• Islet transplantation;
  • transplantation cell therapy;
  • transplantation research;
  • type I diabetes


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This prospective phase 1/2 trial investigated the safety and reproducibility of allogeneic islet transplantation (Tx) in type I diabetic (T1DM) patients and tested a strategy to achieve insulin-independence with lower islet mass. Ten C-peptide negative T1DM subjects with hypoglycemic unawareness received 1–3 intraportal allogeneic islet Tx and were followed for 15 months. Four subjects (Group 1) received the Edmonton immunosuppression regimen (daclizumab, sirolimus, tacrolimus). Six subjects (Group 2) received the University of Illinois protocol (etanercept, exenatide and the Edmonton regimen). All subjects became insulin- independent. Group 1 received a mean total number of islets (EIN) of 1460 080 ± 418 330 in 2 (n = 2) or 3 (n = 2) Tx, whereas Group 2 became insulin- independent after 1 Tx (537 495 ± 190 968 EIN, p = 0.028). All Group 1 subjects remained insulin free through the follow-up. Two Group 2 subjects resumed insulin: one after immunosuppression reduction during an infectious complication, the other with exenatide intolerance. HbA1c reached normal range in both groups (6.5 ± 0.6 at baseline to 5.6 ± 0.5 after 2–3 Tx in Group 1 vs. 7.8 ± 1.1 to 5.8 ± 0.3 after 1 Tx in Group 2). HYPO scores markedly decreased in both groups. Combined treatment of etanercept and exenatide improves islet graft function and facilitates achievement of insulin-independence with less islets.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Although major advances in insulin treatment have improved glycemic control, hypoglycemic unawareness remains a major obstacle for optimal glycemic control in many diabetic patients (1). Even under strict study conditions, the incidence of severe hypoglycemia remains high; this complication not only impairs quality of life but also is life-threatening (2,3).

Islet transplantation can eliminate severe hypoglycemia and restore almost normal glycemic control (4). However, there are several limitations to the widespread application of this procedure (4). These limitations include the need for several donors for sufficient islet yield, lifelong immunosuppression and its attendant risks, and loss of islet function over time due to incompletely defined reasons.

At present, clinical trials in islet transplantation face stringent federal regulations that define islets as a biological drug and islet transplantation as an experimental procedure. Limited resources make fully powered, larger-scale trials prohibitive. Despite limitations, collaborative efforts and single-center experiences are improving results and expanding the body of knowledge as a whole. Herein, we present a phase 1/2 trial investigating whether islet transplantation utilizing a protocol designed at the University of Illinois at Chicago is safe and reproducibly successful. The study explored traditional therapy in the first arm, and then tested a regimen of a TNFα receptor antagonist (etanercept, Enbrel®) and a glucagon-like peptide-1 (GLP-1) analogue (exenatide, Byetta®) to explore whether we could improve islet engraftment with a marginal islet dose (5). Exenatide is FDA-approved for treatment of type II diabetes mellitus (6–8). Recently, exenatide was used to decrease need for exogenous insulin and restore first-phase insulin release in islet recipients who lost insulin-independence after transplantation (9). Herein, we explored the de novo use of a combination of etanercept and exenatide to improve allogeneic islet transplant outcomes in brittle type I diabetic patients.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Study design

This prospective Phase 1/2 trial was conducted under FDA IND 11 807 and approved by the Institutional Review Board. In this single-center, open label, uncontrolled trial we performed 1–3 allogeneic pancreatic islet transplants per subject and followed up for 64 weeks (∼15 months) after transplant. This follow-up period was determined to avoid exceeding the maximum acceptable number of blood draws per individual at any time point and to include all safety and metabolic parameters reported herein. Thereafter, subjects were enrolled in a 5-year follow-up study.

The first objective was to demonstrate the safety of allogeneic islet transplantation in type I diabetic patients performed at the University of Illinois at Chicago (UIC). Therefore, the first four patients (Group 1) were transplanted according to the Edmonton protocol. The second objective was to implement a strategy to achieve insulin-independence with a minimal quantity of islets. Therefore, the next six consecutive patients (Group 2) were transplanted using the UIC protocol detailed below.

Primary efficacy endpoint

The primary efficacy endpoint was independence from insulin injections with adequate control of blood glucose. Insulin-independence was defined as freedom of exogenous insulin injection while achieving fasting glucose levels not exceeding 140 mg/dL more than three times in a week, and not exceeding 2-h postprandial values of 180 mg/dL more than four times in a week. The proportion of insulin-independent subjects meeting the criteria for glucose control was determined at 2 weeks and 1, 3, 6, 12 and 15 months after their final transplants.

Subjects who had reduced insulin requirements but did not achieve insulin-independence and presented reduced HbA1c and number of hypoglycemic episodes were considered to have partial islet graft function.

Absence of measurable levels of C-peptide (<0.03 ng/mL) after transplantation was considered islet failure.

