Pancreas and islet transplantation in diabetes mellitus

INTRODUCTION — The goals of pancreas whole organ and islet transplantation are to enable effective, stable glycemic management (often with insulin independence), to improve quality of life, and to reduce secondary complications. Pancreas transplantation is most often performed with simultaneous kidney transplantation in selected patients with diabetes and end-stage kidney disease, who will already be required to take immunosuppressive therapy for the kidney graft [1]. Pancreas after kidney (PAK) and pancreas transplant alone (PTA) are performed less commonly. Islet cell transplantation has received regulatory approval from the US Food and Drug Administration (FDA) for a deceased donor islet product from a single commercial cell isolation facility [2]. In regions of Canada, Europe, Australia, and Asia, islet cell transplantation is a standard care procedure provided by academic medical institutions for selected patients with diabetes [3].

This topic will briefly review the history, techniques, and clinical results of pancreas and pancreatic islet transplantation in patients with longstanding type 1 diabetes mellitus, with a focus on transplantation of islet or pancreatic tissue alone. Combined pancreas-kidney transplantation is discussed separately. (See "Pancreas-kidney transplantation in diabetes mellitus: Patient selection and pretransplant evaluation" and "Pancreas-kidney transplantation in diabetes mellitus: Benefits and complications".)

INDICATIONS FOR TRANSPLANTATION — The goals of transplantation are to restore glucose-regulated endogenous insulin secretion, arrest the progression of diabetes-related complications, and improve quality of life. Both pancreas and islet transplantation require lifelong immunosuppression to prevent rejection of the graft. Patients with end-stage kidney disease receiving simultaneous pancreas-kidney (SPK) or pancreas after kidney (PAK) transplants are already required to take immunosuppressive therapy for the kidney graft, and therefore the incremental effect of immunosuppressive therapy on quality of life is small. However, for patients receiving pancreas transplant alone (PTA), who have not yet developed advanced nephropathy, the benefit of preventing or slowing the progression of secondary complications must be balanced with the adverse effects of the immunosuppressive agents used in transplantation (eg, diarrhea, neutropenia, anemia, fatigue, hypertension, bone loss, susceptibility to infections and secondary cancers).

The American Diabetes Association (ADA) criteria for transplantation are as follows (figure 1) [4,5]:

●SPK or PAK – Patients with type 1 diabetes and end-stage kidney disease who have had or plan to have a kidney transplant are candidates for pancreas transplantation. The successful transplantation of a pancreas will improve overall glycemia, ameliorate problematic hypoglycemia, usually result in insulin independence, and possibly improve kidney survival [6,7]. Most pancreas transplants are performed in patients with diabetes and end-stage kidney disease. The majority of these patients receive SPK rather than PAK. (See "Pancreas-kidney transplantation in diabetes mellitus: Patient selection and pretransplant evaluation", section on 'Patient selection'.)

●PTA – PTA is generally considered only in patients with serious, progressive complications of diabetes in whom the quality of life is unacceptable. Such complications include:

•A history of frequent, acute, severe metabolic complications (hypoglycemia, marked hyperglycemia, ketoacidosis)

•Incapacitating clinical and emotional problems with exogenous insulin therapy

•Consistent failure of insulin-based management to prevent acute complications

Pretransplant screening for cardiovascular disease in PTA is the same as for combined kidney-pancreas transplantation. (See "Pancreas-kidney transplantation in diabetes mellitus: Patient selection and pretransplant evaluation", section on 'Pretransplant evaluation'.)

●Islet – Islet transplantation provides a less invasive alternative to PTA and PAK transplantation and is much less commonly performed simultaneously with a kidney transplant. The successful transplantation of isolated islets will improve glycemia, ameliorate problematic hypoglycemia, and often result in insulin independence [8,9]. Individuals who are not candidates for pancreas transplantation or wish to avoid its risks may be considered for islet transplantation. (See 'Type 1 diabetes' below.)

PANCREAS VERSUS ISLET TRANSPLANTATION — There are no direct, randomized trials comparing the outcomes from whole organ versus islet transplantation. In addition, there are few observational studies comparing pancreas and islet transplantation procedures. In a report from a single center that performed 33 pancreas transplant alone (PTA) and 33 islet transplantation procedures, patients undergoing PTA had a higher rate of insulin independence (75 versus 57 percent) at an average of one-year follow-up [10]. The rate of long-term adverse events (length and frequency of hospitalization, reintervention for acute surgical or immunological complications, infections) was higher in the patients who received PTA. In a second retrospective study comparing 15 PTA and 10 islet transplantation procedures, the rate of insulin independence at three years was similar (64 versus 70 percent) [11].

Data from pancreas and islet transplantation registries show a higher rate of insulin independence with pancreas transplantation (85 versus 50 percent at one year), but also a higher morbidity due to general surgery [12,13]. The islet transplantation procedure is less invasive, and therefore, it is associated with lower morbidity. However, rates of long-term success (defined by insulin independence) are lower. (See 'Outcomes' below and 'Metabolic outcomes' below.)

