Outcomes Following Extrahepatic and Intraportal Pancreatic Islet Transplantation: A Comparative Cohort Study : Transplantation

Journal Logo

Original Clinical Science—General

Outcomes Following Extrahepatic and Intraportal Pancreatic Islet Transplantation: A Comparative Cohort Study

Verhoeff, Kevin MD1; Marfil-Garza, Braulio A. MD1,2,3; Sandha, Gurpal MD4; Cooper, David MD, PhD5; Dajani, Khaled MD, PhD1; Bigam, David L. MD, MSc1; Anderson, Blaire MD1; Kin, Tatsuya MD6; Lam, Anna MD6,7; O’Gorman, Doug6; Senior, Peter A. MD, PhD6,7; Ricordi, Camillo MD8; Shapiro, A.M. James MD, PhD1,6

Author Information
doi: 10.1097/TP.0000000000004180
  • Free
  • SDC
  • Infographic



Clinical pancreatic islet transplantation (ITx) has evolved considerably since the groundbreaking Edmonton Protocol was established 22 y ago.1 Optimization of isolation techniques and clinical care has led to 10 y graft survival rates of nearly 80%, coupled with near-complete abrogation from severe hypoglycemia, sustained improvements in glycemic control, and substantial reductions in insulin requirements.2-8 Islet infusion into the intraportal hepatic circulation remains the gold standard for clinical ITx. However, evaluation of alternative implantation sites continues to be explored, with promising preliminary experimental data supporting the gastric submucosa,9,10 omentum,11-14 and subcutaneous space.15,16 As ongoing research with stem cell–derived β cell replacement progresses, recent interest in extrahepatic transplant sites has expanded due to its increased accessibility for monitoring of potential off-target growth, that simultaneously facilitates graft recovery, if required. Understanding comparative outcomes after extrahepatic ITx in humans is valuable before extrahepatic sites can be considered for use with stem cell therapies. Although preliminary case reports suggest a degree of success,11 evidence remains scarce. Moreover, no comparative studies with intraportal ITx have been published to date.

Benefits from intraportal ITx include direct blood contact, which maximizes graft oxygenation, and insulin release into the portal circulation, which may facilitate a more physiologic glycemic response. However, caveats include islet damage from the instant blood mediated immune reaction, restrictions in packed cell volume, and, rarely, procedural complications including portal venous thrombosis and bleeding.17-21 Conversely, access for limited graft biopsy in more localized sites such as the gastric submucosa or skin, and the ease of complete graft retrieval at least in the subcutaneous site may have some advantages over the liver. The omentum has been proposed as an attractive site because of easy operative accessibility, lack of volume restriction, and dense vascular supply with portal drainage, although it still involves a surgical (minimally invasive) procedure and its own unique risks including adhesive small bowel obstruction.11,13,19 Similarly, gastric submucosal implantation allows for graft portal venous drainage, a large capacity for implantation, and offers the possibility to biopsy islet grafts endoscopically but has the least evidence evaluating efficacy in patients to date.9,10 Finally, the subcutaneous space offers procedural safety, technically easy graft implantation, and facilitates ongoing graft monitoring;15,16 unfortunately, this space releases insulin systemically and is substantially more hypoxic, which requires prevascularization strategies to support islet engraftment.16,22

Herein, we report a large single-center experience with extrahepatic ITx and compare outcomes compared with intraportal ITx. We aim to evaluate graft survival, and glycemic outcomes for patients receiving extrahepatic ITx, including gastric submucosal, omental, and subcutaneous implants within devices, as compared with patients receiving intraportal ITx.


Study Design and Patient Selection

This is a single-center retrospective cohort study comparing individuals with type 1 diabetes (T1D) receiving allogeneic extrahepatic ITx with intraportal ITx between March 1999 and October 2018. The study protocol has been approved by the University of Alberta Health Research Ethics Board (PRO00001120) and all patients have consented to use their data for research purposes. All adult (≥18 y old) patients diagnosed with T1D undergoing allogeneic ITx were included. Patients receiving pancreas transplants, autologous ITx, stem cell–based ITx, and with type 2 diabetes were excluded. Both patients receiving islet alone and islet after kidney transplantation were included; these were grouped because only short-term outcomes were evaluated and prior kidney transplantation was deemed unlikely to be a substantial contributing factor.

