INTRODUCTION
Improved metabolic control, hypoglycemia awareness, quality of life, and prevention of severe hypoglycemia can be consistently obtained after islet transplantation (ITx) in patients with type 1 diabetes (T1D), even with partial graft function requiring exogenous insulin to maintain glycated hemoglobin (HbA1c) in the target range.1-6 Although islet allograft survival at 1 y is approximately 80%, graft function decreases progressively with time.7 Allosensitization develops in approximately 40%–70% of ITx recipients following discontinuation of immunosuppression after graft failure.8-11 Allosensitization is defined as the presence of circulating antibodies against HLAs and can be measured by different techniques and expressed as percentage reactivity of the patient’s serum to a panel of lymphocytes with known HLA phenotypes or purified HLA antigens in solid-phase assays % PRA).12 The calculated PRA (cPRA) is obtained using a calculator by the United Network for Organ Sharing (UNOS).13 The cPRA provides a more accurate estimate of allosensitization since it includes both class I and class II HLA specificities. Use of traditional PRA (% of class I and class II HLA) underestimates the degree of allosensitization.14 In fact, there is 50% concordance in the group with PRA values of 1% up to 20%, while there is 90% concordance in the group with PRA values ≥80%.14 Allosensitization after transplantation of solid organs and cellular grafts is a common problem, potentially decreasing the donor pool for a given transplant candidate or prolonging the wait time for subsequent organ transplantations. In the case of pancreatic islets, the need for a relatively large number of beta cells to achieve adequate metabolic control has led to the use of multiple donors per recipient in most cases. This may increase the risk for allosensitization to multiple HLAs and represents a significant concern after graft failure as those patients who become allosensitized and subsequently require another transplant will likely experience increased wait time for identification of a suitable donor limiting the chances of future transplantation.
The duration of persistence of PRA positivity after discontinuation of immunosuppression in ITx recipients still remains unknown. Therefore, the purpose of this study was to evaluate the frequency and persistence of allosensitization postgraft failure and upon discontinuation of immunosuppression in ITx recipients and to identify associated factors, which may predict the development of allosensitization.
MATERIALS AND METHODS
Study Population
Fifty-six subjects received islet transplants at our institution from 2000 to 2017. Nine subjects were lost to follow up and 1 subject died. Of the remaining 46 subjects, 28 subjects had graft failure (22 IA and 6 IA+BM). Of the 22 IA, 13 subjects (IA group) signed consent to enroll in the follow-up postgraft failure protocol. Nine did not agree to participate. Of the 6 subjects receiving ITx+BM group, 4 subjects (IA+BM group) signed the consent for the follow-up after graft failure. Two subjects did not sign consent. The remaining 18 subjects constitute the graft function group. In total, 35 subjects were included in this analysis (Figure 1).
;)
FIGURE 1.: Study population diagram representing our retrospective study design on a cohort of patients who received islet transplantation. IA, islet transplantation alone; IAK, islet after kidney transplantation; ITx, islet transplantation; IA + BM, islet transplantation plus bone marrow-derived CD34+ hematopoietic stem cells.
These 35 subjects with T1D enrolled in ITx protocols at the University of Miami were retrospectively evaluated. Study procedures were reviewed and approved by the Institutional Review Board (IRB) at the University of Miami. All participants provided written informed consent and were enrolled in different protocols, namely: (1) follow-up of ITx recipients postgraft failure (ClinicalTrials.gov Identifier: NCT02000687, NCT01999361; IRB Protocol Numbers: 20080127, 20071058) and (2) follow-up of ITx recipients with persistent islet graft function (ClinicalTrials.gov Identifier: NCT01999374, NCT00306098; IRB Protocol Numbers: 20130034, 20000196). The study population and participant demographics are detailed in Figure 1 and Table 1, respectively.
