Years of work has moved technology to the point of large-scale testing that is required before FDA approval and commercialization.
Devices that substantially automate the treatment of type 1 diabetes mellitus (T1DM) are still years away from the market, but Aaron Kowalski, PhD, says that several academic and commercial projects have reached a key inflection point a shift from device development to the sort of large-scale testing that must precede commercialization.
“When we started supporting this idea, there were a few insulin pumps on the market, some experimental glucose monitors, and a small number of secretive efforts to combine the 2 technologies with intelligent software,” said Kowalski, chief mission officer and vice president for research at JDRF.
“We now have nearly a dozen major academic research teams, half a dozen device makers, and many more smaller organizations doing serious research and development. A large number of small trials have validated various approaches, and a handful of organizations are ready to launch pivotal trials. Patients won’t be able to buy an artificial pancreas this year, but they may well be able to buy the first artificial pancreas system that begins to dose insulin in 2017.”
FIGURE
What constitutes an artificial pancreas? The goal is a system that main-tains blood glucose levels and insulin delivery without the need for the user to make adjustments for meals or exercise, according to summary from the JDRF. From the start of JDRF’s funding for the project, it was anticipated that progress would come in steps, with the “closed loop” system coming after 5 earlier phases (see ).1
Medtronic Inc hopes to be the first company to bring such a product to market. The California-based company has long sold the only system that combines a pump and continuous glucose monitor (CGM) with software that automates any aspect of insulin delivery. (This advance led the FDA to dub Medtronic’s MiniMed 530G an “artificial pancreas device system,”2 but limitations in the automation which begins and ends with the ability to pause insulin delivery after blood sugar falls below a preset threshold—underwhelmed some opin-ion makers in the T1DM community). Now, Medtronic is beginning a pivotal study of a far more ambitious device.
The company’s Hybrid Closed Loop system uses data from a CGM and information that patients provide to pump the right amount of insulin at the right time. Patients still need to calibrate the CGM sensor with periodic finger sticks, tell the system how many carbohydrates they eat, and perform maintenance tasks such as replenishing the insulin in the pump, but the machine takes care of all the calculations and automatically adjusts basal insulin.
Medtronic announced this month that it would enroll 150 patients, aged 14 to 75 years, in a yearlong trial that would, if successful, provide all the data the company would need to ask the FDA to approve the system, which uses a Medtronic pump, a third-generation Enlite monitor, and proprietary software that performs the calculations.3
The company has already conducted a number of feasibility studies of its Hybrid Closed Loop system. Most of these studies have tested prototypes on small numbers of patients over short periods of time. Medtronic has not published complete data from any of these tests, but the research has helped the company optimize the underlying technology for the pivotal trial it just announced.
“The trial will measure not only A1C [glycated hemoglobin] levels but also the device’s ability to deliver small packets of insulin every few minutes to keep the patient’s blood glucose at or near a target. Maintaining blood sugar in this narrow band mimics what the body naturally does and may reduce the complications associated with diabetes to a greater degree than just reaching long-term A1C targets,” said Francine R. Kaufman, MD, chief medical officer of Medtronic Diabetes.
“We’ll also be using questionnaires to measure patient satisfaction with the system, and we expect enthusiastic responses. The Hybrid Closed Loop cer-tainly doesn’t automate everything and leave the patient to forget about diabetes, but it does transform the condition from something that’s primarily man-aged by patients to something that’s pri-marily managed by the device. As some-one who started out teaching patients to treat themselves with urine tests and animal insulin, I find the progress amazing. It’s truly exciting to be launching this trial.”
If all goes according to plan, Medtronic’s Hybrid Closed Loop system will reach American patients in the first half of 2017, just about a year after the introduction of a simpler Medtronic system that uses intelligent software to reduce hypoglycemia.
The MiniMed 640G system is a successor to the 530G. Instead of waiting for blood sugar to drop below a preset level before pausing insulin delivery, the 640G uses software to predict when blood sugar is likely to drop below the threshold and pauses insulin delivery in advance. The MiniMed 640G is already available in Europe and Australia, where it could come to market without trial data, and it is undergoing the pivotal trial required for FDA approval.
