Current Assessment of Clinical Pancreatic Islet Allotransplantation

Preeti Chhabra ^†, Kenneth L. Brayman ^†, *

Department of Surgery, Transplantation Division, University of Virginia, Charlottesville, VA, USA

† These authors contributed equally to this work.

* Correspondence: Kenneth L. Brayman

Academic Editors: Kenneth L. Brayman and Preeti Chhabra

Special Issue: Islet Cell Transplantation

Received: September 20, 2020 | Accepted: November 4, 2020 | Published: Novemeber 16, 2020

OBM Transplantation 2020, Volume 4, Issue 4, doi:10.21926/obm.transplant.2004125

Recommended citation: Chhabra P, Brayman KL. Current Assessment of Clinical Pancreatic Islet Allotransplantation. OBM Transplantation 2020; 4(4): 125; doi:10.21926/obm.transplant.2004125.

© 2020 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.

1. Introduction

Pancreas or islet transplantation is currently the only known cure for Type 1 Diabetes (T1D) [1,2], a chronic, progressive autoimmune disease that affects an estimated 1.25 million American children and adults [3,4]. Classic symptoms of the disease include polydipsia, polyuria, and polyphagia, while diagnosis is usually confirmed by hyperglycemia, low or undetectable serum C-peptide levels, elevated glycosylated hemoglobin (HbA1c), and the presence of one or more positive autoantibody markers [5]. The current treatment for T1D consists of intensive diet management coupled with exogenous insulin therapy [6,7]. Multiple daily injections of insulin, continuous glucose monitoring systems and insulin pumps, or an artificial pancreas device such as Medtronic's MiniMed 670G system constitute the latter [7,8,9,10]. Although the tightened glycemic control resulting from insulin therapy has proved highly successful in reducing secondary complications, unfortunately the treatment often renders patients susceptible to increased risk of hypoglycemic episodes and unawareness [11]. To date, allotransplantation of beta cells remains the most reliable and consistent approach for curing T1D.

Pancreatic islet transplantation is a beta cell replacement strategy for clinically restoring glycometabolic control in type 1 diabetic patients that suffer from complications of recalcitrant glycemic lability or debilitating incidences of hypoglycemic episodes and unawareness [12,13,14]. The procedure first gained credibility as a viable cure in the year 2000, when Shapiro et al demonstrated that transplantation of an adequate mass of islets combined with glucocorticoid-free immunosuppression could result in insulin independence with excellent metabolic control [15]. Since then, incremental advances in a number of parallel fields such as donor selection, islet product quality, anti-rejection/anti-inflammatory agents and state-of-art technology helped evolve islet transplantation into an efficacious clinical means for routinely achieving normoglycemia. Currently, select centers (University of Minnesota (UMN), University of Alberta, Edmonton (UAlberta), University of Geneva (UNIGE), University of Lille (ULille) and University of California San Francisco (UCSF)) have demonstrated achievement of insulin independence in nearly 50% of recipients at the five years post-transplantation mark using T-cell depletional induction therapies, bringing it on par with the results observed with whole pancreas transplantation [14,16,17,18,19,20,21,22]. The procedure also offers substantial benefits when compared to pancreatic transplantation, being minimally invasive while improving the patient’s overall metabolic homeostasis as well as quality of life. However, despite the significant progress that has been made in this field during the past decade, it is still not considered as a mainstream treatment option for patients with T1D. Factors such as the requirement of multiple donors, graft rejection, the recurrence of autoimmunity, the harmful side-effects of chronic diabetogenic and nephrotoxic immunosuppression, as well as financial and regulatory approval issues still need to be overcome in order to allow its inclusion into the clinical repertoire as a cure for T1D. This review explores the current status of islet transplantation and the steps that are being taken to expand the availability of this procedure as a fundable clinical option for curing T1D.

1.1 Historical Perspective of Islet Transplantation

The use of insulin as a treatment for T1D was first reported in 1922 by Banting and colleagues [23]. However, the theory that pancreatic extracts contained a vital substance that could relieve diabetic symptoms had already been a topic of debate before this report. As early as in 1892, von Merring and Minkowski demonstrated that mortality associated with total pancreactomy could be prevented by subcutaneous transplantation of autologous pancreatic fragments into a diabetic canine recipient [24]. Soon after in 1894, Watson-Williams and Harsant transplanted fragments of sheep pancreas into a teenage boy with end-stage diabetes and demonstrated temporary improvement in glycosuria [25]. In 1916, Frederick Charles Pybus transplanted human allogeneic cadaveric pancreatic fragments subcutaneously into two diabetic patients with no success [26].

The concept of physically separating exocrine and endocrine pancreas had first been contemplated by Leonid Ssobolew [27] as early as in 1902, and continued to gain momentum in the years that followed as numerous techniques to isolate islets were investigated. Moskalewsky was the first to use a discontinuous gradient to purify pig pancreatic islets with Ficoll in 1965 [28]; and the technique was subsequently perfected in Paul Lacy's laboratory [29]. They used a novel technique to isolate rat islets that involved distending the pancreas with saline followed by enzymatic digestion. Ground-breaking experiments by Ballinger and Lacy in rodents followed in the period 1972-1973 that demonstrated reversal of chemically induced diabetes in rats by intact islet transplantation [30]. These early preclinical studies paved the way for using the technique as a potential clinical treatment option for T1D. The same team also demonstrated the vital impact of the site of transplantation on transplantation outcomes, demonstrating islet embolization into the liver through the portal vein, a clinically relevant site that still remains the preferred site for islet transplantation [31]. Between 1979 to 1980, Sutherland and Najarian performed both autologous and allogeneic islet transplantations using dispersed pancreatic islet microfragments following pancreactomy [32,33,34]. They performed intraportal islet transplantations in 10 patients with surgical induced diabetes using islet autografts that were infused back following islet isolation. Three of these patients achieved insulin independence for 1, 9 and 38 months respectively.

