Human Pancreatic Islets and Diabetes Research

The goal of this article is to highlight the use of human pancreatic islets as a critical resource for investigators studying diabetes mellitus. The national Islet Cell Resource (ICR) Center Consortium¹ is the first and largest organized cooperative effort in the world to provide human islet preparations to researchers while simultaneously addressing the need to improve isolation and transplantation technologies. According to the National Institutes of Health Research Portfolio Online Reporting Tool,² in fiscal years 2001 through 2008, $67.2 million was invested in establishing and maintaining the ICR consortium. Data collected by the ICR's Administrative and Bioinformatics Coordinating Center (ABCC) offer information on the scope, direction, and challenges of diabetes research involving human pancreatic islets from basic laboratory science and clinical perspectives.

Possibly the greatest value of islet research is in the prospect of better understanding the uniqueness of human islet biology in the native pancreas, and other environments, to expand therapeutic options for diabetes patients and prevent development of the disease. Herein, we discuss the need for human pancreatic islets, including the various approaches and tools used to study the production, regulation, and endocrine action of islets, as well as the scientific impact of the research. The major limitations in using human islets are discussed, including the supply and demand dilemma and factors limiting consistent and widespread availability of these preparations for medical research purposes. The sustained support of islet distribution centers as a resource for diabetes investigators is considered. We conclude by examining alternatives to using human islets and discussing relevant clinical applications. The use and support of human pancreatic islets in the study of diabetes is a timely and critical topic for scientific and therapeutic advances.

Information available through PubMed and Computer Retrieval of Information on Scientific Projects databases shows that there is an increasing reliance by investigators worldwide on the use of human pancreatic islets to conduct research studies. Perhaps the best way to demonstrate this need is to examine the requested volume, quality, and frequency of islet preparations. These data are available through the ICR ABCC and show that, for ICR-approved investigators, requests for human pancreatic islets for use in basic science research have increased substantially over the past several years, with a demand of 52.3 million islets equivalents (IEQs) in 2008 (Figure). (The word islet is used interchangeably with the term IEQ. One IEQ represents an islet of a typical size equal to 150 μm in diameter. This scale was introduced nearly 20 years ago to account for islets of different sizes within a pancreas and adopted by all islet isolation laboratories as a standard convention.³ )

Although there was at least a 15% increase in production each year in response to the burgeoning demand, the ICR centers decreased from supplying 81% of the requested total in 2005 to only 46% in 2008. In early 2007, an automated Web-based optimization algorithm for the matching and distribution of pancreatic islet preparations to waiting approved investigators was implemented to improve allocation and demonstrate the efficiency of a central islet coordinating center,⁴ yet the disparities between supply and demand remain.

From its inception in 2001, a total of 14 academic institutions across the United States have been named as ICR centers with 8 currently active and funded. Prior to providing human pancreatic islets, each ICR had to demonstrate compliance with current good-manufacturing-practice facility requirements via an external audit; although all centers certified as having current good manufacturing practices were allowed to provide islets to approved investigators for basic research purposes, only a limited number of highly qualified centers, who demonstrated additional stringent clinical proficiency, supplied islets for clinical transplantation.

The ICR laboratories are responsible for providing human pancreatic islets to qualified investigators for use in transplantation protocols approved by institutional review boards and the Food and Drug Administration, or for basic biological research studies. Eligibility for receipt of ICR islets is determined through an application process designed to minimize the unnecessary review of mature research studies while helping to constructively develop those that may require additional assistance. Applications for islets can be found on the ICR consortium Web site.⁵

Through August 2008, a total of 182 applications (156 basic science, 26 clinical transplant) were processed and approved by the ICR consortium, representing an increase from only 10 applications in 2001. Among the clinical protocols, 15 involved the transplantation of islets alone into eligible type 1 diabetic patients, in 5 studies islets were transplanted into patients following a kidney transplant, 3 protocols involved simultaneous islet and kidney transplantation, and 3 were double–intervention group studies comparing more than 1 approach (Table 1). Clinical trials outcome data from these and other islet transplantation protocols have been recently published in the most complete collection of human islet transplants reported to date.⁶

The approved basic science applications using human islets provided by the ICR centers have been categorized into major areas and subareas of research (Table 2). Categorization of proposals was self-reported, and multiple categories could be chosen. The fact that 59 of these proposals (38%) spanned more than 1 major area or subarea of research highlights the interdisciplinary nature of these studies. Of 182 ICR-supported studies, 119 (65%) were sponsored by the US government, internationally recognized foundations, or both, including the National Institutes of Health (n = 65), Juvenile Diabetes Research Foundation (n = 36), American Diabetes Association (n = 4), National Aeronautics and Space Administration (n = 2), National Science Foundation (n = 1), or a combination of 1 or more sponsors (n = 11). Publicly available funding data^{2 ,7 - 8} were available for 93 of 119 studies and exceeded $83 million.

