Validation of Islet Transport From a Geographically Distant Isolation Center Enabling Equitable Access and National Health Service Funding of a Clinical Islet Transplant Program for England

Abstract

Islet transplantation has become established as a successful treatment for type 1 diabetes complicated by recurrent severe hypoglycemia. In the UK access has been limited to a few centrally located units. Our goal was to validate a quality-assured system for safe/effective transport of human islets in the UK and to successfully undertake the first transplants with transported islets. Pancreases were retrieved from deceased donors in the north of England and transported to King’s College London using two-layer method (TLM) or University of Wisconsin solution alone. Islets were isolated and transported back to Newcastle in standard blood transfusion or gas-permeable bags with detailed evaluation pre- and posttransport. In the preclinical phase, islets were isolated from 10 pancreases with mean yield of 258,000 islet equivalents. No significant differences were seen between TLM and University of Wisconsin solution organ preservation. A significant loss of integrity was demonstrated in islets shipped in gas-permeable bags, whereas sterility, number, purity, and viability were maintained in blood transfusion bags. Maintenance of secretory granules and glucose-stimulated insulin secretion was confirmed following transport. A Standard Operating Procedure enabling final pretransplant quality control from a simple side-arm sample was validated. Moreover, levels of insulin and cytokines in transport medium were low, enabling transplant without centrifugation/resuspension at the recipient site. Six clinical transplants of transported islets were undertaken in five recipients with 100% primary graft function and resolution of severe hypoglycemia. Safe and clinically effective islet transport has been established facilitating sustainable NHS funding of a clinical islet transplant program for the UK.

INTRODUCTION

Successful islet transplantation for the treatment of type 1 diabetes complicated by recurrent severe hypoglycemia has been established for several years (15) with significant advances being achieved in the past decade in both islet isolation and transplant management (8). However, wide-spread adoption of and equitable access to islet transplantation has remained restricted due to a range of factors including shortage of suitable donor tissue (14) and the limited number of accredited clinical isolation units.

Islet isolation facilities are complex laboratories that require extensively trained and highly skilled technical staff with 24-h on-call availability in addition to strict compliance with legislatory standards for production of clinical grade tissue according to current Good Manufacturing Practice (cGMP) (12). Establishing programs for transporting islets isolated at a central isolation facility to a geographically distant transplant site would enable significantly more people to benefit from this treatment with considerable overall cost savings. The potential for successful islet isolation at a distant site, in the University of Minnesota facility 1,500 miles away from the transplant center, was initially demonstrated in the autotransplant setting following pancreatectomy for chronic pancreatitis (11). Since then, various models of islet transport for allotransplantation have been validated in centers in the US (5) and Europe (3) but not in the UK. Our goals were to develop and validate a quality-assured system for safe and effective transport of human islets in the UK; to implement and obtain legislatory approval for Standard Operating Procedures underpinning this process; and to undertake the first UK transplants with transported clinical grade islets.

MATERIALS AND METHODS

Ethical Approval

National ethical and Trust Research and Development approval was obtained for all preclinical studies using deceased pancreases donated for this research with the specific consent of donor relatives.

Pancreas Retrieval and Transport

Pancreases were retrieved from deceased donors after brain death (DBD) in the north of England by the specialist multiorgan retrieval team according to standard procurement protocols for both vascularized pancreas and islet transplantation. This included cooling of the lesser sac with ice at the time of cold vascular perfusion with University of Wisconsin solution (UWS) without vascular perfusion of the portal venous system (13). The retrieved pancreas was perfused briefly with cold UWS on the back table and then placed in a 1-L polyethylene transport chamber (VWR, Leicestershire, UK) with either 600 ml of UWS (ViaSpan, DuPont Pharma Ltd, Herts, UK) alone or 300 ml UWS mixed with 300 ml freshly oxygenated perfluorodecalin (F2 Chemicals, Lancashire, UK). In the case of the latter two-layer method (TLM), a retention frame was employed to maintain the pancreas within the perfluorodecalin layer. The transport chamber was buried in ice inside an insulated transport cool-box (MGD Coolers, Staffordshire, UK) supplied with data-logger set to record temperature every 5 min (Hobo pendant data-logger, Tempcon Instrumentation Ltd, West Sussex, UK) and transported by road to the islet isolation facility at King’s College Hospital NHS Foundation Trust, in London 300 miles away, by a dedicated organ courier (Lifeline Medical Transport Service, Northumberland, UK). In the clinical phase of the program, locally retrieved organs were transported in UWS to the Oxford Diabetes Endocrinology and Metabolism clinical islet isolation facility in addition to the King’s College Hospital facility.