Secondary endpoints

  • • 
    HbA1c was compared between the last value before transplantation and 3, 6, 12 and 15 months after islet transplant. HbA1c reaching normal values of less than 6.5% by 3, 6, 12 and 15 months after first transplant and continuing for 12 months after the final transplant was considered a success.
  • • 
    Fructosamine was compared between the last value before transplantation and 3, 6, 12, and 15 months after islet transplant.
  • • 
    Oral glucose tolerance tests (OGTT) were conducted 6 and 12 months postislet transplant. OGTT was judged normal if blood glucose was below 140 mg/dL after 2 h, impaired if blood glucose was between 140 and 199 mg/dL, and diabetic if blood glucose was above 199 mg/dL.
  • • 
    Mixed meal tests (MMT) were conducted before transplantation and 6 and 12 months posttransplant. Acute C-peptide response and blood glucose levels to a standard mixed meal test were compared.
  • • 
    Glucagon stimulation test (GST) values were measured before transplantation and 6 and 12 months posttransplant. The fold-increase in C-peptide levels 6 min after glucagon injection was evaluated.
  • • 
    Intravenous glucose tolerance test (IVGTT) values were taken at 6 and 12 months post transplantation. We calculated acute insulin response to intravenous glucose challenge (AIR-IVGGT) as an indicator of islet mass (10).
  • • 
    Hypoglycemia was evaluated by the Ryan HYPO score (11) before and 12 months after transplantation. To calculate the HYPO score, subjects performed blood sugar self-assessments for 4 weeks during the 6 months pretransplant and 12 months after transplantation. Points were awarded for each occurrence of documented hypoglycemia and for neuroglycopenic symptoms.
  • • 
    Glycemia status: Subjects measured and recorded blood sugar values seven times per day before and after transplantation: fasting, morning, postprandial, noon, afternoon postprandial, evening and evening postprandial.

Inclusion criteria

Subjects were selected according to the Edmonton protocol criteria (12). Enrolled subjects had type I diabetes mellitus for more than 5 years, complicated by either hypoglycemic unawareness, metabolic lability with documented severe hypoglycemia, or ketoacidosis despite state-of-the-art-intensive insulin therapy (13).

Exclusion criteria

Subjects were excluded from the study if they had coexisting cardiac disease, history of nonadherence to prescribed regimens, BMI >26 kg/m2 or body weight >70 kg at screening visit, creatinine clearance <80 mL/min/1.73 m2 by 24-h urine collection, insulin requirement >0.7 IU/kg/day or HbA1c >12%.

Screening of the study subjects

From November 2004 to May 2006, 88 individuals completed phone screenings, of whom 60 presented with exclusion criteria and 15 withdrew after screening because of the complexity of the study or potential risks involved (Table 1). Thirteen subjects were listed to receive an islet transplant, but three were excluded with unremediated medical conditions when recruitment closed, and the remaining 10 subjects received islet transplants.

Table 1.  Screening outcomes and subject exclusions*
Candidate/Subject statusPhone screen n = 88Pre-screenPhase I ScreenPhase II ScreenTotal
Excluded by screening4412 4 060
Withdrew from screening 8 5 0 215
Subjects excluded after listing for transplantNANANANA 3
Subjects active3619151310
*Reasons for exclusion (60) 
 Age > 65 years1
 Diabetes manageable with intensive insulin therapy6
 Daily insulin requirement >0.7 IU/kg7
 Diabetes <5 years2
 Kidney disease7
 Previous organ transplant2
 Weight >70 kg or BMI >2631 
 Reported psychiatry disease/nonadherence to therapy1
Referral to a center closer to home1
Reasons for withdrawal after phone screening (15) 
 Desire for children1
 Potential risk of procedure/immunosuppression7
 Family illness1
Reasons for exclusion after listing for transplant (3) 
 Unable to control weight1
 New onset kidney1
 New onset heart disease1

Donor selection

Potential donors were screened at the time of organ offer. Donors with positive serology results (except cytomegalovirus), diabetes or suspected diabetes, sepsis, cancer or methanol toxicity were ruled out. Safety screening included criteria defined in the suitability determination for donors of human cellular and tissue-based products final rule and the Centers for Disease Control high-risk criteria for blood-borne pathogens. Quality screening exclusion criteria included donor age < 25 or > 75 years, warm ischemia, cold ischemia > 12 h, non-heart beating donors, and BMI < 19. Negative cross-match with the donor was required for transplant. We routinely performed flow cytometeric T- and B-cell cross-match with a cut-off at 20 channel shifts for T and 40 channel shifts for B cells.

Islet isolation

The pancreata were trimmed, distended with cold Liberase solution (Liberase-HI, Roche, Indianapolis, IN), and digested using a modified Ricordi automated method (14). After collection and wash, the tissue was incubated in UW solution (DuPont Pharma, Bad Homburg, Germany) for 30 min. Islets were purified by a continuous UIC-UB gradient (15) on a cell separator (Cobe 2991, Cobe, Lakewood, CO) and then cultured in Culture Media (Mediatech, Herndon, VA) at 37°C for up to 12 h.

Islet evaluation

Islet yield:  We assessed quantity and purity of the preparations by Dithizone staining (16).

Viability:  The percentage of dead and live cells was estimated by fluorescent staining with Syto-Green/Ethidium Bromide (17,18).

In vitro function:  Islet function was expressed as a stimulation index (SI) after static incubation with low and high glucose conditions (19).

Islet transplantation

After the portal vein was accessed percutaneously under fluoroscopic and ultrasound guidance, the islets were resuspended into 60 mL syringes and slowly injected into the intraportal catheter. During infusion, syringes were turned constantly to avoid sedimentation or clumping of islets. Heparin was administered throughout the procedure for a total dose of 5000 units per transplant followed by enoxaparin (Lovenox®) 30 mg subcutaneously twice daily for 1-week posttransplant.

