Abstract
Inconsistent islet isolation is one of the issues of clinical islet transplantation. In the current study, we applied ductal injection to improve the consistency of islet isolation. Seven islet isolations were performed with the ductal injection of ET-Kyoto solution (DI group) and eight islet isolations were performed without the ductal injection (standard group) using brain-dead donor pancreata. Isolated islets were evaluated based on the Edmonton protocol for transplantation. The DI group had significantly higher islet yields (588,566 ± 64,319 vs. 354,836 ± 89,649 IE, p < 0.01) and viability (97.3 ± 1.2% vs. 92.6 ± 1.2%, p < 0.02) compared with the standard group. All seven isolated islet preparations in the DI group (100%), versus only three out of eight isolated islet preparations (38%) in the standard group met transplantation criteria. The islets from the DI group were transplanted into three type 1 diabetic patients and all three patients became insulin independent. Ductal injection significantly improved quantity and quality of isolated islets and resulted in high success rate of clinical islet transplantation. This simple modification will reduce the risk of failure of clinical islet isolation.
Introduction
Failure to consistently obtain a high quantity and quality of islets is one of the major obstacles for clinical islet transplantation. Even advanced islet centers barely achieved 50% success of clinical islet isolations (1,2,4). Recently we demonstrated that our modification of the Ricordi islet isolation method enabled us to achieve more than 80% success rate of clinical islet isolation with non-heart-beating donors (NHBDs) (6,10). This modified islet isolation method consists of rapid cooling of the pancreas after cardiac arrest, ductal preservation with modified Kyoto solution, two-layer pancreas preservation, Ricordi method for pancreas digestion, and density-adjusted continuous islet purification with iodixanol and Kyoto solution (5). In this study, among those procedures, we introduce pancreatic ductal injection for brain-dead donors (BDDs) in order to clarify the usefulness of this technique. We demonstrated that introduction of pancreatic ductal injection enabled us to achieve seven consecutive successful clinical islet isolations.
Materials and Methods
Ethical Guidelines
This study was approved by the institutional review boards and the application of investigational new drug of clinical islet transplantation is approved by U.S. Food and Drug Administration.
Donor Background
Donor selections were performed based on the Edmonton protocol for clinical-grade pancreas (16). Seven pancreata from BDDs were procured through either Southwest Transplant Alliance (Dallas, TX) or LifeGift (Fort Worth, TX) between February 2007 and May 2008 for the ductal injection (DI) group. For historical control, eight pancreata from BDDs were procured through Southwest Transplant Alliance (Dallas, TX) between April 2005 and December 2006.
Pancreata Procurement, Islet Isolation, and Purification
All pancreata were procured by transplant surgeons of Baylor Regional Transplantation Institute (Dallas & Fort Worth, TX). For the DI group, we removed the duodenum and spleen from the pancreas at the procurement site. This process was performed by the Baylor islet team. A cannula was immediately inserted into the procured pancreas through the main pancreatic duct from the direction of the pancreatic head. Approximately 1 ml/g pancreas of ET-Kyoto solution (Otsuka Pharm Factory Inc., Naruto, Japan) was administered intraductally (5,6,10). For the standard group, the ductal injection process was not performed. All pancreata in both the DI group and the standard group were preserved by the oxygen static charged two-layer (oxygenated perfluorocarbon/UW solution) method for less than 6 h (8).
Islet preparations were performed according to Good Manufacturing Practice (GMP) at the Baylor Research Institute cell processing facility in Dallas, Texas. Islet isolation was performed according to the Ricordi method using the same standard operation procedure (13). Briefly, after the pancreas was decontaminated, the duct was perfused in a controlled fashion with a cold enzyme solution. Liberase HI (Roche Molecular Biochemicals, Indianapolis, IN) was used for two cases of the DI group and all cases of the standard groups. Collagenase NB with neutral proteases (Serva Electrophoresis GMbH, Heidelberg) was used for the remaining five cases of the DI group. The distended pancreas was then cut into pieces and transferred to a Ricordi chamber. The pancreas was digested by repeatedly circulating the enzyme solution through the Ricordi chamber at 37°C. The phase I period was defined as the time between placement of the pancreas in the Ricordi chamber and the start of collection of the digested pancreas. The phase II period was defined as the time between the start and the end of the collection.
The islets were purified with a continuous density gradient using Biocoll in a chilled apheresis system (COBE 2991 cell processor, Gambro Laboratories, Denver, CO) (7,14).
Islet Evaluation
Islet evaluation was independently judged by two investigators. Islet yield was determined using dithizone staining (Sigma Chemical Co., St. Louis, MO) (2 mg/ml) under optical graticule and converted into a standard number of islet equivalents (IE, diameter standardizing to 150 μm) (5,7,12). Purity was assessed by comparing the relative quantity of dithizone-stained tissue to unstained exocrine tissue. Islet viability was evaluated using fluorescein diacetate (FDA) and propidium iodide (PI) staining to visualize live and dead cells simultaneously (5,7,12).