TABLE 1. -
Demographic characteristics of the participants with failed or functioning islet grafts
| Demographic characteristics |
ITx recipients who developed graft failure (n = 17) |
ITx recipients with current graft function (n = 18 [IA], n = 15; IAK, n = 3) |
Pa |
| IA group (n = 13) |
ITx+BM group (n = 4) |
| Female, n (%) |
7 (54) |
3 (75) |
12 (67) |
0.663 |
| Age at the first transplant (y) |
44 ± 10.5 |
41 ± 1.9 |
44.6 ± 5.9 |
0.697 |
| Duration of T1D at transplant (y) |
17 (14–37) |
37 (25–40) |
38 (29–43) |
0.114 |
| No. islet infusions |
2 (1–3) |
1 |
2 (1–2) |
0.059 |
| Total IEQ/kg body weight |
13 185 (6311–15 590) |
7289 (7042–10 009) |
14 067 (11 291–17 179) |
0.181 |
| Body weight (kg) |
68.6 (60–74) |
59 (55.8–76) |
60 (54–67) |
0.188 |
| Immunosuppression Edmonton protocol |
11 (85) |
4 (100) |
9 (50) |
0.024 |
| Immunosuppression T-cell depletion |
2 (15) |
0 |
9 (50) |
0.024 |
| Duration of graft function (y) |
3.3 (1.9–5.0)b |
1.25 (0.8–1.3)b |
12.35 (7.5–14.8) |
<0.001 |
| Years of graft failure at enrollment |
4 (2–5) |
8 (8–9) |
n/a |
0.001 |
| Years of graft failure at the last follow-up |
9.85 ± 2.6 |
15 ± 0.81 |
n/a |
<0.001 |
| Years to graft failure after the first infusion |
3 (1–5) |
0.5 (0–1) |
n/a |
0.023 |
| Years to graft failure after the last infusion |
1 (1–4) |
0.5 (0–1) |
n/a |
0.079 |
| CMV IgG-positivity, n (%) |
8 (61.5) |
3 (75) |
5 (27.8) |
0.075 |
| DSA, n (%) |
6 (46.2) |
3 (75) |
2 (11.1)d |
0.014 |
| DSA toward how many donors |
2 (1–3) |
1 |
1 |
0.104 |
| DSA toward HLA I, n (%) |
6 (46.2)b,c |
3 (75)b,c |
1 (5.6) |
0.003 |
| DSA toward HLA II, n (%) |
3 (23.1)b,c |
3 (75)b,c |
1 (5.6) |
0.013 |
| HLA I mismatches |
8 (6–10) |
5 (4–6)d |
6 (5–7)d |
0.035 |
| HLA II mismatches |
3 (2–4) |
3 (3–4) |
2 (2–5) |
0.789 |
| Autoantibodies 0/1/2/converter,e n (%) |
n = 5/5/1/1 (42/42/8/8)f |
n = 1/1/1/1 (25/25/25/25) |
n = 6/3/2/3 (43/21/14/21)g |
0.848 |
aP value between groups and are according to χ2 test, ANOVA, independent t-test, U Mann–Whitney, or Kruskal–Wallis test as appropriate.
bSignificant difference compared to ITx recipients with current graft function (P value < 0.05).
cChi-square with significant difference after Bonferroni correction.
dSignificant difference compared with the IA group (P value < 0.05).
eAutoantibodies: 0, no antibodies; 1, 1 antibody developed; 2, 2 antibodies developed; converter, individual who became autoantibody-positive during follow-up.
fAutoantibody assessment was performed only in 11 of 13 subjects.
gAutoantibody assessment was performed only in 14 of 18 subjects.
CMV, cytomegalovirus; DSA, donor-specific anti-HLA antibodies; IA, islet transplantation alone; ITx + BM, ITx plus bone marrow-derived CD34+ hematopoietic stem cells; IAK, islet after kidney transplantation; IEQ, islet equivalents; IgG, immunoglobulin G; ITx, islet transplantation; n/a, not applicable; No., number; T1D, type 1 diabetes.