Study medication

Edmonton protocol group 1:  Based on the Edmonton protocol, subjects received an immunosuppressive regimen of daclizumab, sirolimus and tacrolimus (12). Subjects received daclizumab 1 mg/kg intravenously immediately before transplant and 75 mg at 2, 4, 6 and 8 weeks after transplant. Sirolimus was administered at a loading dose of 0.2 mg/kg immediately pretransplant, and continued at 0.1 mg/kg/day each morning. Tacrolimus 1 mg was administered immediately before transplantation, and then adjusted to maintain target trough levels of 3–6 ng/mL throughout the study (Figure 1).


Figure 1. Immunosuppression drug levels for both groups during 15 months follow-up. Sirolimus dose was adjusted to maintain target trough serum blood levels of 12–15 ng/mL for the 3 months following the most recent islet infusion and lowered to 7–10 ng/mL subsequently. Target trough level for tacrolimus was 3–6 ng/mL throughout the study. Average AUC for tacrolimus level was 62.4 ± 14.1 in Group 1 versus 48.4 ± 8.1 in Group 2 (p = 0.14). *This patient developed sirolimus complications and switched to mycophenolate mofetil.

Download figure to PowerPoint

UIC protocol group 2:  In addition to the Edmonton protocol immunosuppression regimen, the UIC protocol included the TNF-alpha receptor antagonist, etanercept and the glucagon-like peptide-1 (GLP-1) analogue, exenatide. Patients received etanercept 50 mg intravenously before islet transplantation and 25 mg subcutaneously at 3, 7 and 10 days after transplant. Exenatide 5 μg was administered subcutaneously twice daily for 1 week within 60 min before or after the morning and evening meals. If tolerated well, the dose was increased to 10 μg twice daily for a total of 6 months after the last islet transplant.

Autoantibody measurement

Antibodies were measured before and after transplant. Alloantibodies were assessed by flow cytometry measurement of class I and II panel reactive antibodies (One Lambda Inc, Los Angeles, CA). Anti-islet antibodies were assessed by indirect immunofluorescence (normal titers <1:4), and anti-GAD antibodies by radioimmunoassay (normal values 0–70 mGAD-U/mL).

Challenge test with or without exenatide

In Group 2 subjects, we measured frequently sampled intravenous glucose tolerance tests (FSIVGTT), C-peptide, proinsulin, amylin and glucagon levels with and without exenatide. Four subjects underwent FSIVGTT at least 12 h after withdrawal of exenatide. Subject 5 was not tested because of illness from infection subsequent to myonecrosis, and subject 9 did not tolerate exenatide. In a second session, the same subjects injected exenatide before testing. Group 1 subjects underwent only IVGTT without exenatide. We calculated acute insulin response to glucose as the mean of the 3-, 4-, and 5-min insulin values following the glucose injection subtracting the basal value (20). Insulin sensitivity index was calculated using the computer MINMOD mathematical model for FSIVGTT (21).

Statistical analysis

Data were expressed as mean ± SD. Comparisons were performed using two-tailed t-test, Paired t-test or Fischer's exact test where applicable. SPSS 13.0 (Chicago, IL) was used to run analyses. A p-value < 0.05 was considered statistically significance. Unless otherwise indicated, all data represent 15 months follow-up after the first islet transplant.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References


Recipients:  Ten C-peptide negative diabetic subjects (nine females, one male) received 18 islet allo- transplantations over 2 years (Table 2). Subjects in Groups 1 and 2 had similar age distribution (49.2 ± 11.3 and 44.8 ± 10.0 years, p = 0.6), diabetes duration (30.2 ± 9.2 and 27 ± 10.8 years, p = 0.6), body weight (61.03 ± 4.27 and 63.8 ± 5.30 kg, p = 0.2), BMI (22.4 ± 1.08 and 22.35 ± 1.01, p = 0.4) and pre-transplant insulin requirements (39.3 ± 2.9 and 32.1 ± 8.6 U/day, p = 0.5). HbA1c was significantly higher in Group 2 than Group 1 (6.5 ± 0.7 and 7.8 ± 1.1 p = 0.046). GAD autoantibody levels were also more elevated in Group 2 than Group 1 (41.8 ± 60.3 vs. 2.3 ± 2.1, p = 0.24). It should be noted that 2/6 patients in Group 2 had GAD levels greater than 100.

Table 2.  Pretransplant characteristics for the 10 C-peptide-negative islet transplant recipients with hypoglycemic awareness
CharacteristicRecipient No.Mean(SD)
Group 1Group 2
  1. BMI = body mass index; PRA = panel reactive antibodies.

Age, years6156433548.8 (11.9)39463762364744.5 (9.7)
Body weight, kg54.96660.761.760.8 (3.9)71.461.8606266.358.663.4 (4.3)
Body mass index22.921.020.823.021.9 (1.0)22.322.518.623. (1.8)
Diabetes duration, years3637321730.5 (9.3)29342110142422.0 (9.1)
Daily insulin, U/Kg3543414440.8 (4.1)44264626384237.0 (8.9)
HbA1c5. (1.0)
Hyposcore2,668207218274842 (1217.9)1892,576228717931,020812.8 (941.7)
Alloantibodies (class I & II PRA)00000037000
 • Anti-GAD655.441.11.03.342.7 (1.8) (74.4)
 • Anti-ICA512/++/

Donors and islet graft characteristics

Donor and graft characteristics for both groups are outlined in Table 3.