Patients in the extrahepatic ITx group included those receiving gastric submucosal (n = 2), omental (n = 4), and subcutaneous device islet implantation (n = 3).23 Demographics, primary, and secondary outcomes were compared between groups to determine any differences. Patients receiving intraportal or extrahepatic ITx were analyzed as “intention-to-treat” from their first procedure. Patients in the extrahepatic ITx group were further analyzed after they received subsequent intraportal ITx to assess the effect of extrahepatic grafts on the effectiveness of subsequent intraportal islet infusions. Data for patients receiving prevascularized subcutaneous ITx have previously been reported by our group and are included in aggregate form in the current study.23 Additionally, a secondary analysis comparing extrahepatic ITx with intraportal ITx occurring between January 2012 and October 2018 was completed to enable the evaluation of outcomes from contemporary groups. This was done to ensure any effects seen were not due to changes in treatment over time including changes in immunosuppression, transplant technique, or patient selection. All extrahepatic transplants were completed during the January 2012–October 2018 timeline.

Patient demographics were collected at time of first transplant and included sex, age at T1D diagnosis, T1D duration, age, and body mass index. Measures of pretransplant diabetes control including HbA1c, insulin dose (units/kg/d), and fasting C-peptide levels (nmol/L) were also collected, as were markers of glycemic lability (Lability Index), and hypoglycemia awareness (Clarke score).24 Transplant characteristics were also evaluated including number of islet infusions, timing of infusions, and total islet equivalents (IEQs)/kg of body weight received.

Outcome Variables

The primary outcome of this study was stimulated C-peptide levels 1–3 mo after first ITx measured at 90 min after a mixed meal tolerance test.1,25 Secondary outcomes include fasting plasma glucose (FPG), and BETA-2 score. The BETA-2 score incorporates insulin dose (insulin units/kg/d), FPG (mmol/L), HbA1c (%), and fasting C-peptide levels (nmol/L) and has been validated as a predictive tool for glycemic control and insulin independence.26,27

Additionally, we evaluated graft survival measured by fasting C-peptide levels over time. In the immediate postinfusion period, fasting C-peptides levels were measured every 2–5 d for the first 60 d and reported as 10 d means with SEM. Subsequent C-peptide values for 5 y after first infusion were collected over 6 mo intervals, and reported as means with SEM. Continuous data are described as medians and interquartile range (IQR), with discrete data reported as absolute frequencies and percentages. To further assess the cohorts, rate of primary nonfunction or early graft failure was determined for each group. Primary nonfunction was defined as C-peptide >0.1 nmol/L, and early graft failure was defined as a return to C-peptide values <0.1 nmol/L (or baseline) before a subsequent infusion or within 60 d of first infusion. Multivariable logistic regression analyzed the entire cohort for patient and transplant factors independently associated with a composite variable of primary nonfunction and early graft failure. Finally, we evaluated allosensitization to extrahepatic grafts defined as any calculated panel reactive antibody increase or any de novo donor-specific antibody development following transplant.

Transplant Procedures

Intraportal transplantation involved ultrasound and fluoroscopy-guided percutaneous cannulation of portal venous circulation and islet infusion as described previously.28 Islet isolation procedures and release criteria have also been described. Omental transplant was completed via the biological scaffold “sandwich” technique previously described by researchers from the Diabetes Research Institute in Miami, Florida.11,12 Patients were brought to the operating room and underwent general anesthesia and laparoscopy. The omentum was laid out flat and islets suspended with the recipient’s own plasma were dripped onto the omentum. Recombinant thrombin was then used to cover each of the islet droplets. The omentum was folded over to cover the implantation site and secured in place with ligaclips. Gastric submucosal transplants were completed as previously described by Echeverri et al.10 Patients underwent conscious sedation with subsequent gastroscopy to evaluate the stomach; islets were then infused through a 19-gauge Boston Scientific Expect Slimline needle in 8 submucosal locations throughout the stomach under direct vision. Procedures were performed by a gastroenterologist with advanced training in therapeutic endoscopy. Finally, prevascularized subcutaneous space ITx was completed with islet implantation into a prevascularized non-immunoisolating polymer chamber device as previously described by Gala-Lopez et al.23

All extrahepatic transplants were completed as clinical trials intended to evaluate the potential of novel transplant sites. These sites were selected because of promising preliminary outcomes from others. Registered clinical trial protocols can be reviewed as follows: omental NCT02821026, gastric submucosal NCT02402439, and subcutaneous NCT01652911. In all cases, we collaborated with investigators who initially reported promising outcomes to optimally replicate their technique.


Patients receiving intraportal ITx received various induction, anti-inflammatory, and maintenance immunosuppression regimens. Induction was primarily alemtuzumab (n = 288 infusions, 47.1%), followed by daclizumab (n = 170 infusions, 27.8%), basiliximab (n = 77 infusions, 12.6%), and antithymocyte globulin (n = 77 infusions, 12.6%). A total of 287 infusions (47.0%) used etanercept and anakinra. All patients received tacrolimus (100%) during follow-up and most had mycophenolate (n = 234, 91.4%) as a secondary maintenance immunosuppressant with the remainder receiving sirolimus combined with tacrolimus. Comparatively, induction immunosuppression for infusions into extrahepatic sites was primarily alemtuzumab (n = 6, 60%) and the others received anti-thymocyte globulin (n = 4, 40%). Most infusions were accompanied by both etanercept and anakinra (n = 6, 60%), and 4 (40%) received only etanercept. All patients with extrahepatic ITx had tacrolimus and mycophenolate for maintenance immunosuppression.