Islet Isolation

Islet isolations were carried out as previously described employing purified collagenase [Liberase HI (Roche Diagnostics Ltd, West Sussex, UK) for the preclinical program, and Collagenase NB 6 GMP Grade (SERVA Electrophoresis GmbH, Heidelberg, Germany) for the clinical program] infusion for enzymatic digestion and mechanical dissociation in a Ricordi chamber, followed by purification on a continuous density gradient using a COBE 2991 processor (4).

Islet Packaging and Transport to Newcastle

Purified islets were resuspended in CMRL medium (PAA Laboratories Ltd, Somerset, UK) supplemented with 2% human serum albumin (ZENLAB20: Bio Products Laboratories, Herts, UK) and 2 mM HEPES (PAA Laboratories Ltd, Somerset, UK) and packaged in either 100 ml (for 300-ml capacity bags) or 400 ml (for 1000-ml capacity bags) transport medium in gas-permeable bags (Permalife, Origen Biomedical Europe, Helsingborg, Sweden) or blood transfusion bags (Baxter Healthcare Ltd, Berkshire, UK). For clinical implementation islets were packaged in 500-ml capacity bags in 200 ml medium. Twelve hours after packing was set as the latest time that dispatched islets could be transplanted clinically. Islet bags were placed horizontally in a shelving unit inside the transport cool-box containing ice packs on the base, avoiding direct contact with the cooled packs. Islets were transported by road to Newcastle using the Lifeline courier. Continuous temperature monitoring throughout the journey was carried out using the Hobo pendant data-logger. Reassessment of islets was undertaken within the Newcastle Biomanufacturing Cellular Therapies GMP suite.

Islet Quality Assessment

Gram staining and bacterial culture was carried out in the clinical microbiology laboratories at the isolation centre and in the recipient center at the Freeman Hospital, Newcastle.

Islet equivalent count (IEQ) corrected for a standard islet diameter of 150 μm and percentage purity was determined following dithizone staining (Sigma-Aldrich, Pool, UK) at a final concentration of 25 μg/ml for 2–4 min (4). Mean percentage integrity was determined in 50 randomly selected dithizone-positive islets according to the intactness of the pseudocapsular border. Mean percentage viability was assessed in 50 randomly selected islets by fluorescein diacetate (FDA) (0.5 μM final concentration)/propidium iodide (PI) (15 μM final concentration) (Sigma-Aldrich, Poole, UK) staining and inverted fluorescence microscope imaging using a standard protocol (2), with the addition of an extra PBS washing step following 2-min incubation with the stains to minimize false-positive PI staining induced by acetone (the FDA solvent).

Static assays of glucose-stimulated insulin secretion (GSIS) were undertaken in a water bath at 37°C. Following preincubation for 1 h in DMEM containing 2 mM glucose and washing in PBS, 6 aliquots of 15 islets were incubated in DMEM containing 2 mM glucose and 6 aliquots were incubated in DMEM containing 20 mM glucose each for 2 h. Supernatant insulin concentration was determined by Dako human insulin ELISA (Bio-Stat, Inverness, UK) with proinsulin concentration determined by human proinsulin ELISA (Mercodia, through Diagenics, Milton Keynes, UK). Intracellular insulin and proinsulin concentrations were assayed in islets lysed by sonication in PBS.

To determine glucose-stimulated insulin secretion in perifused islets, an equal number of islets (70–100) was dispensed into each of four perfusion chambers. Islets were perifused for 60 min using G&G buffer (0.18 M KCl, 1 mM MgCl₂, and 10 mM NaHCO₃) supplemented with 2 mM d-glucose. Perifusion with 2 mM d-glucose was continued for a further 20 min with sample collection from each chamber every 2 min. Islets were then perifused with G&G buffer supplemented with 18 mM d-glucose and samples were collected every 2 min for 20 min (9). Collected samples were assayed by homogeneous time resolved fluorescence (hTRF) insulin assay (Cisbio Bioassays, Bagnols-sur-Cèze, France) following the manufacturer’s instructions. Fold change in insulin release in each chamber was calculated from mean insulin concentration over the 10 min before and after the change from 2 to 20 mM glucose.

