Benchmark Cambridge Trial In Quest For Ambulatory Artificial Pancreas

I had reported earlier that once perfected and approved by regulators, safe and robust ambulatory artificial pancreas ‒ or to use the scientific term ‘closed loop insulin delivery system’ ‒ has the potential to greatly improve the health and lives of people with type 1 diabetes. The idea itself is not new but the old generation closed loop insulin delivery systems were cumbersome and unsuitable for long term or outpatient use.

Artificial pancreas concept

The newer systems link a continuous glucose monitor and a subcutaneous insulin infusion pump via a control algorithm, which retrieves continuous glucose monitoring data in real time (for example, every five minutes) and uses a mathematical formula to compute insulin delivery rates that are then transmitted to the insulin pump.

However, artificial pancreas that can be worn by diabetics on their person as they go about their daily lives is still in development, with the first in-clinic studies now being reported. Preliminary results have been promising ‒ the most notable improvement is in overnight control of type 1 diabetes, with improvements in safety and a reduction in nocturnal hypoglycemia being reported.

These improvements result from the fine adjustment of insulin delivery provided by closed loop control overnight being superior to a generally fixed basal rate and less likely to cause hypoglycemia. The first application of closed loop control is therefore likely to be in glucose regulation overnight, a step that has the potential to improve dramatically the safety of insulin delivery during crucial, generally unsupervised, periods.

Now a University of Cambridge research tem led by Roman Hovorka has demonstrated the safety and efficacy of overnight closed loop insulin delivery with conventional insulin pump therapy in adults with type 1 diabetes.

Artificial Pancreas Best Hope For Diabetics In Near Term

The trial group consisted of 24 adults (10 men and 14 women) aged 18-65, who had used insulin pump therapy for at least three months and the research team used two protocols ‒ a medium sized meal (60 g carbohydrate) and a large size meal (100 g carbohydrate + alcohol) ‒ to see whether artificial pancreas were effective in overcoming nocturnal hypoglycemia.

As in previous studies carried out by Boris Kovatchev and others in the U.S. and France, the Cambridge closed loop system significantly increased the time that plasma glucose was in the target range (70-144 mg/dl), reduced incidence of hypoglycemia, and better overnight control.

But what makes the Cambridge study important is that the randomized crossover trial design is virtually unique in the field of closed loop control. Because this design is the gold standard for clinical research, the results set a benchmark for future studies.

The only other randomized controlled trial of closed loop control was recently presented by the University of Virginia research team led by Kovatchev at the 4th International Conference on Advanced Technologies and Treatments for Diabetes. This study recruited 24 adults and adolescents with type 1 diabetes in the United States and in France and achieved results similar to those reported by Hovorka and colleagues ‒ more time within the target range of 70-180 mg/dl and a threefold reduction in hypoglycemia.

Dr Roman Hovorka

Moreover, the control algorithm used by Hovorka and colleagues belongs to an advanced class of closed loop control technologies known as “model predictive control”. Algorithm designs for artificial pancreas have generally used either “proportional-integral-derivative control” or “model predictive control”.

Proportional-integral-derivative control algorithms are reactive, responding to changes in glucose levels with adjustment in insulin delivery. Model predictive control algorithms are built over a model of the human metabolic system and are therefore proactive, delivering insulin in anticipation of changes in glucose concentrations.

This compensates partially for the time delays inherent in subcutaneous glucose control (the time delay in insulin action, which can amount to 60 minutes or more). For this reason, model predictive control has become the approach of choice more recently.

The algorithm developed by Hovorka and colleagues has certain distinct features, such as real time adaptation of the underlying model to changing patient parameters implemented as a selection from several predefined models. However, this potential advantage remains to be evaluated.

Most importantly, this is one of the first studies to test realistic meal scenarios and challenge the participants with a large dinner that included alcohol. As such, the study is a clear advance in the quest for an artificial pancreas that can be used by a diabetic while performing normal daily activity.

However, as the authors admit, one limitation is the exclusivelymanual control of the artificial pancreas used relied on study personnel to transmit data manually from the continuous glucose monitor (CGM) to the computer running the closed loop control, and to transmit insulin injection recommendations from the computer to the insulin pump because of technological and regulatory barriers

In fully automated systems ‒ which is what researchers and medical device makers are hoping to make a reality for diabetics ‒ these processes are handled by data transmission and pump control devices, respectively. However, Cambridge method limited the investigation to testing only the control algorithm, not the artificial pancreas as a whole. The testing of other key components, such as sensor-pump communication and error mitigation, would require much more effort and thorough system validation.