Islet Use. High-volume distribution of human pancreatic islets is necessary to establish an adequate supply for repeated experimentation. Data from the ICR ABCC show that 297.6 million islets were produced by 14 ICR laboratories between September 1, 2001, and August 31, 2008 (Table 3). Of those, 199.4 million islets (67%) were used for basic science research, 91.6 million islets (31%) for clinical purposes, and 6.5 million (2%) were not used for various reasons ranging from consent limitations to poor-quality pancreata or islets.

Data from the ICR distribution program indicate that 71.1 million of the 199.4 million islets were distributed within 2593 shipments for use in ICR-approved studies requiring human islets for basic research (Table 2). The remaining 128.3 million islets were used for ICR intramural studies to optimize human islet production and study the parameters that affect the quality and performance of islets for transplantation and research. The 91.6 million islets produced for clinical purposes represented 201 islet isolations used for 170 human transplants. Although a majority of these transplants were performed in hospitals with an ICR facility, 5 of these isolations were processed and sent to geographically remote facilities to perform 3 clinical transplants.

Scientific Importance. The importance of human pancreatic islets, clinically or for basic science research, is substantiated by the number and quality of studies being performed that rely on these preparations. Data available through the ICR as of August 2008 indicate that a total of 151 national and international scientists received human islets for use in both intramural research performed by the consortium as well as 182 clinical and basic science projects submitted to the consortium for support. These studies have resulted in 201 publications directly attributable to the use of human islets supplied by ICR laboratories (maintained on the consortium Web site).⁵ An Institute for Scientific Information Web of Knowledge search⁹ yielded information on 186 of these citations and showed that articles appeared in more than 50 different journals, which had a median impact factor of 22, and were cited 2771 times from 2002 to 2008. According to all available data on human islet–based research, these preparations are clearly an essential and critical resource for medical and scientific investigators.

Pancreas Donations. The production of human islets is contingent on the availability of pancreata. According to information from the Organ Procurement and Transplantation Network database, there were 56 888 deceased organ donors in the United States between January 1, 2001, and November 30, 2008.¹⁰ Of those, 11 204 of the 15 197 recovered pancreata (74%) were used for whole organ transplantation. Although information was not specified on the use of the remaining 3993 pancreata, ICR ABCC data for the same time frame indicate that at least 1212 (30%) were used to isolate human islets.

Table 4 and Table 5 outline the general organ donor characteristics available for 1142 of 1212 ICR-processed pancreata. Several articles have described what constitutes optimal donor criteria for human pancreata used by isolation laboratories.^{11 - 12} A wide range of values exist for many of these parameters, which may not have been previously appreciated by medical professionals responsible for referring organ donors or those in charge of coordinating the donation process.

Facility Operations. The production of human pancreatic islets requires a pancreas-processing step prior to their use in clinical or basic laboratory research. This creates a unique set of challenges,¹³ akin to assembly line manufacturing, that require standardization and optimization to minimize disruption of the islet supply for therapeutic and translational research purposes. Each isolation facility must have an infrastructure in place to address regulatory compliance and other factors, including personnel qualifications and training; facility design; reagent, equipment, and supply verification; production and product controls; process validation; adequate documentation; and environmental monitoring. In the ICR model, with the exception of reagents for clinical transplantation, all of the infrastructure expense associated with islet production was paid for by consortium funds.

Moreover, isolation centers manufacturing human islets for clinical purposes must do so under current good-manufacturing-practice regulatory guidelines.¹⁴ The cost to establish a current good-manufacturing-practice–compliant facility can range from $1 million to $7 million and annual maintenance from $0.8 million to $3 million,¹⁵ highlighting the need for regional laboratories. Even in a facility compliant with current good manufacturing practices, unforeseen circumstances can compromise a laboratory, as was the case in 2007 when nearly all human islet transplants were halted because of concern over the risk of bovine spongiform encephalopathy in Liberase HI (Roche Diagnostics, Indianapolis, Indiana), a critical and specialized collagenase enzyme blend used by nearly all islet isolation facilities to digest pancreata.¹⁶

Organ Acquisition Costs. The cost to procure a pancreas represents a formidable fiscal barrier to islet isolation laboratories.¹⁷ Data from the ICR consortium show that for 665 pancreata acquired from 2001 to 2008, standard acquisition charges ranged remarkably from a low of $600 to a high of $39 800 (Table 6). Several actions have been suggested to address the discrepancy in cost¹⁷ ; however, this remains an important consideration that limits the ability of facilities wanting to isolate human pancreatic islets.