PANCREAS TRANSPLANTATION — Pancreas transplantation was first used for the treatment of diabetes in humans in 1966 [14]. The rates of graft and patient survival were low; as a result, very few procedures were performed in the early to mid-1970s. The subsequent introduction of better immunosuppressive regimens (particularly cyclosporine and anti-T-cell antibodies), new surgical techniques, and the selection of healthier recipients resulted in markedly improved outcomes. As a result, the number of pancreas transplantations steadily increased each year in the United States, peaking at 1484 in 2004 [15,16]. In the United States during 2016, 791 simultaneous pancreas-kidney (SPK), 73 pancreas after kidney (PAK), and 73 pancreas transplant alone (PTA) transplants were performed [17]. The decrease in the number of pancreas transplants is probably due in part to advances in insulin-based therapy since the Diabetes Control and Complications Trial (DCCT) that have led to lower rates of secondary complications of chronic hyperglycemia. The number of solitary pancreas transplants (PTA and PAK) in particular has diminished, as increased use of continuous glucose monitoring (CGM) and automated insulin delivery systems has reduced the incidence of problematic hypoglycemia (figure 1).

Outcomes — The mortality, morbidity, and results of transplantation vary with operative experience and patient selection.

Patient survival

●Based on 2004 to 2015 data, patient survival rates for SPK, PAK, or PTA ranged from 96 to 99 percent at one year, 89 to 91 percent at five years, and 70 to 80 percent at ten years postoperatively [17-19]. The majority of deaths in the first three months post-transplant and subsequently were due to cardiovascular or cerebrovascular disease.

●There are few data on the survival benefit for transplanted compared with waitlisted patients. The following findings are based on retrospective analyses of transplantation registries from 1995 to 2003:

•Survival for SPK recipients was much better than that of waitlisted patients who continue to receive dialysis [20]. The decreased mortality is due in part to the well-established survival benefit conferred by kidney transplantation alone (KTA; even without pancreas transplantation) compared with dialysis. (See "Pancreas-kidney transplantation in diabetes mellitus: Benefits and complications", section on 'Patient survival'.)

•For PTA or PAK recipients, survival at four years was equivalent to that of patients on the waiting list (PTA 90.5 versus 87.3 percent for waitlisted patients, PAK 88.3 versus 81.7 percent) [20].

•In another retrospective study of 11,572 patients with diabetes and preserved kidney function who were on a waiting list for pancreas transplantation, survival at four years was significantly worse for patients receiving PTA compared with waitlisted patients receiving conventional therapy (85.2 versus 92.1 percent) [21]. However, these data should be interpreted with caution as there could be important differences between the group transplanted with a pancreas alone and those on the waiting list (who, for example, may turn down a pancreas if feeling well) [22]. In addition, some patients from the waiting list group were counted more than once because they were registered on more than one list, and 8 percent of patients on the pancreas transplant waiting lists were removed from the waiting lists and underwent a kidney transplant first because of deteriorating kidney function. These differences could have biased the outcome of the study in favor of those on the waiting list.

Graft survival — Based on 2004 to 2015 data, early graft failure (within 90 days) occurred in approximately 8 to 9.4 percent of patients [17]. Five-year pancreas graft survival for SPK, PAK, and PTA was approximately 73, 65, and 53 percent, respectively [18]. Pancreas graft survival is inversely related to several donor variables, including age, body mass index, and cardiovascular death. Recipients of PTA whose organs came from donors with poor donor risk indices had a lower rate of graft survival compared with recipients of SPK (77 versus 88 percent at one year) [19,23].

The definition of pancreas graft survival has been variably defined by transplant centers (eg, persistence of complete insulin independence, persistence of C-peptide production) [18]. A consistent definition would strengthen future outcomes studies. In the United States, the most recent United Network for Organ Sharing definition of graft failure is the use of insulin ≥0.5 units/kg/day for 90 consecutive days [18,24]. This definition remains limited by the lack of inclusion of any glycemic metrics.

In 2018, a classification of graft function was developed at a workshop held in Igls, Austria, and proposed by the International Pancreas and Islet Transplant Association and the European Pancreas and Islet Transplant Association with a goal to strengthen and harmonize future outcome studies [25]. The definitions are based on a combination of glycated hemoglobin (A1C), severe hypoglycemic events, insulin requirements, and C-peptide. With this proposed Igls classification, both optimal and good graft function are considered successful clinical outcomes.

●Optimal graft function – Near-normal glycemia (A1C ≤6.5 percent [48 mmol/mol]) without severe hypoglycemia or requirement for insulin or other antihyperglycemic therapy and with an increase in C-peptide compared with pretransplant.

