Has Time Come for New Goals in Human Islet Transplantation?


  • R. Lehmann,

    Corresponding author
    1. University Hospital Zurich, Islet Transplant Program
    2. University Hospital Zurich, Endocrinology, Diabetes, and Clinical Nutrition & Competence Centre for Systems Physiology and Metabolic Diseases
      * Corresponding author: R. Lehmann, roger.lehmann@usz.ch
    Search for more papers by this author
  • G. A. Spinas,

    1. University Hospital Zurich, Endocrinology, Diabetes, and Clinical Nutrition & Competence Centre for Systems Physiology and Metabolic Diseases
    Search for more papers by this author
  • W. Moritz,

    1. University Hospital Zurich, Visceral Surgery and Transplantation
    Search for more papers by this author
  • M. Weber

    1. University Hospital Zurich, Islet Transplant Program
    2. University Hospital Zurich, Visceral Surgery and Transplantation
    Search for more papers by this author

* Corresponding author: R. Lehmann, roger.lehmann@usz.ch


The enthusiasm regarding clinical islet transplantation has been dampened by the long-term results. Concerns about the associated risks of life-long immunosuppression and the striking imbalance between potential recipients and available donor pancreata warrant changes in some of the current goals.

Islet transplantation will never be a cure of type 1 diabetes in the majority of patients with no secondary complications, but is a valid option for a limited number of patients with brittle diabetes waiting for an organ or after organ transplantation. Furthermore, insulin independence should not be the main goal of islet transplantation, but avoidance of severe hypoglycemia and good glycemic control, which can be achieved with a relatively small functional beta-cell mass. Therefore, initially one islet infusion is sufficient. Retransplantation at a later time point remains an option, if glucose control deteriorates.

Efforts to improve islet transplantation should no longer focus on islet isolation and immunosuppression, but rather on the low posttransplant survival rate of islets caused by activation of the coagulation pathway and the limited oxygen delivery to the islets. Transplantation of smaller islets be it naturally small or size tailored reaggregated islets has the potential to facilitate these processes.


The revived enthusiasm regarding clinical islet transplantation as a cure for type 1 diabetes mellitus after the landmark Edmonton Trial (1) has declined over the past few years. As pinpointed in a recent editorial (2) the islet cell transplant world faces a series of concerns in a complex procedure:

  • 1Does the risk of severe hypoglycemia justify an expensive procedure with life-long immunosuppression, which—as reported by the Edmonton group—may put the recipient's kidney or life at risk after islet transplantation, even if the actual risk is low, (3) or jeopardizes the outcome of a future kidney transplantation by sensitization (4)? This concern is particularly relevant when considering the low rate of insulin independence 5 years after islet transplantation (5).
  • 2Is it justified to transplant islets of several organ donors in the face of organ donor shortage (6) and the imbalance between newly diagnosed patients with type 1 diabetes and the number of performed islet transplantation worldwide? These concerns severely limit the general feasibility of the current indications for islet transplantation and warrant changes in at least three current goals (Table 1).
Table 1.  Current and new goals in islet transplantation
Current goalsNew goals
Main indication of islet transplantation: Patients with type 1 diabetes with no or minimal secondary complications in order to prevent hypoglycemia and diabetes complicationsMain indication of islet transplantation: Patients with type 1 diabetes and end stage nephropathy waiting for a kidney transplant or after kidney or other solid organ transplantation
Major goal of islet transplantation: Insulin independence (generally several donor pancreata required)Major goal of islet transplantation: Avoidance of severe hypoglycemia and good metabolic control in order to protect transplanted kidney and to reduce CVD morbidity and mortality (generally one donor pancreas required)
Main research focus of islet transplantation: Improvement of islet isolation outcome and development of new immunosuppressive strategiesMain research focus of islet transplantation: Improvement of survival and better engraftment of transplanted islets
Key factors:
Selection and production of smaller islets to improve oxygen diffusion
Inhibition of coagulation cascade
Strict avoidance of inhibitors of engraftment
Coating of islets or alternative sites for islet transplantation

Islet transplantation will never cure the majority of patients with type 1 diabetes with no secondary complications, but it is a valid option for a limited number of patients with brittle diabetes waiting for or after organ transplantation