While working in Paul Lacy’s laboratory, Camillo Ricordi and colleagues developed the Ricordi® Chamber and perfected the semi-automated method for islet isolation that enabled successful harvesting of sufficient numbers of human islets for transplantation [35,36]. Briefly, the Ricordi® Chamber consisted of lower and upper stainless steel chambers separated by a stainless steel screen (280 um mesh). The lower cylindrical chamber had two inlet ports and a third opening on the side for a temperature probe. The upper chamber was conical with one output opening. Both chambers were held together by three screws and the device was connected to a shaker (Model 75, Burrel, Pittsburgh, PA) by a stainless steel bar. Enzyme infused pancreas was loaded into the lower chamber along with seven glass marbles. The chamber was filled with additional collagenase solution, the screen placed in position and the device closed. The shaker was then activated (320 oscillations/minute) and the peristaltic pump turned on with a flow rate of 40 mL/min. Digestion of the pancreas was conducted in a closed system in which Hanks' solution was aspirated through a filter (94 um mesh) placed in the upper portion of the recirculation cylinder, passed through the pump to a heating circuit and finally reached the lower chamber of the isolator progressively diluting the collagenase solution inside the chamber. The heating circuit allowed the temperature in the digestion chamber to progressively increase (1^oC/min) until 37^oC was reached, following which a heating circuit bypass was switched on and off to maintain a stable temperature. From the lower chamber, the solution then passed through the 280um screen to a cooling circuit that inactivated the collagenase and finally returned to the lower portion of the recirculation cylinder. Both the cooling circuit and the recirculation cylinder were placed in an ice bath in order to inactivate the collagenase and preserve the islets. The solution was then drawn from the top of the recirculation cylinder through the 94um filter while the islets remain settled at the bottom of the cylinder. The islet preparation was sampled every two minutes from a stopcock near the output opening of the digestion chamber to monitor the digestion. When free islets were detected, the recirculation cylinder and the heating circuit were bypassed and the islet separation conducted on an open system in which the temperature progressively decreased, and the collagenase solution was diluted by fresh Hanks’ solution. The islet preparation was then collected in 2-Liter flasks. The flow rate in the open system phase increased up to 200 mL/min, allowing a faster turnover of the digestion chamber content. The isolation procedure was generally terminated after 30-60 min, when islets were no longer detected in the samples [35].

By 1990, the first successful series of clinical human islet allotransplantation was reported by the Pittsburgh group wherein long-term insulin independence could be attained using a steroid-free immunosuppressive regimen that included FK506 (Prograf) [37,38]. These exciting results invigorated efforts in the field. During that decade, several centers including those at University of Miami (UM), UAlberta, Washington University, University of Pennsylvania (UPenn), UMN, and San Raffaele University, Milan (UniSR) incorporated innovative changes to existing protocols to improve islet transplantation outcomes [39,40,41]. However it was in the year 2000 that the islet transplantation group consisting of Shapiro and colleagues at UAlberta demonstrated achievement of insulin independence for >1 year post-transplantation with sustained C-peptide production in all seven consecutive participants who had received islets transplants, using what is now famously known as the “Edmonton protocol” [15]. The protocol briefly consisted of a corticosteroid free immunosuppressive protocol consisting of a combination of two potent immunosuppressants: rapamycin and low-dose tacrolimus, together with an anti-CD25 antibody to protect against rejection and recurrent autoimmunity. Using this protocol, the islet mass was augmented with islet preparations harvested from two or more pancreata, infusing a total islet dose substantially higher than previously used in clinical islet trials (>13,000 islet equivalents (IEQs) per kilogram (kg) recipient body weight) [42]. Although the Edmonton protocol consistently achieved short-term outcomes on par with pancreas transplantation at 1 year follow-up, insulin independence was frequently lost in the graft recipients by the 3-5 year mark, with only 10% demonstrating insulin independence at the five year mark [43]. However, partial graft function (defined as partial insulin dependence) consistently persisted in 80% of graft recipients contributing to decreased episodes of severe hypoglycemia, reduced HbA1c levels, and improved cardiovascular function [44]. Furthermore, stabilization of diabetic retinopathy and neuropathy was observed, resulting in improved quality of life with increased glycemic awareness. From 2006-2012, a spate of activity ensued with several centers worldwide and as well as independent Clinical Islet Transplantation (CIT) centers (for example, UAlberta, UMN, UCSF, Geneva-GRAGIL, and ULille) reporting insulin-independence rates of more than 50% at five years using T-cell depletion induction therapies [21,22,45,46,47,48,49,50]. Importantly, their results closely approximated the International Pancreas Transplant Registry data of solitary vascularized pancreas transplant [21,22,51,52,53,54]. Noteworthy was that these results demonstrated that the procedure was not associated with the incumbent risks that attend a major surgical intervention like pancreas transplantation [21,22,54].

1.2 The Collaborative Islet Transplant Registry

The Collaborative Islet Transplant Registry (CITR) was established in 2001 by the National Institute of Diabetes & Digestive & Kidney Diseases to compile data from all islet transplant programs in North America from 1999 to the present. Additional funding was granted by the Juvenile Diabetes Research Foundation (JDRF) to include JDRF-funded European and Australian centers from 2006-2015. The cumulated data from North American, European and Australian centers are pooled for analyses and are included in the annual reports.

Briefly, data is collected from the following groups of study subjects: A) Islet allotransplantation performed as either islet transplant alone or islet after kidney (IAK) [55]. A small number of cases have been performed as islet simultaneous with kidney or kidney-after-islet; and B) Autologous islet transplantation, performed after total pancreatectomy. The CITR 10th Annual Report focuses on the islet allotransplantation recipients. It states that 1086 islet allotransplant recipients (877 islet alone, 183 islet after kidney, 24 simultaneous islet kidney, and 2 kidney after islet) have received 2,150 islet allograft infusions, representing a 7% increase in the number of recipients, a 12% increase in the number of infusion procedures, and 8% increase in donors, compared to the 9th Report during the period from 1999 to 2015. Also, a total of 827 autologous islet transplant recipients were registered in CITR, as of Sept 30, 2015. In North America 33 out of 42 programs have performed islet allograft transplantations with 669 total recipients and 1,321 total infusions.