Statistical Analysis

For analysis of longitudinal C-peptide data, a mixed-effects model using the maximum-likelihood method was fitted to determine differences over time and between groups, while accounting for missing data. For the analysis of categorical data, the χ2 tests were applied. To compare continuous variables in 2 independent groups, Mann-Whitney U tests were used. A value of P < 0.05 is considered statistically significant.

To evaluate the independent effect of patient and transplant factors on early graft failure and primary nonfunction, a nonparsimonious multivariable logistic regression model was developed using hypothesis-driven selection methods. Variables with statistical significance in the multivariable model (Wald test P < 0.05) were evaluated for multicollinearity using the variance inflation factors. Variables with variance inflation factors >10 were further explored for collinearity diagnostic tests and excluded if deemed collinear.


Overall, 264 patients were included in this study. Of these, 9 (3.4%) patients received extrahepatic ITx for initial islet transplant before undergoing intraportal ITx. These patients were compared with 255 (96.6%) control patients receiving intraportal ITx. At baseline, patients were similar with regard to the age at T1D diagnosis, body mass index, and T1D duration (Table 1). Patients receiving extrahepatic ITx were more likely to be male (41.6% intraportal versus 77.8% extrahepatic, P = 0.032) and were older at time of first transplant (48.8 intraportal versus 59.8 extrahepatic, P = 0.025). Median fasting C-peptide levels, HbA1c levels, and insulin requirements pre-ITx were similar (Table 1).

TABLE 1. - Demographic and baseline characteristics of patients undergoing pancreatic islet transplantation according to implantation site
Variable Intraportal (n = 255) Extrahepatic (n = 9) P  a
Demographics and clinical data at baseline, before 1st transplant
Sex, M/F, n (%) 106 (41.6)/149 (58.4) 7 (77.8)/2 (22.2) 0.031
Age at diagnosis, y (IQR) 14.0 (9–23) 15 (12–26) 0.417
Duration of DM, y (IQR) 30.6 (22.6–40.2) 35.4 (25.3–46.1) 0.374
Age at transplant, y (IQR) 48.8 (41.3–55.8) 59.8 (54.3–60.4) 0.025
Body mass index (IQR) 25.0 (22.9–27.8) 25.4 (25.0–27.3) 0.751
Number of infusions per patient (IQR) 2 (2–3) 2 (2–4) 0.672
Number of extrahepatic infusions per patient (IQR) 1 (1–2)
Time between infusions, mo (IQR)
  Time to 2nd infusion 5.0 (2.1–11.1) 4.3 (3.0–8.5) 0.823
  Time to 3rd infusion 40.4 (16.6–70.9) 5.5 (4.3–16.8) 0.014
  Time to 4th infusion 91.0 (68.5–140.5) 15.1 (5.5–37.7) 0.007
  Time to 5th infusion 165.4 (143.6–181.5) 14.6 (—) 0.157
Total IEQs/kg of body weight, ×1000 (IQR) 14.3 (11.2–18.6) 22.5 (13.3–27.2) 0.097
1st infusion, ×1000 (IQR) 6.1 (4.8–7.0) 7.0 (6.5–9.1) 0.018
Purity (1st infusion) 60 (50–70) 55 (45–65) 0.477
Lability index (IQR) 449 (296–699) 566 (374–608) 0.520
Clarke score (IQR) 5 (4–7) 5 (3–5) 0.052
Laboratory values at baseline, before 1st transplant
C-peptide (nmol/L) (IQR) 0.02 (0.02–0.03) 0.02 (0.02) 0.055
HbA1c % (IQR) 8.2 (7.5–9.0) 8.1 (7.5–8.1) 0.333
Insulin units/kg/d (IQR) 0.54 (0.46–0.68) 0.51 (0.46–0.60) 0.588
Data are n (%) and median (IQR).
aχ2 was used to compare categorical variables, Mann-Whitney’s test was used to compare continuous variables.
bWeighted averages were calculated as follows: weighted average= sum of weighted terms/total number of terms. For example, weighted average = purityinfusion1 (islet numberinfusion1) + purityinfusion2 (islet numberinfusion2) + …/total number of islets infused.
IEQ, islet equivalent; IQR, interquartile range.

