaccines: Bayhill Therapeutics and Diamyd Medical.