2. The Islet Transplantation Process

2.1 Indications for Clinical Islet Transplantation

The main indication for an allogeneic islet transplant is undetectable C-peptide levels (<0.3ng/mL) with severe glycemic lability complicated with recurrent hypoglycemic episodes and impaired hypoglycemic awareness [2,39]. Also included are patients refractory to efforts aimed at stabilizing glycometabolic control such as intensive insulin therapy, improved monitoring, insulin pumps, or continuous glucose monitoring systems. Evidence of T1D (C-peptide <0.3ng/mL) with sufficient duration (>5 years) is required to justify that adequate opportunity for optimized medical management has been provided. Age is restricted to >18 years, and subjects >90kg and/or with daily insulin requirement >1.0 U/kg and/or with a high body mass index (BMI >30kg/m²) are usually avoided to prevent the futility of transplantation in the presence of marked insulin resistance or excessively high insulin requirements [14].

Stringent selection criteria have been put in place to justify replacing injectable insulin with lifelong immunosuppression. Usually an objective selection criteria for islet transplantation alone include a Clark score of ≥4 (0 = no hypoglycemia, ≥4 = hypoglycemia unawareness), HYPO score ≥1000; lability index (LI) score ≥400 mmol/L2/h/wk, or a combination of these variables [2,21,51]. Patients having subcutaneous insulin resistance requiring intraperitoneal or intravenous infusions are also considered. Also included are patients that do not qualify as candidates for whole-pancreas transplantation or are unable to accept the risks associated with major surgery.

According to latest CITR annual report [55], at the time of their first Islet transplant, islet transplant alone recipients were 14-74 years of age (mean 46±10.5 years), had T1D for 2-61 years (mean 29±11.5 years), and 77% had very poor diabetes control including hypoglycemia unawareness. Poor glycemic control was defined as frequent episodes of critically low blood sugar levels (which often result as a reaction to injected insulin, and require the assistance of another person to avert a possibly life-threatening loss of consciousness), wide swings in blood sugar levels (blood glucose lability), or consistently high HbA1c levels (>8% of total hemoglobin). Typically, those selected displayed frequent metabolic instability involving hypo-and hyperglycemia, ketoacidosis despite intensive insulin therapy; hypoglycemic unawareness (<54mg/dL); and severe metabolic lability (mean amplitude of glycemic excursion >11.1mmol/L or 200mg/dL).

There is less agreement on the exclusion criteria, however absolute contraindications is detectable C-peptide levels (>0.3ng/mL). Relative contraindications include insulin requirements (>0.7units/kg/day) or <15units/day, HbA1c >10%, BMI >26kg/m² or weight <50kg, Creatinine >1.5mg/dL and/or albuminuria >300mg/24h or measured glomerular filtration rate (GFR) <80mL/min/1.73m², untreated arterial disease, history of panel reactive anti-human leukocyte antigen (HLA) antibodies, inability to comply with immunosuppression and significant comorbidities, as well as the presence of infections or psychiatric diseases [14,56,57].

2.2 Islet Isolation

The islet transplantation process begins with procurement of the pancreas, usually from a heart-beating, brain-dead organ donor or in the case of autologous islet transplantation, from a recipient undergoing therapeutic pancreactomy [2]. The pancreas is transported in University of Wisconsin (UW) or histidine-tryptophan-ketoglutarate (HTK) preservation solution to a sterile good manufacturing practice (GMP) facility, avoiding prolonged cold storage of the pancreas [12]. Multiple reports indicate that other preservation solutions may be used as effective alternatives to UWS and HTK [58,59]. For example, Celsior is an extracellular, low-viscosity preservation solution that was originally designed for cardiac transplantation. A study comparing Celsior with UWS for pancreas preservation demonstrated that two-year recipient survival rates, two-year graft survival rates, pancreas leakage rates, and clinical graft pancreatitis rates were similar in both groups [58]. However, colloid-free preservation solutions were reported to be inferior to UWS for preservation of pancreases intended to be used for islet isolation. Islet isolation yields from pancreases preserved with Celsior solution were found to be 2.1-fold lower than those obtained when UWS was used [59,60]. Solution de conservation d’organes et de tissus (SCOT), an extracellular solution containing polyethylene glycol (PEG) has been shown to have some immunoprotective effects on islet cells [61]. Cell swelling and pancreas edema were not significant following 12 hours of cold storage with SCOT when compared with UWS, and SCOT could improve islet yield when used during isolation and could prolong islet allograft survival without immunosuppression when used for culture, as compared to control solutions [59,61]. Perfluorohexyloctane/Polydimethylsiloxane 5 (F6H8S5) too has been demonstrated to be a suitable preservation solution for pancreases intended for islet isolation, allowing for extended cold ischemic time without inflicting islet function and numbers [62]. A technique known as the two-layer method (TLM) for pancreas preservation was reported to enhance islet yield and quality by improving pancreas oxygenation during preservation [62]. Following reports in 2002 by UMN [63], UM [64], and UAlberta [65,66] that indicated improved islet yields and increased islet transplantation opportunities using TLM, this technique was widely used by islet transplant centers worldwide. However, several clinical studies since have demonstrated no beneficial effect of the TLM with preoxygenated perfluorocarbon on islet isolation and islet transplant outcomes [67,68,69].