Regarding infusion characteristics of the first ITx, both groups received a similar islet preparation purity (60% intraportal versus 55% extrahepatic, P = 0.499) but patients receiving extrahepatic grafts received more IEQs per kg of body weight (6100 IEQ/kg intraportal versus 7000 IEQ/kg extrahepatic, P = 0.018). Overall, after groups received all of their ITxs, both groups received a similar number of infusions, and there was no difference in total infused IEQs/kg of body weight (14 300 IEQ/kg intraportal versus 22 500 IEQ/kg extrahepatic, P = 0.096). However, patients receiving extrahepatic infusions had a shorter delay between their third and fourth islet infusions (Table 1). Eight patients in the extrahepatic group received 1 extrahepatic implantation and 1 patient received 2; patients receiving extrahepatic ITx were switched to the intraportal route if they failed to achieve clinical benefit from their initial graft including insulin reduction, improved glycemic lability, or reduced hypoglycemia. The decision to relist patients for transplant was made following review by the ITx team and determination that late onset graft function was unlikely.

Primary outcome assessment demonstrated that patients receiving extrahepatic ITx had significantly lower stimulated C-peptide levels 1–3 mo after first ITx compared with patients receiving only intraportal infusions (0.05 nmol/L, IQR 0.02–0.24 extrahepatic versus 1.26 nmol/L, IQR 0.95–1.59 intraportal; P < 0.001; Figure 1A). Secondary outcomes showed statistically higher FPG, and lower BETA-2 scores (Figure 1B and C) early after extrahepatic ITx compared with intraportal (Table 2). Once patients with initial extrahepatic ITx received subsequent intraportal islet infusions, they achieved similar stimulated C-peptide levels and FPG compared with those who initially received intraportal infusions (Figure 1A and B). Notably, BETA-2 scores were higher after intraportal transplant in the patients who initially received extrahepatic grafts (19.1, IQR 13.3–22.7; P = 0.004; Figure 1C). Similar differences in primary and secondary outcomes were observed in subanalyses according to specific extrahepatic implantation sites compared with intraportal ITx with a suggestion that best outcomes may have been observed with the omental site (Figure 1D and F). Secondary analysis of patients receiving extrahepatic ITx to contemporary intraportal ITx (n = 106) also showed similar outcomes (Figure S1, SDC, https://links.lww.com/TP/C438).

TABLE 2. - Primary outcomes following extrahepatic and intraportal pancreatic islet transplantation
Variable Extrahepatic alone Extrahepatic + 1st intraportal Intraportal alone
Primary outcome
Stimulated C-peptide (nmol/L) a 0.05 (0.02–0.24) 1.68 (0.4–1.89) 1.26 (0.95–1.59)
Secondary outcomes
Fasting plasma glucose (mmol/L) 9.33 (8.3–10.44) 6.35 (5.89–8.06) 7.32 (6.39–8.18)
BETA-2 score 0 (0–4.9) 19.1 (13.3–22.7) 11.6 (7.55–15.7)
Data are median (IQR).
aAll measures are 1–3 mo after implantation. A single stimulated C-peptide measure was included at 5 mo in the extrahepatic group, since this patient had a second ITx in the gastric submucosa.
IQR, interquartile range; ITx, islet transplantation.

Stimulated C-peptide and secondary outcomes comparing extrahepatic islet cell transplant, intraportal islet cell transplant, and second (intraportal) transplant in patients who initially received extrahepatic implantation. A, Stimulated C-peptide; (B) fasting plasma glucose; (C) BETA-2 score; (D) stimulated C-peptide for individual extrahepatic sites; (E) fasting plasma glucose for individual extrahepatic sites; (F) BETA-2 score for individual extrahepatic sites. EH, extrahepatic transplant; EH + IP, extrahepatic and intraportal transplant; IP, intraportal transplant. Data are presented as medians with error bars representing interquartile range. All measures are 1–3 mo after implantation. A single stimulated C-peptide measure was included at 5 mo after the patients' second transplant.

Assessment of fasting C-peptide levels over time showed that patients receiving extrahepatic transplants had significantly lower levels after their initial implants compared with those with intraportal infusions (mixed effect model, group effect: P < 0.001; Figure 2A). Of the 9 patients undergoing extrahepatic ITx, 7 (77.8%) elected to proceed with subsequent ITx following failed extrahepatic graft. Only 1 patient proceeded with subsequent extrahepatic ITx, had similar early graft failure, and then proceeded with intraportal ITx. Patients receiving gastric submucosal, omental, or prevascularized subcutaneous transplant failed to produce a median fasting C-peptide level ≥0.2nmol/L in the first 60 d when compared with intraportal infusion; however, 3 of 4 subjects in the omental group had measurable C-peptide ≥0.2nmol/L at some point after extrahepatic transplant. Fasting C-peptide levels following intraportal ITx in patients who initially received extrahepatic implants were similar to those patients receiving whose initial transplants were via intraportal infusions (mixed effect model group effect: P = 0.17; Figure 2B). All patients with extrahepatic ITx responded similarly after receiving subsequent intraportal transplant (Figure S2, SDC, https://links.lww.com/TP/C438). One subject receiving gastric submucosal ITx developed de novo donor-specific antibody, but no other patient had any calculated panel reactive antibody increase after extrahepatic ITx.

