Posttransplant Characterization of Long-term Functional hESC-Derived Pancreatic Endoderm Grafts
Abstract
The paucity of human donors limits broadened application of β-cell replacement therapy. Insulin-producing cells derived from human embryonic stem cells (hESCs) have recently been investigated clinically as a feasible surrogate to primary tissue. Herein, we examine the long-term efficacy of hESC-derived pancreatic endoderm cells (PECs) to maintain normoglycemia posttransplant and characterize the phenotype of the PEC grafts. Mice with chemically induced diabetes were transplanted with PECs into the subcutaneous device-less site. Transplant function was assessed through nonfasting blood glucose measurements, intraperitoneal glucose tolerance testing (IPGTT), and human C-peptide secretion for 517 days. Explanted grafts were assessed for ex vivo function and immunohistochemically. All PEC recipients (n = 8) maintained normoglycemia until graft retrieval. IPGTTs at 365 and 517 days posttransplant did not differ (P > 0.05), however, both demonstrated superior glucose clearance compared with nondiabetic and transplant controls (P < 0.001). Serum C-peptide levels demonstrated significant glucose responsiveness (fasted vs. stimulated) (P < 0.01). Small intragraft cysts were palpable in all mice, which resolved but recurred after aspiration. Cysts showed monomorphic neuroendocrine proliferation and lined by ductal epithelium. Explanted grafts demonstrated similar insulin secretory capacity as human islets and stained positively for endocrine cells. Our results demonstrate the ability of PECs to differentiate in vivo and restore glycemic control while confirming minimal proliferation and absence of neoplastic change within the grafts during the time evaluated.
Introduction
Apart from pharmacological exogenous insulin therapy, β-cell replacement represents an effective therapeutic strategy to restore physiologic glycemic control in patients with type 1 diabetes (T1D). Islet transplantation is a proven and efficacious means to achieve normoglycemia, prevent hypoglycemia, and potentially protect against vascular complications associated with T1D (1–4). Notwithstanding marked progress in clinical islet transplantation, with the maintenance of insulin independence in approximately half of recipients up to 5 years (5), this replacement therapy is limited to those with life-threatening hypoglycemic unawareness or severe glucose lability (6). The shortage of human pancreas donors further restricts the broader application of islet transplantation. Continued progress in stem cell developmental biology with the mimicking of intrinsic human pancreas embryonic development, through the production of human embryonic stem cell (hESC)-derived or induced pluripotent stem cell (PSC)–derived insulin-producing cells, facilitates the possibility of permanently replacing the need for human pancreatic islet donors in future practice.
Pretransplant differentiation (7–12) and alternative transplant techniques (7,9,11,13) are being developed to varying degrees in the laboratory. Nonetheless, host immune responses to foreign encapsulation materials remain limiting, although innovations in biomaterial science have led to promising strategies to minimize the host’s response to microencapsulation of therapeutic cells in mice (13,14). Furthermore, advancements in the manufacture of pancreatic endoderm cells (PECs) for clinical trials (15) and the generation of insulin-producing cells of higher maturity (>stage 5) derived from hESCs or induced PSCs in vitro, with capacity to restore earlier normoglycemia compared with early progenitor cells, have recently been reported (10,11). As a means to minimize potential recipient risk in early clinical trials, the transplant site for hESC-derived cell-based therapies should ideally exploit an approach that is retrievable but provides necessary structural and vascular support (16).
ViaCyte, Inc. (San Diego, CA), has recently initiated the first-in-human pilot phase 1/2 clinical trial to test two alternative approaches, VC-01 (clinical trial no. NCT02239354, ClinicalTrials.gov) and VC-02 (NCT03163511) combination products, in limited cohort T1D clinical trials. These products combine CyT49 hESC-derived PECs (stage 4 [S4]) contained within a macroencapsulated, immune-protective device (VC-01) or nonimmune isolated porous device (VC-02) transplanted subcutaneously, the latter of which requires systemic immunosuppression. Evidently, continued development of device-free transplant sites for stem cell–derived product implantation will likely require further refinements to improve engraftment.
Previously, we demonstrated the short-term efficacy and safety of transplanting S4 CyT49 hESC-derived PECs within a prevascularized subcutaneous device-less (DL) site followed for 24 weeks (17). This approach preconditions the subcutaneous site into a more viable microenvironment to facilitate reversal of diabetes posttransplant of rodent and human islets, as well as hESC-derived PECs across a spectrum of diabetes models (5,17–22). Herein, we further evaluated the long-term efficacy and durability of PECs to maintain recipient normoglycemia while characterizing the PEC graft’s phenotype within the DL subcutaneous space.
