Prolonged Insulin Independence after Islet Allotransplants in Recipients with Type 1 Diabetes

Study Design

From August 2003 through March 2005, we enrolled 6 alloislet transplant recipients in a 1-year single center study. The immunosuppression protocol consisted of ATG and etanercept induction therapy and cyclosporine and everolimus maintenance.

The primary objective of our study was to examine the safety of alloislet transplants with a modified immunosuppressive protocol. The secondary objective was to assess the proportion of recipients who obtained insulin independence in the first year posttransplant, as well as the proportion with full or partial graft function at 1 year after the last islet infusion. Insulin independence was defined by fasting blood glucose levels ≤126 mg/dL and 2-hour postprandial levels ≤180 mg/dL without exogenous insulin. Full graft function was defined by insulin independence with hemoglobin A_1c (HbA_1c) ≤ 7%; partial function was defined by insulin dependence but C-peptide ≥0.5 ng/mL and HbA_1c ≤ 7%.

Beyond 1 year after the last islet infusion, recipients were enrolled into a long-term follow-up protocol for continued monitoring. Due to the risk of nephrotoxicity with calcineurin inhibitor and mTOR inhibitor combination therapy, we substituted an inosine monophosphate dehydrogenase inhibitor (mycophenolic acid or mycophenolate mofetil) for everolimus in 2 recipients due to toxicity before 1-year after final infusion (#2 at day 153 and #4 at day 280 relative to last infusion), and in the remaining 4 at 1–1.4 years after final infusion. During long-term follow-up, 5 recipients continued on cyclosporine with either mycophonelic acid or mycophenolate mofetil (choice based on insurance coverage, recipients # 1, 3, 4, 5, 6); in 1 recipient tacrolimus was substituted for cyclosporine because of gingival hypertrophy (# 2); and 1 recipient received additional treatment with sirolimus for 5 months for concern of new positive PRA results and increasing insulin needs (#4).

Study protocols were reviewed and approved by the University of Minnesota (UM) Institutional Review Board; all participants gave informed consent.

Recipients

Recipients were age 18 years or older and had a history of type 1 diabetes with severe hypoglycemia and reduced awareness of hypoglycemia (Clark hypoglycemia unawareness score ≥ 4 of 7) despite intensive diabetes management. Severe hypoglycemia was defined as an event with hypoglycemic symptoms in which the patient required the assistance of another person to treat and which was associated with a blood glucose <50 mg/dL or prompt recovery after oral carbohydrate, intravenous glucose, or glucagon administration.

Exclusion criteria included the following: body mass index (BMI) >26 kg/m², insulin requirement of > 50 IU per day, positive C-peptide levels, untreated proliferative retinopathy, creatinine clearance < 60 ml/min/1.73 m²for women and <70 ml/min/1.73 m²for men, and panel-reactive anti-HLA antibody levels >10%.

Islet Preparation

Islets were prepared as previously described (2). Briefly, pancreases were procured from brain-dead, ABO-compatible donors; preserved using the 2-layer technique; and dissociated using Liberase HI (Roche Diagnostics Corp, Indianapolis, IN) and the Ricordi method. Free islets were separated from non-islet tissue on continuous iodixanol density gradients, and cultured free-floating in supplemented CMRL 1066 at 37°C for 1 day and 22°C for 1 day pretransplant.

Transplant Procedure

Portal vein access was obtained via percutaneous transhepatic portal vein catheterization (7 infusions in 4 recipients) or minilaparotomy (3 infusions in 3 recipients), and the islet preparation was infused intraportally over 15 to 60 minutes. Portal vein pressure was monitored periodically during the infusion, with injection halted if portal pressures exceeded 30 cmH₂O and resumed if portal pressures dropped below 25 cmH₂O. Prophylactic anticoagulation was with intraportal heparin initially (70 units/ kilogram), followed by continuous intravenous heparin infusion for 48 hours (target, partial thromboplastin time of 60 seconds). Recipients were on subcutaneous enoxaparin prophylaxis from day 2 to day 7 posttransplant.

They received a continuous intravenous insulin infusion from day −2 to day 2 (target, blood glucose levels of 80 to 120 mg/dL).

Subsequent Islet Transplants

Recipients who received ≥ 5,000 IE/kg with the initial transplant but did not obtain insulin independence were eligible for a second transplant, performed within 18 months after the initial transplant. A minimum of 3,000 IE/kg was administered during the second islet infusion.