Medtronic has announced prelimi-nary 640G trial data, and the system’s predictive low glucose management technology appears to significantly reduce hypoglycemia. Indeed, a study of 22 T1DM patients, published in Diabetes Technology & Therapeutics, found that it reduced hypoglycemia associated with manually controlled systems by 26.7%. (The low glucose suspend technology of the 530G reduced hypoglycemia by 5.3%). The predictive technology also reduced the duration of each hypoglycemic episode. The average duration was 101 minutes with the threshold suspend technology but only 58 minutes with the predictive technology (P <.001).4
Medtronic has divulged much more about its plans for artificial pancreas technology than its main competitors, but several of those companies have given investors and patients some guidance.
Kim Blickenstaff, CEO of Tandem Diabetes Care, outlined his company’s plans during an earnings call in February: “In 2014, we initiated the R&D concept phase of our artificial pancreas offering. Our first AP [artificial pancreas] product will utilize Tandem’s propri-etary technology platform and will partially automate insulin delivery based on CGM information and predictive al-gorithms to aid a user. [It will] maintain their targeted blood glucose level and may reduce the frequency and severity of a hyper- or hypoglycemic event. In the second half of 2015, we plan to file an Investigational Device Exemption or IDE with the FDA for a clinical study in rolling out our first AP product.”5
Insulet Corp’s president and CEO Patrick J. Sullivan provided a similar level of guidance concerning his company’s plans when he spoke to investors that same month: “We are putting together strategy so that OmniPod is a very sig-nificant part of the artificial pancreas product offering in the future. We have been working internally on our own CGM development, which continues, but we also have an agreement with Dexcom to use their CGM sensor along with our new [personal diabetes manag-er] to integrate those 2 products togeth-er. I’d also say that we would evaluate and look at opportunities with Abbott and others that would have potential CGM integration opportunities for us. So in the short term we’re looking at other people that have CGM capabilities and algorithm capability, but at the same time we are continuing at a low level our own efforts in our own CGM product development.”6
While Animas released periodic updates about its “HHM” (hypoglycemia-hyperglycemia minimizer) between 2011 and 2013, there haven’t been many recent announcements from this Johnson & Johnson subsidiary, but work has continued.
“Since 2010, Johnson & Johnson Diabetes Solutions Companies has collaborated with the JDRF, industry, academia, and regulatory bodies to develop a first-generation automated insulin delivery device (what some call an “artificial pancreas”) based on predictive closed-loop technology,” wrote Johnson & Johnson spokeswoman Bridget Kimmel in response to inquiries about the project’s progress.
“As part of the program, 3 clinical research center—based studies have been successfully completed which clearly demonstrate the ability of the system to decrease and increase insulin delivery in order to avoid, or mitigate, potential hypo- and hyperglycemic excursions. Two related peer-reviewed articles have been published in the Journal of Diabetes Science and Technology, and a third is due for publication this year. Johnson & Johnson remains committed to intro-ducing this important next generation pump to people living with type 1 diabetes.”
Another company that hopes to be among the first to bring artificial pacreas technology to the US market is a small start-up that has yet to sell any products commercially.
Bigfoot Biomedical is less than a year old, but it raised more than $15 million in venture capital this year to pursue its plan for radically disrupting the diabetes device market. Its initial intellectual property stemmed from software that a company founder had developed to make existing products into the artificial pancreas products that his wife and son have worn for years. In June, Bigfoot bought the pump-making business that Asante had just shut down. Looking for-ward, Bigfoot’s goal is to begin pivotal trials by the end of next year and bring a closed-loop system to market by 2018.
“Everything about using an insulin pump is incredibly complex. Calculat-ing basal rates and bolus infusions several times per day, putting insulin in the pump, ordering new supplies, getting insurance coverage: it’s all complex. Most general practitioners won’t even get involved, so patients who want to try pumps have to go to endocrinologists,” said Bigfoot CEO Jeffrey Brewer.
“We aim to simplify every single step in the process. Current technology won’t let us dispense with some notification at mealtimes, but other than that, the system itself will pretty much run on autopilot. Our pump technology eliminates the need to mix insulin and spend hours each month getting rid of air bubbles. Our supply-order process will be automated because the system will track what patients have ordered in the past and what they have used. As for payments, we plan to charge users a flat monthly fee to simplify that as well. Our goal isn’t to attract the tiny percentage of medical specialists and tech-savvy patients who have always used pumps. It’s to attract the overwhelming majority of doctors and patients who have always thought them too complicated to use.”