Research Design and Methods
Embryonic Stem Cell Culture Differentiation
Research-grade PECs were differentiated from CyT49 hESCs by ViaCyte as previously described (23) with a modification of culture format during differentiation (15). hESCs were expanded in adherent culture in DMEM/F12 (cat. no. 10565; Life Technologies) containing GlutaMAX (10565; Life Technologies), supplemented with 10% v/v xeno-free KnockOut Serum Replacement (12618-001; Life Technologies), 1% v/v nonessential amino acids (11140-050; Life Technologies), 1% v/v penicillin/streptomycin (15070-063; Life Technologies), 10 ng/mL heregulin-1b (100-03; PeproTech), and 10 ng/mL activin A (338-AC; R&D Systems). hESCs were plated at 50,000 or 33,000 cells/cm2 for 3- and 4-day growth cycles, respectively. Suspension aggregates of hESC were formed in roller bottles in StemPro medium. The next day, differentiation was initiated in suspension in roller bottles as previously described (23). At differentiation day 12, PECs were harvested and cryopreserved as previously described (24). For the studies described herein, PECs were thawed from cryopreservation and cultured in sixwell plates in S4 media as previously described (23). Once PEC production was completed at ViaCyte, free aggregates were shipped at 18–25°C to the University of Alberta for transplantation.
Human Islets
Control human islets were prepared by the Clinical Islet Laboratory at Alberta Health Services. Deceased donor pancreata were processed for islet isolation, with appropriate ethics approval and consent obtained from next of kin of the donor. Islets were isolated implementing a modified Ricordi technique as previously described (25). Permission for these studies was granted by the Health Research Ethics Board of the University of Alberta.
Preparing a Prevascularized Subcutaneous Transplant Site
One month prior to PEC transplant, a prevascularized subcutaneous DL transplant site was created in 10- to 12-week-old, 20- to 25-g male B6.129S7-Rag1tm1Mom(B6/Rag−/−) immunodeficient mice and C57BL/6 mice (The Jackson Laboratory, Bar Harbor, ME) as previously described (17,19,20). All mice were housed at a maximum of five per cage under conventional conditions with access to food and water ad libitum. Animal care was in accordance with the guidelines approved by the Canadian Council on Animal Care.
Diabetes Induction and Cell Transplantation
One week before transplantation, normoglycemic (<11.1 mmol/L) mice were chemically rendered diabetic through administration of an injection of streptozotocin (Sigma-Aldrich Canada, Oakville, Ontario, Canada), 180 mg/kg i.p., in acetate phosphate buffer, pH 4.5. Induced diabetes was confirmed when nonfasting blood glucose levels exceeded 15 mmol/L for two consecutive days post–streptozotocin administration. Only animals meeting this inclusion criterion were selected for transplantation. At the time of transplantation, 20 μL (5.0 × 106 cells) PECs (ViaCyte), 2,000 human islets, or 500 syngeneic C57Bl/6 mouse islets were infused into the DL site as previously described (17). Prior to recovery, recipients received a 0.1 mg/kg subcutaneous bolus of buprenorphine.
Evaluation of Graft Function
Graft function was assessed through nonfasting blood glucose measurements, using a portable glucometer (OneTouch Ultra 2; LifeScan Canada) twice weekly following transplantation. Periodic fasting blood glucose values were also measured. Reversal of diabetes was defined as two consecutive readings <11.1 mmol/L and maintained until graft retrieval. Exogenous insulin therapy (LinBit pellet; LinShin Canada) was administered subcutaneously peri-transplant and removed at 1 month.
One-year posttransplant serum from recipient mice were assayed for human C-peptide at basal and post–glucose stimulation. Mice were fasted overnight, and blood was collected via the tail vein for basal analysis prior to receipt of a glucose bolus (3 g/kg i.p.). Blood was collected again from the tail vein 60 min post–glucose injection to determine stimulated human C-peptide levels. Serum human C-peptide concentrations were determined by ELISA (Mercodia, Uppsala, Sweden) and expressed as nanograms per milliliter.