●Good graft function – A1C <7 percent (53 mmol/mol) without severe hypoglycemia and with a significant reduction in insulin requirements (>50 percent reduction, which should also be <0.5 units/kg/day) and an increase in C-peptide compared with pretransplant.

●Marginal graft function – A1C ≥7 percent, the occurrence of any severe hypoglycemia, or less than a 50 percent reduction in insulin requirements, when there is an increase in C-peptide compared with pretransplant.

●Failed graft – Absence of any evidence for clinically significant C-peptide production (eg, fasting C-peptide <0.3 ng/mL [0.1 nmol/L]).

This Igls classification has been further validated against CGM-derived metrics of glycemia that align with outcome measures used in studies of automated insulin delivery systems, thus enabling comparison between different insulin replacement strategies [26]. In the absence of insulin independence, thresholds of clinically significant C-peptide production may assist with graft monitoring, as exogenous insulin requirements may fluctuate due to factors other than graft function.

Metabolic effects — Both cross-sectional and prospective studies have shown that pancreas transplantation can result in independence from exogenous insulin therapy and improvements in glucose metabolism, A1C values, acute insulin responses to intravenous glucose, and counterregulatory responses of serum glucagon and glucose to insulin-induced hypoglycemia [27-29]. These beneficial effects on glucose regulation (the result of restoring pancreatic islet function) are maintained for many years. Persistent improvement for 15 years is common in some centers [30].

●Insulin responses – Patients with well-functioning grafts have normal insulin responses to oral and intravenous glucose stimulation as well as to intravenous arginine and intravenous secretin [28,31-37]. However, owing to the systemic (rather than portal) venous drainage of the allograft, basal and stimulated peripheral insulin concentrations are two to three times higher than normal [33,38]. Thus, the hepatic first-pass uptake and metabolism of insulin secreted into the portal vein are bypassed; in individuals without diabetes, 50 to 90 percent of the insulin in portal venous blood undergoes first-pass hepatic degradation.

●Serum lipids – Post-transplantation hyperinsulinemia does not appear to have adverse effects on cardiovascular risk factors. Serum triglyceride and low-density lipoprotein (LDL) cholesterol concentrations tend to fall and serum high-density lipoprotein (HDL) cholesterol concentrations tend to rise in recipients of pancreas transplants [39-41].

●Glucose counterregulation and hypoglycemia – Glucose counterregulation in response to hypoglycemia is improved by pancreas transplantation [42-44]. This is an important benefit because patients who have had diabetes for many years before transplantation typically have abnormal glucose counterregulation due to absent glucagon and decreased epinephrine responses to hypoglycemia. The improvement in glucose counterregulation in recipients of pancreas transplants is due to normalization of the glucagon counterregulatory response and improvement in the epinephrine counterregulatory response [44]. Recognition of symptoms of hypoglycemia also improves [44].

Hypoglycemia as a complication of pancreas transplantation has been reported [45], but it is usually a mild form of alimentary hypoglycemia that responds to dietary modification [46]. Compromised glucose counterregulation in individuals with diabetes is reviewed in detail separately. (See "Physiologic response to hypoglycemia in healthy individuals and patients with diabetes mellitus".)

Effects on the chronic complications of diabetes — The impact of successful pancreas transplantation and normalization of glycemia on the secondary complications of diabetes is reviewed briefly below and in detail elsewhere. Many of these studies involved patients who had undergone SPK transplantation. (See "Pancreas-kidney transplantation in diabetes mellitus: Benefits and complications", section on 'Other potential benefits'.)

Following successful pancreas transplantation:

●Recurrent and de novo diabetic nephropathy is prevented. PTA may reverse established diabetic lesions in patients with early diabetic nephropathy.

●There is stabilization and, in some cases, improvement in peripheral and autonomic diabetic neuropathy.

●The effect on diabetic retinopathy is not clear. While some studies have found no benefit in terms of halting or reversing the progression of advanced retinopathy after pancreas transplantation, other reports have noted stabilization or occasional regression of retinal lesions following successful pancreas transplantation.

These findings regarding the effects of pancreas transplantation on diabetic microvascular disease must be interpreted in light of the fact that most patients undergoing pancreas transplantation have already had diabetes for over two decades and have advanced complications. Data are not available from prospective trials comparing patients undergoing pancreas transplantation with control populations matched for complication stage.

Other effects — Quality-of-life studies consistently demonstrate benefits after pancreas transplantation, including return to work. Successful pregnancies have been reported with no notable adverse effects on the fetus or the mother.

Technique — The technique used for pancreas transplantation is the same whether or not a kidney is transplanted into the pelvic area at the same time. In the most common procedure, a whole pancreas, still attached to a small portion of the duodenum containing the ampulla of Vater, is taken from a deceased donor. The pancreas is placed laterally in the pelvis, with arterial anastomosis to a branch of the iliac artery and venous anastomosis to a branch of the iliac vein or inferior vena cava, which results in secreted insulin appearing first in the systemic (rather than the hepatic portal) circulation. A modification to this approach includes venous anastomosis to the superior mesenteric vein providing portal rather than systemic drainage of the endocrine pancreas.