Maintenance Immunosuppression
In IA and IA+BM subjects, induction immunosuppression consisted of anti-IL2 blockade with either daclizumab or basiliximab and anti-TNF blockade with either infliximab or etanercept (Edmonton-like protocol).2,15-17 Maintenance immunosuppression consisted of a dual combination strategy with either sirolimus, tacrolimus, or mycophenolate mofetil (MMF). MMF was primarily used in the case of intolerance/adverse events to either sirolimus or tacrolimus. Target trough levels for sirolimus were 12–15 ng/mL for 3 mo then 7–10 ng/mL thereafter; tacrolimus trough levels were kept at 3–6 ng/mL (or 8–10 ng/mL if on Tacrolimus-MMF). MMF target dose was 1000 mg PO twice daily with the dose adjusted depending on subject tolerance. In the group of subjects with graft function, 50% underwent induction with T-cell depleting agents (alemtuzumab or antithymocyte globulin).1,18,19
Subjects With Islet Graft Failure
Seventeen ITx recipients who ultimately experienced graft failure were enrolled. Among them, 13 patients received ITx alone (IA group) and 4 received the same donor CD34+ bone marrow (BM)-derived hematopoietic stem cells (HSCs; a minimum of 2 × 106 HSC/kg of recipient body weight) in combination with ITx (ITx+BM group). In the ITx+BM group, BM cells were used to induce hematopoietic chimerism. As per the study design, all subjects were required to stop immunosuppression at 1 y independent of graft function status. Therefore, no subject remained on immunosuppression after 1 y. Apart from 1 subject who lost function 5 mo after islet transplantation, all other subjects maintained graft function at the 1-y time point when immunosuppression was discontinued. After discontinuation of immunosuppression, all grafts failed. Immunosuppression dosing was managed by the senior author of this study to ensure that all patients remained on full immunosuppression for the duration of the study. No subject was on monotherapy at any given time. It is expected that immunosuppression levels would had been undetectable after graft failure as cessation of immunosuppression preceded graft failure.
Subjects in the IA group and ITx+BM group were enrolled into the follow-up protocol 3.6 ± 1.9 and 8 ± 0.7 y after graft failure, respectively (Table 1). Laboratory workup was performed every year following graft failure. Blood samples were collected for determination of alloantibodies and cPRA calculated. For the purpose of data analysis, we established the following time points: (1) pre-ITx; (2) half-life graft survival (1/2 GS), defined as the midpoint between the first islet infusion and occurrence of graft failure; (3) graft failure, defined as the time at which serum levels of stimulated C-peptide (assessed by mixed meal tolerance test) were <0.3 ng/mL15; and (4) yearly follow-up postgraft failure.20
Subjects With Persistent Islet Graft Function
Eighteen subjects with persistent islet graft function at 10.6 ± 4.8 y follow-up were included in the study for comparison analysis. Among them, 15 received IA and 3 received islet after kidney transplantation (IAK). Alloantibody data for these patients were evaluated pre-ITx and at 1, 5, 7, and 10 y posttransplant.
Alloantibody Determination
Subjects belonging to IA and IA+BM groups were tested for alloantibodies using single-antigen bead solid-phase assays for HLA class I and II (LabScreen Single Antigen Class I and LabScreen Single Antigen Class II from One Lambda [Thermofisher]). Values of cPRA were calculated using the cPRA calculator by UNOS.13,14 A mean fluorescence intensity exceeding 3000 was reported as positive and subsequently compared with donor HLA to determine donor-specific anti-HLA antibodies (DSA)-positivity in the recipient’s serum. cPRA was used as a continuous measure of allosensitization.13,14,21 HLA-DP was not included in either the cPRA calculations and DSA analyses. In the subjects with graft function, alloantibodies were evaluated pre-ITx and at 1, 5, 7, and 10 y posttransplant. HLA mismatches were evaluated between donor and recipient using HLA-A, -B, -DR, and DQ loci.
Autoantibody Determination
Autoantibodies to 65-kDa isoform of glutamic acid decarboxylase (GAD65) and islet antigen-2 (IA-2) were measured using radioimmunoassays validated in the proficiency workshops of the Immunology of Diabetes Society and Centers for Disease Control and Prevention.22,23 Autoantibody levels are expressed as the ratio of the autoantibody index levels of the patient over the cutoff index of each assay. A ratio of ≥1 indicates a positive result.
Statistical Analysis
Results are expressed as mean ± SD, median (25th percentile; 75th percentile) or percentages. Descriptive characteristics were analyzed between groups through 1-way ANOVA, unpaired 2-tailed T-test with Welch correction, Kruskal–Wallis 1-way, U Mann–Whitney, or chi-square test, as appropriate. Post hoc analyses were performed with Tukey and Bonferroni correction. To determine differences of cPRA within IA graft failure group, we chose to compare the following time points: (1) pre-ITx; (2) graft failure; and (3) 6-y follow-up. For this comparison, we used related-samples Friedman’s with Dunn-Bonferroni post hoc tests. Statistical analysis was not performed for evaluating time points in the ITx+BM group due to the small sample size. Data were analyzed using IBM SPSS v22.0 (NY). A P value <0.05 was considered to indicate statistical significance.