Fasting C-peptide levels following extrahepatic and intraportal pancreatic islet cell transplantation. A, Fasting C-peptide 10 d medians for the first 60 d after implantation. B, C-peptide after intraportal transplant and after intraportal transplant in patients who initially received extrahepatic implantation (extrahepatic group). *Data are presented as mean (solid lines) and SEM (shaded area).

Primary nonfunction and/or early graft failure occurred significantly more following extrahepatic ITx than following initial intraportal transplant (88.9%, n = 8 of 9 extrahepatic versus 2.0%, n = 5 of 255 for intraportal, P < 0.001, Figure S3, SDC, https://links.lww.com/TP/C438). Extrahepatic ITx was independently associated with graft primary nonfunction (odds ratio 1.709, confidence interval 73.8-39 616.0, P < 0.001). No other patient, transplant, or immunosuppression factors were independently predictive of primary nonfunction (Table 3). Evaluating patients with either early graft failure or primary nonfunction we see that only 14% (n = 1 of 7) with extrahepatic grafts experienced early graft failure after subsequent intraportal ITx; comparatively, in patients receiving initial intraportal ITx who experienced either graft primary nonfunction, 80% (n = 4 of 5) experienced similar early graft loss following second intraportal transplant.

TABLE 3. - Outcomes from multivariable logistic modeling evaluating factors independently associated with graft primary non-function (median C-peptide ≤0.1 nmol/L)
Risk factor Odds ratio 95% confidence interval P
Age 0.97 0.89-1.06 0.549
Male gender 1.23 0.21-7.22 0.816
IEQ per kg 1.00 1.00-1.00 0.292
Tacrolimus level (months 0–3) 0.76 0.48-1.20 0.236
BMI (kg/m2) 1.01 0.98-1.02 0.108
Extrahepatic transplant site 1709 73.80-39 616.00 <0.001
BMI, body mass index; IEQ, islet equivalent.


This study demonstrates that the administration of large numbers of high-quality islets by an extrahepatic route failed to result in significant production of basal or stimulated C-peptide within the first 3 mo posttransplant. Conversely, recipients of primary intraportal ITx demonstrated markedly superior C-peptide production in the first 3 mo, as well as demonstrating sustained graft survival, and improved glycemic-related outcomes compared with extrahepatic ITx. However, individuals who had received extrahepatic ITx were able to achieve similar stimulated and fasting C-peptide levels, and similar glycemic outcomes once they subsequently received intraportal ITx, compared with those receiving initial intraportal ITx. Overall, when compared with intraportal ITx, extrahepatic implantation failed to show islet engraftment or improved diabetes outcomes in patients who subsequently received successful intraportal grafts, suggesting that the extrahepatic site and not patient or graft factors, was the cause of these outcomes. Before future clinical evaluation of these extrahepatic sites, ongoing optimization of these innovative techniques is required.

The study’s primary outcome analysis showed that patients with intraportal ITx had significantly higher stimulated C-peptide over the first 3 mo after implantation compared with those receiving extrahepatic islet grafts. Notably, extrahepatic grafts produced a median stimulated C-peptide <0.1 nmol/L despite receiving a greater islet mass. C-peptide production remains a primary outcome measure that correlates with glycemic control, insulin independence, and resolution of glycemic lability, particularly hypoglycemia.29-32 Differences in secondary outcomes were also evident, with significantly higher FPG after extrahepatic ITx, and worse graft function measured by BETA-2 scores. This is in contrast to some cases reported previously wherein substantial C-peptide production was observed in 2 of 3 patients undergoing omental ITx,11,14 preclinical models demonstrating successful gastric submucosal ITx in large animal models,10 and promising results for the prevascularized subcutaneous approach in mice.15,16 However, although not clinically significant, C-peptide levels appeared higher in the omental when compared with the 2 other extrahepatic sites. Together, our data raise concerns about the feasibility of extrahepatic ITx, and emphasize a need to further optimize oxygenation, neovascularization, and protection from fibrosis or other deleterious processes in extrahepatic sites to achieve clinical outcomes equivalent to intraportal ITx.

Importantly, patients who initially received extrahepatic transplant without success, and who subsequently underwent intraportal ITx achieved similar stimulated C-peptide levels, glycemic outcomes, and graft function to those who underwent initial intraportal transplant. Overall, similarities in our primary outcome and most secondary outcomes support the notion that a failed extrahepatic ITx does not impact the success of subsequent intraportal ITx, and that improvement in techniques for extrahepatic ITx is necessary. However, an unplanned finding of this study demonstrated that patients receiving intraportal ITx who experience early graft failure or graft primary non-function may be at risk of subsequent graft failure and studies evaluating this patient cohort would be of interest. Additionally, although allosensitization was uncommon in patients receiving extrahepatic ITx, this risk remains a potential consideration in this patient population that is at risk of requiring future islet or solid organ transplants. The risk of allosensitization may have been mitigated in this series because maintenance immunosuppression was continued while on the waitlist for a subsequent intraportal islet infusion.