Immunosuppression

Induction immunosuppression for the first islet infusion was with ATG (Thymoglobulin, Genzyme) given on day −2 (0.5 mg/kg), day −1 (1.0 mg/kg), and days 0 through 2 (1.5 mg/kg/day) relative to transplant. Premedications included acetaminophen, diphenhydramine, and pentoxifylline, and with the first ATG infusion, methylprednisolone (2 mg/kg). For subsequent islet transplants, induction therapy was with basiliximab (Simulect, Novartis; 20 mg intravenously 2 hours pretransplant and 4 days posttransplant). Etanercept (Enbrel; Amgen) was given for peritransplant antiinflammatory therapy (loading dose, 50 mg intravenously 1 hour pretransplant, then 24 mg subcutaneously on days 3, 7, and 10 posttransplant).

Cyclosporine was initiated on day 1 after initial islet transplant (initial dose 3 mg/kg/day divided twice daily, then adjusted to these target blood levels: 2-hour post-cyclosporine level 350 to 500 ng/mL for the first 3 months posttransplant, 200 to 350 ng/mL thereafter). Everolimus was initiated on day-2 (3 mg orally once on day −2, then 1.5 mg orally twice daily and adjusted as needed to these target trough levels: 3 to 15 ng/mL for the first 3 months posttransplant, 3 to 12 ng/mL thereafter; mean levels maintained at 7.2±1.8 ng/mL at 3 months, 6.3±3.4 ng/mL at 6 months, 7.3±2.3 ng/mL at 12 months). At any time, if deemed clinically appropriate, mycophenolate mofetil (320 to 720 mg orally twice daily) replaced everolimus.

Concomitant Therapy

All recipients received antimicrobial prophylaxis. Filgrastim was used as clinically indicated for absolute neutrophil count < 1000/µL. Intravenous insulin was administered in the peritransplant period and subsequently transitioned to subcutaneous therapy. Subcutaneous insulin therapy was withdrawn if the recipient’s blood glucose levels were maintained at < 126 mg/dL fasting and <180 mg/dL 2-hours postprandial.

Safety and Efficacy

Patients were admitted to the General Clinical Research Center at UM for days −2 through 2 and were seen in outpatient follow-up at regular intervals for 12 months posttransplant. Following enrollment into the long-term islet survival study (1 year after the last islet infusion), participants underwent laboratory testing locally and submitted blood glucose levels and insulin logs regularly. Recipients and their primary care physicians were asked to complete questionnaires annually regarding current medical health, insulin use, hypoglycemia episodes, and diabetes complications.

Laboratory safety assessments included complete blood counts, coagulation studies, hepatic panel, creatinine, 24-hour urinary albumin excretion, glomerular filtration rate (GFR), and immunosuppressive medication levels measured at regular intervals.

Metabolic follow-up testing was defined by time since last islet infusion. Self-monitored blood glucose levels, documentation of insulin requirements, and hypoglycemia history were reviewed at each follow-up visit. HbA_1c levels, measured by high-performance liquid chromatography (BioRad, Hercules, CA), were obtained at regular intervals. Participants underwent a 2-hour 75-gram oral glucose tolerance test (OGTT) on days 90 and 365 and a 5-gram arginine stimulation test (AST) on days 90, 180, and 365 posttransplant. The acute insulin response (AIR_a) and acute C-peptide response (ACR_a) to arginine stimulation testing were defined by the mean of the highest 3 post-arginine values between 2 and 5 minutes with the basal value subtracted. Insulin levels were measured by chemiluminescent immunoassay, and C-peptide levels were measured by double-antibody radioimmunoassay.

Alloimmunity and Autoimmunity

Titers of anti-glutamic acid decarboxylase (GAD65) antibodies, insulin autoantibodies (IAA), and anti-islet cell antibodies (ICA) were measured by radioassay (12,13). Panel-reactive antibody (PRA) levels were measured using ELISA pre-transplant and flow luminex posttransplant.

Statistical Analysis

Data are expressed as mean ± standard deviation. Because of the low number of study participants, we did not perform formal statistical analysis.

Recipient and Donor Characteristics

Our 6 recipients (table 1) all had a history of severe hypoglycemia in the year preceding their first islet infusion and a Clark score ≥ 4. Pretransplant, they required a mean of 23.7 ± 4.4 units per day of insulin and had a mean HbA_1c of 7.3% ± 0.7%.

The 6 recipients received a total of 11 islet grafts (Table 2). Recipients who failed to reach insulin independence with a single infusion underwent a second infusion 54 to 476 days after the first. One recipient required only 1 islet graft to obtain insulin independence, 1 received 2 islet grafts simultaneously, and 4 underwent 2 consecutive islet infusions. Recipients received a mean of 11,872 ± 2541 islet equivalents/ kilogram body weight (IE/kg).