The current benchmark for islet transplantation is whole pancreas transplantation; a procedure which nowadays results in a 3-year insulin independence rate of 80% (http://www.iptr.umn.edu/IPTR). While initially the results of pancreas transplantation alone were clearly inferior to simultaneous pancreas kidney transplantation, recent results have improved to an encouraging 5-year graft survival rate of 78% (7). With islet transplantation, on the other hand, insulin independence is not sustainable over time in the majority of patients (7), although a 5-year insulin independence rate of up to 50% is achieved in some centers. Since isolation efficacy is comparable in most centers, higher insulin independence rate is achieved at the cost of highly selective donor and recipient criteria, thus reducing the number of suitable donor organs. If all the pancreata that are being used for pancreas transplantation in the USA were used for islet transplantation in patients with severe recurrent hypoglycemia, which is currently the main indication for islet transplantation, only every 30th to 40th patient, that is 3–5% of 1.5 million people with type 1 diabetes in the US would benefit from this treatment, not to mention the 30 000 new cases of type 1 diabetes each year. On the other hand, for patients who are listed for or already bear a kidney transplant the risks associated with immunosuppression are acceptable, because immunosuppression is needed for the kidney transplant. Each year, about two-third of the ∼4000–5000 patients with type 1 diabetes and end-stage renal disease receive a kidney transplant in the United States (6). It would be feasible to offer a combined islet-kidney transplant for almost all these patients. Therefore, selecting the ideal candidate is crucial. In our opinion, patients with brittle diabetes and end-stage kidney disease may represent the best candidates, as this group of patients evidently benefits from kidney transplantation and will require life-long immunosuppression. An important caveat is the high risk of sensitization after failed islet alone transplantation, which might preclude from being considered for kidney transplantation in the future (4). A randomized controlled trial by the CIT group of the NIH will compare islet-kidney transplantation with kidney transplantation alone plus medical therapy to assess the beneficial aspects of islet transplantation in this group of patients.

Insulin independence is not the main goal of islet transplantation, but glucose stability, as defined by avoidance of severe hypoglycemia and near-normal glycemic control

In spite of encouraging 1-year graft survival data, long-term insulin independence can apparently not be achieved with islet transplantation based on the Edmonton protocol (1,8) Yet, even partial graft survival was associated with a reduction of severe hypoglycemia. Thus, a realistic goal for islet transplantation could be the conversion from a ‘brittle’ diabetes state into a more easily manageable disease. Such a view is supported by a quality of life assessment in patients after islet transplantation. Even though insulin independence was mentioned as desirable, it was not a fundamental element in the patients' well-being. The most frequently reported beneficial effect was stability of glucose control and absence of hypoglycemic episodes resulting in a feeling of independence and reliability, not experienced before the transplant (9). So far, a careful risk-benefit assessment of islet transplantation including C-peptide positivity (achieved in 80% of patients at 5 years), good metabolic control, the long-term consequences and side effects of life-long immunosuppression, has not been done. In most published islet trials a proper control group, that is intensively state of the art treated patients (experienced diabetes specialist, nutritionist, diabetes educators, CSII, glucose sensors) is missing, and therefore a direct comparison of positive and adverse effects of both treatment options is not possible. Nevertheless, the interim results of the Vancouver group comparing patients with type 1 diabetes receiving islet transplantation or intensive insulin treatment have shown that median HbA1c was significantly better in the islet group (6.6% vs. 7.3%). The rate of severe hypoglycemic episodes was, however, not assessed in that study (10).

A recent comparison between islet and pancreas transplantation in combination with a kidney transplant revealed that the glycated hemoglobin levels that can be achieved with islet transplantation are almost the same as with pancreas transplantation, even if a majority of patients have to inject small amounts of insulin (11). The required insulin dose over the whole follow-up period is directly correlated with the transplanted islet volume, whereas the glycated hemoglobin is not (Figure 1). A minimal amount of endogenous insulin production is required for glucose stability, however. Thus, we think that retransplantation should be considered only, if the primary goal, stability of glucose control, defined as near-normal HbA1c and absence of hypoglycemia, is no longer met. Such a strategy would spare pancreata for other patients waiting for an islet transplant, and seems justified as the 3- and 5-year insulin independence rate is not different if one, two or three islet infusions were applied (5). Yet, this is somehow at odds with the CITR data (web.emmes.com/studies/isl) suggesting a longer survival of islet grafts with a greater initial islet mass that is sufficient for insulin independence. Korsgren et al. suggest that the reduction in endocrine function might well depend on the side effects of immunosuppression, but is also reinforced by glucose and lipid toxicity, once the islet mass is reduced to 20–40% normal (12). The choice of strategy is inherently dependent on goal 3.

Figure 1.

Relationship between mean HbA1c or insulin dose and number of transplanted islets in 18 consecutive simultaneous islet-kidney transplant recipients at the University Hospital Zurich (2000 to 2007). Glucose stability, defined as a near-normal glycated hemoglobin (blue squares) without severe hypoglycemia, was maintained over the whole posttransplant follow-up period (mean: 3.7 years; range 0.7–7.5) either by transplanting a large islet number (= insulin independence) or a smaller number with additional exogenous insulin (= partial graft function), reflected by the significant correlation (r =−0.62; p = 0.006) between transplanted islet number and exogenous insulin (red dots) with no correlation (r =−0.14; p = 0.6) being found between mean HbA1c and islet number (blue squares).