Once received, the pancreas is cleaned and cannulated and infused with enzyme in the GMP facility. The distended pancreas is cut into seven to ten pieces and subjected to continuous, temperature controlled, mechanically enhanced enzymatic digestion in a dissociation or filtration chamber known as the Ricordi® Chamber that allows immediate cooling and removal of the dissociated islets to avoid overdigestion, thus preserving their morphologic and functional islet integrity [2]. Liberase H1 was the commonly used enzyme until the discovery of potential safety risks regarding the contraction of bovine spongiform encephalopathy [2,12,70]. Consequently, collagenase NB1 supplemented with neutral protease (NP) (SERVA Electrophoresis GmbH, Heidelberg, Germany) was adopted by many centers globally as an alternative despite its lower islet isolation outcomes [2,12,70,71]. The latter was attributed to degradation of collagenase C1 into various breakdown products with reduced activity [72]. The combination of CIzyme collagenase HA (containing nondegradable class I [60%] and II [40%]) and NP although effective in isolating islets for transplantation, however were not GMP grade products [73]. Currently, liberase mammalian tissue-free collagenase/ thermolysin (MTF C/T) (Roche Diagnostics) is being used routinely for human islet isolation for clinical islet transplantation [73].

Pooled tissue containing islets are purified on a Ficoll, Biocoll, Iodixanol, or UW-Biocoll density gradient in a COBE2991 processor (Terumo BCT, Gambro BCT, Inc., Lakewood, CO) under cold conditions, which separates them from exocrine and other non-islet tissue [2,12,14,21]. The purified islets are washed, pelleted and resuspended, following which an aliquot is removed to quantify islet yield (dithizone-stained islets), evaluate islet size and perform islet count (using light microscopy), assess viability (fluorescein diacetate-propidium iodide staining), sterility and pyrogenicity (endotoxin assay, gram stain and tests for aerobic and anaerobic microbial, fungal, and mycoplasmal contamination), purity (visual morphometric examination of dithizone-stained acinar-free islets using light microscopy) and function (perifusion, glucose-stimulated insulin secretion assay) [2]. Within 24 hours of isolation, the resuspended islets are loaded into a gravity-fed transfer bag for intraportal vein infusion into the eligible transplant recipient [2].

A healthy human pancreas contains up to 4-15 million islets [57,74]. However, the yield and purity of isolated purified islets are highly variable with islet equivalents count ranging from 100,000 to 300,000 per pancreas [57,75]. The isolated islets are then infused into the hepatic portal system of the recipient. A target islet mass of approximately 11,000 islet IEQs per kg of recipient body weight is aimed for, translating into two or three islet infusions from multiple donors [21]. However the minimum requirements include >30% purity of the preparation being islets, membrane integrity-viability ≥70%, a final enriched islet cell product in <5 cm³ of tissue pellet, negative Gram stain, content ≤5 endotoxin U/kg (recipient body weight) and most importantly, adequate number of islets i.e. ≥5,000 islet equivalent recipient body weight for the first infusion and ≥3,000 for further infusions [76,77,78,79]. According to the latest CITR annual report [55], the total cell volume infused has declined appreciably over the eras (4.0±0.1 in 1999-2002 to 3.1±0.1 in 2011-2014), while total IEQs and IEQ/kg recipient have remained remarkably stable. Total beta cells and beta cells/kg were higher for IAK (5.3±0.6 vs. 3.6±0.2) and have increased over the eras (3.0±0.3 to 4.5±0.4). Endotoxin (both total and /kg) has declined sharply over the eras (0.5±0.1 to 0.1±0.05) as has the Stimulation index (3.6±0.3 to 2.8±0.2). The eras consist of 1999-2002, 2003-2006, 2007-2010, 2011-2014 and 2015-2018.

2.3 Intra-hepatic Portal Vein Islet Transplantation

The hepatic parenchyma through the portal system of the recipient is the implantation site that is routinely used although other sites are being explored [76]. The final islet-cell product suspended in transplant media containing 70 units heparin/kg recipient body weight is loaded into a gravity-fed transfer bag for infusion following percutaneous catheterization of the portal vein [14]. This is a minimally invasive procedure performed under local anesthesia in a radiology suite by an interventional radiologist under ultrasonographic and fluoroscopic guidance [14,80]. The procedure begins with measurement of portal pressure to ensure absence of portal hypertension. The islets (~5 mL; up to 10 mL tissue pellet) are suspended in 200 to 400 mL transplant media with heparin and human albumin contained in a plastic infusion bag [78]. Briefly, a small (7-French) catheter is inserted into the main portal vein and the islets are dripped slowly through the intravenous line connecting the harvested islet bag with the portal catheter. Both, at the midpoint and at the conclusion of the infusion which typically takes 30 to 45 minutes, the portal pressure is measured. If percutaneous access via the portal vein fails, then the procedure can also be performed under general anesthesia during a limited mini-laparotomy with intravenous delivery through the mesocolonic vein [78] or by recanalization of the obliterated left umbilical vein to enable access to the left portal system. In these instances, a dual-lumen 9 French Broviac-type central-line catheter is advanced into the main portal vein for simultaneous monitoring of portal pressure (through the smaller lumen while islets are infused [14,81,82].

Following transplantation, subcutaneous heparin and insulin are administered for at least two weeks to promote islet engraftment [78]. The rationale being that patients receiving peri-transplant insulin and heparin infusions are more likely to become insulin independent and have a significantly greater percent reduction in insulin requirement [48]. This could be attributed to protection from the instant blood-mediated inflammatory reaction (IBMIR) combined with islet rest and other anti-inflammatory effects [21,48]. IBMIR is a thrombotic reaction characterized by the activation of coagulation and complement cascades and by platelet consumption that results in the disruption of islet morphology and function [13]. Heparin may be beneficial due to its ability to reduce inflammation via inhibition of complement activation, reduce the generation of reactive oxygen species and secretion of proinflammatory cytokines, inhibit transcription factor nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB), and limit macrophage migration [2,48]. The benefits of injecting insulin would include protection of the grafted islets from hyperglycemia, resulting in rest and gain vascular supply [2,48,83]. Insulin may also suppress the generation of reactive oxygen species and the binding of transcription factor NFκB [2,48]. By reducing plasma levels of tissue factor and plasminogen early activator inhibitor-1, insulin may also exert an antithrombotic effect, protecting against IBMIR [48].