As a means to further assess the PEC graft’s metabolic capacity in response to a glucose bolus, glucose tolerance tests were conducted 365 and 517 days posttransplant. Nondiabetic mouse and human islet transplant recipients and nontransplanted naive mice were tested concurrently, serving as physiological controls. Animals were fasted overnight prior to receiving a glucose bolus (3 g/kg i.p.). Blood glucose levels were monitored at 0, 15, 30, 60, 90, and 120 min postinjection, allowing for area under the curve (AUC) to be calculated and analyzed between transplant groups.
Cellular Graft Retrieval
For confirmation of graft-dependent euglycemia, all recipients had their PEC transplants explanted intact by subcutaneous graft excision as previously described (17). Excised grafts were perifused ex vivo (vida infra) or preserved for immunohistochemistry in 10% formalin. Nonfasting blood glucose measurements were monitored for 7 days subsequent to PEC-graft removal to observe a return to hyperglycemia (pretransplant), confirming posttransplant graft function.
Glucose-Stimulated Insulin Secretion
The glucose-responsive insulin secretory capacity of human islets and intact PEC excised grafts was assessed by dynamic perifusion, as described by Cabrera et al. (26). PEC graft specimens were harvested 365 days posttransplant and immediately cut into sections (1 by 3 mm) and placed in individual perifusion chambers. Glucose solutions were perfused through the system at 100 μL/min at 37°C in 16-min intervals with an initial concentration of 2.8 mmol/L followed by an interval of 28 mmol/L, concluding with a final 6-min interval of 2.8 mmol/L. Subsequent to glucose infusion, terminal depolarizing KCl (30 mmol/L) was administered over 6 min. With respect to sampling, during the initial low-glucose phase (1–16 min) effluent fractions were collected 8 min apart. During the subsequent high- and low-glucose phases, elution samples were collected at 1-min intervals (16–32 min), and throughout the KCl depolarization samples were collected 2 min apart (32–44 min). Insulin concentrations (mU/L) in the effluent supernatant were measured by ELISA (Mercodia).
Histological Assessment
Immunofluorescence with anti-insulin and anti-glucagon antibodies was used to identify the presence of pancreatic β-cells and α-cells, respectively. Briefly, following deparaffinization and antigen heat retrieval, graft sections were washed with PBS supplemented with 1% goat serum, followed by blocking with 20% goat serum in PBS for 30 min. The sections were treated with a primary antibody of guinea pig anti-pig insulin (A0564; Dako) diluted 1:100 and rabbit anti-glucagon (ab43837; Abcam) diluted 1:200 for 24 h at 4°C. All primary antibodies were diluted in PBS with 1% goat serum. Samples were rinsed with PBS with 1% goat serum followed by secondary antibody treatment consisting of goat anti–guinea pig (Alexa Fluor 568) diluted 1:500 (PBS with 1% goat serum) and goat anti-rabbit (Fl-1000 Vector Laboratories) diluted 1:500 (PBS with 1% goat serum) for 1 h at room temperature. Samples were rinsed with PBS and counterstained with DAPI in antifade mounting medium (ProLong; Life Technologies). With use of a fluorescent microscope, the resulting microphotographs were taken using the appropriate filter with AxioVision imaging software. Graft cellular composition was analyzed using ImageJ software (National Institutes of Health, Bethesda, MD).
Cystic Lesion Characterization
Cystic lesions were aspirated where present using a 27-gauge needle at ∼200 days posttransplant. Clear fluid aspirates were analyzed for the presence of carcinoembryonic antigen (enzyme immunoassay) and amylase content (Beckman reaction). Furthermore, aspirates were assayed for human C-peptide measured by ELISA (Mercodia). Carcinoembryonic antigen and human C-peptide results are expressed as nanograms per milliliter, whereas amylase concentrations are expressed as units per liter.
Long-term grafts embedded in paraffin were sectioned and 3,3′-Diaminobenzidine stained for neuroendocrine antibody markers CD56 and chromogranin A, ductal epithelium via CK19 and CK7 antibodies and the cellular proliferation antibody MIB1.
Statistical Analysis
Nonfasting blood glucose and human C-peptide data are represented as the mean ± SEM. Human C-peptide analysis was conducted through paired two-tailed t test. Blood glucose AUC (mmol/L per 120 min) analysis for glucose tolerance tests and histological analyses were conducted through parametric one-way ANOVA using GraphPad Prism (GraphPad Software, La Jolla, CA). Newman-Keuls and Tukey post hoc tests were used following the ANOVAs for multiple comparisons between study groups. Kaplan-Meier survival function curves were compared using the log-rank statistical method. P < 0.05 was considered significant. All data fell within the limit of detection for each specific assay or measurement; therefore, no data are missing.