Our findings contrast somewhat with the promising preliminary results with omental ITx from Baidal et al, with outcomes that are similar to subject 2 in their study.14 Our omental transplants did demonstrate sustained but low C-peptide production over time but remained inferior to intraportal infusion in our hands. Our experience is limited to only 4 omental ITxs and variance in technique or islet quality could explain this difference. Alternatively, it is likely that human-to-human variation in vascularity and fatty infiltration in the omentum between subjects, or age- and species-specific differences in omental anatomy or immunologic response could explain the discrepancy between preclinical experience and clinical translation of this approach in murine13 and macaque12 models. Although omental cell composition remains similar between species,33 fat density and vascular distribution are highly variable with animal models often having substantially less adiposity, potentially increasing diffusion capacity;34 the latter remains crucial to islet engraftment and may further contribute to our findings.35 Two recent studies evaluating omental ITx in humans have found similar results to ours, with >50% of patients having early graft failure, and others achieving marginal clinical benefit.36,37 Of note, the Van Hulle et al36 group evaluated graft biopsies and demonstrated that substantial foreign body reaction may have led to their outcomes.

Similarly, our results contrast with promising findings of allotransplantation of porcine islets within the gastric submucosa in immunosuppressed pigs reported by Echeverri et al.10 Importantly, the gross and microscopic gastric anatomy differs between humans and porcine models, with the porcine stomach being 2–3 times larger and having much more cardiac mucosa than humans.38 These cardiac cells produce primarily mucus, whereas the human gastric submucosa contains parietal and chief cells that produce hydrolytic enzymes and acid.38 Again, these small interspecies differences may account for the findings in our study.

Finally, our results also diverge from those shown with ITx in the prevascularized subcutaneous space in mice.15,22,23 However, a recent oral presentation by Dr Witkowski’s group suggests that the prevascularized subcutaneous space (using similar techniques to the one employed in this study) can achieve engraftment and sustained C-peptide production and insulin dose reduction in humans when a lower islet tissue concentration is transplanted in the subcutaneous device within the rectus muscle fascia.39 Our experience here only includes 3 patients and it remains possible that our observed failure with subcutaneous ITx is due to device capacity overload from the high islet masses that we implanted.23 Alternatively, improved vascularization within intramuscular sites may offer potential improvements to current techniques.40 Small iterative modifications may enable success of subcutaneous and other extrahepatic sites, and encourage ongoing refinements to further optimize these techniques.

Of note, for the gastric submucosal, subcutaneous, and omental extrahepatic islet transplants we were more discriminatory in the selection of higher quality, higher purity preparations that should have lent favorably to improved islet survival and engraftment in these sites. Our aim was to reduce the amount of exocrine contamination in grafts placed in sites with more limited physical restraints. This may or may not have been wise in retrospect. We also selected young healthier recipients to optimize the conditions, which again should have lent favorably for extrahepatic sites.

A major limitation to the current study is the very small cohorts included in each of the extrahepatic sites, and the potential variability in our adoption of these new techniques. Although this study presents only 9 subjects receiving extrahepatic grafts, it still represents the largest compilation of extrahepatic ITx to date. Additionally, the lack of substantial C-peptide production in every included patient suggests that even if additional patients were included, differences compared with intraportal infusion would still remain. Similarly, although we only present early outcomes following extrahepatic ITx, with the limited and relatively poor function observed in the extrahepatic sites, it is unlikely that these grafts would spontaneously gain further function beyond the 1–3 mo timeframe evaluated in this study; the fact that these patients were promptly given intraportal islet infusions also precluded a longer term analysis. We cannot, however, completely rule out that late onset graft function would have occurred. Comparison of extrahepatic ITx that occurred since 2012 to intraportal from 1999 to 2018 also introduces the potential of era-related effects. To limit that risk, we performed a secondary analysis comparing era matched cohorts and showed similar results. It should also be noted that this is a single-center experience and subject bias cannot be ruled out. It remains to be determined whether technical aspects have led to our negative results with extrahepatic ITx. Specialists assisted and trained our group with their omental transplant technique to minimize the likelihood that our findings are due to technical variability.11,14 All endoscopic procedures were performed by a single experienced interventional gastroenterologist following previously described techniques that are easily reproducible.10 As previously described, surgeons who were successful with animal model participated in the human procedures to ensure technical consistency.16,22,23 Therefore, although technical differences could explain our findings, substantial efforts were made to reduce that likelihood. As discussed above, an additional variable that could contribute to the differences in our study compared with others is variance in islet cell preparations and transplanted islet tissue concentration. Herein, we report the IEQ/kg and purity of the extrahepatic and intraportal transplants but no data are available to compare our islet preparations with other centers. Mechanistic evaluation of the reasons for extrahepatic graft outcomes shown here is limited and may be beneficial to further improve these techniques. Nevertheless the successful outcomes with intraportal transplantation in both groups are consistent with the high quality islet preparations consistently provided by our islet isolation team. Ongoing optimization of the islet preparation and transplant techniques may enable future success of extrahepatic sites and ongoing work is encouraged.