Table 3.  Donor graft characteristics for the 10 C-peptide-negative islet transplant recipients with hypoglycemic awareness
TransplantGroup 1Group 2
  1. CVA = cerebrovascular accident; MVA = motor vehicle accident.

 Age, years593736304737434657513064303543414960
 Body weight, kg10010074918676110143.413082.110557.593.11249410811898
 Body mass index29312830.430.5243245.337283924.737.534.539.1383534.8
 Blood glucose 
   • Admission15016114625719521015715222121013112112218097138150148
   • Peak167252378257195227233187221477451261159190155150162148
   • Preharvest16720314222710722718112417615845126113817912515095126
 HLA-A/B matches010011201211100001
 HLA-DR matches02101100120020101
 Cold ischemia time, hr8.5876.2129.2549.211110126.349.587.1110.1156.2
 Tissue volume, mL7.986.
 Total islet equivalent652 000378 000386 000405 000569 000450 000372 000716 0001 094 000921 000417 000377 000635 000633 000450 968883 000482 000462 000
 Islet equivalent/kg11 24167986942613686217413612811 60417 73114 9275840618010 58310 550727313 31872697884
 Islet purity,%503055605070505060653070609070406045
 Islet viability,%908090958587908095908090959290909090
 Stimulation index8.
 Endotoxin, EU/kg0.910.460.

Safety of procedure and medications

Procedure-related events:  Two bleeds occurred in 18 islet infusions (11% of total islet infusions, 20% of total subjects). One subject experienced an intraperitoneal bleed during withdrawal of the intraportal catheter, which was self-limiting and did not require transfusions or surgical intervention. A second subject with situs viscerum inversus, left isomerism, variant portal venous anatomy and several small hemangiomas developed an intrahepatic hematoma probably resulting from inadvertent puncture of a small hemangioma coupled with peri-procedural heparin. This subject received two units of packed red blood cells for low posttransplant hemoglobin, and received an additional three units after an unlucky blunt abdominal trauma provoking an intrahepatic rebleeding 6 weeks posttransplant. No interventions were needed for any of the islet transplant recipients. No portal vein thrombosis was observed.

Medication-related events:  All subjects lost weight with a mean decrease from 62.3 ± 4.5 kg to 59.3 ± 5.6 (p = 0.2) after islet transplantation. BMI was reduced significantly from 22.5 ± 1.2 to 21.6 ± 1.5 (p = 0.04). Group 2 lost insignificantly more weight than Group 1 (3.4 ± 2.2 vs. 2.5 ± 3.1 kg, p = 0.6).

Most subjects could tolerate the immunosuppressive regimen. One recipient in Group 2 developed viral stomatitis, severe anemia and elevated creatinine that resolved after switching from sirolimus to mycophenolate mofetil (22). As a cohort, there was no significant change in kidney function (0.78 ± 0.17 at base line to 0.93 ± 0.29 mg/dL after 15 months for Group 1, p = 0.11 and 0.97 ± 0.2 to 1.07 ± 0.41 for Group 2, p = 0.35). However, one subject in Group 1 and two subjects in Group 2 showed some increase in creatinine from the baseline (Figure 2). All recipients (10/10) developed transient anemia.


Figure 3. HbA1c levels for both groups during 15 months follow-up (top panels). Exogenous insulin requirement (U insulin/day) during 15 months follow-up for both groups (lower panels). All patients became insulin- independent when adequate islets were provided.

Download figure to PowerPoint

All subjects (6/6) in Group 2 experienced temporary nausea and/or vomiting and weight loss. Two subjects in Group 2 could not tolerate exenatide, one with preexisting diabetic autonomic gastroparesis, and the other with persistent vomiting and weight loss. Another subject in Group 2 developed myonecrosis of the right scalene muscle attributed to methicillin sensitive staph aureus (MSSA) vertebral osteomyelitis (C7-T1) that resolved with antibiotic therapy and reduced immunosuppression. This subject developed partial graft failure and resumed insulin therapy. Ultimately, she withdrew from the study because of medication side effects. One subject in Group 1 underwent an abdominal hysterectomy for irregular menstrual bleeding and ruptured ovarian cyst possibly related to immunosuppressive medications.

Primary efficacy endpoint

Insulin-independence:  All subjects completed the 15 months posttransplant follow-up and became insulin-independent after one or more islet transplants (Figure 3). In Group 1, all subjects remained insulin-free for the 15-month posttransplant observation period. Group 1 subjects received a mean total number of islets (EIN) of 1460 080 ± 418 330 in 2 (n = 2) or 3 (n = 2) islet infusions to achieve primary insulin-independence (Figure 4). All patients in Group 2 became insulin-independent after the first transplant (537 495 ± 190 968 EIN, p = 0.028 compared to Group 1). Four of the six Group 2 subjects remained insulin-independent at the end of the 15 months follow-up. In Group 2 two subjects resumed insulin. One subject had partial graft failure at 19 weeks posttransplant when immunosuppression was reduced during an infectious complication (scalene myonecrosis and MSSA vertebral osteomyelitis). The second subject, who had severe preexisting gastroparesis and ongoing nausea, vomiting and weight loss, discontinued exenatide and resumed insulin 17 weeks post transplant. This subject received a second transplant and again became insulin-independent, but resumed insulin 5 months posttransplant. One subject experienced an increase in HbA1c to 6.5% and received a second islet preparation to avoid resuming insulin injections. Ultimately, subjects in Group 2 received a mean total EIN of 723 328 ± 461 446 in 1 (n = 4) or 2 (n = 2) islet infusions (Figure 4). During the 15 months observation period, none of the 10 patients in either group presented with complete graft failure.


