Diabetes mellitus (DM) is increasing worldwide¹ and now affects up to 5% of the population in the UK. Roughly 10% of these patients have type I DM, caused by insulin deficiency secondary to autoimmune destruction of pancreatic islet cells. DM is associated with life-threatening metabolic or vascular complications in 30% of patients². According to data from the UK Renal Registry, 20% of all new patients in the UK under the age of 65 years requiring dialysis treatment have end-stage renal failure secondary to DM. In addition to renal care, patients with diabetes require a diverse range of services including cardiology and cardiac surgery, vascular surgery and ophthalmology.

The Diabetes Control and Complications Trial (DCCT)³ showed that tight glycaemic control delays and reduces diabetic complications. However, intensive insulin treatment is poorly tolerated by many patients and will decrease the number of patients who develop microvascular complications by no more than 30-40%. Further, a small but substantial number of patients have life-threatening hypoglycaemic episodes despite scrupulous attention to their insulin regimens. Therefore, to improve diabetic care, the need is for treatments that achieve metabolic stability and prevent microvascular complications. Reports from Professor Shapiro's team describe major improvements in the early clinical outcome of patients with type I DM treated with human islet cell transplantation by newly developed protocols. Here we discuss the key areas they report that contribute to improvements in the outcome of islet cell transplantation, particularly the use of novel immuno-suppressive strategies.

Until recently, islet cell transplantation was less successful than whole pancreatic transplantation, which achieves insulin independence at one year in 80% of patients. However, the latter is an invasive procedure with a substantial morbidity⁴. In the UK it is performed in only a few centres and usually in patients with end-stage renal failure who receive a simultaneous kidney transplant. As a result, less than 10% of available cadaveric pancreases are transplanted. In 1999 just 78 patients were on the waiting list for pancreatic transplantation, and 69 of these were listed to receive simultaneous kidney and pancreas transplants⁵.

Results from the International Islet Transplant Registry⁶, held in Giessen, Germany, show why, until recently, there has been little enthusiasm for establishing islet cell transplantation in routine clinical practice. 445 adults with type I DM have received transplants, 355 in the past 10 years. Of patients transplanted between 1990 and 1996, 70% lost all islet cell function within the first year. 12.4% of patients remained insulin independent beyond 1 week, 8.2% beyond 1 year and only one recipient for longer than 5 years. The reasons for these poor results have been reviewed by Hering and Ricordi⁷.

In June 2000 Shapiro and colleagues reported their results with islet cell transplantation by percutaneous transhepatic portal venous embolization and a new clinical protocol. They had treated seven patients (median age 44 years; range 29-54; type I DM for a median of 35 years) who were troubled by repeated episodes of severe hypoglycaemia. At the time of reporting, follow-up was 11.9 months (range 4.4-14.9). All seven patients became insulin independent, although two briefly required insulin during intercurrent illnesses. To achieve insulin independence, all had required a second islet infusion from another donor pancreas a median of 29 days (range 14-70) after the first procedure and one, the most obese, had required a third. The patients were in hospital for a median of 2.3 days (range 0.5-14.7). After transplantation, glycosylated haemoglobin levels did not rise and serum C-peptide concentrations did not decrease over time. No patient experienced further episodes of life-threatening hypoglycaemia.

The Edmonton group have performed islet cell transplants according to this protocol in sixteen patients with a one-year insulin independence rate of 80% (Shapiro J, personal communication). In addition four have received single-donor transplants with a protocol including infliximab. There is some evidence that this anti-TNFα antibody improves engraftment; all four of these patients are positive for C-peptide.

Why should these early results from this group be so much better than those in the registry data? Several changes were introduced simultaneously, so their individual contributions cannot be assessed, but Shapiro and colleagues believe the key factors to be as follows.

Daclizumab (Zenapax) is a humanized anti-CD25 monoclonal antibody used for prophylaxis against acute rejection in solid organ recipients. Cell surface expression of CD25, the alpha chain of the IL-2 receptor, is a key event in the activation of T cells following alloantigen recognition. Blocking of this target in solid organ transplantation leads to a 40% reduction in the incidence of acute allograft rejection without increasing rates of cytomegalovirus or other tissue invasive infections or post-transplant lymphomas. A chimeric anti-CD25 antibody called basiliximab (Simulect) matches the efficacy of daclizumab in solid organ transplantation. Daclizumab is given immediately before transplantation and every two weeks after transplantation for four more doses¹⁰. The Edmonton group repeats this dose regimen for subsequent islet transplants. In the randomized control trials with anti-CD25 antibodies there are no reports of a cytokine release syndrome.

Tacrolimus (FK 506) binds to an intracytoplasmic immunophilin to produce a complex that inhibits calcineurin, a key intracellular signalling protein activated after T cell receptor ligation. This interaction inhibits IL-2 gene expression and subsequent T-cell activation. Solid organ transplant recipients treated with tacrolimus have a reduced incidence of acute rejection and possibly a reduced incidence or delayed onset of chronic graft rejection¹¹. The Edmonton group used low-dose tacrolimus to maintain a trough concentration at 12 hours of 3-6ng/mL. In combination with daclizumab and sirolimus this provided adequate immunosuppression whilst avoiding dose-related complications such as hypertension, diabetes (due to diminished beta cell function), and nephrotoxicity.

Sirolimus (rapamycin) binds to intracytoplasmic binding proteins to form an intracellular complex that prevents T cell proliferation¹² by inhibiting a key regulatory kinase. In kidney transplantation the use of sirolimus substantially reduces the occurrence of acute rejection¹³. Preclinical studies in islet cell transplantation point to extended allograft survival¹⁴. In cyclosporin-based regimens, the use of sirolimus increases serum triglycerides and cholesterol in 40-50% of patients. However, of the patients transplanted in Edmonton, only one had an increase in cholesterol (mild). Probably the avoidance of steroids and the use of tacrolimus in low dosage lessens the risk of dyslipidaemia. Although there is concern about nephrotoxicity with sirolimus, it has not been evident with this regimen.