We present a comparative cohort study evaluating patients receiving extrahepatic and intraportal ITx at a single islet transplant center. Patients who received extrahepatic ITx failed to achieve substantial C-peptide production when compared with intraportal transplantation. However, these patients did subsequently achieve similar graft function with a secondary intraportal ITx, suggesting that inadequate engraftment after extrahepatic transplant rather than graft or patient factors is implicated in graft failure after extrahepatic ITx. As we consider alternative sites for engraftment of islets or stem cell-islets, it is clear from our preliminary experience that more refinements will be needed to substantially improve cellular engraftment and survival if these sites are to match the current efficiency of the intraportal approach. Intraportal ITx, despite all of its limitations, prevails as a current gold-standard as the only implantation site to have consistently demonstrated the capacity to support long-term islet engraftment, glucose-responsive C-peptide production, glycemic outcome improvements, and sustained insulin independence. Although the concept of achieving clinical success with extrahepatic ITx remains attractive, substantial work is required to transform this concept into a reality.


1. Shapiro AMJ, Lakey JRT, Ryan EA, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med. 2000;343:230–238.
2. Verhoeff K, Henschke SJ, Marfil-Garza BA, et al. Inducible pluripotent stem cells as a potential cure for diabetes. Cells. 2021;10:278.
3. Verhoeff K, Marfil-Garza BA, Shapiro AMJ. Update on islet cell transplantation. Curr Opin Organ Transplant. 2021;26:397–404.
4. Markmann JF, Rickels MR, Eggerman TL, et al. Phase 3 trial of human islet-after-kidney transplantation in type 1 diabetes. Am J Transplant. 2021;21:1477–1492.
5. Marfil-Garza BA, Shapiro AMJ, Kin T. Clinical islet transplantation: current progress and new frontiers. J Hepatobiliary Pancreat Sci. 2021;28:243–254.
6. Vantyghem M-C, Chetboun M, Gmyr V, et al. Ten year outcome of islet alone or islet after kidney transplantation in type 1 diabetes: a prospective parallel-arm cohort study. Diabetes Care. 2019;42:2042–2049.
7. Lemos JRN, Baidal DA, Ricordi C, et al. Survival after islet transplantation in subjects with type 1 diabetes: twenty-year follow-up. Diabetes Care. 2021;44:e67–e68.
8. Marfil-Garza BA, Lam A, Bigam D, et al. 116-OR: Comparison of pancreas vs. islet transplantation outcomes from a large single center. Diabetes. 2020;69(Supplement 1):116–OR.
9. Fujita M, McGrath KM, Bottino R, et al. Technique of endoscopic biopsy of islet allografts transplanted into the gastric submucosal space in pigs. Cell Transplant. 2013;22:2335–2344.
10. Echeverri GJ, McGrath K, Bottino R, et al. Endoscopic gastric submucosal transplantation of islets (ENDO-STI): technique and initial results in diabetic pigs. Am J Transplant. 2009;9:2485–2496.
11. Baidal DA, Ricordi C, Berman DM, et al. Bioengineering of an intraabdominal endocrine pancreas. N Engl J Med. 2017;376:1887–1889.
12. Berman DM, O’Neil JJ, Coffey LC, et al. Long-term survival of nonhuman primate islets implanted in an omental pouch on a biodegradable scaffold. Am J Transplant. 2009;9:91–104.
13. Berman DM, Molano RD, Fotino C, et al. Bioengineering the endocrine pancreas: intraomental islet transplantation within a biologic resorbable scaffold. Diabetes. 2016;65:1350–1361.
14. Baidal D, Ricordi C, Berman DM, et al. Long-term function of islet allografts transplanted on the omentum using a biological scaffold. Diabetes. 2018;67(Supplement 1):140–OR.
15. Pepper AR, Pawlick R, Bruni A, et al. Transplantation of human pancreatic endoderm cells reverses diabetes post transplantation in a prevascularized subcutaneous site. Stem Cell Reports. 2017;8:1689–1700.
16. Pepper AR, Gala-Lopez B, Pawlick R, et al. A prevascularized subcutaneous device-less site for islet and cellular transplantation. Nat Biotechnol. 2015;33:518–523.
17. Senior PA, Kin T, Shapiro J, Koh A. Islet transplantation at the University of Alberta: status update and review of progress over the last decade. Can J Diabetes. 2012;36:32–37.