The duodenal segment is connected to a loop of bowel or, less commonly, the urinary bladder, which receives the pancreatic exocrine secretion. Bladder drainage has the advantage of providing a means to monitor urine amylase concentration to help detect allograft rejection but can also be complicated by metabolic acidosis, hematuria, and recurrent urinary tract infections. With either exocrine drainage approach, the native pancreas is left intact.

Donors — The pancreases of deceased donors (either after declaration of brain death or circulatory death) may be utilized for pancreas transplantation. If present, fatty infiltration of the pancreas increases risk for reperfusion pancreatitis and possible early graft loss, so these organs are usually declined for whole pancreas transplantation. In regions with an available islet isolation facility, these organs instead can provide large quantities of high-quality islets if the donor had normal glucose homeostasis (as assessed by A1C). (See 'Islet transplantation' below.)

In the past, patients received a segment of pancreas donated by a living-related donor who is willing to undergo a hemipancreatectomy [47]; however, this approach is seldom performed due to elevated risk for future diabetes affecting both hemipancreatectomy donors and recipients [48,49].

Immunosuppression — Pancreas transplantation requires lifelong immunosuppression to prevent rejection of the graft and, in individuals with underlying type 1 diabetes, to prevent recurrence of autoimmune diabetes. Conventional maintenance regimens consist of a combination of immunosuppressive agents that differ by mechanism of action. This strategy minimizes morbidity and mortality associated with each class of agent while maximizing overall effectiveness. Despite a diversity of protocols, most recipients of pancreas transplants receive monoclonal or polyclonal anti-T-cell antibodies at the time of surgery and chronic immunosuppression therapy consisting of a calcineurin inhibitor (cyclosporine or tacrolimus) and an antimetabolite (mycophenolate mofetil or azathioprine). Glucocorticoids are usually administered at high doses at the time of transplant and tapered to a near-physiologic dose of prednisone (approximately 5 mg) in the first days to weeks post-transplant. Since 2007, most patients have received tacrolimus and mycophenolate mofetil, usually in combination with prednisone [17].

With the availability of new immunosuppressive drugs [50-53], some centers perform pancreas transplants without glucocorticoids and complete taper off glucocorticoid therapy is increasingly common [54-58]. Details concerning immunosuppressive regimens are discussed elsewhere. (See "Pancreas-kidney transplantation in diabetes mellitus: Surgical considerations and immunosuppression", section on 'Immunosuppression'.)

Causes of graft loss — The causes of pancreas transplant loss vary with the time after transplantation. Early graft loss, defined as loss occurring within hours or days after surgery, usually results from thrombosis, leaks, bleeding, infection, and pancreatitis (collectively referred to as technical failures). In one report of 211 patients undergoing pancreas transplant, technical graft failure occurred in 23 patients (11 percent), most commonly due to thrombosis [59]. Risk factors for technical failure included donor and recipient obesity and increased preservation time of the donor organ. With careful selection of donor organs, transplantation performed at experienced centers carries a technical graft failure rate of 5 percent. Later graft loss after several weeks is more common and is most often caused by immunologic rejection [19,60].

Rejection — A transplanted pancreas can be rejected within days or after years of successful transplantation. The incidence, diagnosis, and treatment of pancreas allograft rejection are reviewed in detail separately. Methods for detection are briefly summarized below. (See "Pancreas allograft rejection", section on 'Epidemiology and risk factors' and "Pancreas allograft rejection", section on 'Treatment'.)

Laboratory testing is performed routinely to detect rejection. Given the wide interindividual variation in the monitored laboratory tests, routine measurement is important for establishing an individual's baseline, and rejection is considered only with significant deviations [61]. (See "Pancreas allograft rejection", section on 'When to suspect rejection'.)

The following laboratory findings are possible indicators of rejection:

●Increasing serum amylase and/or lipase (more sensitive)

●Increasing fasting plasma glucose

●Decreasing serum C-peptide (usually interpreted in the context of a simultaneously measured fasting glucose concentration)

●Decreasing urinary amylase excretion (only patients with bladder drainage of the donor exocrine pancreas)

●Increasing serum creatinine (SPK recipients only)

Increasing serum pancreatic enzymes, such as amylase and lipase, are nonspecific indicators of rejection [62,63]. Serum amylase, in particular, may be higher than the normal range due to the presence of two pancreases, and cross-sectional imaging (abdominal computed tomography [CT]) may be helpful in evaluation. A rise in fasting glucose is a relatively late indicator of pancreas allograft destruction, and a decline in C-peptide may not be evident initially as increasing glucose levels provide a greater stimulus for insulin secretion. (See "Pancreas allograft rejection", section on 'Diagnosis'.)