Insulin and Proinflammatory Cytokine Quantification in Transport Medium

Medium samples from bags were collected upon arrival in Newcastle and were frozen at −20°C. Insulin concentration was assayed by ELISA (Dako) and proinflammatory cytokines [interferon-γ (INF-γ), interleukin-1 (IL-1), IL-6, tumor necrosis factor-α (TNF-α] were quantified by multiplex assay (Meso-Scale Discovery, Gaithersburg, MD, USA).

Clinical Islet Transplant Criteria and Protocols

Inclusion criteria were C-peptide-negative type 1 diabetes complicated by recurrent severe hypoglycemia despite optimized conventional management, including a trial of continuous subcutaneous insulin infusion pump therapy or suboptimal glycemic control in the presence of a functioning renal graft. Contraindications included insulin resistance [body mass index (BMI) >28 or total daily insulin requirement >0.7 U/kg] and isotopic glomerular filtration rate <60 ml/min/1.73 m²/macroalbuminuria (albumin excretion rate >300 mg/24 h)—unless previous renal graft. Induction was with either daclizumab or alemtuzumab (single 15 mg subcutaneous dose administered immediately after completion of islet infusion preceded by 100 mg hydrocortisone and 10 mg chlorpheniramine intravenously and followed by 1 g paractetamol orally every 6 h for 48 h) with maintenance immunosuppression comprising mycophenolate mofetil and tacrolimus (0.1 mg/kg) (target trough range 8–10 μg/L). Islets were infused under local anesthesia into the hepatic portal vein following ultrasound-guided percutaneous transhepatic cannulation.

Statistical Analysis

Data are reported as mean ± SD with static and perifusion GSIS reported as a stimulation index (insulin secretion rate at high glucose/insulin secretion rate at low glucose). Differences were compared by paired/unpaired Student’s t-test and values of p < 0.05 considered statistically significant.

RESULTS

Pancreas Preservation and Islet Isolation Outcome

Ten pancreases were evaluated in the preclinical phase. All fulfilled the inclusion criteria, in place at the time, for acceptance of an organ into the clinical islet transplant program but with no suitable recipient on the active transplant waiting list. These included donors aged 20–65 years with no history of diabetes, recurrent pancreatitis, chronic renal disease, malignancy, Intensive Therapy Unit stay >10 days, alcohol or drug abuse. Serology was negative for hepatitis B, hepatitis C, Treponema, and HIV with no evidence of nodularity, fibrosis, or severe fatty infiltration on macroscopic examination of the pancreas.

Six donors (60%) were male. Cause of death was intracranial hemorrhage (including subarachnoid hemorrhage) in nine (90%) and middle cerebral artery infarct in one (10%). Macroscopic appearance was normal in six (60%) with moderate fatty infiltration in four (40%).

Six organs (60%) were transported to the King’s College Hospital isolation facility using TLM and four (40%) in UWS alone. Mean donor age was 48 years (range 30–54) and mean BMI 27 kg/m² (Table 1). Mean cold ischemic time (defined as cross-clamp until pancreas perfusion) with collagenase was 8.3 ± 4.1 h. Mean duration for islet isolation was 6.2 ± 0.9 h. Sterility was confirmed in all preparations. Islet equivalent (IEQ) yield varied widely from 60,000 to 580,000. Mean purity was 69% and viability 72%. There were no significant differences between donor organs preserved with the TLM or with UWS (Table 1).

Table 1.

Comparison of Pancreas Preservation Methods on Assessment at Isolation Center

	N (%)	Age	BMI	CIT	IEQ × 1000 (Range)	Purity	Viability
TLM	6 (60)	46 ± 9	27 ± 5	7.3 ± 1.6	300 (150–580)	65 ± 16	63 ± 24
UWS	4(40)	52 ± 7	27 ± 4	9.8 ± 6.5	196 (60–350)	75 ± 13	85 ± 9
Total	10	48 ± 9	27 ± 5	8.3 ± 4.1	258 (60–580)	69 ± 15	72 ± 22
p-value		0.3	0.9	0.5	0.3	0.3	0.1

Values are mean ± SD or mean (range). BMI, body mass index; CIT, cold ischemic time; IEQ, islet equivalent; TLM, two-layer method; UWS, University of Wisconsin solution.