18. Ryan EA, Paty BW, Senior PA, et al. Risks and side effects of islet transplantation. Curr Diab Rep. 2004;4:304–309.
19. Cantarelli E, Piemonti L. Alternative transplantation sites for pancreatic islet grafts. Curr Diab Rep. 2011;11:364–374.
20. Shapiro AM, Pokrywczynska M, Ricordi C. Clinical pancreatic islet transplantation. Nat Rev Endocrinol. 2017;13:268–277.
21. Li X, Meng Q, Zhang L. The fate of allogeneic pancreatic islets following intraportal transplantation: challenges and solutions. J Immunol Res. 2018;2018:2424586.
22. Pepper AR, Pawlick R, Gala-Lopez B, et al. Diabetes is reversed in a murine model by marginal mass syngeneic islet transplantation using a subcutaneous cell pouch device. Transplantation. 2015;99:2294–2300.
23. Gala-Lopez BL, Pepper AR, Dinyari P, et al. Subcutaneous clinical islet transplantation in a prevascularized subcutaneous pouch – preliminary experience. CellR4. 2016;4:e2132.
24. Ryan EA, Shandro T, Green K, et al. Assessment of the severity of hypoglycemia and glycemic lability in type 1 diabetic subjects undergoing islet transplantation. Diabetes. 2004;53:955–962.
25. Shapiro AM, Ricordi C, Hering BJ, et al. International trial of the Edmonton protocol for islet transplantation. N Engl J Med. 2006;355:1318–1330.
26. Forbes S, Oram RA, Smith A, et al. Validation of the BETA-2 score: an improved tool to estimate beta cell function after clinical islet transplantation using a single fasting blood sample. Am J Transplant. 2016;16:2704–2713.
27. Senior PA, Rickels MR, Eggerman T, et al. 360-OR: BETA-2 score is highly correlated with acute insulin response to intravenous glucose: an analysis of the clinical islet transplantation consortium trials. Diabetes. 2020;69(Supplement 1):360–OR.
28. Owen RJ, Ryan EA, O’Kelly K, et al. Percutaneous transhepatic pancreatic islet cell transplantation in type 1 diabetes mellitus: radiologic aspects. Radiology. 2003;229:165–170.
29. Wahren J, Kallas A, Sima AA. The clinical potential of C-peptide replacement in type 1 diabetes. Diabetes. 2012;61:761–772.
30. Wahren J, Ekberg K, Johansson J, et al. Role of C-peptide in human physiology. Am J Physiol Endocrinol Metab. 2000;278:E759–E768.
31. Palmer JP, Fleming GA, Greenbaum CJ, et al. C-peptide is the appropriate outcome measure for type 1 diabetes clinical trials to preserve beta-cell function: report of an ADA workshop, 21-22 October 2001. Diabetes. 2004;53:250–264.
32. Leighton E, Sainsbury CA, Jones GC. A practical review of c-peptide testing in diabetes. Diabetes Ther. 2017;8:475–487.
33. Wilkosz S, Ireland G, Khwaja N, et al. A comparative study of the structure of human and murine greater omentum. Anat Embryol (Berl). 2005;209:251–261.
34. Chaffanjon PC, Kenyon NM, Ricordi C, et al. Omental anatomy of non-human primates. Surg Radiol Anat. 2005;27:287–291.
35. Bochenek MA, Veiseh O, Vegas AJ, et al. Alginate encapsulation as long-term immune protection of allogeneic pancreatic islet cells transplanted into the omental bursa of macaques. Nat Biomed Eng. 2018;2:810–821.
36. Van Hulle F, De Groot K, Hilbrands R, et al. Function and composition of pancreatic islet cell implants in omentum of type 1 diabetes patients. Am J Transplant. 2022;22:927–936.
37. Saudek F, Hladiková Z, Hagerf B, et al. Transplantation of pancreatic islets into the omentum using a biocompatible plasma-thrombin gel: first experience at the institute for clinical and experimental medicine in Prague. Transplant Proc. 2022. 2022;54:806–810.
38. Kararli TT. Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm Drug Dispos. 1995;16:351–380.
39. Bachul P, Generette GS, Perez-Gutierrez A, et al. 307.5: Modified approach allowed for improved islet allotransplantation into pre-vascularized sernova cell pouchtm device - preliminary results of the phase i/ii clinical trial at University of Chicago. Transplantation. 2021;105:S25.
40. Espes D, Eriksson O, Lau J, et al. Striated muscle as implantation site for transplanted pancreatic islets. J Transplant. 2011;2011:352043.

Supplemental Digital Content

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.