S. J. Sullivan

Islet Transplantation With Immunoisolation

Immunoisolation is a potentially important approach to transplanting islets without need for immunosuppressive drugs. Immunoisolation systems have been conceived in which the transplanted tissue Is separated from the immune system of the host by an artificial barrier. These systems offer a solution to the problem of human islet procurement by permitting use of Islets Isolated from animal pancreases. The devices used are referred to as biohybrid artificial organs because they combine synthetic, selectively permeable membranes that block immune rejection with living transplants. Three major types of biohybrid pancreas devices have been studied. These include devices anastomosed to the vascular system as AV shunts, diffusion chambers, and microcapsules. Results In diabetic rodents and dogs indicate that biohybrid pancreas devices significantly improve glucose homeostasis and can function for more than a year. Recent progress made with this approach is discussed, and some of the remaining problems that must be resolved to bring this technology to clinical reality are addressed.

Treatment of Severely Diabetic Pancreatectomized Dogs Using a Diffusion-Based Hybrid Pancreas

Long-term survival of dog islet allografts implanted in diabetic pancreatectomized dogs was achieved by islet encapsulation inside cylindrical chambers fabricated from permselective acrylic membranes (nominal Mr exclusion of 50,000–80,000). Dog islets were isolated from the pancreases of outbred mongrel dogs by collagenase digestion. Chambers containing mean ± SE 316 ± 63K islet equivalents (mean islet volume, 558 ± 111 mm3, purity 90–95%) were peritoneally implanted into six totally pancreatectomized dogs. The dogs were monitored for glycemic control by fasting and postprandial blood glucose determinations, and responses to both intravenous glucose (intravenous glucose tolerance test 0.5 g/kg) and oral glucose (oral glucose tolerance test 1 g/kg). All of the dogs required appreciably lower dosages of exogenous insulin therapy for control of fasting blood glucose levels, with the mean daily insulin dose dropping from 38 ± 7 to 5 ± 1 U/day during the 1st wk. Three recipients required no insulin for >82, >68, and 51 days. Intravenous glucose tolerance test K values (decline in glucose levels, %%min) at 1 and 2 mo postimplantation were 2.7 ± 0.4 and 2.0 ± 0.5, respectively compared with 3.5 ± 0.5 before pancreatectomy. The glucose values during oral glucose tolerance tests at 2 wk, although returning to < 125 mg/dl (< 7.0 mM) by 2 h, exceeded the normal range, with peak values of 174 to 202 mg/dl (9.7 to 11.3 mM). These preliminary results are encouraging, and represent an important step in determining the feasibility of using this type of diffusion-based hybrid artificial pancreas as treatment for diabetes mellitus in humans.

Transplantation of islet allografts and xenografts in totally pancreatectomized diabetic dogs using the hybrid artificial pancreas

Previously the authors reported on a Hybrid Artificial Pancreas device that maintained patent vascular anastomoses in normal dogs and, when seeded with allogeneic canine islets, maintained normal fasting blood sugars (FBS) in diabetic pancreatectomized dogs. Eventual failure of these devices was believed to be related to loss of islet viability and/or insufficient islet mass. The current study evaluates the effect of increased islet mass produced by implantation of two islet-seeded devices in pancreatectomized dogs and compares the results with those from dogs that received a single device. Twelve of fifteen dogs receiving single devices showed initial function as determined by elimination or reduction of exogenous insulin requirement; four showed initial function and seven showed extended function (100 to 284 days). Excessive weight loss (more than 20%), despite normal FBS and insulin dependence, required that four animals in this latter group be killed. Devices seeded with xenogeneic islets have met with limited success. One dog that received two bovine islet-seeded devices achieved function for more than 100 days; the remaining bovine-seeded devices (n = 8) functioned for only 3 to 16 days. Porcine islet-seeded devices were assessed by intravenous glucose tolerance tests (IVGTT). Recipients of two devices seeded with allogeneic islets demonstrated improved IVGTT results when compared to those from pancreatectomized dogs and recipients of single devices but were abnormal when compared to intact animals. Histologic examination of device and autopsy material from all failed experiments was performed and showed no mononuclear cell infiltration of the islet chamber or vascular graft material, only a few incidence of device thrombosis, and varying degrees of islet viability as judged by morphologic and immunohistochemical evaluation. The authors believe they have demonstrated progress toward the development and clinical applicability of the Hybrid Artificial Pancreas.