Figure 2. Creatinine levels for both groups during 15 months follow-up.

Download figure to PowerPoint


Figure 4. Total number of islets expressed as Islet Equivalent required to achieve insulin-independence in both groups. (A) Total number of islets transplanted in both groups during the entire study period (B).

Download figure to PowerPoint

Secondary endpoints

HbA1c:  HbA1c reached normal range in both groups, decreasing from 7.2 ± 1.1% at baseline to 5.9 ± 0.4% (p = 0.001) after transplant. HbA1c decreased from 6.5 ± 0.6 at baseline to 5.6 ± 0.2 after at least two transplants in Group 1. HbA1c reached normal levels for subjects in Group 2 after one islet transplant, decreasing from 7.8 ± 1.0 at baseline to 6.1 ± 0.3 (Figure 3). It should be noted that the baseline HbA1c was significantly higher in Group 2 than Group 1 (p = 0.046).

Fructosamine:  Fructosamine reached normal levels after transplant for subjects in both groups, decreasing from 296.7 ± 44.2 μmol/L at baseline to 244.8 ± 36.6 μmol/L (P = 0.01) after transplant.

Oral glucose tolerance test:  At 12 months posttransplant, one subject in Group 1 showed diabetic oral glucose tolerance, one subject had borderline glucose intolerance, and two subjects showed a normal glucose tolerance. In Group 2, three subjects had impaired glucose tolerance, and two had diabetic glucose tolerance. One subject resumed insulin and was not tested.

Mixed meal test:  The MMT showed a nondiabetic response to glucose stimulation testing with appropriate C-peptide response in all Group 1 subjects and in 4/5 of Group 2 subjects.

Glucagon stimulation test:  At entry into the study, all subjects demonstrated absence of C-peptide (< 0.3 ng/mL) in response to GST and MMT. Posttransplant, all subjects became C-peptide positive and maintained normal levels of C-peptide for the entire study duration. In Group 1 all subjects showed significant increase of C-peptide after glucagon stimulation. In Group 2, 4 subjects underwent GST. One subject had a low response, though initial glucose level was low, which is known to reduce C-peptide response. All subjects demonstrated increased insulin production in response to glucagon stimulation.

Intravenous glucose tolerance test:  IVGTT showed a similar glucose disappearance pattern in all patients. However, insulin response varied among individuals, with most presenting an intermediate acute insulin response (10).

Hypoglycemia:  All subjects entered the study with multiple episodes of severe hypoglycemia without awareness. During the follow-up period, there was no recurrence of severe hypoglycemia. We applied Ryan's formula to quantify pretransplant scores. After islet transplantation, we calculated actual HYPO scores from subjects' daily blood sugar records. Pretransplant HYPO score was 841.8 ± 1217.9 and 812.8 ± 941.7 for Group 1 and 2 respectively. Posttransplant HYPO score was 0 and 33.5 ± 50.0 for Group 1 and 2 respectively (p < 0.01, compared to pretransplant values, Table 4 and Figure 5C). Two subjects in Group 2 presented with mild hypoglycemia after exenatide injection.

Table 4.  Outcome of the study 15 months posttransplantation
VariableRecipient No.Mean (SD)
Group 1Group 2
1234Mean (SD)56789*10*
  1. BMI = body mass index; PRA = panel reactive antibodies; OGTT = oral glucose tolerance test; IGT = impaired glucose tolerance; NL = normal; DM = diabetes mellitus. *These two subjects did not tolerate exenatide (<2 months taken).

Body weight, kg51.368.957.663.060.2(6.5)69.454.956.761.858.555.859.5(4.9)
BMI (kg/m2)21.322.420.023.121.7(1.2)
Total islet mass/kg (No. of transplants)24 981 (3)14 757 (2)13 541 (2)44 262 (3)24 385 (14 209)5840 (1)6180 (1)21 133 (2)7273 (1)20 587 (2)7884 (1)11 483(7302)
Daily insulin, U/kg00002000018.30
HbA1c 5.4 5.4 5.8 5.8 5.6(0.2) 5.4 6.0 6.1 6.1 6.5 6.2 6.1(0.3)
Alloantibodies (class I & II PRA)00000051000
 • Anti-GAD651 6.2 1.0 1.0 1.6 2.5 (2.2)117.2  1.1 1.2 6.3100.7  1.137.9 (50.5)
 • Anti-ICA512++

Figure 5. Glycemic control before (Panel A) and after (Panel B) transplantation in both groups. The top, bottom and line through the middle of the box correspond to the 75th percentile, 25th percentile, and 50th percentile respectively. The whiskers on the bottom extend from the 10th percentile and top 90th percentile, • represents the arithmetic mean. Panel C shows the frequency and severity of hypoglycemia before and after transplantation assessed by Ryan HYPOscore. HYPOscore was significantly decreased after transplant for both groups (p < 0.01). Two subjects in Group 2 experienced occasional, nonsevere, postprandial hypoglycemia while under exenatide.

Download figure to PowerPoint

Glycemia:  As shown in Figure 5A, before transplant subjects in both groups exhibited wide variability in daytime glucose levels. After transplantation, both groups clearly showed tighter, more regular glucose control (Figure 5B).

