Artificial pancreas will soon be tested on diabetics

TRENTON, N.J. -- An experimental artificial pancreas will soon be tested in diabetes patients, potentially sparing them the hormonal disorder's most dangerous complications and frequent blood sugar checks and insulin injections.

If all goes well, a commercial model could be on the market in four years, said Aaron Kowalski, research director of the Juvenile Diabetes Research Foundation's artificial pancreas project.

The device, being developed by the foundation and health care giant Johnson & Johnson, potentially could help about 6 million diabetics in the U.S. alone who use insulin, Kowalski said.

It could prevent the life-threatening seizures that can occur when blood sugar drops too low, as well as blindness, amputations and organ damage caused by years of too-high blood sugar. And it would end having to track carbohydrate intake and then calculate the insulin doses needed all day long. That's because the device would constantly measure blood sugar and its computer would decide when to give the patient more insulin.

The research partnership, which also includes glucose monitor maker DexCom Inc., was announced Wednesday.

"It's been a goal for 40 years" at the foundation, Kowalski said. "The pieces of the puzzle are in place ... I'm confident this is going to work."

The foundation has been working with researchers at major universities in recent years testing the components of what will be the artificial pancreas -- different types of insulin pumps, continuous blood glucose monitors and software -- in various combinations to see what works best, Kowalski said in an interview Tuesday.

Johnson & Johnson's Animas Corp. unit, which sells the OneTouch Ping insulin pump, and DexCom will use data from that research to produce a commercial model. A device about the size of a cell phone would be worn outside the body, linked to a tube of about three-eighths of an inch under the skin to hold insulin for release as needed. The first patient testing could begin in less than a year, Kowalski said.

"Four years is probably doable," given the project's "running start" and J&J's skill and experience in developing new products and working with regulators, said Erik Gordon, a professor and analyst at University of Michigan's Ross School of Business. "It sounds promising."

Analyst Steve Brozak of WBB Securities also expects approval because the Food and Drug Administration has been "exceedingly friendly" in reviewing diabetes products because the complications are so severe and better treatments to prevent them are needed.

"You're looking at a multibillion-dollar-a-year product," Brozak predicted, unless some "landscape-changing development" such as an insulin pill comes along.

Gordon gave a more conservative estimate, pegging sales at perhaps $500 million in the first year and growing to $1 billion a year or more over time.

The foundation is providing $8 million in funding, and J&J will spend an undisclosed amount, including providing equipment and personnel.

Currently, patients with Type 1, or insulin-dependent diabetes, and many with advanced Type 2 diabetes must inject themselves with insulin in the stomach two, three or more times a day to keep their blood sugar in a safe range. Some patients have an insulin pump tethered to their waist so they can avoid the injections. But even those patients must calculate how much insulin they need before each meal, based on how many grams of carbohydrates they expect to eat.

In Type 1 patients, their immune system is attacking and killing cells in the pancreas that produce insulin, the hormone that helps the body convert food into energy. In Type 2 diabetes, which comprises about 85 percent of cases, the body doesn't make enough insulin or doesn't use it efficiently.

The experimental artificial pancreas would not cure diabetes, which has exploded along with obesity into an epidemic -- dubbed diabesity -- in the U.S. and other developed countries. But the device could enable patients to control diabetes much better, preventing complications that can lead to frequent hospitalizations and sometimes early death.

That's important because research has shown the average Type 1 diabetes patient has blood sugar too high or too low about 70 percent of the time, Kowalski said.

About 25 million people in the U.S. and millions more elsewhere have diabetes. As the number using insulin grows with expanding waistlines, more and more Type 2 patients will be candidates for the product, Brozak noted.