TREATMENT OF SEVERE DIABETES MELLITUS FOR MORE THAN ONE YEAR USING A VASCULARIZED HYBRID ARTIFICIAL PANCREAS

We report the successful application of a hybrid artificial pancreas device for the treatment of severe diabetes mellitus induced by total pancreatectomy in two dogs. Control of the blood sugar was achieved for more than 1 year in these two animals without any immunosuppressive therapy. Although exogenous insulin was required therapy. Although exogenous insulin was required during the latter part of the study period, removal of the devices resulted in a rapid increase in the fasting blood sugar levels and the exogenous insulin requirements (P < 0.001 versus weeks 1-52 in both dogs). Metabolic studies, postexplant in vitro studies, and histologic analyses confirmed islet cell survival and insulin production by the devices. This hybrid artificial pancreas has a clear clinical potential for islet cell transplantation without immunosuppression.

Successful treatment of diabetes with the biohybrid artificial pancreas in dogs.

We have investigated a new hybrid artificial pancreas device to transplant islet allografts without immunosuppression. The device consists of a chamber through which passes a copolymer membrane connected to standard vascular grafts. Islets are placed inside the chamber but are outside of the blood stream. Nominal molecular porosity of 80,000 daltons permits free diffusion of nutrients and insulin across the membrane but inhibits the entry of immunoglobulins and immunocytes from the blood stream into the chamber. Initial studies focused on the technical feasibility of implanting the unseeded (no islets) devices. In 12 normal mongrel dogs, the arterial limb of the device was anastomosed end-to-end to the common iliac artery and the venous limb end-to-side to the common iliac vein. Vascular patency was monitored by an audible bruit over the device. Two devices currently remain patent at 388 and 421 days. The remaining experiments failed due to thrombosis and membrane rupture, with 2 failing as late as 170 and 279 days. In a second series, both arterial and venous anastomoses were done end-to-side and dogs were placed on low-dose aspirin therapy. All 8 dogs are currently maintaining patent unseeded devices (96-226 days postimplantation). Subsequent studies determined the function of devices seeded with isolated canine pancreatic islet allografts in totally pancreatectomized, severely diabetic dogs. Diabetes was controlled by once-a-day insulin injection. After 2-3 weeks of diabetic control, a seeded device was implanted. Diabetic control was monitored by fasting blood levels and postprandial and intravenous glucose tolerance tests, and vascular patency by the loudness of the bruit. In the first series of 6 dogs given seeded devices without aspirin, no significant function was discernible, with failure attributable to thrombosis, poor islet viability, and surgical complications. In the second series of 13 dogs given aspirin, 8 dogs have required an appreciably lower dose of injected insulin to maintain fasting blood glucose at acceptable levels. Of note are 4 dogs that required virtually no exogenous insulin for at least 3 weeks. One dog lost function on day 74 and another still requires no insulin at 267 days postimplantation. However, despite normal fasting glucose levels, the glucose tolerance tests showed delayed return to normal levels. Weight lost following pancreatectomy was rapidly regained in the presence of a functioning seeded device. Histologic examination of the removed devices revealed no signs of rejection.

Islet transplantation using an immunoprotective membrane in dogs that have undergone a pancreatectomy

The use of a selectively permeable membrane to transplant nonsyngeneic tissue without accompanying immunosuppressive therapy has been investigated using two approaches. The first hybrid artificial pancreas is implanted as a vascular shunt in which blood circulates through the lumen of a tubular membrane. The islet tissue is distributed within a chamber surrounding the membrane enclosed by an acrylic housing. Studies with diabetic dogs that have had pancreatectomies have demonstrated that these devices could replace exogenous insulin therapy for at least 6 months in five animals. This report presents data on two of these dogs, demonstrating viability and function of the transplanted tissue after 1 year. As an alternative to the vascular device, islets sealed within cylindrical permselective membrane chambers have been implanted in the peritoneum. Preliminary data from three dogs indicate that the nonvascular implants can also regulate fasting glucose levels in the diabetic dog model.