Reversal of Diabetes by Pancreatic Islet Transplantation into a Subcutaneous, Neovascularized Device Antonello Pileggi, 1,2 R. Damaris Molano, 1 Camillo Ricordi, 1,2,3 Elsie Zahr, 1 Jill Collins, 1 Rafael Valdes, 4 and Luca Inverardi 1,3,5 Background. Transplantation of pancreatic islets for the treatment of type 1 diabetes allows for physiologic glycemic control and insulin-independence when sufficient islets are implanted via the portal vein into the liver. Intrahepatic islet implantation requires specific infrastructure and expertise, and risks inherent to the procedure include bleeding, thrombosis, and elevation of portal pressure. Additionally, the relatively higher drug metabolite concentrations in the liver may contribute to the delayed loss of graft function of recent clinical trials. Identification of alternative implan- tation sites using biocompatible devices may be of assistance improving graft outcome. A desirable bioartificial pan- creas should be easy to implant, biopsy, and retrieve, while allowing for sustained graft function. The subcutaneous (SC) site may require a minimally invasive procedure performed under local anesthesia, but its use has been hampered so far by lack of early vascularization, induction of local inflammation, and mechanical stress on the graft. Methods. Chemically diabetic rats received syngeneic islets into the liver or SC into a novel biocompatible device consisting of a cylindrical stainless-steel mesh. The device was implanted 40 days prior to islet transplantation to allow embedding by connective tissue and neovascularization. Reversal of diabetes and glycemic control was monitored after islet transplantation. Results. Syngeneic islets transplanted into a SC, neovascularized device restored euglycemia and sustained function long-term. Removal of graft-bearing devices resulted in hyperglycemia. Explanted grafts showed preserved islets and intense vascular networks. Conclusions. Ease of implantation, biocompatibility, and ability to maintain long-term graft function support the potential of our implantable device for cellular-based reparative therapies. (Transplantation 2006;81: 1318–1324) I n type 1 diabetes mellitus (T1DM) the insulin-producing -cells in the pancreatic islets are destroyed by an autoim- mune process. Tight glycemic metabolic control by intense exogenous insulin treatment can reduce or prevent the pro- gression of diabetes complications (1). Restoration of -cell function by transplantation of allogeneic pancreatic islets represents a promising therapeutic option for patients with T1DM, providing a more physiologic metabolic control than exogenous insulin (2–6). Numerous studies have attempted to identify the most suitable site of implant for islet grafts (7, 8), and clinical islet transplantation is currently performed by intrahepatic embo- lization through the portal system (2–6). Even if the safety of this technique has been established, specific infrastructure and expertise are needed. Risks inherent to the procedure are bleeding, thrombosis and elevation of portal pressure (5, 9, 10). Repeated infusions of islets isolated from more than one donor pancreas might increase the risks associated with the transplantation procedure. The inherent hyperglycemic envi- ronment and the relatively higher concentrations of drug me- tabolites in the liver (11, 12) may also contribute to the de- layed loss of graft function observed in recent clinical trials (4, 5), possibly related to -cell toxicity (13). Development of alternative implantation sites for pancreatic islets may be of assistance to reduce risks and improve the success rate of islet transplantation. Tissue engineering approaches for the development of bioartificial organs, including the pancreas, have included in- vasive surgical procedures that are difficult to implement in the clinical setting, such as subcapsular kidney space, spleen (8), intra-abdominal vascularized bio-hybrid devices (14, 15), omental pouch (16), excluded intestinal segments (17) and intramuscular (18), amongst others. The subcutaneous site seems advantageous because of the potentially less inva- sive procedure that could be performed under local anesthe- sia. However, subcutaneous islet implantation has been ham- pered so far by inherent limitations of this site (lack of early vascularization, induction of local inflammation, and me- chanical stress on the graft) resulting in primary non-func- tion (7, 19 –22). Subcutaneous implantation of biocompat- ible materials (i.e. diffusion chambers, hollow fibers) for tissue and cellular transplantation has been performed, but disadvantages included degradation of biomaterials and gen- eration of intense avascular fibrotic reactions limiting oxygen A.P. and R.D.M. equally contributed to this work. This study was supported by the Diabetes Research Institute Foundation (Hollywood, FL). 1 Diabetes Research Institute, University of Miami Leonard M. Miller School of Medicine, Miami, FL. 2 Department of Surgery, University of Miami Leonard M. Miller School of Medicine, Miami, FL. 3 Department of Medicine, University of Miami Leonard M. Miller School of Medicine, Miami, FL. 4 Hospital Infantil de Me´xico Federico Go´ mez, Departamento de Cirugı ´a y Trasplantes, Facultad de Medicina UNAM, Me´xico DF, Me´xico. 5 Address correspondence to: Luca Inverardi, M.D., Cell Transplant Center, Diabetes Research Institute, University of Miami Leonard M. Miller School of Medicine, 1450 NW 10th Avenue (R-134), Miami, FL 33136. E-mail: linverar@med.miami.edu Received 13 September 2005. Accepted 22 December 2005. Copyright © 2006 by Lippincott Williams & Wilkins ISSN 0041-1337/06/8109-1318 DOI: 10.1097/01.tp.0000203858.41105.88 1318 Transplantation • Volume 81, Number 9, May 15, 2006