Islet Sheet Research

We believe that a therapy that produces euglycemia without risk of hypoglycemia and without immune suppression drugs will be widely accepted as a cure for diabetes.

The chart below shows identified approaches that might accomplish such a cure. First, a mechanical artificial pancreas comprises a glucose sensor, an insulin pump with a reservoir and a computer to determine the pump rate. The chief difficulty with this approach has been that no glucose sensor is sufficiently sensitive, stable, accurate and safe. A significant risk of hypoglycemia remains.

Second, transplantation, of "naked" islets of Langerhans is a proven cure with the "Edmonton protocol" (at the cost of lifetime immune suppression). Transplantation of islets protected by some sort of membrane barrier is called the bio-artificial pancreas because it combines living tissue with an implanted device.

Bio-artificial pancreases are categorized by the means that nutrients get to the islets inside. A vascular shunt device conducts blood through the device. Such devices have a poor safety record. Diffusion devices derive nutrients passively from adjacent vascularized tissues. Our thin sheet is a particular form of a diffusion-based device.

We have identified the approaches we believe are likely to succeed in green. The Edmonton protocol is already proven. Three others, the bio mechanical pancreas, macroencapsuled islets and the Islet Sheet, should be proven in the coming five years.


Transplantation of Islets of Langerhans

Replacement of the cells that naturally produce insulin in physiologically appropriate amounts is obviously an attractive approach. The July 2000 New England Journal of Medicine published a breakthrough study from the University of Alberta showing that transplantation of islets of Langerhans using improved immune suppression eliminates the need to inject insulin with a success rate of 100%. Because of the known toxicity of anti-rejection drugs the procedure was limited to use on patients who had hypoglycemia unawareness, a dangerous complication of diabetes.

Extension of the University of Alberta's "Edmonton Protocol" to the vast majority of people with diabetes is impossible because of the side effects of chronic immune suppression. (A further limitation of the "Edmonton Protocol" is the availability of human pancreases.) Either improved immune therapy or a physical barrier to protect the implanted islets is needed. The principal authors of the Edmonton work, Drs. Shapiro and Lakey, are working with Islet Sheet Medical to investigate our physical barrier, the thin sheet bioartificial pancreas.

Physical Barrier: The Bio-Artificial Pancreas

The Islet Sheet bioartificial pancreas contains live, functional islets in an artificial polymer matrix. The advantages of the bioartificial pancreas include:

  • No Immune Suppression Required
  • Xenografts (use of animal tissue) Possible
  • More Types of Genetically Engineered Tissue Possible
  • Retrievable
  • Replaceable

Comparison of Various Transplantation Approaches

Each method for treating diabetes with islet transplantation requires (1) insulin-secreting tissue and (2) a means to prevent rejection of the insulin-secreting tissue. Whether islets are freshly prepared from a donor pancreas (primary islets), expanded through cell culture or made through genetic engineering, all must be combined with either a protective polymer ("Islet Sheet") or toxic immune suppression drugs.

The highly successful "Edmonton Protocol" uses primary islets with immune suppression drugs. The Islet Sheet protects islets (including primary islets) without immune suppression drugs. All islet transplants must use either immune suppression drugs or a physical barrier such as the Islet Sheet.

Islet Sheet Medical is investigating implants of primary animal islets into diabetic animals. These studies can be used to predict clinical success. The market for implanting Islet Sheets into people with diabetes will be limited by availability of human donor organs. The Company anticipates using other sources of tissue such as pig islets or cultured human islets to meet the demand for islet transplantation without immune suppression.

Developing the Thin Sheet Bioartificial Pancreas

Many ventures have been founded in the past 20 years to attempt profiting from the opportunity for the bioartificial pancreas. More than $200 million has been invested. To date, no product is on the market.

The technical requirements for a bioartificial pancreas are exacting and have proven very difficult to satisfy. Why has the problem proven so difficult to resolve? Because the islets inside these previously designed devices die. Often the surface of the bioartificial pancreas provokes a foreign body response. The resulting fibrotic reaction walls off the device, and the islets cannot receive nutrients. Another problem has been that the dimensions of most bioartificial pancreases do not permit oxygen to penetrate to the core of the device and the islets there die. The process used to make the device can damage or destroy islets. And sometimes islet cells are exposed to the host and then trigger a proliferating immune response.