Results
Durable Long-term Function of PEC Grafts
We observed the ability of PECs to differentiate in vivo and reestablish recipient glucose homeostasis for 517 days posttransplant. All recipients (8 of 8) of PECs transplanted using the DL technique demonstrated sustained biologically relevant nonfasting blood glucose profiles and insulin independence, which were preserved until the cellular grafts were excised on day 517 (Fig. 1A). Of note, mean ± SEM nonfasting blood glucose measurements (4.3 ± 0.1 mmol/L [n = 355]) were significantly lower than periodic fasting blood glucose measures (3.0 ± 0.2 mmol/L [n = 19]; P < 0.001, two-tailed t test). Exogenous insulin pellets were implanted into recipients for 30 days posttransplant, at which time they were surgically retrieved. Thereafter, mice did not receive additional medical or therapeutic intervention. Graft-dependent glycemic function was confirmed by prompt reversion to hyperglycemia post–graft retrieval (Fig. 1A). Rigorous glycemic control and graft glucose responsiveness were confirmed through human serum C-peptide collected 1 year posttransplant. This is reflected by the significant increase in stimulated C-peptide detected compared with fasting circulating quantities (P < 0.01, paired two-tailed t test [n = 5]) (Fig. 1B).
PECs restore euglycemia after transplant into diabetic B6/Rag−/− mice. A: Nonfasting blood glucose measurements demonstrate that recipients of PECs transplanted into the subcutaneous DL site (n = 8) maintained normoglycemia posttransplant until the grafts were retrieved (arrow), at which point they became hyperglycemic, similar to pretransplant levels. Dashed line represents time frame in which recipients were implanted with insulin pellets. Data points represent blood glucose mean ± SEM. At 365 days posttransplant (*), glucose tolerance testing was conducted while three recipients were sacrificed for graft characterization. B: Serum human C-peptide levels were collected 365 days posttransplant after an overnight fasting period and 60 min postintraperitoneal glucose bolus. Recipients (n = 5) demonstrated significant glucose-responsive C-peptide secretion (**P < 0.01, paired two-tailed t test). Data points represent C-peptide concentration (ng/mL) (mean ± SEM) and are presented as a scatterplot.
Long-term functional graft characterization was further evaluated and exemplified through intraperitoneal glucose tolerance tests (IPGTTs), conducted at 365 and 517 days posttransplant (n = 6 and n = 3, respectively). In parallel, IPGTTs were also performed on naive, nontransplanted, nondiabetic mice representing positive controls (n = 12). In addition, we have included the IPGTT glycemic profiles of mice transplanted with both syngeneic mouse islets (n = 10) and human islets (n = 6). PEC DL recipients rapidly became normoglycemic following glucose challenge at both time points while demonstrating glucose clearance profiles superior to those of naive controls (Fig. 2A). This was evident by significantly smaller blood glucose AUCs, for glucose clearance, in PEC graft recipients compared with healthy, nontransplanted, nondiabetic mice and both human and mouse islet recipients (P < 0.001, ANOVA) (Fig. 2B). We postulate that this observation could be accounted for by the transfer of a lower physiological glucose set point of human islets and their precursors compared with that of rodents (27), confirming our previous observations (17). We must note that control mice were not age matched to the 517 days posttransplant PEC graft recipients.
Glucose clearance profiles from long-term functional PEC, mouse and human islet transplant recipients. IPGTT of transplant recipients 365 days posttransplant (purple, n = 6 mice), 517 days posttransplant (green, n = 3), human islets (red, n = 6), mouse islets (blue, n = 10), and naive control mice (black, n = 12). A: Blood glucose after dextrose bolus. B: PEC recipients’ AUC did not differ between posttransplant time points (P > 0.05, two-tailed unpaired t test) but was significantly reduced compared with naive controls (P < 0.001, ANOVA). Mice were administered 3 g/kg 50% dextrose i.p. Blood glucose measurements were monitored at 0, 15, 30, 60, 90, and 120 min. Data points represent blood glucose mean ± SEM (**P < 0.01; ***P < 0.001).