Challenge tests with and without exenatide:  To further study the metabolic effect of exenatide on islet performance, we compared frequently sampled intravenous glucose tolerance tests (FSIVGTT), C-peptide, proinsulin, amylin and glucagon with and without exenatide.

Frequently sampled intravenous glucose tolerance tests:  In Group 2 the AIRg (postglucose acute insulin response) increased from 2.2 ± 1.2 to 2.7 ± 1.6 after taking exenatide (p = 0.4). In Group 1 the AIRg in was 2.4 ± 0.9 (p = 0.8) without exenatide. In our subjects, we did not find any intrapatient relationship between the response to IVGTT and OGTT (Table 5).

Table 5.  Oral glucose tolerance test (OGTT) and Intravenous glucose tolerance test (IVGTT) results 12 months after the first transplant
 Recipient No.
Group 1Group 2
 Acute insulin response (mU.L−1.min)33.333.67 18.6610.333.6617.6615.37.3318.66
 Sensitivity index ((mU/L)−1.min−1) 3.963.2213.6 0.42 0.014 3.58 3.6 0.042 1.65

Intravenous glucose tolerance testing:  On IVGTT conducted after subjects injected 5 μg of exenatide, C-peptide increased from 1.9 ± 0.2 to 2.5 ± 0.6 (p = 0.3). Proinsulin increased from 5.7 ± 1.6 to 7.8 ± 2.3 (p = 0.3) and amylin increased from 7.2 ± 2.3 to 14.7 ± 6.6 (p = 0.1).

Proinsulin to insulin ratio:  The proinsulin to insulin ratio decreased after exenatide in all subjects (7.1 ± 2.1 to 4.3 ± 1.9, p = 0.02), pointing toward more efficient proinsulin to insulin processing with exenatide.

Glucagon:  Glucagon levels decreased significantly in all subjects. The delta decrease was 18.7 fold more significant after exenatide in all subjects (Δ change 8.9 ± 18.4 vs. 165.7 ± 46.4 w/o and with exenatide, respectively; p = 0.04).

Autoantibodies:  Two subjects in Group 1 and 5 subjects in Group 2 tested positive for anti-GAD 65. Of the 10 subjects, only 2 (both from Group 2) were positive for anti-ICA 512. Neither subject remained insulin-independent.

Long-term follow-up

The study was originally designed for a 15-month posttransplant follow-up. We offered subjects an additional 5-year follow-up which 8/10 subjects accepted. At the time of writing this report, two of the four subjects from Group 1 remain insulin-independent with an average follow-up of 30.3 ± 0.9 months. These two subjects maintain HbA1c values between 5.5 and 5.7 at almost 3-year follow-up. Of the two subjects taking insulin, one was withdrawn from immunosuppression because of lobular breast carcinoma after 19 months. Subject 1 in Group 1 developed breast cancer after 19 months follow-up and insulin-independence. This patient was 64 years old at the time of diagnosis and underwent surgery and chemotherapy and she is doing well with recovery and no signs of distant disease and no local recurrence more than 1 year later. Subject 4 in Group 1, experienced graft failure 18 months posttransplant for unknown reasons, requires 27–30 units of insulin daily, and maintains HbA1c at 6.7. All subjects in Group 2 maintain HbA1c values between 5.8 and 6.3 with average follow-up of 21.2 ± 4.1 months. Four of the six subjects in Group 2 remain off insulin. Of the two subjects on insulin, subject 5 withdrew from the study and subject 9 was unable to tolerate exenatide and resumed 15–20 U of insulin daily by pump while awaiting a third transplant.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This study demonstrates that the addition of exenatide and etanercept to the Edmonton protocol is associated with a significantly lower number of islets required initially to achieve insulin-independence. In our cohort, recipients of the Edmonton protocol received either two or three sequential islet transplantations at least 2 weeks apart to achieve insulin-independence. By comparison, the subsequent six subjects who were treated under the UIC protocol with the addition of exenatide and etanercept achieved insulin-independence after only one islet preparation, though during the follow-up, two subjects required an additional transplant to maintain insulin-independence and one of these subjects resumed insulin 5 months after the second transplant.

The salutary effect of exenatide on decreasing initial islet requirement is largely unknown. However, evidence exists that exenatide can increase insulin synthesis under elevated glucose levels (23) and has antiapoptotic and β-cell proliferative effects in both rodent and human β-islets in in vivo and in vitro experiments (24,25). Moreover, the antiinflammatory effects of etanercept (26) likely contributed to early islet function and improved engraftment. In fact, the University of Minnesota group has successfully used the peri-operative administration of etanercept to achieve insulin-independence after transplant of a marginal dose of islets from a single donor (5). In addition, another recent report has confirmed the efficacy of IL-1 receptor antagonist in preventing the inflammatory β-cell destruction in diabetes type II (27). We found that acute insulin response to glucose was increased, although not significantly, in recipients of the UIC protocol considering they received fewer islets than recipients of the Edmonton protocol. It is likely that a larger number of islets survived in subjects treated with the UIC protocol as compared to the amount of the islets in recipients of the Edmonton protocol. This suggests that the addition of etanercept and exenatide may improve islet engraftment. It would be interesting to see if administering exenatide for a longer period of time could enhance the ‘trophic’ effect of the drug. None of the patients treated with the UIC protocol achieved normal OGTT, showing IGT or DM responses, while normal OGTT was observed in 50% of the subjects treated with two to three transplants and the Edmonton protocol. The islet mass in UIC treated subjects is presumably adequate under normal condition. However, under stress and excess demand during OGTT, the islets in these subjects could not meet the demand, resulting in IGT or DM. This indicates that UIC protocol-treated patients have lower islet mass, which apparently did not increase under exenatide treatment. An alternative speculation may be considered as well: the finding of decreased pro-insulin to insulin ratio after exenatide in all subjects tested can point toward a more efficient pro-insulin to insulin conversion. We expected a more significant increase in C-peptide, proinsulin and amylin after exenatide due to the proposed effect of the GLP-1 analogues on augmentation of pancreatic β-cell response to glucose stimulation via receptors on the β-cells (28,29). However, in our subjects there was no significant increase, and only one subject, as well as a healthy, nondiabetic control showed an increase in the acute insulin response to an intravenous glucose challenge when exenatide was given before the test.