ISLET TRANSPLANTATION — Islet transplantation has been performed in patients with type 1 diabetes and in patients with chronic pancreatitis. In contrast to patients with type 1 diabetes, patients with chronic pancreatitis undergo islet transplantation in conjunction with total pancreatectomy, with infusion of their own islets (auto-islet transplantation), which does not require treatment with immunosuppressive drugs. (See 'Chronic pancreatitis' below.)

Type 1 diabetes — For individuals with type 1 diabetes, islet transplantation offers a less invasive alternative to pancreas transplantation (figure 1) [64].

Procedure — Much effort has been expended to establish techniques to maximize the yield and quality of islets isolated from various sources. This has resulted in improved techniques for organ procurement and standardization of methods for islet isolation [65]. Typically, ≥325,000 islets isolated from a deceased donor pancreas are infused either via a percutaneous catheter that is introduced into the liver and advanced retrograde to the portal vein of the recipient or via a mesenteric vein catheter placed during mini laparotomy to reach the portal circulation. Protocols for islet isolation and transplantation have been developed and shared throughout the world by academic institutions.

Availability — Within national health services in many countries, islet cell transplantation is an established alternative to pancreas transplant alone (PTA) and pancreas after kidney (PAK) transplantation, and the procedures are regulated similarly.

In the United States, each islet isolation facility is considered by the US Food and Drug Administration (FDA) as a unique drug manufacturing entity requiring its own biologic licensure. Only one facility has received regulatory approval for its deceased donor islet product (labeled donislecel for donor islet cell therapy). Donislecel is approved for adults with type 1 diabetes who experience recurrent episodes of severe hypoglycemia despite intensive diabetes management and education [66,67]. The clinical use of donislecel is uncertain. Some UpToDate experts would not use this product until published, peer-reviewed data are available to support its efficacy and safety.

Metabolic outcomes

●Insulin independence – More effective and less toxic immunosuppressive drug regimens and continued improvements in islet harvesting techniques have contributed to a significantly increased rate of insulin independence after islet transplantation [68]. For example, in a report that included information on 1011 islet allograft recipients (819 islet transplant alone [ITA], 192 islet after kidney [IAK]) with transplantation performed between 1999 and 2013, approximately 50 percent of adults with type 1 diabetes who received islet transplantation (alone or after kidney) were insulin independent at one year with declining rates over time (30 percent ITA and 20 percent IAK at five years) [12].

In a subsequent cohort study, the rate of insulin independence five years after final islet infusion was 53 percent among individuals with type 1 diabetes who received ≥325,000 islet equivalents and optimized induction and maintenance immunosuppression therapy [69]. A similar rate of long-term insulin independence was demonstrated in Clinical Islet Transplantation (CIT) Consortium trials of ITA and IAK; in these trials, insulin independence was achieved by approximately 75 percent of islet recipients and maintained over eight years of follow-up in more than 50 percent in both cohorts [70].

●Glycemia and severe hypoglycemic events – Even in the absence of insulin independence, islet transplantation can markedly improve glycemia and reduce hypoglycemic events. In a series of 677 islet allograft recipients, near-normal glycemia (indicated by A1C <6.5 percent) was maintained by approximately 60 percent of recipients over three to five years of follow-up [68]. Further, whereas >90 percent of patients were experiencing severe hypoglycemia prior to transplant, >90 percent remained free of severe hypoglycemic events through five years of follow-up [68].

In a cohort of 398 individuals with type 1 diabetes complicated by severe hypoglycemia who underwent ITA and were analyzed five years after final islet infusion, patient age, transplanted islet number, and immunosuppression regimen influenced metabolic outcomes. Among those who were aged ≥35 years at transplant and received ≥325,000 islet equivalents, induction immunosuppression with T cell depletion and/or tumor necrosis factor [TNF]-alpha inhibition, and maintenance immunosuppression with both mechanistic (mammalian) target of rapamycin (mTOR) and calcineurin inhibitors (n = 126), 95 percent were free of severe hypoglycemic events and 76 percent had A1C <7 percent [69]. In a subsequent retrospective analysis in 1210 ITA and IAK recipients, measurement of primary graft function by a validated composite index (BETA-2 score) 28 days after final islet infusion predicted the metabolic outcomes of adequate glycemic management (A1C <7 percent), absence of severe hypoglycemic events, and islet graft survival (C-peptide ≥0.3 ng/mL) over five years of follow-up [71]. These findings suggest that an initially high engrafted islet mass is important for mitigating islet cell exhaustion and optimizing long-term outcomes.

In CIT Consortium trials of both ITA (n = 48) and IAK (n = 24) in individuals with type 1 diabetes complicated by impaired awareness of hypoglycemia and severe hypoglycemia, islet transplantation led to near-normal glycemia (A1C v6.5 percent for ITA and A1C <7 percent for IAK) and freedom from severe hypoglycemia in 87.5 and 62.5 percent of recipients, respectively, at one year after transplant [8,9]. With long-term follow-up over eight years, >90 percent of patients in both cohorts remained free of severe hypoglycemia, and A1C <7 percent was maintained by 56 percent of patients in the ITA cohort and 49 percent in the IAK cohort [70].