Studies using fully automated systems have already been reported by the Artificial Pancreas Project and offer hope for the future of ambulatory systems i.e. devices that be worn by diabetics on their person in their daily lives.

Lastly, despite the sophistication of the control algorithm and the significant reduction in nocturnal hypoglycemia, four episodes of severe hypoglycemia (<70 mg/dl) occurred, three of which the authors thought were attributable to the preceding prandial insulin dose and could not be prevented by the artificial pancreas suspending insulin delivery.

This finding reinforces the recently proposed idea that a dedicated hypoglycemia safety system ‒ a separate algorithm responsible solely for the assessment and mitigation of the risk of hypoglycemia ‒ may need to accompany closed loop control. Such safety systems already exist, and have proved useful.

Based on ‘Boris Kovatchev: Closed Loop Control For Type 1 Diabetes (BMJ 2011; 342:d1911)

Related Post: Killer Apps That Are Revolutionizing Diabetes Care

Individualized Care Plans Important for Treating Diabetes, Says AACE

The American Association of Clinical Endocrinology (AACE) on April 14 released new clinical practice guidelines for developing comprehensive care plans for patients with type 1 and type 2 diabetes mellitus, developed by a panel of 23 of the leading diabetes experts in the U.S.

Debunking one-size-fits-all care plans, the guidelines emphasize the importance of achieving a treatment plan that avoids hypoglycemia, now considered to be a continual and pressing concern for many patients with diabetes. The implications of the new guidelines for practicing physicians, as well as new data on low blood sugar in patients with diabetes, are being discussed at the AACE 20th Annual Meeting and Clinical Congress, now in session in San Diego.

Caring Doctors Improve Patients' A1c, LDL Scores

The new AACE guidelines are also published in supplement 2 of the March/April issue of the association's official medical journal, Endocrine Practice.

The guidelines emphasize a personalized approach to controlling diabetes and achieving blood glucose targets with care plans that take into account patients' risk factors for complications, comorbid conditions, and psychological, social, and economic status. Although the guidelines recommend a blood glucose target of an HbA1c level of 6.5%, if it can be achieved safely, a treatment plan should take into account a patient's risk for the development of severe hypoglycemia.

CGM is the New Standard

The new guidelines also provide information on the appropriate use of new technologies such as insulin pumps and continuous glucose monitoring, as well as managing conditions that may not be immediately obvious to treating physicians, such as sleep and breathing disturbances and depression.

Millions Risk Early Death Because of Poor Diagnosis and Ineffective Treatment

In a statement, Yehuda Handelsman, MD, AACE president-elect and co-chair of the AACE Diabetes Guidelines Writing Committee, said that it was crucial for physicians to address not just hyperglycemia in patients with diabetes but also associated cardiovascular risk factors. "These state-of-the-art guidelines provide the most up-to-date evidence-based answers to real-life (clinical) questions," Dr. Handelsman said.

Tight Cholesterol, BP Control Does Little Good for Diabetics

In the guidelines, AACE recommends comprehensive diabetes lifestyle management education at the time of diagnosis, as well as throughout the course of diabetes. The importance of medical nutrition therapy, physical activity, avoidance of tobacco products, and adequate quantity and quality of sleep should be discussed with patients who have prediabetes, as well as type 1 and type 2 diabetes, according to the new guidelines.

Related Posts:
Killer Apps That Are Revolutionizing Diabetes Care

Diabetes: Controlling Blood Sugar Is Not Enough

Aggressive Diabetes Therapy May Raise Death Risk

Even Telephonic Intervention Improves Diabetes Control

Efficacy of HbA1c Results Questioned; CGM is the New Standard, Claim Experts

WHILE HbA1c tests have been considered the gold standard for years, these tests do not necessarily correlate to good diabetes control. But HbA1c is only an average of glucose levels over time. A person can have major highs and lows every day, yet still have a "good" HbA1c level.

This is one reason why recent clinical trials reveal that HbA1c tests alone do not necessarily correlate to good diabetic control. It is also important to know that the more often you are outside your target range, the higher your risk of diabetes-related complications.