In contrast to prior approaches, the Islet Sheet has been designed from the "islet's point of view" for maximum islet viability. The first problem in designing a workable device was identifying a suitable polymer for the "artificial" part of the device. In the early 1990's several researchers (including Randy Dorian, a founder of Islet Sheet Medical) concluded that the best polymer is ultra-purified alginate. Alginate solutions quickly gel when contacted with calcium salts. However, only with great effort are all impurities removed so that alginate is "bio-invisible." One group, including the founders of Islet Sheet Medical, developed methods for small microcapsules using a combination of electrostatic micro-drop formation, overcoating with a spinning disc atomizing device, and crosslinking with imaginative ion exchange chemistry. A animal treated with these capsules was off insulin for nearly seven years.

Beginnings of Islet Sheet Research

To address several shortcomings of microencapsulation, in 1995 the founders of Islet Sheet Medical began work on a thin, retrievable sheet intended to be a second-generation device. We manufactured prototype sheets and studied their properties, including strength, thickness and diffusion characteristics. We filed our first patent application on October 13, 1995.

Islet transplant studies require large animals such as animals and pigs, and are expensive. To keep costs low all animal research is done at academic institutions with licenced animal research facilities.

In June 1998 we raised $300,000 in "angel" money (after spending $80,000 from the founders) to do experiments designed to show that we could make sheets that fit the criteria for the perfect passive diffusion bioartificial pancreas. Experiments began at the University of Chicago in September 1998, and moved to the University of Cincinnati and the University of Alberta in 2000. The first clear demonstration of the efficacy of the Islet Sheet took place in 2000. Large animal studies are the company's principal activity at this time.


Design Objectives

We set out to make a better bioartificial pancreas, and determined the essential design objectives tabulated below. The diagram illustrates the design concept which satisfies these objectives.

1. Keep the Islets Alive
  • Material Contacting Islets Is Biocompatible
  • Process for Fabrication must Not Damage Islets
  • Dimensions Permit Rapid Diffusion of Nutrients
2. Prevent Destructive Host Response
  • Outer Surface Is Totally "Bioinvisible" (Provoking No Fibrotic Response)
  • All Islets must Be Completely Covered (To Prevent Immune Sensitization)
  • Permeability Is Controllable
3. Assure Practical Surgical Implantation
  • Islet Density High (Sheet Size Not Too Large)
  • Device Retrievable
  • Bio-stable Long-term
  • Chemically & Physically Durable

Islet Sheet Fabrication

This electron micrograph, taken in April 2001 at the University of Alberta shows the perfected Islet Sheet. All of the components are clearly visible: the islets, the core alginate, the coating alginate, and the mesh. In the micrograph one can see two large islets and several small ones. The mesh is colored blue and is just under the alginate outer layer on the bottom of the Sheet. This photo and others show that the outer layers are perfect: not one islet or cell was exposed. (During processing for SEM the layers separate.)

Pilot Studies in Large Animals

In June 1998 we began large animal studies with the Islet Sheet. Evaluation of biocompatibility took place at commercial laboratories in California and the University of Minnesota. In September 1998 at the University of Chicago our collaborator Prof. Rilo isolated islets and implanted test sheets into large animals for the first time. Several significant problems came to light and were overcome. For example, the best material for the reenforcing mesh proved elusive. Our first choice, a nylon mesh, was too stiff and caused the sheets to delaminate. We tried no reenforcement at all, and the sheets rolled up like a jelly roll. Finally, after implants into five large animals and many rodents, Islet Sheets implanted into a large animal in May 1999 were shown to fulfill the design criteria. Sheets retrieved after 12 days in the diabetic animal produced insulin in response to stimulation, proof that the islets were alive in the retrieved sheets.

In 1999 Prof. Rilo moved to the University of Cincinnati, and we worked in our San Francisco laboratory to further improve sheet fabrication. We resumed our large animal research in 2000 with him at the University of Cincinnati Medical School. In an important experiment a large animal had a total pancreatectomy (to induce diabetes) and six sheets with canine islets from two unrelated donors were attached to the large animal's omentum with four sutures each. (The omentum is a well-vascularized apron of tissue hanging over and in front of the stomach. We had chosen the omentum because of its accessibility and because of experiments showing that conventional islet transplants work when the islets are secured in an omental purse or "taco.") Eighty-four days after implantation the sheets and adjacent omentum were removed. Examination of the sheets before fixation showed viable islet cells containing insulin. From the day of implantation through the day of explantation the fasting (morning) blood sugars were all less than 140. This indicates that the Islet Sheet was producing enough insulin to cure the large animal's diabetes. After the sheets were removed, the glucose levels rose gradually until full diabetes was evident a week post explant.