Islet Shipment. Investigators using human islets are often located far from the pancreas processing facility producing them, necessitating standardized shipping strategies that preserve the mass, quality, and function of the preparation while avoiding contamination. Although shipment protocols for clinical transplantation purposes are still being developed,¹⁸ those created for basic science islet distribution have been successfully standardized and are available through the ICR ABCC.⁵ The fact that only 3 ICR centers have been prepared to ship islets to remote locations for clinical use,¹⁹ and only 1 has established an ongoing collaboration with a distant transplant facility, underscores in part the complex requirements for both the islet isolation facility producing the islets and the transplant center using the graft.²⁰ In contrast, 2593 basic science islet shipments have been made by the consortium. Data available on 2124 shipments show that investigators found the quality of islets, by subjective evaluation, to be excellent in 681 shipments (32%), good in 1118 (53%), fair in 218 (10%), and poor in 117 (5%).

Sustained Production. While the number of studies requiring human islets is at a 5-year high, so are the observations that human islets differ considerably from their nonhuman counterparts.^{21 - 22} Nearly 20 years after automating the isolation procedure for human pancreatic islets,²³ and through the collaborative efforts of government and private foundation sponsorship, an islet distribution infrastructure has been built and is increasingly in demand by investigators worldwide for both basic science research and clinical transplantation. Responding to this demand will require confronting the escalating costs and complexities of maintaining and improving an islet infrastructure capable of responding to the changing needs of the diabetes community. For the ICR consortium, this has meant addressing facility and production challenges, such as those with personnel, reagents, and equipment, as well as attempts to continually improve islet isolation and transplantation technologies, such as pancreas preservation, organ procurement and processing, predictive islet potency testing, and shipping conditions of human islets for either clinical or basic research purposes.

For more than 30 years, pancreatic islets have been used as experimental clinical and laboratory models in diabetes research.^{24 - 25} Islets have been obtained from and examined in diverse animal models, including monkey, pig, rat, mouse, hamster, rabbit, and dog.^{26 - 28} In addition, islet-like insulin-producing cells have been generated and studied from germ-cell precursors and mature tissues and organs other than the pancreas. Considerable molecular, cellular, and genetic data have been collected from these studies.²⁹ The various approaches pursued are described here.

Surrogates for Human Beta Cells. Several strategies to restore human beta-cell function have been explored to reduce or abolish the need for pancreas-derived isolated human islets.^{30 - 31} The first focuses on provoking beta-cell regeneration in vivo through the use of pharmacological and other beta-cell stimulatory agents. This approach is supported by the observations that beta-cell mass is increased in human obesity³² and pregnancy³³ and that beta cells can be stimulated to grow in mice.³⁴

The second and third strategies, respectively, seek to induce ex vivo expansion of pancreatic beta cells by either exploiting the proposed models of human beta-cell mass maintenance through differentiation of a stem cell or progenitor population or by beta-cell self-replication. Literature supporting these mechanisms primarily have relied on the use of rodent and other nonhuman models, resulting in controversy when attempting to validate these findings using human pancreata.³⁵

A fourth approach is based on the transformation of one cell to another using mature tissues or organs other than the pancreas.³⁶ Such transdifferentiation has been performed using bone marrow, intestinal, and liver cells and results in cells that are capable of producing insulin.³¹ A recent study combining several of these approaches demonstrated for the first time in vivo the ability to convert pancreatic exocrine tissue from adult mice into cells that mimic beta cells by introduction of 3 key transcription factors into the splenic lobe of the dorsal pancreas.³⁷ Yet many fundamental clinical questions remain unanswered regarding surrogate beta cells, such as the pleiotropic effects, oncogenic potential, autoimmune properties, and long-term metabolic function.

Cell Lines. Given the innate variability of biological samples and the cost of human islet isolation, a common renewable reference standard is desirable for investigational and therapeutic purposes. Unlike the cancer model where tumor-derived immortal cell lines are used for preliminary experimentation, no such acceptable substitute yet exists for human islets. A limited number of human beta-cell lines, together with numerous rodent-derived insulinoma cell lines, have proven useful as models for certain aspects of human beta-cell maturation and function.^{38 - 39} Although promising, these models have not yet reached the point of therapeutic potential. The main obstacles are similar to those for surrogate beta cells,⁴⁰ namely long-term cell survival, responsiveness to glucose, and oncogenicity.