The biggest limitations on artificial pancreas development in recent years have probably been the speed with which the insulin acts and the accuracy of glucose monitors. Many research-ers are working on a faster-acting (and shorter-acting) liquid insulin, but progress has been slow. Glucose monitors, on the other hand, have been improving rapidly. Dexcom recently won FDA approval for an algorithm that improves the accuracy of its Gen 4 Platinum moni-tor from a mean absolute relative differ-ence (MARD) of 13% to an MARD of 9%. Dexcom Chairman Terry Gregg says that level of accuracy is right on the edge of making CGMs a true replacement for finger sticks, if the product could maintain it without recalibration.7
The company is working with the FDA to demonstrate sustained accuracy, and of course, it is also working on new monitors. Dexcom has already filed for approval of its Gen 5 Platinum monitor, which should reach consumers later this year. It’s unclear whether that product will offer improved accuracy, as it uses the same sensor as the Gen 4. The Gen 6 monitor, on the other hand, is being developed with new sensor technology that should reduce (or even eliminate) the need for recalibration after initial installation at a new site.8
Dexcom expects to have this device to market as early as late 2016,9 so sub-sequent artificial pancreas systems that use Dexcom monitors (and most aspiring pancreas makers other than Medtronic do use Dexcom monitors) could greatly reduce or entirely eliminate the need for routine finger sticks. Moreover, Medtronic is also working to improve its CGM technology, though it has released less information than Dexcom about current performance and up-coming releases. (The technology used in the pivotal trial of Medtronic’s Hybrid Closed Loop asks users to recalibrate their sensors every 12 hours.)
Other companies are working on glucose-sensing technology as well. Abbott, for example, has already won approval from European regulators for a CGM that can be worn for 14 days at a time with no calibration.10 Some companies are also reportedly working on monitors that would never need to be moved because they detect blood glucose through the skin. Indeed, some observers expected the Apple Watch to launch with such technology.11
As commercial device makers such as Medtronic and Bigfoot work to bring their devices to market, academic re-search groups are working to develop and test the next generation of artificial pancreas technology. Significant efforts are under way in England, France, Israel, Australia, the United States and other places that have won substantial fund-ing from governments and research advocates such as JDRF.
That said, the most advanced and best known of the academic development efforts—at least to American readers—are probably those ongoing at the Univer-sity of Virginia and Boston University-Massachusetts General Hospital.
The Virginia project has published several studies over the past few years, all with positive results, with the most recent unveiled at the American Diabe-tes Association (ADA) meeting in Boston. At ADA, one of the project’s leaders announced that in a study on 36 adults with T1DM, overnight use of the Virginia system controlled glucose far better than a conventional insulin pump in both outpatient and inpatient settings12 The current iteration of the Virginia system has also been tested in patients who wear the system 24 hours a day in outpatient settings. It requires users to enter information about meals but is otherwise automated for users who choose to take a hands-off approach. (Users who wish to take some control over the system can do so.)12
The Virginia system, like every other system described above, uses nothing but insulin to control blood sugar. The Boston University-Massachusetts General system uses both insulin (to lower blood sugar) and glucagon (to raise it). Many believe this 2-hormone solution is impractical for reasons that range from the instability of glucagon to the extra expense of 2-hormone pumps, but the team in Boston believes it gives them much finer control over blood sugar lev-els and that it better mimics how a func-tional body controls its own blood sugar.
Trial results have been very impressive, but the only results published to date have come from small, short studies.13 Results from larger trials should appear soon, as the trials had been scheduled to be completed this spring,14 and the leaders of the Bionic Pancreas team say they hope to bring the technology to market by 2017.15
“The exciting thing is that it isn’t just 1 group that has produced great results in early trials. It’s a whole bunch of groups that have demonstrated technology that can control blood sugar better than current standards of care and do it with a lot less effort than current standards of care,” said Kowalski.
“If it were just 1, then there’d always be the worry that it would fall through, but what we see now shows that the technology is ready.” References
1. Artificial pancreas project. JDRF website. http://www.jdrf.ca/our-research/treat/artificial-pancreas-project/. Accessed June 30, 2015.
2. Types of artificial pancreas device systems. FDA website. http://www.fda.gov/Medical Devices/ProductsandMedicalProcedures/ HomeHealthandConsumer/ConsumerProducts/ ArtificialPancreas/ucm259555.htm. Published December 10, 2014. Accessed June 28, 2015.