Indeed, there are many diabetics who experience hypoglycemia and hyperglycemia so often that they no longer realize when they have symptoms. Large-scale studies have shown, over time, these glycemic excursions cause major complications to the vascular system and organs.

Writing in his Endocrine Today blog, Michael Kleerekoper, MD, MACE wonders whether HbA1c is a reliable measure of glycemic control in an individual patient. His answer is: “It depends.”

As far as I can gather, the HbA1c is reliable in patients with fairly stable home blood glucose values, but too few patients measure blood glucose often enough to know this. It’s not the patient’s fault — their insurance carrier is very niggardly about the number of times a day a patient is allowed to measure blood sugar. If only the carrier would recognize the longer-term health care costs in those patients! All too often the patient leaves the glucose meter at home — that’s like going to the pediatrician without the baby.

Serial HbA1c is far less reliable in patients with widely fluctuating blood glucose values. Continuous glucose monitoring (CGM) is being utilized more frequently to document widely fluctuating blood glucose values, and patients learn a lot from the graphical illustration of how well or not well they are taking care of their diabetes.

Point-of-care HbA1c instruments are available and are also being used more often in the clinic. It makes sense to have the data available while the patient is still in the clinic rather than wait 24 hours for the lab result to come back.

Caveat emptor! If you are of a mind to get a handheld HbA1c device for your clinic, I encourage you to first read the editorial and article published in Clinical Chemistry about 1 year ago. Only two of the eight available devices were reliable!

Kleerekoper’s reservations bring us back to the question posed in the headline of this article: Are HbA1c results reliable enough?

Goals of Diabetes Self-Management
Basically, the goal of diabetes self-management is to help patients optimally manage their diabetes based on:

• Individual needs

• Customized goals

• Personal lifestyle considerations

So, do diabetics stick with the old fingerstick monitoring or shift tocontinuous glucose monitoring (CGM)?

Improving Diabetes Management
Over a decade ago, a landmark study called the Diabetes Control and Complications Trial — the largest study of its kind — revealed the importance of good glucose control. New, large-scale studies are revealing even more improvements in glucose control with the use of CGM technology.

In a recent study, people who were considered to be in "good control" showed distinct improvements in time spent within their target zone while using a CGM device. And subjects who were considered to be in "poor control" before the study, had even more impressive improvements with a 70% increase in time they spent in their target zone.

A recent study by the Juvenile Diabetes Research Foundationshowed that using a CGM, for at least 6 days a week can lead to significant decreases in HbA1c and can provide a greater ability to reach the ADA recommended goal of 7%.

Revealing Unnoticed Highs and Lows
A CGM device can call attention to highs and lows that you may often experience, but not recognize. Sometimes people have"hypoglycemia unawareness" — they no longer feel their symptoms of going low. And some people are also unable to feel symptoms of high glucose. Over time, frequently swerving outside of your target range can lead to severe consequences, such as damage to your nerves, eyes, kidneys, and other organs.

Fingerstick vs. CGM: Revealing Unnoticed Highs & Lows

Recent studies reveal that fingerstick testing alone does not provide enough information to stay within their target range — even people who monitor frequently. In contrast, the use of a CGM device does a better job of helping people to stay between the lines. Indeed, CGM, especially in conjunction with HbA1c, helps to show a more complete picture of glucose control.

Real-Time Trending = 288 Fingersticks!
Fingerstick monitoring alone does not provide enough information for patients to act preemptively to avoid hyper- and hypoglycemic events. A major benefit of CGM is that it can help patients identify fluctuations and trends that would otherwise go unnoticed with standard fingerstick testing.

A CGM measures glucose levels and sends a glucose value every 5 minutes from the Transmitter to the Receiver. This is equivalent to a patient taking 288 fingerstick readings daily. However, unlike fingersticks, patients see real-time trending of glucose levels and patterns.

A CGM system shows patients their current glucose value, as well as their glucose trend over the past several hours (1-, 3-, 6-, 12-, and 24-hour trends). It also tells them how quickly their glucose is changing. And when glucose levels go above or below the set target range, alerts notify patients so they are able to take corrective action before going too low or too high — helping them stay between the lines.

Damage Caused By Glucose Variability
There is a growing body of clinical evidence showing that glucose variability itself — independent of HbA1c — initiates a cascade ofphysiological damage. Over time, hyper- and hypoglycemic events significantly increase the risk of diabetic complications, such as retinopathy and general microvascular pathogenesis, at least partially due to increases in oxidative stress and pro-inflammatory cytokines that are triggered by hyper- and hypoglycemic swings.