This metabolic result was exciting, but it was clear from the condition of the Islet Sheets and the omentum that the omental site is not good: there was a significant tissue reaction. Sections of the sheets that remained flat did not provoke a fibrotic response. It was encouraging that islets performed well in the sheets – well enough to cure diabetes for weeks – but such a foreign body reaction would make the Islet Sheet clinically worthless.

Most materials provoke a fibrotic response, but highly purified alginate does not, so the exterior surface of the Islet Sheet is highly purified alginate. Also, our methods were refined so that the sheet was smooth on a microscopic scale. Sheets placed in the highly fibrogenic rat subcutaneous space produced minimal fibrosis. It appeared that the foreign body reaction we had seen on the large animal omentum was caused by the polyester mesh becoming exposed in cracks where the sheet folded. We decided to test this hypothesis in large animals by comparing the omentum with other sites.

In October 2000, in our first large animal experiment at the University of Alberta, four non-diabetic animals were implanted with six sheets each, on the omentum, mesentery, pancreas, liver, diaphragm and under the skin. Two large animals were examined at 2 ½ weeks and two at nine weeks. The experiment showed what we had hoped to find: if the sheet remained flat, it produced minimal foreign body reaction. All four sheets in the omentum produced masses like the one seen in the previous metabolic large animal. The less flexible sites performed well.

For example, sheets sutured to the diaphragm produced little fibrotic response. The sutures (blue) can be seen. The sheet is so transparent that blood vessels can be clearly seen. After nine weeks the sheet is clean (below). Electron microscopy confirmed that the first step of the tissue reaction, deposition of fibrin and collagen, did no occur when the sheet remained flat.

We would like to emphasize how extraordinary this result is. We are not aware of any material other than highly purified alginate that would produce substantially no fibrosis at nine weeks in the large animal peritoneal cavity. If a fibrotic response occurs, it starts within two weeks and is well advanced in two months. These results considerably boost our confidence that the Islet Sheet will function for a long time, if it is kept flat.

Having demonstrated that the liver, pancreas, diaphragm and perhaps the subcutaneous space all accepted aceuular sheets with minimal fibrosis, in March 2001 we went to Edmonton to do an allograft using the islets from two donor animals. The Edmonton team made a lot of islet tissue for us; sheet fabrication went well as did the surgery.

The sites where sheets remain flat and free of fibrosis are the liver, pancreas, diaphragm and perhaps under the skin. Each represents a unique surgical challenge. No surgical methods exist for sheet implantation because no such device has been available before. We anticipate that during 2001 a series of animal experiments will determine which of these sites is suitable for sheet implantation and how the sites compare.

Preclinical and Clinical Studies

For the FDA and Canadian authorities we will need to demonstrate the safety and stability of the product, individual components, and the final formulation. We will also be required to develop a set of standards to measure the reproducibility of different batches, including islet function.

The Islet sheet will be regulated as a combination biologic/device, and will be characterized as a biologic, so an Investigational New Drug (IND) application will be required. It will make most sense to have separate IND's for different cell types (human allograft islets vs. xenograft islets vs. cell lines etc.). Following IND approval (expected 30 days from submission) clinical trials will begin. Islet trials have an excellent surrogate end point, namely euglycemia. It will not be necessary to prove that vascular problems improve. We believe that the Pre-Market Approval (PMA) will be submitted after two years' clinical trials.

The FDA has previously approved clinical investigation of encapsulated islets, blazing the trail for these types of devices. We believe, because of the nature of our product, that the preclinical and clinical programs should run together, especially if we use human islets, because of their proven safety. Therefore, initial studies will be with sheets protecting allografts, that is, human primary islets.

We are developing a detailed clinical-regulatory plan in collaboration with Peter Hammonds, our regulatory consultant.

Other Sources of Insulin-Producing Tissue

Primary human islets encapsulated in the Islet Sheet should be our first profitable product, but will be limited by the availability of human pancreases (approximately 5000 per year). Because of this, research on cells grown in culture will be pursued as well as genetically engineered cells as well as animal cells.

Biotechnology methods have the potential to produce fully functional, "immunologically human" islets in larger quantities and at a lower cost than primary human islets. We are already in collaborative studies with various organizations producing such islets and cells.