Long-term PEC Graft Characterization
Small intragraft cysts were detectable in all eight recipients by ∼200 days posttransplant. These cysts were readily aspirated but recurred over ∼2 weeks (Fig. 3A–C). Aspiration did not disrupt endocrine function (Fig. 1A). The volume of the collected aspirate was 245 ± 31 µL (mean ± SEM) (n = 8). The clear cystic fluid contained carcinoembryonic antigen levels consistent with benign serous-type pancreatic cysts (Table 1). In addition, aspirate contained negligible amylase yet significantly greater human C-peptide levels compared with serum concentrations collected concurrently (P < 0.05, two-tailed unpaired t test) (Table 1). Histological assessment of grafts collected 1 year posttransplant exhibited cystic lesions, which were lined with insulin- and glucagon-positive endocrine tissue, but without evidence of neoplastic or excessive proliferative changes (Fig. 3D–F).
Long-term hESC-derived PEC grafts transplanted into the DL space present with cyst-like formations. Gross photograph of fluid-filled lesion prior to (A) and after (B) fine needle aspiration. C: Photograph of the clear fluid aspirate. D: Representative images, with hematoxylin-eosin, of a PE graft cross-section in DL site 365 days posttransplant, demonstrating cyst-like voids within pancreatic endocrine tissue (E: higher magnification [scale bars = 100 μm]). F: Representative immunofluorescent staining of PE engrafted in the DL site harvested 365 days posttransplant. Grafts were stained with insulin (Ins) (red), glucagon (Glu) (green), and nuclear DAPI (blue) (scale bars = 100 μm).
Cyst aspirate characterization
Sections of excised grafts, harvested at 365 days posttransplant, were immediately assayed by perifusion and demonstrated human islet–like ex vivo insulin secretory profiles (Fig. 4). The rapid secretion of insulin in response to high glucose exposure and subsequent to depolarization with KCl, mimicking human islets, highlights the degree of posttransplant differentiation of the PECs into physiologically regulated endocrine tissue.
PEC grafts excised 365 days posttransplant possess similar dynamic glucose-regulated insulin secretory capacity compared with adult human islets. In vitro secretion of human insulin adult human islets (blue, n = 5 donors, 3 replicates/donor, 50 islets samples per replicate) (A) and excised pancreatic endoderm DL grafts (purple, n = 3 grafts, 3 replicates/graft) (B) during glucose perifusion assay. The system exposes cell preparation to low glucose (2.8 mmol/L), high glucose (28 mmol/L), and low glucose and KCl (30 mmol/L). Data are presented as means ± SEM.
PEC Grafts Possess Both Neuroendocrine and Ductal Tissue
Histological examination of grafts collected at days 365 and 517 posttransplant confirmed their composition of partially solid and dilated ducts, with minimal proliferation and absence of neoplastic change, located subdermally within the subcutaneous tissue just above skeletal muscle (Fig. 5A). Furthermore, histological sections revealed neuroendocrine cells of monomorphic type containing round nuclei in conjunction with cyst-like dilated ducts structures lined by bland epithelium (Fig. 5A). At high power, the neuroendocrine cells depicted a typical stippled chromatin pattern, while the dilated ducts were focally lined by benign, metaplastic mucinous-producing epithelium (Fig. 5A). Strong membranous (CD56+) and cytoplasmic (chromogranin A+) staining confirmed the neuroendocrine nature of the solid component (Fig. 5B and C). Moreover, cyst-like dilated ducts stained positive for CK7 and CK19, confirmed a ductal epithelium phenotype (Fig. 5D and E). Notably, only a few nuclei (estimated <3%) within the entire excised grafts stained positive for MIB1, illustrating the low proliferative turnover of long-term PEC grafts transplanted subcutaneously (Fig. 5F). There was no evidence of local teratoma or other concerns relating to unchecked growth. Recipient necropsy was void of gross organ lesions including liver, spleen, and lungs, confirming, in part, absence of any metastatic change.
Histological characterization of excised pancreatic endoderm grafts. Long-term functional grafts presented with neuroendocrine proliferation and cyst-like structures lined by ductal epithelium as evident through hematoxylin-eosin (H&E) staining (A). Grafts were largely comprised of neuroendocrine tissue staining positive for CD56 (B) and chromogranin A (Chrom A) (C). Cystic lesions stained positive for ductal epithelium markers CK19 (D) and CK7 (E). Minimal cellular proliferation within pancreatic endoderm grafts was confirmed by MIB1 staining (F). Large photomicrographs, original magnification ×40; inserts, original magnification ×100. Scale bars = 100 μm.