A CGM device can call attention to dangerously low overnight glucose levels that often go undetected, reveal previously unnoticed hyperglycemia spike trends between meals, show early morning highs in glucose, clarify the way diet and exercise affect your patients' glucose levels and provide a long-term comprehensive assessment of the effects of adjustments in diabetes management.

It is not surprising, therefore, that in recent years CGM has gained acceptance among diabetes experts as an effective tool for helping people achieve their diabetes goals:

• Achieve HbA1c targets without adding hypoglycemia

• Reduce glucose highs and lows

• Reduce hypoglycemia

Indeed, when using a continuous glucose monitoring (CGM) device, study results demonstrate that the patients were actually experiencing extensive fluctuations in glucose levels. CGM, especially in conjunction with HbA1c, helps to show a more complete picture of glucose control.

Diabetes: Artificial Pancreas Best Hope For Diabetics In Near Term

The realization that breakthroughs in biology-based therapies for diabetes are not imminent is spurring tech-based innovation in insulin delivery mechanisms
The days when ‘the only choice open to diabetes sufferers was that between death by coma and death by starvation’ passed unmourned into history in 1922, when insulin was first used therapeutically.

Even today, however, diabetes has lost none of its fearsomeness, because even today diabetics live in constant fear of overdosage or underdosage of their medicines, especially insulin, and of consequent hypoglycemic episodes and late complications that can result from inadequate treatment and prolonged elevation of blood glucose level.

Artificial Pancreas Project concept

Patients with diabetes whose blood glucose levels are kept close to normal by means of suitable therapeutic measures avoid the risk of dangerous hypoglycemic episodes and develop complications of diabetes considerably less frequently and later than their less successfully treated counterparts. But there is realization that breakthroughs in biology-based therapies for diabetes are not imminent. Sure, there is hope that they’re going to happen, but more long-term research is needed.

In the meanwhile, technology that can meet this need is available and is spurring innovation in insulin delivery mechanisms. As a result, diabetes technology, and particularly the artificial pancreas, has become an area of very rapid academic and industrial development.

A precondition for this success is close monitoring of blood glucose levels. Therefore, a great deal of research activity has been directed towards the development of sensors that permit near-painless, continuous measurement of blood glucose level. The objective is to develop a system that pairs continuous blood glucose monitoring with an insulin pump and thus acts as an ‘artificial pancreas’.

Why Continuous Blood Glucose Monitoring Is Desirable
It is difficult to achieve good metabolic control in diabetics. Especially in patients on intensive insulin therapy, good metabolic control calls for frequent blood glucose determinations by patients themselves. The timing and dose of insulin injections have to be adapted to a variety of factors that influence blood glucose level, such as carbohydrate intake, physical exertion, sporting activities, stress (including operations, injuries and infections) and also rest periods such as periods spent asleep.

Insulin Pump

In addition to being painful and unpleasant, individual determinations of blood glucose by patients themselves using the conventional invasive techniques provide no more than a snapshot of the patient’s blood glucose level at the moment the blood sample was taken.

Continuous glucose monitoring (CGM), by contrast, would detect fluctuations in blood glucose level over a prolonged period and indicate when major deviations from the normal range occur. Every diabetic could benefit from continuous monitoring of their blood glucose level.

What Is An Artificial Pancreas?

An artificial pancreas is essentially a device that would both measure sugar levels and dispense appropriate amounts of insulin to keep blood sugar levels in optimal range. It would take much of the guesswork out of daily management of the disease and in the long-run, controlled sugar levels will help to lessen or avert the devastating complications from diabetes.

An artificial pancreas will integrate two currently available technologies ‒ continuous glucose monitors and insulin pumps ‒ with an algorithm that provides the right amount of insulin at the right time. It will enable people with diabetes to achieve tight blood glucose control avoiding both highs and dangerous lows, thereby significantly reducing the risk of the disease's devastating complications.

Why Is An Artificial Pancreas Needed?
The current diabetes treatment market comprises three related but distinct submarkets that address different aspects of the condition. Products in the market are currently comprised of blood glucose monitors, lancets & test strips, continuous blood glucose monitors, insulin, insulin pumps, syringes, and other insulin delivery devices & anti-diabetic drugs. Right now, the most significant growth in the U.S. market is in continuous blood glucose meters, insulin pumps, and anti-diabetic drugs.