Islet Transport and Reassessment at Satellite Site

In the preclinical phase, islets were packaged immediately following postisolation assessment without interim culture. Mean time from packaging in King’s College Hospital to reassessment in Newcastle was 7.5 ± 3.3 h. Maintenance of stable temperature in the sealed transport box was confirmed with mean temperature of 7.8 ± 1.0°C.

Reassessment of islet sterility, number, purity, integrity, and viability was undertaken in the Newcastle Biomanufacturing Cellular Therapies GMP suite. Sterility was maintained in all preparations confirmed by negative gram staining and absence of growth in culture. In early studies, each isolation batch was divided into semi-spermeable and standard blood transfusion bags for transport. Viability was comparable in both types of bag, but a significant loss of integrity was demonstrated in islets shipped in semipermeable bags (Table 2). This was believed to be due to shear stresses induced by islet movement during the journey in these more brittle bags. Moreover, on one occasion a semipermeable bag ruptured during the journey, precluding further consideration for transport of clinical preparations. No clear cause for the rupture was identified. However, possible friction with the side wall of the plastic shelving unit during the journey may have contributed. Standard blood transfusion bags remained intact throughout all transport journeys and were thus adopted for the clinical program.

Table 2.

Comparison of Islet Preparations in Semipermeable and Standard Blood Bags

	Viability (%)		Integrity (%)
Isolation Batch	Blood Bag	Permalife Bag	Blood Bag	Permalife Bag
1	90	90	80	60
2	90	90	80	60
3	50	55	50	40
Mean ± SD	77 ± 23	78 ± 20	72 ± 19	53 ± 12
t-Test (p)	0.2		0.02

For definitive comparisons of the 10 islet preparations transported in standard blood transfusion bags, six preparations were evaluated following centrifugation and resuspension confirming no significant decrease in islet equivalent number (Table 3). In parallel, a Standard Operating Procedure was validated enabling a 4-ml “side-arm” sample to be taken from the bag port under sterile conditions without the requirement of a GMP clean room or manipulation of the product in vitro. The robustness and utility of this method for maintaining islet sterility, purity, integrity, and viability was demonstrated, with no evidence of deterioration during the transport journey (Table 3). Preservation of normal β-cell morphology without degranulation was confirmed before and after transport (Table 3, Fig. 1).

Table 3.

Evaluation of Islet Quality Following Transport

	King’s	Newcastle	p-Value
IEQ (×1000)	181 ± 16	180 ± 36	0.6
Purity (%)	69 ± 12	69 ± 13	1.0
Integrity (%)	58 ± 16	58 ± 21	1.0
Viability (%)	72 ± 22	76 ± 18	0.1
Secretory granules/β-cell	177 ± 63	207 ± 66	0.2
GSIS stimulation index: static		1.3 ± 0.2	0.006
GSIS stimulation index: perifusion		3.4 ± 1.2	0.0002
Proinsulin/insulin processing (%) (intracellular)			0.03
2 mM glucose		70 ± 3
20 mM glucose		74 ± 3
Proinsulin/insulin processing (%) (secreted)			0.007
2 mM glucose		96 ± 0.4
20 mM glucose		97 ± 0.2
Insulin content in transport bag (U/100,000 IEQ)		0.6 ± 0.2
IFN-γ (pg/ml)		7.3 ± 1.8
IL-1β (pg/ml)		1.1 ± 0.1
IL-6 (pg/ml)		1.1 ± 0.05
TNF-α (pg/ml)		5.3 ± 0.7

Values are mean ± SD. GSIS, glucose-stimulated insulin secretion; IFN, interferon; IL, interleukin; TNF, tumor necrosis factor.

Figure 1.

Electron microscopic images of islets at King’s College islet isolation facility (left) and in Newcastle (right) demonstrating the presence of numerous secretory granules in both.

Preserved β-cell function in transported islets was confirmed by significant increases in insulin secretion in the presence of high glucose in both static incubation and dynamic perifusion studies (Table 3). Greater than 95% processing of proinsulin to insulin was demonstrated in secreted (pro)insulin, with a minor but significant increase in proinsulin processing on glucose stimulation.