Discussion
Herein, we assessed function and neoplastic potential in immune-deficient mice with a research-grade CyT49 stem cell–derived PEC graft transplanted into a prevascularized subcutaneous site for >500 days. The clinical derivative of this important cell line, PEC-01, is currently being tested in first-in-human studies. The long-term risks of such cells have yet to be fully defined. Previously, we showed that hESC-derived PECs can survive and differentiate into glucose-responsive human insulin–producing cells while restoring glycemic control (>100 days) in diabetic recipients upon transplantation into a safely retrievable DL site. Indeed, the DL space was found to confer superior metabolic functional reserve compared with grafts placed within the unmodified subcutaneous space or inguinal fat pad (17).
In the current study, when PECs were transplanted into the DL site all recipients became euglycemic, which was maintained until recovery graft retrieval at 517 days posttransplant. These observations validate graft-dependent euglycemia and illustrate that when transplanted subcutaneously, PEC grafts can be safely removed should this be indicated. Recipient mice demonstrated durable normoglycemia, reflecting the ability of PEC grafts to induce biologically relevant glucose homoeostasis. Again, similar to previous observations (17), PEC grafts transferred a glycemic set point to the mouse recipients that is typical of human islet engrafted in mice (27). These data and the IPGTT profiles of various islet and PEC recipients support the findings in the work by Rodriguez-Diaz et al. (27), that species-specific glycemic set points are transferred with the transplanted islets—and in this case human PECs. We did note a significantly lower fasting blood glucose profile in PEC recipients that which could be considered borderline hypoglycemic in patients with type 1 diabetes. Since we did not transplant bona fide human islets (or β-cells), the glycemic set points established posttransplant, albeit human islet–like, may be entirely unique to the phenotype of the PECs prior to transplant. Glycemic control by insulin secretion relies on the paracrine input from α-cells (27), which we have previously demonstrated to be in higher abundance in PEC grafts compared with human islet grafts, in mice (17). Stage and rate of in vivo differentiation, graft α-cell composition, and recipient factors plausibly influence the unique human islet–like glycemic set point carried by the PEC transplanted, which will be the focus of future experimentation. Substantial glucose-regulated serum human C-peptide was observed 1-year posttransplant, which exceeded that measured in previous work 24 weeks posttransplant (17).
At both 365 and 517 days posttransplant, recipient mice were metabolically challenged to determine their physiologic response to a glucose tolerance test, which yielded superior clearance profiles compared with those of nondiabetic control mice. These data further echo the proficiency and durability of PECs to subcutaneously engraft long-term, their endogenous human islet–like metabolic phenotype, and the capacity to induce glycemic homeostasis in a physiologically relevant fashion.
As previously described, transplants of PEC into the epididymal fat pad can give rise to cysts in addition to endocrine tissue (23). Therefore, the palpable cyst-like lesions observed ∼200 days posttransplant were not surprising. We performed detailed characterization of these long-term endocrine grafts and cystic lesions. The cysts were small but palpable and readily aspirated via fine needle but recurred after ∼2 weeks. Importantly, this procedure did not impair graft function. The clear viscous fluid contained similar levels of carcinoembryonic antigen, consistent with benign mucinous pancreatic cysts (28). Notably, no appreciable amylase was detectible, excluding a pseudocyst (29). The aspirate did indeed contain high concentrations of human C-peptide—higher than basal serum samples collected concurrently. This observation was not unforeseen, as many of the ductal formations within the PEC graft were lined with insulin- and glucagon-positive cells when examined histologically. If we are to assume there is an equal ratio of C-peptide to insulin within the exudate of the cysts, the risk of inherent hypoglycemia, should a cyst spontaneously rupture, would be negligible. Based on the concentration of C-peptide and cystic value, ∼60 µU insulin would be released. This quantity is exceedingly lower compared with the 1 unit/day recipient mice were receiving during the 1st month posttransplant.