●Glucose counterregulation – Glucose counterregulation in response to hypoglycemia is also improved by islet transplantation due to partial recovery of the glucagon counterregulatory response and improvement in the epinephrine counterregulatory response [72]. By six months following islet transplantation, the glycemic thresholds normalize for activation of epinephrine and autonomic symptom responses to insulin-induced hypoglycemia [73], demonstrating recovery from hypoglycemia-associated autonomic failure. In the CIT Consortium trials, longitudinal follow-up over two years demonstrated essentially no time spent in hypoglycemia by continuous glucose monitoring (CGM) and complete normalization of the epinephrine response to insulin-induced hypoglycemia [74]. (See "Physiologic response to hypoglycemia in healthy individuals and patients with diabetes mellitus".)

Other outcomes — Quality-of-life studies conducted over two and three years of follow-up consistently demonstrate benefits including significant reductions in diabetes distress and hypoglycemia fear [9,75]. Successful pregnancies also have been reported following islet transplantation with no notable adverse effects on the fetus or mother [76].

Adverse events — At least 50 percent of islet recipients experience at least one adverse event [8,12]. Adverse events are related to immunosuppression (neutropenia, elevated liver function tests, elevated serum creatinine) and procedural complications (intraperitoneal bleeding requiring transfusion with possible laparoscopic or interventional radiology intervention).

In the CIT consortium trials, at least one adverse event was experienced by 60 percent of islet recipients (43 of 72 patients) over long-term follow-up with no reported deaths [8,9,70]. The following adverse events were noted in these two trials:

●Bleeding – Five procedural bleeding events occurred in a total 114 islet infusion procedures (<5 percent) [70].

●Infection and malignancy – Immunosuppression was associated with 15 episodes of infection, including two opportunistic infections (one with Pneumocystis jiroveci and one with systemic histoplasmosis). Four patients in the cohort were diagnosed with a malignancy, one each with lung (associated with smoking history), breast, prostate, and small intestine carcinoma (associated with Celiac disease) [70].

●Decline in kidney function – Kidney function may decline more after ITA than IAK, likely due to the interim initiation of calcineurin inhibitor-based immunosuppression. Kidney function as measured by estimated glomerular filtration rate (eGFR) declined by 6.92 mL/min/1.73 m² during the first post-transplant year in ITA recipients and then stabilized at an expected slope of -1.27 mL/min/1.73 m² per year up to eight years post-transplant [70]. In IAK recipients, a modest reduction in eGFR of 0.7 mL/min/1.73 m² was observed during the first post-transplant year, after which the eGFR improved by 0.55 mL/min/1.73 m² per year during the remainder of follow-up [70].

●Donor-specific antibody formation – Another potential consequence of islet transplantation is sensitization (development of donor-specific antibodies) [8,77]. Some recipients may receive more than one islet infusion obtained from multiple donors, so islet transplant recipients may be exposed to multiple human leukocyte antigen (HLA) mismatches. Multiple mismatches can result in antibody formation, which may preclude the ability to undergo future transplantation (islet, kidney, pancreas) due to a decreased likelihood of finding a compatible graft. In the CIT consortium trials, donor-specific antibodies developed in 3 of 48 ITA and 5 of 24 IAK recipients (approximately 10 percent of all recipients) [70], although higher rates have been observed in ITA recipients following discontinuation of immunosuppressants [77]. Thus, the potential for an islet cell transplant to compromise the ability to receive a future transplant must be discussed routinely with all potential recipients.

Immunosuppression — Like pancreas transplantation, islet transplantation usually requires lifelong immunosuppression to prevent rejection of the graft and, in individuals with underlying type 1 diabetes, to prevent recurrence of autoimmune diabetes. Immunosuppressive therapy is not needed for individuals with chronic pancreatitis who undergo auto-islet transplantation. (See 'Chronic pancreatitis' below.)

In IAK transplantation, the maintenance immunosuppression used to support the kidney is usually continued, including a low-dose glucocorticoid. ITA generally avoids the need for glucocorticoids, and an mTOR inhibitor (sirolimus or everolimus) may be used in place of an antimetabolite, enabling use of lower tacrolimus doses and avoidance of calcineurin-inhibitor toxicity [78]. Immunosuppressive regimens otherwise are similar to those used after pancreas transplantation. (See 'Immunosuppression' above and "Pancreas-kidney transplantation in diabetes mellitus: Surgical considerations and immunosuppression", section on 'Immunosuppression'.)

In islet transplantation, TNF-alpha inhibitors have been administered during the peri-transplant period to prevent cytokine-mediated beta cell apoptosis during engraftment [79]. This strategy may be especially important when T cell-depleting monoclonal antibodies are used for induction.