2.4 Intra-portal Vein Infusion-related Complications

Complications that could arise as a result of intra-portal vein infusion are a transient elevation of transaminase, post-infusion portal vein thrombosis and hepatic infarct [14,78,80]. With the optimization of the access technique wherein smaller catheters are used and the puncture tract sealed off, the risk of bleeding has significantly decreased [78,84,85]. This in turn allows for a more aggressive use of heparin, which combined with limited tissue pellet volume infused (packed cell volume <5 mL) further reduces the risk of portal vein thrombosis close to zero. Complications arising due to partial portal vein thrombosis are also very rare [85,86]. A theoretical increased risk of allosensitization also exists since islet allotransplantation requires islet infusions from 2 to 3 donors. It has been clinically observed that discontinuation of immunosuppression due to graft failure or side effects, results in approximately 70% of the patients developing donor-specific antibody and high panel-reactive antibody (PRA) (>50%) [87,88,89]. However, in the presence of adequate immunosuppres­sion, this risk remains very small even after their fourth and fifth islet infusion [88,90]. Moreover, the risk of sensitization remains the same as after one islet infusion despite increased number of HLA class I mismatches with each infusion [91]. In fact, exposure to repeat HLA class I mismatch in subsequent islet infusions results in less frequent development of de novo HLA class I antibodies compared with new class I mismatch. Poor graft outcome and increased risk for donor-specific antibodies is associated with >20% PRA [78]. Posttransplant monitoring of intraportal islets has proven difficult. Positron emission tomography combined with computed tomography may have the potential to assess the early posttransplant loss of islets [92].

3. Limitations of Clinical Islet Transplantation and Strategies to Address the Limitations

Widespread application of clinical islet transplantation is hindered by a shortage of donor pancreata, the harmful side effects associated with chronic immunosuppressive therapy, innate and alloimmune graft rejection, and the recurrence of autoimmunity.

3.1 Shortage of Donor Pancreata

Usually, islet transplant recipients require islets from two or more pancreata to regain permanent glycometabolic control, which becomes a problem due to the shortage of donor pancreata. In order for clinical islet transplantation to prove cost effective, and to increase availability of islets for transplantation and minimize the risk of donor allosensitization, the best scenario would be to use islets from a single donor similar to whole-pancreas transplantation. While several groups have met with limited success using single-donor islet transplants, routine single-donor islet engraftment success has not been observed universally [2,51,57,93].

Several strategies have been investigated to address the issue of shortage of donor tissue. These include amongst others, obtaining islets from donors after cardiac death (DCD), from partial pancreactomized living donors, as well as from alternate, sustainable sources of islets such as porcine islets. Another avenue that has been explored is the directed differentiation of stem or progenitor cells (e.g. embryonic stem cells) or induced pluripotent stem cells into insulin-producing beta cells [2,94,95,96]. Apart from these obvious stem cell choices, others such as mesenchymal stem cells, adipose tissue-derived stem cells, umbilical cord blood-derived stem cells, as well as organ-specific facultative progenitor cells from the liver and the pancreas, have also been tested for their ability to differentiate into bona fide insulin-secreting beta cells. Additionally, the generation of insulin-producing beta cells by transdifferentiation of alpha cells, non-islet pancreatic ductal cells, acinar cells, hepatic cells, and bile duct epithelial cells into insulin-producing beta cells as well as agents such as Exenatide that stimulate regeneration of beta cells have also been studied [2,94,95,96].

Apart from the overall shortage of cadaveric donor pancreata, the quality of a donor pancreas also plays an important role in determining transplant outcomes, with factors such as donor age and BMI, cause of death, duration of hospital stay, organ procurement team and cold ischemia time acting as major contributors [97,98,99,100,101,102]. Currently, the islet transplantation procedure is hindered by the quality of pancreas allocated. For example, donor pancreases unsuitable for whole-pancreas transplantation such as pancreas from older, overweight donors that are associated with decreased survival when utilized in whole-organ pancreas transplant are allocated for islet transplantation [2,103]. In order to better long-term islet transplantation outcomes, this disparity in organ allocation will definitely need to be addressed.

Another important point is the ischemic, mechanical, and enzymatic stress that islets are subjected to during the isolation process [12]. Efforts to protect islets by either tweaking isolation techniques or by treating the islets are ongoing. Currently many clinical trials are focusing on cytoprotective strategies to enhance transplantation outcomes. Some of these are listed below:

1. lPhase I/II clinical trial # NCT02520076 studies the use of Serine protease inhibitor Alpha-1 Antitrypsin (AAT) in islet transplant to decrease the amount of cell death caused by general inflammation and improve single donor success rate in clinical islet transplantation. #NCT02464878 is another multicenter trial that focuses on the effect of AAT on islet transplant engraftment and durability after renal transplant. In the clinical trial #NCT02713997, following removal of pancreas from patients with chronic pancreatitis, autoislets are transplanted back into patient’s liver along with three inflammatory therapies arms: Tumor necrosis factor (TNF)-alpha, AAT and standard care in order to see if anti-inflammatory therapy improves outcomes after total pancreatectomy with islet autotransplantation (TPIAT).
2. lIn clinical trial #NCT03073577, the use of supplementing islet preservation medium during islet isolation process with anti-aging glycopeptide (PKX-001) is being studied. Islets are maintained for minimal 6-72 h in supplemented CMRL1066-based media containing PKX-001 until the time of transplant.
3. lIslet Isolation using MnTE-2-PyP (BMX-010) is a phase I pilot study (#NCT02457858) that has been completed. The investigators hypothesized that addition of antioxidant BMX-010 to perfusion solution, digestion solution and culture medium during the islet isolation process could lead to greater preservation of islet mass and metabolic function, such as improved islet yield, viability and functional potency.
4. lClinical trial # NCT00678990 is a phase I, open study (not yet recruiting) to evaluate safety and efficacy of allogenic islet transplantation using islets coated with immobilized heparin. In this study, prior to transplantation, islets will be surface modified to carry immobilized heparin. The primary objective is to investigate safety and efficacy of islet allotransplantation using islets coated with immobilized heparin. The modification with heparin is anticipated to protect islets from blood (coagulation and inflammation) as a means to preserve a larger portion of the islets in the initial transplant phase.
5. lThe effect of selective complement inhibition using complement inhibitor Eculizumab combined with standard anticoagulation during and after transplantation, in reducing the extent of early tissue loss after portal infusion of islets, has also been studied in completed clinical trial # NCT02727608. 
6. lThe C-X-C Motif Chemokine Ligand 8 (CXCL8) plays a key role in the recruitment and activation of polymorphonuclear neutrophils in post ischemia reperfusion injury after organ transplantation. Reparixin is a low molecular weight blocker of CXCL8 biological activity in clinical development. Reparixin can be used to immunomodulate and control of nonspecific inflammatory events surrounding the early phases of pancreatic islet transplantation. A phase II/III clinical trial (#NCT01967888) is being conducted to assess whether reparixin leads to improved transplant outcome as measured by the proportion of insulin-independent patients following TPIAT. 
7. lIn a terminated clinical trial #NCT00853944, the effect of Sitagliptin on graft function following islet transplantation was being studied to evaluate whether administration of Sitagliptin to islet transplant recipients could enhance engraftment when administered up to one-year post transplant. The study was terminated due to lack of enrollment.
8. lInvestigator initiated, Phase I/II clinical trial #NCT01653899 studies Caspase inhibition in islet transplantation. In this study, type 1 diabetic participants receive a 14-day oral treatment of the investigational caspase inhibitor drug IDN-6556 following their first islet transplant.
9. lCombinatorial cellular immunotherapy approaches are also being investigated in the field of islet autotransplantation. For example, clinical trial # NCT02384018 is a mesenchymal stem cell and islet co-transplantation study aimed to evaluate safety and tolerability of autologous bone-marrow derived mesenchymal stromal cells (BM-MSCs) in chronic pancreatitis patients who concurrent to receiving autoislets following total pancreatectomy, are also infused with autologous BM-MSCs.

3.2 Immunosuppression

In any allogeneic organ transplant, immunosuppression acts as a double-edged sword, preventing or delaying the graft rejection by the host, but simultaneously increasing the risk for infection and cancer [104,105,106]. Routinely, during both peri- and post-allotransplantation, multiple induction and maintenance immunosuppressive agents are administered to the graft recipient to prevent graft rejection. The most commonly used component of immunosuppression regimens are the Calcineurin-Inhibitors (CNIs) Cyclosporine or Tacrolimus. Interestingly, both have been implicated in renal dysfunction and beta cell toxicity [107,108,109,110]. Therefore alternative CNI-free immunosuppression methods that prevent rejection while avoiding the toxicity of CNIs have been thoroughly investigated in islet allotransplantation [111,112]. Following the spectacular results obtained by the Edmonton group using the corticosteroid-free immunosuppressive regimen, more than 90% of islet programs adopted their protocol [15] i.e Rapamycin-based therapy with low doses of tacrolimus and anti-interleukin-2 receptor (IL2R) antibody (Daclizumab) as induction therapy. This approach was designed to improve the survival and function of islets while protecting the kidney function by minimizing CNIs. Induction with IL2R antagonists only, which comprised about 54% of all initial infusions in 1999-2002, was subsequently replaced or supplemented with regimens that included T-cell depletion with/without TNF antagonists in about 68% of new infusions performed by 2011 to 2014 [55]. However, the most successful immunosuppressive regimens involved T-cell depletion induction therapy in the form of Thymoglobulin, anti-CD52 antibody (Alemtu­zumab), or anti-CD3 humanized antibody [78].

Another trend that emerged to improve outcomes, tolerability and patient compliance was to replace Sirolimus with Tacrolimus and Mycophenolate Mofetil (MMF) [84]. According to the latest CITR annual report, the maintenance immunosuppression in 1999-2002 consisting predominantly of (65%) CNI plus mammalian target of rapamycin (mTOR) inhibitors has been increasingly replaced or supplemented throughout the eras by a CNI and Inosine Monophosphate Dehydrogenase (IMPDH)-inhibitor combination [55]. In the most recent era (2015-2018), CNI plus mTOR inhibitors were used in 15% of new infusions while CNI plus IMPDH inhibitors were used in about 56% [55]. Table 1-2 list active and completed clinical trials that focus on immunosuppressive approaches aimed at improving graft outcomes. Table 3 lists clinical trials that are actively recruiting. Despite their limitations, these studies confirm that it is possible to achieve single-donor insulin independence without using CNI and steroids.

Table 1 IMMUNOSUPPRESSION FOR ISLET TRANSPLANTATION. CLINICAL TRIALS: ACTIVE, NOT RECRUITING.

Table 2 IMMUNOSUPPRESSION FOR ISLET TRANSPLANTATION. CLINICAL TRIALS: COMPLETED.

Table 3 IMMUNOSUPPRESSION FOR ISLET TRANSPLANTATION. CLINICAL TRIALS: RECRUITING.

More recently, Salk Institute scientists have generated the first immune-evasive, glucose responsive human islet-like organoids (HILOs) from induced pluripotent stem cells that provide a promising alternative to cadaveric and device-dependent therapies in the treatment of diabetes [113]. They demonstrated that non-canonical WNT4 signaling drives the metabolic maturation of HILOs necessary for robust ex vivo glucose-stimulated insulin secretion. HILOs are able to re-establish glucose homeostasis in diabetic immunodeficient NOD/SCID mice following transplantation. Furthermore, overexpression of programmed death-ligand 1 (PD-L1) promotes the survival of HILO xenograft, restoring glucose homeostasis in Streptozotocin-induced, immunocompetent diabetic mice for 50 days. In vitro stimulation with interferon-γ (IFN-γ) induces endogenous PD-L1 expression and restricts T cell activation and graft rejection. Extensive research and preclinical testing to confirm the durability, safety and efficacy of this system for use in humans is still required to advance this system to clinical trials.