To evaluate the potential of transplanting islets from the transport bag without a washing and resuspension step, insulin and proinflammatory cytokine levels in the media were assayed following the journey. Total insulin content was less than 2 units in all cases with cytokine concentrations low (Table 3).

Clinical Transplantation

Successful completion of the preclinical phase and approval of the validated Standard Operating Procedures provided sufficient evidence for National Health Service commissioning of a fully funded clinical islet transplant program in the UK for those with recurrent hypoglycemia despite optimal conventional therapy or suboptimal glycemic control with a functioning renal transplant. This was implemented on April 1, 2008, configured around central clinically certified islet isolation facilities in London and Oxford with the potential for transport of purified islets to additional satellite transplant centers in Newcastle, Manchester, and Bristol to ensure equity of access throughout England.

To date, five patients have received a total of six transplants in Newcastle with islets isolated in King’s College Hospital and Oxford, transported and evaluated according to the protocols developed in the preclinical phase. Islet yields of 224,000–600,000 IEQ were attained with maintained sterility, purity, integrity, and viability confirmed on arrival by ‘side-arm” analysis (Table 4). Four female recipients have received single islet transplants alone (ITA), with a single female islet after kidney (IAK) recipient receiving two transplants (the second 3 months after the first). Induction for the first recipient was intravenous basiliximab, with all other recipients receiving subcutaneous alemtuzumab immediately after each islet infusion.

Table 4.

Islet Preparations Employed in the Clinical Transplant Program

Isolation Site	CIT	Time From Islet Packaging Until Transplant	Transplant Mass (IEQ/kg)	Viability (Pretransport)	Viability (Posttransport)	Recipient No.	Age at Transplant (Years)	Primary Graft Function	Loss of Graft Function	Prevention of Severe Hypoglycemia
King’s	4.1	7.5	5,200	90	80	1	54	yes	6 months	yes
King’s	4.3	9.5	5,368	90	90	2	44	yes	—	yes
King’s	8.5	9.0	6,700	90	90	2	44	—	12 months	yes
Oxford	3.0	8.0	12,000	90	85	3	60	yes	no	yes
King’s	3.1	9.8	6,349	90	90	4	64	yes	no	yes
Oxford	9.0	7.1	4,058	90	92	5	49	yes	no	yes
Mean ± SD	5.3 ± 2.7	8.5 ± 1.1	6,613 ± 2,799	90 ± 0	88 ± 5		54 ± 8

Alemtuzumab was well tolerated without adverse effects other than for transient asymptomatic pyrexia (single episode 37.5–38°C for less than 40 min) 24 h following administration. Target trough tacrolimus levels were maintained in all with no infusion- or immunosuppression-related serious adverse events.

Primary graft function (defined as positive C-peptide (>50 pmol/L) at 28 days after first transplant) in addition to prevention of recurrent severe hypoglycemia was achieved in all. In the first recipient, there was evidence of loss of islet integrity on a second in vitro assessment of the side-arm sample following transplant. This recipient rapidly developed donor-specific antibodies with loss of detectable C-peptide at 6 months posttransplant. However, the patient has maintained improvement in hypoglycemia awareness and reduction in severe hypoglycemia. In all subsequent transplants, an initial period of overnight culture at 37°C in CMRL medium supplemented with 5% human serum albumin in a 5% CO₂ incubator was added prior to islet transport, to enable any loss of viability or integrity to become manifest.

For the second transplant in the second recipient (performed at 3 months post-first transplant), although overall viability was maintained at 90% posttransport, occasional PI-positive cells were seen in the core of the largest islets (>300 μm diameter). This recipient incrementally lost graft function over the subsequent months, becoming C-peptide negative at 12 months after first graft. This was despite an HbA1c of 4.7% at 6 months. Any evidence of central necrosis has since been added as an exclusion criterion for clinical islet transplantation in the UK program.

Good graft function after a single transplant has been maintained in all other recipients with absolute prevention of further severe hypoglycemia and HbA1c <7%, fulfilling National Institute for Health and Clinical Excellence (NICE) criteria for transplant success (10). Insulin requirements have been reduced, with insulin independence maintained for >24 months in recipient 3, who received >10,000 IEQ/kg (Table 4).