Experimental techniques — Islet transplantation is an evolving therapy but has achieved its potential to reliably treat patients with severe hypoglycemia or labile type 1 diabetes. Techniques are being perfected to improve islet harvest, enhance engraftment, decrease apoptosis, allow less toxic immunosuppressive regimens, induce immune tolerance, and noninvasively monitor islet cell fate after transplantation.

Single islet donors — One of the limitations of islet transplantation is the need for multiple islet donors in many recipients. An alternative approach is the use of a single islet donor with an induction immunosuppression regimen that includes antithymocyte globulin and etanercept (a TNF-alpha inhibitor) [80]. With islet transplant from a single donor pancreas, additional interventions introduced to the peri-transplant period include intensive antithrombotic and insulin therapy, which have independently been associated with a higher rate of insulin independence [81]. Rates of insulin independence following single-donor islet transplantation have varied across studies [8,82]. With an initial islet infusion, transplantation of >6000 islet equivalents per kilogram recipient body weight is necessary but not always sufficient to achieve insulin independence after single-donor islet transplantation [64].

Alternative islet sources — Harvesting sufficient numbers of healthy human pancreatic islets from deceased donors is a major barrier to successful islet transplantation. Islets make up only 2 to 3 percent of the mass of the adult human pancreas. Research is ongoing to identify alternative sources for beta cells. Many studies, both in vitro and in vivo, have focused on identifying islet-producing stem cells and protocols to generate differentiated islets.

Human embryonic stem cells have been induced to differentiate in vitro into pancreatic endodermal cells [83,84]. After implantation into immunodeficient mice, these cells generated glucose-responsive, insulin-producing cells capable of reversing diabetes [85,86], including when transplanted under the skin using a macroencapsulation device [87]. In immunocompetent humans, encapsulation devices elicit a foreign-body response with surrounding fibrosis that prevents oxygen delivery to contained islets and represents a major barrier to achieving immunosuppression-free transplantation. Cell-permeable macroencapsulation devices intended to allow direct graft vascularization are under investigation in early stage trials but require systemic immunosuppression [88,89].

Human embryonic stem cells have also been induced to differentiate in vitro into islet-like cell clusters that contain maturing endocrine cells with the capacity for glucose-dependent insulin secretion before transplantation [90,91]. Transplantation of these stem cell-derived islets in humans with type 1 diabetes is under investigation.

Xenotransplantation — Xenotransplantation of islets has also been evaluated. Intraportal transplantation of islets isolated from pigs into diabetic nonhuman primates (macaques) treated with monoclonal antibodies to suppress T cell activation can lead to prolonged reversal of diabetes [92,93]; however, there is significant morbidity associated with the required immunosuppression. Novel genome editing approaches may enable the breeding of pigs for islets with significantly lower immunogenicity [94], although the clinical potential for xenograft islet transplants remains uncertain.

Engraftment site — The majority of islet transplantations are performed via infusion into the portal vein with subsequent engraftment in the liver. In addition to the risk of bleeding during intraportal catheterization, the major limitation of the liver as a site for islet transplantation is the inability for biopsy to reliably acquire islet tissue necessary for pathologic evaluation [95].

Alternative engraftment sites (such as the omentum, subcutaneous space, and gastric submucosa) have been attempted; however, limited available clinical data suggest that these may not be preferable alternatives to intrahepatic transplantation [96].

Monitoring islet engraftment — Optimal methods of monitoring islets after transplantation would provide information on the accuracy of infusions, success of early engraftment, and long-term graft survival. In a small number of patients, scanning with positron emission tomography (PET) combined with CT allowed visualization of islet survival and distribution after transplantation [97]. This technique may be useful for evaluating alternative sites of implantation or for developing strategies to improve intrahepatic transplantation.

In human studies of auto-islet transplantation, laboratory measurements of beta cell mass prior to islet implantation correlated highly with measures of insulin secretory reserve post-transplantation using the clinical technique of glucose potentiation of arginine-induced insulin secretion [98]. This technique has also been used to demonstrate improvements in engrafted islet beta cell mass in allo-islet transplantation [82].

Chronic pancreatitis — Common consequences of chronic pancreatitis include severe chronic abdominal pain, weight loss, diarrhea, poor quality of life, and narcotic use. Progressive inflammation of acinar tissue may affect endocrine tissue function, thereby progressively damaging the islets of Langerhans, resulting in diabetes. The course of the disease is often punctuated by repeated pancreatic duct stenting and/or partial pancreatectomy. Some patients undergo total pancreatectomy for pain relief, which leads to immediate and total insulin-deficient diabetes. In the 1980s, surgeons at the University of Minnesota reasoned that the resected pancreas could be used for islet isolation and infusion of the islets into a patient's liver to effect auto-islet transplantation that would prevent onset of diabetes [99]. Auto-islet transplantation does not require treatment with immunosuppressive drugs.