Efforts to improve islet transplantation should focus less on islet isolation and immunosuppression, but rather on improvement of survival rate of the transplanted islets

The isolation procedure is straightforward and almost standardized worldwide (8). Yet, changes of critical components of islet isolation, such as the switch from one collagenase brand to another can severely influence clinical outcome. It is obvious that with the limited number of islet transplantations performed worldwide, that is <150 per year, the 30 currently proposed immunosuppressive protocols can never be successfully tested. The success of the Edmonton protocol is based rather on the high volume of infused islets and avoidance of steroids, than on the particular choice of immunosuppressive drugs. Yet, despite a high number of transplanted islets, that is 12 000 IEQ/kg recipient weight corresponding to a normal beta-cell mass (13), the functional capacity of the transplanted islets was only 20–40% of that in nondiabetic persons (14). The original hope that the diabetogenic effects of steroids and the toxic effects of calcineurin inhibitors can be avoided by the use of sirolimus as key immunosuppressant in the Edmonton protocol, has been offset by many adverse effects of sirolimus. Of particular concern are its antiangiogenic effects linked to impaired VEGF production (15), which occur at concentrations that correspond to the currently recommended sirolimus levels (16).

Thus, in order to improve the outcome of islet transplantation, measures that promote islet engraftment or prevent the rapid apoptosis/necrosis after transplantation need to be envisaged. The fate of islets in the immediate posttransplant period is not only determined by the instant blood-mediated immune response (IBMIR), but also by the hypoxic state in the portal field. Normally, pancreatic islets have a blood perfusion that is 10 times higher than in the exocrine pancreas (17) resulting in a significantly higher oxygen tension. During the process of isolation and in vitro culture the islet vasculature dedifferentiates or degenerates and, immediately after transplantation, the islets are supplied with oxygen and nutrients solely by diffusion. Even after revascularization the oxygen tension of transplanted islets in the portal blood is 10 times lower than in islets in the native pancreas (17). Avoidance of culture has been shown to increase vascular density, at least in a syngeneic transplantation model (18). This might be due to the presence of donor endothelial cells, which disappear during culture (19).

We propose a new, simple, and up to now neglected parameter to improve islet engraftment, namely islet size. Islet size has been shown to be of importance for in vitro and in vivo function (20). Also with respect to vascular engraftment, small transplanted rat islets (<100 μm) had twice the vascular density of large islets (21) and survive better after cryopreservation (22). Based on our data we suggest an optimal islet size of 50–100 μm to improve the chance for survival in a hypoxic milieu. As, evidently, large islets cannot be discarded, they may be transformed into small ‘pseudoislets’ by the hanging drop technology, as recently shown in a feasibility study (23). The simple parameter, islet size, might influence many other factors such as more distal or more proximal embolization in the portal field with different diffusion distances or a more favorable site for revascularization (12), higher percentage of beta-cells, better oxygenation and, therefore, more favorable outcome (20,24). In fact, islet size might prove to be a key factor in quality assessment. It must be noted, however, that assessment of both size and volume of islets is far from being standardized. A prospective trial to compare the assessment of the islet volume in the form of islet equivalents as compared to counting the islet number by state-of-the-art technology (e.g. Coulter Technique, COPAS macrosorter, image analyzing systems) is warranted. Replacing islet equivalents by islet numbers might have significant consequences: by using islet isolations with an islet volume, which so far has been considered insufficient for a good clinical outcome, but with an adequate islet number, more isolations could be used for transplantation, which is a contribution towards reducing the burden of organ donor shortage and may result in economic savings. Presently, new technologies are being developed to transform islets into the desired size range (23), to coat islets with heparin (25) or to find alternative sites (omental pouch (26), intramuscular site (27)) for clinical transplantation that may improve islet transplantation outcome considerably.


The scope of this personal viewpoint was to challenge current goals in islet transplantation. With the present expectations in beta-cell replacement the importance of islet transplantation for the treatment and cure of type 1 diabetes is overestimated as is the engrafted islet volume. Thus, the results and clinical outcome of this treatment modality appear to be disappointing. We, therefore, propose a reappraisal of three current goals in islet transplantation and propose patients with type 1 diabetes, who wait for or already bear a transplanted organ as primary candidates for organ transplantation. As long as islet engraftment is not considerably optimized to consistently achieve insulin independence with one islet transplant, insulin independence should not be the primary goal of islet transplantation, but glucose stability and good glucose control. The major focus of research should be to improve islet engraftment and long-term survival of islets. Thus, one islet transplant might be all that is required for glucose stability and insulin independence in the future.


We gratefully acknowledge the support by the Hans-Paul Waelchli Foundation.