Comprehensive histopathology of long-term posttransplant grafts revealed the presence of partially solid and cystic proliferative tissue confined within the subcutaneous space. The grafts were mainly composed of neuroendocrine tissue presenting with characteristic stippled chromatin and a monomorphic cell population with round nuclei. The endocrine phenotype within long-term grafts was confirmed by membranous CD56 and cytoplasmic chromogranin A–positive staining—traditional neuroendocrine cell markers (30,31). These data confirm our previous findings that PECs transplanted into the DL site differentiate into insulin-, glucagon-, somatostatin-, pancreatic polypeptide–, and ghrelin-secreting endocrine cells (17). In conjunction, cyst-like structures were lined by ductal metaplastic mucinous-producing epithelium, confirmed by ductal keratin markers CK7 and CK19, which are expressed in the gastroenteropancreatic and hepatobiliary tracts (32,33). Mucinous differentiation is encountered routinely in the normal pancreatic duct, and it has been suggested that ductal mucinous hyperplasia represents a metaplastic response to a variety of unidentified proliferative stimuli (34). Ductal mucinous hyperplasia has been found in >60% of pancreata from patients with no history of pancreatic disease and with no difference in degree of ductal mucinous hyperplasia found in patients with cancer compared with noncancer patients (34). Allen-Mersh (34) concludes that a likely explanation is that ductal mucinous hyperplasia and pancreatic carcinoma are unrelated responses to proliferation stimuli. Despite the presence of these small cystic lesions, the expression of nuclear proliferation protein Ki-67 stained by an MIB1 antibody within excised grafts was exceedingly low (<3%). MIB1 staining is a frequently used histological method for assessing cellular proliferation and to classify tumors based on proliferation rate cutoffs with a minimum 25% value adequately identifying highly proliferative neoplasms (35). Confirming the low proliferation capacity of the PEC grafts, there was no evidence of local teratoma and there were no concerns related to unchecked growth. In addition, necropsy was void of gross organ lesions, thus demonstrating, in part, the absence of any metastatic change. A conjecture that is often evoked when conducting in vivo experimentation with pancreatic progenitor-derived cell lines is regarding whether transplanting later-stage hESC-derived insulin-producing cell products avoids cystic differentiation compared with transplantation at early stages. Since S4 and later-stage S6 populations are both heterogenous, each is comprised of both pancreatic progenitor (PP) and endocrine cell fractions. A potential advantage of transplanting more differentiated cell populations (S6 or later) is reducing the PP cell ratio by two- to threefold compared with S4. However, if it is assumed that the cystic structures derive from the PP fraction, S6 cell populations still contain ∼30% PP, facilitating ample opportunity to make these cystic lesions from either S4 or S6. It remains unclear whether these structures are derived exclusively from the PP fraction. Likewise, both S4 and S6 populations will contain some level of nonpancreatic cells—mostly other endoderm lineages (ranging in anterior/posterior axis, typically from lung to midgut) and most likely in the same frequency (2–10%) in both cases, S4 or S6. Cystic lesions could potentially arise from these fractions as well. Lack of international consensus on performance, relevance, and control of tumorigenicity assays remains a challenge to the field of human PSCs. A recent report from the International Alliance for Biological Standardization comments that until a critical breakthrough linking what genetic changes specifically lead to a dangerous product that causes tumors, the “gold standard” for the field remains 9-month tumorigenicity studies in immune-suppressed rodents where there is evidence of cell product remaining at the end of the study (36). If grafts do not show evidence of tumorigenesis in that time frame, it is generally considered safe to begin clinical evaluation from the perspective of toxicity and tumor risk. Collectively, the lack of tumorigenicity evident in the current study, with 43 months of follow up, supports a negligible risk of the observed benign cystic lesions progressing to premalignant or malignant potential, at least in mice.
The degree of stem cell–derived β-cell differentiation will impact the cells’ functional resemblance to mature human β-cells, e.g., with glucose sensitivity (10,11). Previously, we demonstrated the modest glucose-stimulated insulin- secreting capability of ViaCyte’s research-grade PECs prior to transplant in response to glucose perifusion (17). This observation was expected, as it is known that S4 PECs are phenotypically and functionally naive compared with mature human β-cells (8) and thus require a period of in vivo differentiation before the establishment of glucose-dependent graft insulin secretion (9,23). In light of the recent work by Rodriguez-Diaz et al. (27), who demonstrated that the control of glycemia by insulin sections from islet grafts relies on paracrine input from α-cells, it may be advantageous to transplant less differentiated pancreatic progenitor cells (S4) with the capacity to differentiate into all pancreatic endocrine cells rather than than using more differentiated β-like cells alone (S7 and S8), as this cellular product may lack all the necessary α-cell input required for a systemic glucostat. In the current study, we examined the state of graft differentiation 1 year posttransplant by assaying the ex vivo insulin secretory profiles of small excised graft specimens compared with mature human islets through perifusion. We have demonstrated that PEC graft fragments have the ability to rapidly secrete human insulin in response to high glucose exposure, recapitulating human islet responses. While these data are indeed preliminary, and many questions remain pertaining to dynamic PEC graft insulin secretion compared with human islets and grafts, we do feel the data serve as a validation of proof of principle for future experimentation. Furthermore, we believe the data highlight the physiological state of posttransplant differentiation of the PECs into functional endocrine tissue. We believe the ability to functionally and phenotypically characterize PEC grafts posttransplant, to evaluate in vivo differentiation in the presence of enhancers and detractors, will be critical to further optimize engraftment and improve clinical outcomes.