Decrease in glucagon is a known effect of GLP-1 and its analogues that was recently validated by Ionut et al. (30), who tested the effect in dogs. They infused glucose in combination with GLP-1 similar to the meal response. Results showed no augmentation of insulin stimulation. The failure of GLP-1 to augment the insulin secretory response in their experimental setting suggested a possible indirect effect such as reduction in glucagon and is consistent with our results. In fact, we found a significant decrease of glucagon levels in subjects receiving exenatide.

It is important to note that the majority of subjects receiving the UIC protocol did not demonstrate any evidence of a potential ‘exhaustion effect’ after prolonged administration of exenatide on β-cell capacity to secrete insulin.

The adverse effects of exenatide include mainly nausea and vomiting. Nausea is common with initial doses and tends to diminish over time. As a result of persistent nausea and vomiting, two of our subjects did not tolerate exenatide longer than 2 months. The side effects of etanercept include serious nervous system disorders such as multiple sclerosis and seizures, blood disorders, increased risk of lymphoma and injection site reactions. However, these have been reported in long-term treatment for autoimmune diseases such as rheumatoid arthritis, psoriasis and ankylosing spondylitis; Group 2 subjects received only four doses of etanercept during the first 10 days posttransplant.

In our study, two subjects developed serious adverse events: a 64-year-old patient in the first group, treated with Edmonton protocol, developed breast cancer 19 months posttransplant and another subject receiving the UIC protocol developed diabetic myonecrosis on her neck, which was complicated by muscle and bone infection. The pathophysiology of diabetic myonecrosis is not well understood. Thromboembolism superimposed on diabetic small-vessel disease and subsequent ischemia-reperfusion injury has been suggested as the major underlying mechanism. A few cases of diabetic muscle infarction have been reported after simultaneous pancreas-kidney transplants. Underlying diabetic microangiopathy and hypercoagulability have been proposed as contributing factors in development of myonecrosis in these patients. Furthermore, tacrolimus and cyclosporine have been shown to be associated with thrombotic microangiopathy and impaired endothelial function. In our patient, the immunosuppression therapy may have had a contribution to this late infectious complication.

As this was a pilot study, its major weakness is the relatively small sample size and potential for both type II and type I error. In addition, the lack of a concurrent control group as opposed to the sequential study design is less than ideal. Furthermore, while four of six subjects in the UIC protocol required fewer islets than those in the standard Edmonton protocol, two subjects did need to resume insulin therapy. Since two subjects in Group 2 did not tolerate exenatide, its effects may have been underestimated in this pilot study.

Despite these caveats, the results presented are extremely promising. The islet equivalents utilized to achieve insulin-independence in the UIC protocol arm of the study were lower than the sequential Edmonton protocol group and among the lowest reported in the literature. Few adverse long-term effects were seen, and while subjects on exenatide did experience high rates of nausea, this usually abated within 1 to 2 weeks.