Islet graft function

Recipients have been followed for 1173 ± 270 days since their last islet infusion (range, 969 to 1685 days). Insulin use and results of metabolic testing for all recipients are displayed in Table 3. Insulin independence was attained in 5 recipients; 1 recipient (#6) did not attain insulin independence after 2 islet infusions. This insulin-dependent recipient (#6) had a 60% reduction in insulin needs and met the definition of partial graft function at 1 year posttransplant. Of 6 recipients, 5 were insulin-independent with full graft function at 1 year posttransplant, and 4 were insulin-independent with full graft function at 2 years posttransplant. These 4 recipients continue to be insulin-independent at a mean of 3.4 ± 0.8 years posttransplant.

All 6 recipients, including those on insulin, showed evidence of C-peptide production at their most recent follow-up (≥ 2.6 years). All 6, including the 1 on insulin, had an HbA_1c <6% at 1 year posttransplant. None experienced recurrence of severe hypoglycemic episodes posttransplant.

At 2 years posttransplant, mean HbA_1c levels were 6.0% ± 0.7% for all recipients and 5.6% ± 0.2% for insulin-independent recipients, indicating good glycemic control. At 2 years posttransplant, insulin-independent participants had a mean fasting blood glucose level of 102 ± 11 mg/dL; mean C-peptide, 1.9 ± 1.3 ng/mL.

At their most recent follow-up (mean 3.4 ± 0.8 years), of the 4 insulin-independent recipients, mean HbA_1c levels were 5.7% ± 0.3%; mean fasting glucose level, 106 ± 17 mg/dL. Of the 2 insulin-dependent recipients, 1 (#4) demonstrated euglycemia on insulin with a HbA_1c level of 5.8% and a fasting glucose level of 85 mg/dL; the other had poor metabolic control with a HbA_1c level of 9.3% and no recent fasting laboratory glucose level.

Only 2 recipients had a 2-hour blood sugar level consistent with diabetes on an OGTT performed at 1 year posttransplant—namely, 1 on insulin and 1 who resumed insulin shortly after the 1 year follow-up. Of the 4 currently insulin-independent recipients, 2 had a normal response to an OGTT and 2 had impaired glucose tolerance.

The lowest ACR_a and AIR_a at 6 months and 12 months were in the recipient on insulin and the recipient who later resumed insulin.

Auto- and alloimmunity

Results of pretransplant and posttransplant PRA and autoantibody testing are displayed in Table 4. No recipient had evidence of allosensitization pretransplant. All had evidence of autoimmunity pretransplant: 2 of 3 antibodies were positive in 5 recipients and all antibodies positive in 1 recipient.

On most recent posttransplant PRA (mean 2.4 ± 0.8 years posttransplant), 4 recipients had positive results—class II only in 2 and both class I and class II positive in the other 2, although 1 of the recipients with both positive had a low class I PRA at 4%. Of these 4 recipients, 3 had donor-specific antibodies. In the 2 recipients on insulin, 1 was positive for class II donor-specific antibodies (#4); 1 was positive for both class I and class II donor-specific antibodies (#6).

Adverse Events

We noted no episodes of opportunistic infections, posttransplant lymphoproliferative disease, or other malignancies. All recipients experienced aphthous ulcers, leukopenia, and transient elevations of liver enzymes in the early posttransplant period. Two recipients transplanted by percutaneous approach experienced symptomatic cholelithiasis or acute cholecystitis (days 8 and 72 posttransplant) and required cholecystectomy; this was possibly related to the transplant procedure, although no known injury to the gallbladder occurred during portal vein catheterization. We noted no other procedure-related adverse events.

Serum creatinine levels increased in all recipients: from a mean of 0.8 ± 0.1 mg/dL at baseline to 1.1 ± 0.2 mg/dL at 1 year posttransplant. At the most recent follow-up, levels were relatively stable at 1.1 ± 0.3 mg/dL.

Measured GFR decreased from pretransplant to 1 year posttransplant in 5 of 6 recipients: from a mean of 110 ± 21 mL/min/1.73m² at baseline to 83 ± 19 mL/min1.73m² at 1 year posttransplant. Two recipients had >25% decrease in GFR, with an overall mean decrease of 28 ± 19% among all participants (table 5).

Mean urinary microalbumin increased from 12 ± 12 mg/g Cr pretransplant to 23 ± 19 mg/g Cr at 1 year posttransplant.

Pretransplant, 1 recipient required antihypertensive medications; posttransplant, 4 did, with 2 of these requiring ≥ 2 medications for adequate control of hypertension. One recipient required lipid-lowering medication pretransplant; posttransplant, 3 did.