3.3 Graft Function

Currently, approximately 30-50% of original islet mass can be recovered from a cadaveric pancreas for transplantation [78]. Of this, approximately half of the transplanted islets are further destroyed by the IBMIR, a serious obstacle to successful islet engraftment. First achievement of insulin independence measured from initial islet infusion with or without subsequent infusion, is an indicator of the rate of engraftment under real-time conditions that include early graft loss, islet resource availability, patient/doctor decisions and myriad other factors [55]. The first islet infusion usually results in partial islet function correlating with detectable serum C-peptide, lowering of insulin requirements in 40-60% of the recipients, immediate improvement in glucose control with HbA1c at <7% and prevention of severe hypoglycemic episodes and hypoglycemic unawareness, resulting in an instant improvement in quality of life [14,78,114]. Using the original Edmonton Protocol, a steep loss of insulin independence was observed by the third year, but in most of the transplanted patients a persistent C-peptide and continuous longitudinal protection from hypoglycemia was observed [43]. Apart from allograft rejection and the recurrence of autoimmunity exacerbated by marginal initial islet mass engraftment, a number of factors played a contributory role in the long-term loss of insulin independence. However since then, advances in donor selection, islet isolation, antirejection and anti-inflammatory agents have yielded clinical islet transplant outcomes that mirror whole-organ pancreas transplantation outcomes, achieving greater than ~70% and ~50% insulin independence rates in select centers at one year and five years post-transplantation [14,16,17,45,47,56,94,115,116,117,118,119,120,121]. According to the latest CITR report, it is notable that the cumulative rate of achievement of insulin independence follows the general shape of engraftment curves for solid organs, but with a slower initial slope, indicative of multiple infusions [55]. While the overall rate of first achievement of insulin independence is, remarkably, nearly identical between islet transplantation alone and islet after kidney recipients, the most predictive factors of this endpoint for islet transplantation alone were, immunosuppression with IL2R antagonists and recipient age ≥35. Factors that result in optimized outcomes post last infusion included, induction immunosuppression with T cell depletion and/or TNF-alpha inhibitor, maintenance immunosuppression with mTOR inhibitor and CNI, IEQ’s ≥325,000 [55]. The CITR report also demonstrates that even moderate amounts of endogenous C-peptide significantly improves HbA1c which in turn contributes to reducing the risk of long-term complications such as those affecting the vasculature and peripheral nerves. Additionally, a complete abolition of severe hypoglycemic episodes and impaired awareness of hypoglycemia is observed in most cases indicating that islet transplantation offers an alternative form of cure for patients who cannot achieve adequate glycemic control using intensive exogenous insulin treatment and suffer from debilitating and recurrent hypoglycemic events [56].

3.4 Islet/beta Cell Encapsulation Strategies

Islet encapsulation strategies involve encapsulation of human or pig islets with a semi-permeable barrier consisting of a semi-porous device (micro- or macro-encapsulation/bioartificial pancreas) [5]. The device needs to support the exchange of nutrients and oxygen as well as allow the encapsulated islets to sense fluctuations in blood glucose and respond by secreting insulin. It should also physically isolate the islets from the host’s immune responses potentially eliminating the requirement for toxic systemic immunosuppression [5,102]. Despite the progress in biomaterial strategies for encapsulation, the host’s immune responses to foreign materials remains a major limitation hindering their potential for clinical transplantation. Clinical trials of porcine islets conducted in New Zealand and Argentina [122,123,124,125,126,127] have met with limited success, however there is still a considerable amount of work involving optimization of encapsulation methods and materials that needs to be accomplished before the technique can be used in large clinical trials in the US.

Clinical trials that are currently looking into islet encapsulation strategies are listed below:

1. lClinicalTrial # NCT03513939 is a Phase I/II trial that studies the safety, tolerability and efficacy of implanting the Sernova's Cell Pouch™ subcutaneously. The Cell Pouch™ is designed as a scaffold made of non-degradable polymers, formed into small cylindrical chambers which, when placed in the subcutaneous site, becomes incorporated with tissue and microvessels to the circumference of removable plugs within as early as two weeks as demonstrated in preclinical studies. After tissue incorporation, the plugs are removed, leaving fully formed tissue chambers with central void spaces for the transplantation of therapeutic cells including islets. The Cell Pouch™ forms a natural environment, rich in microvessels that allows the transplanted islets to engraft with intravascular microvessels. It is believed that this engraftment will enable long-term survival and function of transplanted islets. Immunosuppression is initiated after a minimum of three weeks after Cell Pouch™ implantation, and optimized for another three weeks. This allows for proper vascularization of the Cell Pouch™ chambers and the patient to be stabilized on immunosuppression prior to transplantation. A mass of highly purified islets (>3,000 IEQs per kg of patient body weight) can be transplanted in the Cell Pouch™.
2. lClinical trial # NCT00790257 is a phase I clinical trial to study the safety and efficacy study of encapsulated human islet allotransplantation to treat T1D. With the view to avoid immunosuppression, this protocol tests the safety and the efficacy of encapsulated human islets in a "Monolayer Cellular Device" for islet allotransplantation in type 1 diabetic patients, performed at the University clinical hospital Saint-Luc, Brussels. Encapsulated human islets will be transplanted in the subcutaneous tissue as a “safety procedure” for patients.
3. lClinical trial # NCT01379729 is a phase I clinical trial that studies an alginate embedded human beta cell graft in a "therapeutic" dose in the intraperitoneal cavity of type 1 diabetic patients under immunosuppression with Tacrolimus/MMF.
4. lClinical trial # NCT02064309 is a phase I/II clinical trial to assess the safety and efficacy of transplantation of macro-encapsulated human islets within the Bioartificial Pancreas Beta-Air implanted subcutaneously in patients with T1D without immunosuppression.