Islet Transplantations Into Type 1 Diabetic Patients
Once islet preparations met the criteria of the Edmonton protocol for transplantation, those isolations were considered successful. Our current criteria for the approval of clinical transplantation are that islet yields are more than 4000 IE/kg body weight, purity more than 30%, viability more than 70%, tissue volume less than 10 ml, endotoxin level less than 5 EU/kg body weight, and a negative Gram stain based on the Edmonton protocol (16).
Recipient selections were performed based on the Edmonton protocol (16). Patients were sedated and a percutaneous transhepatic approach was used to gain access to the portal vein for all patients. Once access was confirmed, the Seldinger technique was used to place the Kumpe catheter within the main portal vein. Islets were infused by gravity and using the bag technique (6).
Assessment of Transplanted Islet Function
Islet functioning was assessed in terms of daily serum glucose levels, serum C-peptide, amount of insulin requirement, and HbA1C before and after islet transplantation.
Statistic Analysis
Values for the data collected represent means ± SE. Two groups were compared using unpaired t-test. Ratio between two groups was compared using Fisher's exact test. Values of p < 0.05 were considered significant.
Results
Donor and Islet Characteristics
Donor-related variables were shown in Table 1. There were no significant differences in the ratio of gender, age, body mass index, peak blood levels of glucose, alanine aminotransferase (ALT), and creatinine.
Donor-Related Variables of Human Pancreas
Values are expressed as mean ± SE. The p-value was calculated using Student's t-test except for gender. The p-value for gender was calculated using Fisher's exact test.
Islet isolation variables were shown in Table 2. There were no significant differences in pancreas weight, cold ischemic time, phase I period, and undigested tissue volume. All pancreata were preserved less than 6 h. Phase II period was significantly longer in the DI group.
Islet Isolation Variables
Values are expressed as mean ± SE. The p-value was calculated using Student's t-test.
Prepurification islet yield was significantly higher in the DI group (DI vs. standard: 902,350 ± 139,397 vs. 497,457 ± 89,414 IE; p < 0.03) (Fig. 1, right). After islet purification, islet yields were also significantly higher in the DI group (DI vs. standard: 588,566 ± 64,319 vs. 354,836 ± 89,649 IE; p < 0.01) (Fig. 1 left). Islet variables are shown in Table 3. Viability was significantly higher in the DI group. Purity was significantly lower in the DI group.
Islet yields before and after purification in the ductal injection group (DI) and the standard group (standard). Islet yields were significantly higher in DI group both before and after islet purification.
Islet Variables
Values are expressed as mean ± SE. The p-value was calculated using Student's t-test.
Success of Islet Isolation
All isolated islet preparations were qualified for transplantation in the DI group (Table 4). Three out of eight isolated islet preparations were qualified for transplantation in the standard group; the other five had an insufficient islet yield. We attempted to transplant all seven islet preparations in the DI group, but in one case the radiologist could not gain access to portal vein and the preparation was not transplanted. Therefore, only six preparations were transplanted into three type 1 diabetic patients. Each patient received two islet preparations. In the standard group, two successful preparations were transplanted into two type 1 diabetic patients. The other one case was not transplanted due to lack of full preparation in clinical side.
Qualification for Transplantation
The p-value was calculated using Fisher's exact test.
Clinical Outcome in the DI Group
In the DI group, fasting blood glucose of all three patients improved after single islet transplantation and further improved after the second islet transplantation (Fig. 2). Importantly, after the second islet transplantation, no patients experienced severe hypoglycemia thereafter.
Fasting blood glucose levels before and after islet transplantation of three patients in the DI group. All patients improved glycemic control after islet transplantation.
All three patients became insulin independent (Fig. 3). HbA1C before transplantation was 8.3% (first patient), 8.3% (second patient), and 7.4% (third patient) and after transplantation were 6.0%, 5.8%, and 5.8%, respectively. Fasting C-peptide levels were all undetectable before transplantation. The current fasting C-peptide for the first patient was 2.2 ng/dl, 3.2 ng/dl for the second patient, and 2.1 ng/dl for the third patient.
Daily insulin doses before and after islet transplantation of three patients in the DI group. All three patients became insulin independent.
Discussion
To our knowledge, this is the first study of pancreatic ductal injection at the donor site for clinical islet transplantation using brain-dead donors (BDDs). This modification enabled us to have seven consecutive successful clinical islet isolations. Failure of islet isolation is one of the major issues for clinical islet transplantation because of the loss of donor pancreas, waste of money and efforts (3). Therefore, this simple modification is of great value for clinical islet transplantation.