The recent initiation of first-in-human phase 1/2 clinical trials testing ViaCyte’s VC-01 combination product has propelled stem cell–derived replacement strategies to the forefront of T1D intervention. Importantly, from the safety profiles garnered from the VC-01 PEC-Encap trials, the U.S. Food and Drug Administration and Health Canada have provided ViaCyte “allowance” to proceed with a phase 1/2 investigational clinical trials agreement to test their VC-02 non-immune-isolating device. The commencement of ViaCyte’s VC-02 (PEC-Direct) trial in the summer of 2017 is particularly noteworthy, as this open device facilitates the direct vascularization of implanted PECs with host vascularization. It has been well documented that human PSC-derived insulin-producing cells display assorted mature β-cell characteristics dependent upon the degree of in vitro differentiation (7,8,10–12). Therefore, current VC-02 grafts will be susceptible to the recipient’s auto- and alloimmune responses, thus necessitating pharmacological systemic immunosuppression. The impact of these antirejection and antiproliferative drugs on in vivo PEC differentiation remains ill-defined and will indeed pave the way for future stem cell–based regenerative and reparative therapies. Furthermore, these trials may validate that PECs overcome the limitations of cadaveric human islet transplantation into the liver by providing a reproducible and ubiquitous source of regulatory approved glucose-responsive cells, delivered in a safe and retrievable manner under the skin. We continue to believe that further long-term stem cell–derived PEC graft characterization and alternative engraftment strategies will facilitate improved transplant outcomes.
Herein, we have shown that hESC-derived PECs can survive, differentiate, and restore stable and durable long-term glycemic control to diabetic recipients upon transplantation into a safe, retrievable, and device-free subcutaneous site, for the natural life span of these mice. We found these studies reassuring, in that over 500 days there was tightly controlled glucose homeostasis, absence of neoplastic change, and minimal proliferative cell turnover encountered within the PEC grafts. Our transplant methodology meets a prerequisite of PSC-derived replacement therapy (16) by providing a means to safely retrieve the graft should unanticipated complications develop. Taken together, the safety data generated in the current study complement the current clinical investigation of the VC-01 and VC-02 combination products and offer additional preclinical methodology for safety testing as current and future stem cell–derived insulin-producing cell therapies continue to evolve into the clinic.
Article Information
Acknowledgments. The authors thank Dr. Patrick MacDonald and Richard Yan-Do (Department of Pharmacology, University of Alberta) for technical assistance with the ex vivo graft characterization experiments.
Funding. This work was supported in part by the Diabetes Research Institute Foundation Canada; Collaborative Research and Innovation Opportunities, Alberta Innovates Health Solutions (AIHS); and Canadian Institutes of Health Research (Proof of Principle). A.M.J.S. is supported through a Canada Research Chair in Transplantation Surgery and Regenerative Medicine and through a Senior Clinical Scholarship from AIHS. A.R.P. is supported by an AIHS postdoctoral fellowship. A.B. is supported by scholarships from the University of Alberta and the Alberta Diabetes Institute. The authors thank the Clinical Islet Laboratory and Alberta Health Services for providing human islet research preparations. The authors also thank ViaCyte, Inc., for kindly providing the pancreatic endoderm cells.
Duality of Interest. A.M.J.S. serves as a paid consultant to ViaCyte, Inc. No other potential conflicts of interest relevant to this article were reported.
Author Contributions. A.R.P. and A.M.J.S. initiated and designed the experiments. A.R.P. and A.M.J.S. analyzed data. A.R.P., A.B., R.P., and D.O. performed the experiments. T.K. isolated and provided human islets. A.T. characterized and performed histological analysis of excised pancreatic endoderm grafts. A.R.P., A.B., R.P., D.O., T.K., A.T., and A.M.J.S. provided input in the manuscript writing and discussion. A.M.J.S. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Footnotes
See accompanying article, p. 901.
- Received July 18, 2018.
- Accepted November 6, 2018.
- © 2019 by the American Diabetes Association.
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.
References
In this Issue