DISCUSSION

This study confirms the safety and utility of islet allo-transplantation in the UK employing accredited clinically proven central isolation facilities and satellite transplant centers, ensuring equity of access regardless of geographical location. While a small number of transplant programs internationally use transported islets, protocols for these programs remain specific to the local setting. Transport of both whole organ pancreas and isolated islets was carried out using air chartered flights in the Miami–Houston program with cold ischemic times of less than 7 h (5), while the Groupe Rhin-Rhône-Alpes-Genève pour la Transplantation d’Ilots de Langerhans (GRAGIL) Swiss/French collaborative network employed road services with cold ischemic time of up to 8 h (3,7). In the UK the most suitable transport method was road ambulance, enabling target cold ischemic time of less than 9 h. Pancreases in the Miami–Houston program were packaged either in TLM or UWS (5), while in the GRAGIL program pancreas packaging was carried out using UWS (3,7). In the preclinical phase of the UK program, both TLM and UWS alone were evaluated with no significant differences in islet quality. Therefore, UWS, which is well established and currently used in the retrieval Standard Operating Procedure for whole vascularized pancreas transplantation, was used to simplify the procedure for national adoption.

Islets for the Miami–Houston program were packaged in gas-permeable bags in variable but relatively large volumes of MM1 medium at room temperature. Islet transport journey to transplantation center was by air and lasted <5 h (5). Upon arrival islets were washed and reassessed again prior to transplantation. In the GRAGIL program islets were initially packaged in 50-ml syringes in X-vivo medium but later changed to purpose made bags (7) in CMRL-1066 medium. Islets were transported by road at room temperature and the transport journey lasted <5 h. Islets were transplanted upon arrival directly without repeated quality assessments.

In the preclinical phase of our program both gas-permeable and blood bags were evaluated and transport was carried out at a relatively cool temperatures to reduce biological activity and thus enable transplantation without a further wash step at the receiving center. Within this setting, no meaningful advantage of gas-permeable bags on islet quality could be detected. Indeed, there was evidence of increased islet fragmentation and therefore only standard blood bags were used in the clinical program. Repeated assessment at Newcastle confirmed maintained sterility, number, purity, integrity, and function, together with the utility of using a sample obtained directly from the bag via a side arm to confirming quality immediately prior to transplantation. Moreover, safety of transplanting islets directly without a washing step at the receiving centre was supported by low levels of insulin and inflammatory cytokines in the transport/transplant medium.

As journey times of <10 h can always be achieved by road between all of the UK centers with no evidence of product deterioration over this period, there was no need to consider shipping by air. Equally there was no evidence that oxygen concentration or nutrient consumption were limiting at the relatively cool temperature employed with sufficient media volume and gas headspace. Semipermeable bags have been successfully used to maintain islets in longer transport journeys (6) but decreased number and viability in conventional versus semipermeable bags has only been confirmed in incubations of 18 h or more at 20°C (5). The possibility for puncturing of semipermeable bags and inadequate protection against hypoxia at the islet core, certainly at temperatures above 20°C, has been reported (1). Islet integrity following transport by road in semipermeable bags has not previously been reported to our knowledge. In the present studies, all journeys were associated with decreased integrity, and careful reevaluation of islet integrity and fragmentation following transport using semipermeable bags in other programs is recommended.

Preclinical evaluation underpinned approval, funding, and implementation of a national program for islet transplantation in England providing equity of access. This enabled the first transplants of islets in the UK transported from a geographically distant isolation facility. Success was confirmed by attainment of primary graft function and resolution of severe hypoglycemia in all recipients.

Relatively early loss of C-peptide positivity in the first two recipients, despite maintenance of adequate immunosuppression throughout, underlines the importance of pretransport culture to enable any loss of integrity or decreased viability to become manifest pretransplant (5), in addition to the necessity of careful reevaluation of quality posttransport with criteria precluding transplantation, including any evidence of islet central necrosis.

Administration of single dose alemtuzumab subcutaneously immediately following islet infusion avoided induction of T-cell depletion prior to confirmation of successful completion of the islet transplant procedure. Preceded by steroid/antihistamine “cover,” this was both well tolerated and effective with no evidence of cytokine release beyond transient asymptomatic pyrexia. Peri-transplant serum cytokine profiles are being analyzed in ongoing studies to further characterize any potential for alemtuzumab-induced proinflammatory response during engraftment. This remains the favored induction agent in the UK clinical islet transplant program, with longer term outcome data in a larger number of recipients awaited.