●Insulin independence – Auto-islet transplantation has been successful in adults and children without diabetes who have chronic painful pancreatitis [99-101]. Many islet autograft recipients have normoglycemia and normal serum insulin responses to oral and intravenous glucose and intravenous arginine soon after transplantation if a sufficient number of islets can be isolated from the diseased pancreas. These outcomes can last for many years after transplantation [100,102-104]. In most large series of auto-islet transplantation at a median follow-up of two years, approximately one-third of patients are insulin independent, approximately one-third can be managed with daily long-acting insulin alone, and approximately one-third require intensive insulin therapy with basal-bolus administration by multidose injection or pump delivery [64]. This heterogeneity in transplantation outcome largely reflects the wide variability in islet numbers isolated from diseased pancreases affected by chronic pancreatitis.

The importance of transplanted islet number was highlighted in an initial series in which 10 of 14 patients receiving more than 300,000 islets were insulin independent two years after transplantation [105], a finding verified by a later series in a much larger group of recipients [106]. In the latter series, insulin independence was reported in approximately two-thirds of patients receiving >5000 islet equivalents per kilogram recipient body weight, in approximately one-third of patients receiving between 2500 to 5000 islet equivalents, and rarely in those who received <2500 islet equivalents [106]. The lower requirement of islet numbers for metabolic benefit in auto- versus allo-islet transplantation may be explained by islet isolation from a living rather than deceased donor, avoidance of potentially toxic effects of immunosuppressive drugs, and the absence of both auto- and allo-immune responses. The magnitude of beta cell function after total pancreatectomy and islet autotransplantation (TPIAT) has been shown to be similar to that of normal individuals when the secretion data are normalized to the number of islets transplanted [98].

●Hypoglycemia – Patients who undergo TPIAT very frequently experience bouts of hypoglycemia associated with exercise and meals [107]. One study reported a lack of glucagon responses during insulin-induced hypoglycemia after intrahepatic islet transplantation [108] that was not observed in a subgroup of patients whose auto-islets were placed into both intrahepatic and nonhepatic sites [107]. This impaired glucagon response may result from intrahepatic glycogenolysis that releases free glucose and thereby inhibits glucagon release from intrahepatic islets [109]. However, an impaired glucagon response is not observed in islet allograft recipients with intrahepatic transplantation. An alternative explanation for hypoglycemia after TPIAT is the development of alimentary hypoglycemia related to the Roux-en-Y gastrointestinal reconstruction that follows total pancreatectomy. In subsequent studies of intrahepatic auto-islet recipients, endogenous glucose production failed to rise during moderate exercise [110], and the glucagon response was absent during postprandial hypoglycemia [111].

SUMMARY AND RECOMMENDATIONS

●Goals of transplantation – The goals of transplantation are to restore glucose-regulated endogenous insulin secretion, arrest the progression of the complications of diabetes, and improve quality of life. (See 'Introduction' above.)

●Indications for transplantation – Transplantation is generally considered only in patients with serious, progressive complications of diabetes in whom the quality of life is unacceptable (figure 1).

•Patients with end-stage kidney disease – Patients with type 1 diabetes and end-stage kidney disease who have had or plan to have a kidney transplant are candidates for pancreas transplantation. The successful simultaneous or subsequent transplantation of a pancreas will improve glycemia and may improve kidney survival. (See 'Indications for transplantation' above.)

•Patients without advanced kidney disease – Patients without substantial kidney disease are candidates for pancreas transplantation alone if they have a history of frequent, acute, severe metabolic complications (hypoglycemia, marked hyperglycemia, ketoacidosis); incapacitating clinical and emotional problems with exogenous insulin therapy; and consistent failure of insulin-based management to prevent acute complications. (See 'Indications for transplantation' above.)

●Pancreas versus islet transplantation – Although no randomized trials have directly compared the outcomes from whole organ versus islet transplantation, pancreas transplantation has a higher rate of insulin independence than islet transplantation. Whole organ transplantation is associated with higher morbidity due to general surgery than islet transplantation. (See 'Pancreas versus islet transplantation' above.)

●Islet transplantation – The less invasive procedure of islet transplantation in people with diabetes is safer than pancreas transplantation with current metabolic outcomes approaching that of solitary pancreas transplantation (pancreas transplant alone [PTA] and pancreas after kidney [PAK] transplant), although it often requires infusion of islets from more than one donor pancreas. (See 'Islet transplantation' above.)

Islet transplantation is performed as part of standard practice in many regions of Canada, Europe, Australia, and Asia. In the United States, only one islet isolation facility has received regulatory approval for its deceased donor islet product (donislecel). Donisclecel is approved for adults with type 1 diabetes who experience recurrent episodes of severe hypoglycemia despite intensive diabetes management and education. (See 'Availability' above.)