Previously we have shown that modification of the Ricordi method including ductal injection improved islet yields using NHBDs (6,10). For NHBDs, we used ET-Kyoto solution combined with ulinastatin (6,10); however, we eliminated ulinastatin for this study because ulinastatin is not available in the US. In addition, usefulness of trypsin inhibition for BDDs is controversial (9,14). In this study, we confirmed that the ET-Kyoto solution alone was effective for ductal preservation for BDDs.
Recently we have shown that more than 10% of exocrine tissue suffered apoptic cell death during preservation before islet isolation and the ductal injection of modified Kyoto solution reduced exocrine tissue apoptosis to less than 2% in the porcine model (11). In addition, the ductal injection with both UW solution and modified Kyoto solution improved ATP activity in the cellular component in a porcine model (11). However, UW solution inhibits collagenase activity for human pancreas (11), and therefore we chose ET-Kyoto solution for human islet isolation. Importantly, phase I time and undigested tissue volume was not different between the DI group and the standard group, suggesting that ductal injection of ET-Kyoto solution did not inhibit the collagenase activity during human pancreas digestion.
Purity was significantly lower in the DI group. We speculated that the healthier exocrine tissue survived well during the islet isolation process, causing lower purity of islet preparations. In addition, significant prolonged phase II time in the DI group suggested that healthier exocrine tissue had less autolysis, resulting in a prolonged collection period. Because autolyzed exocrine tissues release several digestive enzymes, less autolysis could be important to prevent overdigestion of isolated islets. It is reasonable to think that the ductal injection prevents exocrine cell death and therefore avoids overdigestion of isolated islets. Low purity of isolated islets can be a concern because a lower purity results in a larger tissue volume. The tissue volume was higher in the DI group, although the difference did not reach statistical significance. However, all islet preparations were adjusted to less than 10 ml and we had no transplant complications related to relatively large tissue volume.
Viability of isolated islets was significantly higher in the DI group. Previously we have shown that DI prevented both exocrine tissue and islets from apoptotic cell death in a porcine model (11). This suggested that DI also improved the quality of isolated islets.
Sawada et al. demonstrated that the ductal injection of small amount of UW solution protected pancreatic duct in rodent model (15). This is another important demonstration of the usefulness of ductal injection because it is essential to maintain good patency of pancreatic duct for collagenase delivery. Ductal preservation at the procurement site allows us to maintain the patency of the pancreatic duct during preservation and transport; it is therefore possible to use only one cannula for collagenase delivery. The single cannulation technique is better than the usual two cannulations because this technique eliminates cutting pancreas for cannulation. Because the pancreas is not cut, there is excellent pancreas distension and minimal collagenase leaking.
In this study, we lost approximately 35% of islets during the purification process. Previously, we used density-adjusted ET-Kyoto and iodixanol solution for purification with NHBDs, resulting in approximately 80% recovery rate (6). If we were able to achieve the same recovery rate with BDDs, we might be able to obtain more than 700,000 IE from a single donor. Currently 10,000 IE/kg recipient body weight is the target for insulin independence (16); therefore, this high yield would enable us to perform single donor islet transplantation in patients up to 70 kg body weight. Introduction of density-adjusted ET-Kyoto and iodixanol solution for purification is currently under investigation at our laboratory.
All three transplanted patients in the DI group became insulin independent after the second infusion, and have improved glycemic control with positive C-peptide. All patients are free from severe hypoglycemia. This clinical outcome shows that DI is not only useful for obtaining high islet yields but is also contributing to the high quality of islets.
We compared with our own historical data to demonstrate the impact of introduction of a new strategy. Strictly speaking, a randomized study is required to draw a definite conclusion. However, the fact of achieving seven consecutive successful clinical islet isolations and all transplanted patients becoming insulin independent is very promising.
Our historical standard group used Liberase HI in all cases and the DI groups used only two cases. To examine whether our result was not due to the difference of the enzyme, we compared the results with Liberase HI and collagenase NB in the DI group. The final islet yields were quite similar between the two groups (570,209 ± 127,542 IE in the Liberase group and 595,895 ± 83,747 IE in the collagenase NB group, p = 0.88). Therefore, the difference seems not due to the difference of collagenase.
In conclusion, the ductal injection of ET-Kyoto solution made it possible to achieve seven consecutive successful clinical islet isolations from BDDs. This simple modification will reduce the risk of failure of clinical islet isolation. A large-scale randomized study will clarify the importance of this modification.
Footnotes
Acknowledgments
The authors wish to thank Dr. Har-rod Carson for reviewing and editing this manuscript. This research is supported by All Saints Health Foundation and Otsuka Pharmaceutical Factory Inc. Shinichi Matsumoto received a research grant from Otsuka Pharmaceutical Factory Inc. Salary of Yasutaka Fujita was supported by Otsuka Pharmaceutical Factory Inc.