In summary, utilization of exenatide and a short course of TNF alpha blockade shows promise in decreasing the need for multiple donors and warrants further investigation. While the presented protocol allowed for consistently achieving insulin-independence with lower islet mass, it did not lead to a measurable increase in functional islet mass over time. Future studies will have to investigate whether the speculated ‘trophic’ effect of exenatide could be enhanced by higher drug levels and prolonged administration. In our opinion, this will require improving the pharmacokinetics of exenatide, avoiding peaks and allowing for extended release. Such pharmacokinetics may decrease the likelihood of the reported side effects, and may allow for achieving higher overall drug exposure. Therefore, we suggest that different treatment regimens with exenatide (e.g. extended release formulations) should be investigated in future studies.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This work was supported by funding from the College of Medicine, University of Illinois at Chicago. J.O. and J.M. were supported by the Islet Cell Resource Center NIH grant (RFA-RR 05–003), the Christopher Foundation and the Efroymson Foundation.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    PE C. The limiting factor in the glycaemic management of Type I and Type II Diabetes. Diabetologia 2002; 45: 937948.
  • 2
    Shu C, He J. The long-term observation of the transplantation of encapsulated rat islets to the diabetic mice. Biomater Artif Cells Immobiliz Biotechnol 1993; 21: 8589.
  • 3
    Group U. Intensive blood-glucose control with sylhonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes. Lancet 1998; 352: 837853.
  • 4
    Shapiro AM, Ricordi C, Hering BJ et al. International trial of the Edmonton protocol for islet transplantation. N Engl J Med 2006; 355: 13181330.
  • 5
    Hering BJ, Kandaswamy R, Ansite JD et al. Single-donor, marginal-dose islet transplantation in patients with type 1 diabetes. JAMA 2005; 293: 830835.
  • 6
    DeFronzo RA, Ratner RE, Han J, Kim DD, Fineman MS, Baron AD. Effects of exenatide (exendin-4) on glycemic control and weight over 30 weeks in metformin-treated patients with type 2 diabetes. Diabetes Care 2005; 28: 10921100.
  • 7
    Poon T, Nelson P, Shen L et al. Exenatide improves glycemic control and reduces body weight in subjects with type 2 diabetes: A dose-ranging study. Diabetes Technol Ther 2005; 7: 467477.
  • 8
    Kendall DM, Riddle MC, Rosenstock J et al. Effects of exenatide (exendin-4) on glycemic control over 30 weeks in patients with type 2 diabetes treated with metformin and a sulfonylurea. Diabetes Care 2005; 28: 10831091.
  • 9
    Ghofaili KA, Fung M, Ao Z et al. Effect of exenatide on beta cell function after islet transplantation in type 1 diabetes. Transplantation 2007; 83: 2428.
  • 10
    Robertson RP. Consequences on beta-cell function and reserve after long-term pancreas transplantation. Diabetes 2004; 53: 633644.
  • 11
    Ryan EA, Shandro T, Green K et al. Assessment of the severity of hypoglycemia and glycemic lability in type 1 diabetic subjects undergoing islet transplantation. Diabetes 2004; 53: 955962.
  • 12
    Shapiro AM, Lakey JR, Ryan EA et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 2000; 343: 230238.
  • 13
    Grading diabetic retinopathy from stereoscopic color fundus photographs—an extension of the modified Airlie House classification. ETDRS report number 10. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology 1991; 98(5 Suppl): S786S806.
  • 14
    Ricordi C, Lacy PE, Finke EH, Olack BJ, Scharp DW. Automated method for isolation of human pancreatic islets. Diabetes 1988; 37: 413420.
  • 15
    Barbaro B, Salehi P, Wang Y et al. Improved human pancreatic islet purification with the refined UIC-UB density gradient. Transplantation 2007; 84: 12001203.
  • 16
    Ricordi C, Gray DW, Hering BJ et al. Islet isolation assessment in man and large animals. Acta Diabetol Lat 1990; 27: 185195.
  • 17
    Yang H, Acker J, Chen A, McGann L. In situ assessment of cell viability. Cell Transplant 1998; 7: 443451.
  • 18
    Avila JG, Tsujimura T, Oberholzer J et al. Improvement of pancreatic islet isolation outcomes using glutamine perfusion during isolation procedure. Cell Transplant 2003; 12: 877881.
  • 19
    Tsujimura T, Kuroda Y, Kin T et al. Human islet transplantation from pancreases with prolonged cold ischemia using additional preservation by the two-layer (UW solution/perfluorochemical) cold-storage method. Transplantation 2002; 74: 16871691.
  • 20
    Teuscher AU, Kendall DM, Smets YF, Leone JP, Sutherland DE, Robertson RP. Successful islet autotransplantation in humans: Functional insulin secretory reserve as an estimate of surviving islet cell mass. Diabetes 1998; 47: 324330.
  • 21
    Pacini G, Bergman RN. MINMOD: A computer program to calculate insulin sensitivity and pancreatic responsivity from the frequently sampled intravenous glucose tolerance test. Comp Methods Prog Biomed 1986; 23: 113122.
  • 22
    Kaplan B, West P, Neely H et al. Use of chronic IL-2 receptor blockade in an islet transplant recipient. Clin Transplant 2007 (in press).
  • 23
    Li Y, Hansotia T, Yusta B, Ris F, Halban PA, Drucker DJ. Glucagon-like peptide-1 receptor signaling modulates beta cell apoptosis. J Biol Chem 2003; 278: 471478.
  • 24
    Tourrel C, Bailbe D, Meile MJ, Kergoat M, Portha B. Glucagon-like peptide-1 and exendin-4 stimulate beta-cell neogenesis in streptozotocin-treated newborn rats resulting in persistently improved glucose homeostasis at adult age. Diabetes 2001; 50: 15621570.
  • 25
    Xu G, Stoffers DA, Habener JF, Bonner-Weir S. Exendin-4 stimulates both beta-cell replication and neogenesis, resulting in increased beta-cell mass and improved glucose tolerance in diabetic rats. Diabetes 1999; 48: 22702276.
  • 26
    Vally M, Seenu S, Pillarisetti S. Emerging peptide therapeutics for inflammatory diseases. Current Pharm Biotechnol 2006; 7: 241246.
  • 27
    Larsen CM, Faulenbach M, Vaag A et al. Interleukin-1-receptor antagonist in type 2 diabetes mellitus. N Engl J Med 2007; 356: 15171526.
  • 28
    Kreymann B, Williams G, Ghatei MA, Bloom SR. Glucagon-like peptide-1 7–36: A physiological incretin in man. Lancet 1987; 2: 13001304.
  • 29
    Mojsov S, Weir GC, Habener JF. Insulinotropin: Glucagon-like peptide I (7–37) co-encoded in the glucagon gene is a potent stimulator of insulin release in the perfused rat pancreas. J Clin Invest 1987; 79: 616619.
  • 30
    Ionut V, Liberty IF, Hucking K et al. Exogenously imposed postprandial-like rises in systemic glucose and GLP-1 do not produce an incretin effect, suggesting an indirect mechanism of GLP-1 action. Am J Physiology 2006; 291: E779785.