An artificial pancreas could potentially revolutionize diabetes care and management, significantly improving the ability of people with diabetes to maintain strict blood glucose control, and ‒ as a direct result ‒ helping reduce kidney disease, heart attacks and stroke, amputations, blindness, and death from severe hypoglycemia.

Extensive research shows that glucose control is the primary factor in avoiding the devastating complications of diabetes. The landmark Diabetes Control and Complications Trial (conducted 1983-1993) showed that intensive diabetes management and improved glycemic control reduces major long-term complications of diabetes.

A later study published in the New England Journal of Medicinefound that intensive diabetes therapy aimed at achieving good control reduced the risk of any heart disease event by 42 percent, and the risk of nonfatal myocardial infarction, stroke, or death from heart disease by 57 percent.

However, clinical research shows that most people with diabetes are not controlling blood glucose levels nearly well enough. The risk of complications ‒ and the economic burden placed on our health care system ‒ could be significantly lowered with devices that improve blood glucose control. And good glucose control will probably enhance the effectiveness of promising new cure therapies such as beta cell regeneration and islet transplantation.

Diabetes Technology To The Rescue
There are several classes of technologies used for diabetes care at home. In general, the devices include monitoring of blood glucose levels and delivery of insulin. The monitoring can happen with finger sticks or with the newer generation of continuous glucose monitoring that permanently attach to the person.

Insulin delivery, the old-fashioned way, is through injections several times a day. The newer devices are insulin pumps. They attach to the person with little needles under the skin that deliver insulin at continuous rates.

But the current problem with even the most advanced treatment of diabetes is that these devices don’t talk to each other. Even the most sophisticated insulin pumps will keep delivering insulin regardless of the blood sugar level of the person because it doesn’t have any information coming from the monitors. That can cause severe reactions.

Dr Boris Kovatchev

Boris Kovatchev, director of the Center for Diabetes Technology at the University of Virginia, has focused on diabetes technology for more than a decade. His contribution to this area is to make these devices talk to each other in a smart fashion, to insert an algorithm that can take the reading from the monitoring device and tell the insulin pump to deliver insulin in a smart way.

The Artificial Pancreas Project is the most advanced application of the device. It connects the most advanced continuous monitoring device to the most advanced insulin pumps available in a continuous fashion — all the time. This is the top of the line integrated technology — and that means continuous monitoring of the person’s blood sugar levels and the reaction to changes.

The system's "smart" algorithm, developed by Kovatchev’s team and collaborators from the University of Padova in Italy, uses existing continuous glucose monitoring and insulin pump technology to automatically regulate a patient's insulin levels, with no action required on behalf of the user. The algorithm is currently being tested in clinical trials at the U-Va. Health System and 10 other centers spanning seven countries.

This project ‒ initiated by the Juvenile Diabetes Research Foundation (JDRF) ‒ has been going on since 2006. Kovatchev’s group was one of the first to join this project. Since then, it has grown significantly. Now, there are government initiatives in several countries and a lot of companies are interested in this technology.

The U-Va group has undertaken some extensive clinical trials on around 60 patients on that system for short periods of time for testing purposes. Besides, there are ongoing clinical trials in eight countries, all using Kovatchev’s patented technology or components of it.

The most challenging aspect of this technology is predicting the future. The reason that the future must be predicted accurately in this particular technology is that insulin delivery under the skin and glucose monitoring under the skin have delays.

“The monitoring of the blood sugar level generally works with a one-hour delay. Imagine you’re driving a car and you’re reacting to oncoming traffic with a one-minute delay. In situations like that, you have outdated data and delayed action. You have to anticipate what is going on in the next hour,” explains Kovatchev.

A device developed by Kovatchev and collaborators to automatically regulate a patient's insulin levels as part of the Artificial Pancreas Project (right) is shown with a continuous glucose monitor (left)

To make his device predict the near future for diabetes management, Kovatchev is developing a class of control strategies called model predictive control. Every human is assigned a mathematical model that mimics the functioning of the metabolic system of that person.

“Based on what that model says, we can predict the future. It’s similar to weather forecasts. But instead of weather models, we have models of a particular human,” he says with confidence.