Based on the futuristic prospect that human stem cell derived islets may be used for islet transplantation thereby addressing the limitation of shortage of donor organs to provide islets, California based company ViaCyte Inc. has successfully differentiated human embryonic stem cells (hESCs) into pancreatic endoderm cells (PEC). The PECs displayed positive C-peptide, pro-insulin as well as key transcription factors that led to regulated insulin production post-transplantation and in-vivo differentiation [16,96,97,127]. Two clinical trials of Viacyte’s VC-01(NCT02239354) and VC-02 (NCT03163511) utilize these hESC-derived PECs encased in a macroencapsulation device and implanted subcutaneously. In the first clinical trial, the encapsulation device has an intact membrane preventing direct immune cell-to-cell contact, whereas in the second trial, the device has microperforations designed to improve neovascularization, but necessitating requisite immunosuppression. Ongoing data are eagerly awaited to validate the safety and preliminary efficacy of these promising approaches.

3.5 Islet Transplantation in Extrahepatic Sites

The current site of transplantation in the liver is not ideal, due to several factors. These include amongst others, significant liver inflammation following islet infusion; potential complications such as massive bleeding and thrombosis, portal hypertension and fatal hepatic necrosis; islet toxicity due to high levels of immunosuppressants and gastrointestinal toxins; inability to retrieve islets post-transplantation; and development of graft dysfunction in a number of intrahepatic allogeneic and autologous islet graft recipients [12,14,77,128,129,130,131]. Furthermore, islets in the liver, are subjected to variable glucose concentrations than they would be exposed to in the pancreas which could be toxic [12,131]. Also compared to the native tissue oxygen tension and parenchymal oxygen tension of 40 and 30 mmHg that islets are exposed to in the pancreas, the liver has significantly lower tensions of <10 mmHg for both [12,130,132]. Free fatty acids are also thought to potentially contribute to islet dysfunction in the liver [133]. Additionally, insulin secretion results in triglyceride deposition into the surrounding hepatic tissue further exacerbating this problem.

Other extrahepatic sites that have been explored include amongst others bone marrow, the gastric submucosa, genitourinary tract, the omentum, anterior eye chamber, testis, thymus, striated muscle, kidney capsule and the peritoneum [5,12,16,131]. Of these, sites favored have been the omental pouch due to its high level of angiogenic factors, the brachioradialis muscle where there has been some success, under the kidney capsule, within a venous sac, immune privileged sites which would allow for the avoidance of rejection, and also the pancreas itself although there is insufficient evidence for this location [12].

Several clinical trials are currently recruiting or active, are studying alternative site of transplantation. For instance, Clinical trial # NCT02213003, NCT02821026 and NCT02803905 study islet allotransplantation in the omentum; Clinical trial # NCT03977662 studies cotransplantation of allogeneic parathyroid glands (PTG) with adult pancreatic islets in the intramuscular site; Clinical trial # NCT02916680 studies islet transplantation in the anterior chamber of the eye; Clinical trial # NCT01722682 studies islet transplantation in the bone-marrow; and Clinical trial # NCT01571817 and NCT02402439 study islet transplantation in the gastric submucosa.

4. Islet Transplantation: Way Past an Experimental Procedure

In Australia, Canada and many Asian and European countries including United Kindgdom, Sweden, Switzerland, France and Italy, islet allotransplantation is approved and covered by national healthcare systems [14]. However in the US, islet allotransplantation is still considered an experimental procedure. Two multicentre phase III Biological Licensure Application (BLA)-enabling islet allotransplantation clinical trials have been conducted that were funded by NIH and coordinated by the CIT Consortium [79,134]. These Phase III trials represent a pivotal milestone since these and similar studies will help pave the way towards reimbursement for islet allotransplantation through Medicare and Medicaid, which will have major impact on its future as a therapeutic strategy for T1D in the USA. Currently patients with brittle T1D associated recurrent hypoglycemic episodes or marked glycemic lability as well as cardiovascular disease form the basis of consideration for coverage as pancreas transplantation is not feasible because of the risk of major surgery [21,81]. Also improved outcomes that have been observed in long-term insulin independence with near normalization of HbA1c in the absence of severe hypoglycemic episodes in select trials, coupled with advances in technical approaches and immunosuppressive therapy as well as the increase in number of clinical islet allotransplantations performed further bolster the chances of BLA approval [45,50]. Costs incurred by this technology such as the costs that are charged by the Organ Procurement Organization for a donor pancreas or costs that are incurred for establishing an Federal Drug Administration (FDA) approved GMP islet isolation facility and maintaining a trained islet isolation team, have also proven to be a major deterrent to its wide spread application. Based on the improvement that has been observed in the quality of life and disease management as well as the reduced risk of diabetes associated complications in patients who have had either an autologous islet transplantation following total pancreatectomy or have had islet allotransplantation for treatment of brittle T1D associated with severe hypoglycemic episodes and unawareness and glycemic lability, the high cost associated with this technology should not play a primary role in the decision-making process to consider third party reimbursement, especially since similar costs are incurred by the already accepted pancreas transplantation [135].

5. Conclusions

Over the last 16 years, incremental advances have been made in several aspects of islet transplantation including amongst others, optimized donor selection, better preservation of donor pancreas, improvements in islet isolation technique and experience of the centers and personnel, novel beneficial islet culture techniques prior to transplantation and potent immunosuppressive therapy that promote longitudinal islet transplant outcomes. Improved islet graft durability at five years post transplantation now parallels that of whole pancreas grafts in select centers, with the added advantage of islet transplantation being a less invasive procedure that is accompanied with an improvement in the quality of life due to prevention of severe hypoglycemic episodes and unawareness and risk of complications. The main hurdles that currently remain relate to procedural availability, the cost of the technology and most importantly sustainable reimbursement. Islet transplantation is still considered an experimental procedure in the USA and recognition by the FDA would lead to third party reimbursement which would help pave the way to make islet transplantation a clinically therapeutic option with wide-spread application.

Author Contributions

Both authors contributed equally to this work. Dr. Preeti Chhabra contributed to literature survey, writing, editing and proof-reading this article. Dr. Kenneth L. Brayman contributed to writing and editing this article.

Funding

American Diabetes Association #1-17-IBS-244.

Competing Interests

The authors have declared that no competing interests exist.