Nonhuman Islets. Nonhuman pancreas–derived isolated islets have been used therapeutically^{41 - 43} and have been in widespread use experimentally for decades. Islets from swine, primates, and rodents have been used primarily with the rodent model predominating in basic research and preclinical studies of diabetes. Benefits of the rodent model include short generational times, control of environment, homogeneity of genetic background, ease of genomic manipulations, ability to selectively breed, and the relative inexpensiveness of husbandry. Compared with human islets, rodent islets can be isolated in less time, at a fraction of the cost, with a robust insulin secretory response, and in sufficient quantities for experimentation.

Limitations of Alternate Models. Although rodent islets have been used extensively to study various aspects of islet function, basic science studies have shown important differences between human and rodent islets. Early work demonstrated that the beta-cell cytotoxic effects of agents seen in rodent islets was not observed in human preparations,²¹ suggesting a species-specific response to beta-cell injury. Independent laboratories established that the proportion of beta cells in a human islet are nearly one-third less than that found in the mouse and that, although cell types within the mouse islet cluster together, the same was not true of the human counterpart.^{44 - 45}

This difference in beta-cell composition accounts for the robust insulin secretion seen in rodent islets, and it raises the possibility that there is an intra-islet mediated concomitant increase of cytokine secretion not seen in human preparations. For example, while vascular endothelial growth factor has been shown to be an important regulator of islet vascularization in rodents,⁴⁶ recent evidence now implicates human duct cells, but not islets, as the main source of production of vascular endothelial growth factor.²² Taken together with the differences in islet architecture between human and rodents, it is not surprising that there exists species-specific models of islet vascularization and blood flow.⁴⁷ Recently, Parnaud et al⁴⁸ showed that, unlike those purified from rats, beta cells from fully differentiated human islets were not capable of proliferating in vitro when grown using extracellular matrix and various growth factor stimulants. These observations illustrate the ongoing critical requirement for human pancreatic islets.

The primary objective of islet-based research is to cure diabetes. Perhaps the most prominent clinical application of this research is currently in the form of cell replacement therapy. With the exception of 1 report in a type 2 diabetic cohort,⁴⁷ islet transplantation has been used exclusively for a subset of individuals with type 1 diabetes mellitus and was shown, at least temporarily, to improve glucose control and, in a few cases, to lead to insulin independence. One-year insulin independence success rates of 10% were seen prior to the introduction of the Edmonton protocol in 2000, subsequently improving to 80% to 90% when this protocol was performed at established centers.⁴⁹ Although the insulin independence rate declined to approximately 10% at 5 years after transplantation, 80% of the recipients continued to have islet graft function with near-normal glycosylated hemoglobin and were protected from severe hypoglycemia, the most common indication for islet transplantation.⁵⁰

Illustrating the complexity of the procedure,⁴⁹ success rates vary greatly across transplantation centers, with 1 facility reporting a 4-year insulin independence rate exceeding 50% in single-donor transplants.⁵¹ A recent study of 325 adult islet transplant recipients concluded that the procedure significantly improved metabolic control for individuals with previously labile disease.⁶

Although islet transplantation has been shown to offer both protection against long-term complications of the disease and significant improvement in quality of life,^{52 - 53} several obstacles remain, such as limited engraftment,^{54 - 57} chronic immunosuppression, and inconsistent supply of human islets. These issues must be addressed if the procedure is to be used as a standard of care for qualified individuals.

The use of human pancreatic islets can serve as a gold standard for the assessment of beta-cell function in the future, as clinical research advances toward a cure for diabetes. Investigators seeking to understand the biology of human islets have approached the problem in a variety of ways. Some have used surrogate beta cells and cell lines while others have focused on pancreas-derived isolated islets from human or nonhuman sources. Islets and islet-like cells have been used experimentally and clinically to increase beta-cell mass. Basic research studies have shown species-specific differences in islets, especially between human and rodent models, when examining proliferative potential, differentiation capacity, response to injury, composition and organization of cell clusters, and type and amount of secreted products. Clinical studies have been promising, but the challenges continue to include limited engraftment of the islet preparation after transplant, chronic immunosuppression of the individual, and availability of human islets.

Islet sharing networks have been established to help address the supply and demand issues faced by islet isolation laboratories and clinical and laboratory scientists. These distribution networks have had a global influence on diabetes research but face economic obstacles in preserving the availability of human islets in a growing research community. In the 5-year history of the ICR basic science distribution program, islets were provided free of charge for all but 11 months of this initiative. Had a subscription fee been maintained throughout the program, it is conceivable that yearly production may have increased further. On the other hand, investigators faced with the prospect of paying for islets, and unable to secure long-term funding to do so, may have reduced the total number of islets requested.

Human pancreatic islets will be critical for restoration of beta-cell function in patients with diabetes. Even given adequate funding levels, the ongoing challenges to supplying human islets must be addressed for the successful exploration of therapeutic options for this chronic and debilitating disease.