3. Medtronic accelerates path to artificial pancreas with new MiniMed 640G data and start of hybrid closed loop trial [press release]. http://newsroom.medtronic. com/phoenix.zhtml?c=251324&p=irol-newsArticle&ID=2056959. Boston, MA: Medtron-ic newsroom; June 6, 2015.
4. Danne T. Tsioli C, Kordonouri O, et al. The PILGRIM study: in silico modeling of a predictive low glucose management system and feasibility in youth with type 1 diabetes during exercise. Diabetes Technol Ther. 2014;16(6):338-347.
5. Tandem CEO Kim Blickenstaff on Q4 2014 results-earnings call transcript. Seek-ing Alpha website. http://seekingalpha.com/ article/2946926-tandems-tndm-ceo-kim-blicken-staff-on-q4-2014-results-earnings-call-transcript. Published February 24, 2015. Accessed June 20, 2015.
6. Insulet’s CEO Patrick Sullivan on Q4 results-earnings call transcript. http://seekingalpha. com/article/2957376-insulets-podd-ceo-patrick-sullivan-on-q4-2014-results-earnings-call-tran-script. Published February 27, 2015. Accessed June 20, 2015.
7. Pomanger J. How Dexcom plans to eliminate the finger stick (and bring CGM to the masses): part 2. MedDevice Online website. http://www. meddeviceonline.com/doc/how-dexcom-plans-to-eliminate-the-finger-stick-and-bring-cgm-to-the-masses-part-two-0001. Published December 15, 2014. Accessed June 20, 2015.
8. Hoskins M. Talking next-gen tools with Dexcom leaders. DiabetesMine website. http://www. healthline.com/diabetesmine/talking-next-gen-fiabetes-tools-with-dexcom-leaders#6. Published March 5, 2015. Accessed June 20, 2015.
9. Pomanger J. How Dexcom plans to eliminate the finger stick (and bring CGM to the masses): part 1. MedDevice Online website. http://www. meddeviceonline.com/doc/how-dexcom-plans-to-eliminate-the-finger-stick-and-bring-cgm-to-the-masses-part-one-0001. Published December 12, 2014. Accessed June 20, 2015.
10. Lawrence S. Abbott glucose monitor eliminates finger sticks, receives CE mark. Fierce Medical Devices website. http://www. fiercemedicaldevices.com/story/abbott-glucose-monitor-eliminates-finger-sticks-receives-ce-mark/2014-09-03. Published September 3, 2014. Accessed June 20, 2015.
11. Heisler Y. Why the iWatch won’t measure glu-cose levels. Network World website. http://www. networkworld.com/article/2226419/wireless/ why-the-iwatch-won-t-measure-glucose-levels. html. Published February 26, 2014. Accessed June 20, 2015.
12. Brown A, Close K. Taking the artificial pan-creas home, 24 hours per day. DiaTribe website. http://diatribe.org/taking-artificial-pancreas-home-24-hours-day. Published May 18, 2015. Accessed June 20, 2015.
13. Russell SJ, El-Khatib FH, Sinha M, et al. Out-patient glycemic control with a bionic pancreas in type 1 diabetes. N Engl J Med. 2014;371:313-325.
14. Bionic pancreas website. The bionic pancreas multi-center study. Boston University website. http://sites.bu.edu/bionicpancreas/clinical-trials/. Accessed June 28, 2015.
15. Brown A, Liu N. Dr Ed Damiano presents next set of bionic pancreas results at ATTD. DiaTribe. http://diatribe.org/issues/62/conference-pearls. Published March 31, 2014. Accessed June 28, 2015.
How English- and Spanish-Preferring Patients With Cancer Decide on Emergency Care
November 13th 2024Care delivery innovations to help patients with cancer avoid emergency department visits are underused. The authors interviewed English- and Spanish-preferring patients at 2 diverse health systems to understand why.
Read More
Geographic Variations and Facility Determinants of Acute Care Utilization and Spending for ACSCs
November 12th 2024Emergency department (ED) visits and hospitalizations for ambulatory care–sensitive conditions (ACSCs) among Medicaid patients constitute almost 40% of all ED visits and hospitalizations, with lower rates observed in areas with greater proximity to urgent care facilities and density of rural health clinics.
Read More
Pervasiveness and Clinical Staff Perceptions of HPV Vaccination Feedback
November 11th 2024This article used regression analyses to quantify how clinical staff perceive provider feedback to improve human papillomavirus (HPV) vaccination rates and determine the prevalence of such feedback.
Read More