RESULTS
Recipients With Islet Graft Failure
IA Group (n = 13)
Among the 13 subjects enrolled, 9 completed >6 y of follow-up (range: 6–15 y) and 4 were lost to follow-up. Patient demographics and baseline characteristics are detailed in Table 1. Subjects were transplanted between 2001 and 2006. Immunosuppression induction included daclizumab, along with an anti-TNF-alpha agent (infliximab, n = 2; etanercept, n = 11). Table 2 shows the demographic characteristics of DSA-positive and DSA-negative recipients in IA group subjects with graft failure. There was no significant association between the number of HLA-I and HLA-II mismatches, development of DSA, and number of donors (Table 3). Regarding islet autoantibodies, 42% of subjects had 1 islet autoantibody (GAD65 or IA-2), 42% were negative for autoantibodies, 8% had both autoantibodies, and the remaining 8% became positive during follow-up (“converter” subjects) (Table 1). Graft failure in these subjects occurred at 3 (1–5) y after ITx and 1 (1–4) y after last ITx. Non-DSA were also detected in 10/13 (77%) subjects (Table S1, SDC, https://links.lww.com/TP/C109).
TABLE 2. -
Demographic characteristics of DSA-positive and DSA-negative recipients in IA group subjects with graft failure (n = 13)
|
DSA-positive recipients (n = 6) |
DSA-negative recipients (n = 7) |
P
|
| Female |
3 (50) |
4 (57.1) |
0.999 |
| Recipient age at the first transplant (y) |
43.7 ± 6.9 |
45.8 ± 13.8 |
0.733 |
| Duration of T1D (y) |
26.5 ± 16.5 |
22.5 ± 13.6 |
0.639 |
| No. donors |
3 ± 1 |
2 ± 1 |
0.629 |
| HLA I mismatchesa |
8 (7–9) |
8 (5–11) |
0.731 |
| HLA II mismatchesa |
4 (2–5) |
2 (1–4) |
0.295 |
| Total IEQ/kg body weight |
13 382 (12 956–29 593) |
8681 (5261–14 144) |
0.534 |
| Body weight (kg) |
67.4 ± 12.8 |
68.4 ± 10.8 |
0.877 |
| Immunosuppression Edmonton protocol,b n (%) |
6 (100) |
5 (71.4) |
0.462 |
| Immunosuppression T-cell depletion,c n (%) |
0 |
2 (28.6) |
0.462 |
| Myfortic protocol,d n (%) |
2 (33.3) |
5 (71.4) |
0.286 |
| On exenatide |
3 (50) |
4 (57.1) |
0.999 |
| Duration of graft function (y) |
3.3 ± 1.5 |
3.1 ± 2.1 |
0.851 |
| Total pregnancies |
0 (0–3; n = 3) |
0 (0–3; n = 4) |
NA |
| CMV IgG-positivity |
4 (66.7) |
4 (57.1) |
0.999 |
| Autoantibody positivitye |
4 (66.7) |
3 (60.0) |
0.999 |
a The maximum number of mismatches per donor = 4.
bThe Edmonton protocol for islet transplantation consists of a glucocorticoid-free immunosuppressive regimen (sirolimus, tacrolimus, and daclizumab).
c Alemtuzumab or thymoglobulin.
d Myfortic = mycophenolic acid.
e Autoantibody assessment was performed in 12 of 13 subjects.
CMV, cytomegalovirus; DSA, donor-specific anti-HLA antibodies; IA, islet transplantation alone; IEQ, islet equivalents; IgG, immunoglobulin G; No., number; T1D, type 1 diabetes.
TABLE 3. -
Values of cPRA in IA group subjects (n = 13) before transplantation, during transplant follow-up and after graft failure
| Subject |
cPRA (%) Pre-ITx |
cPRA (%) 1/2 GS |
cPRA (%) GF |
cPRA (%) 1 y post-GF |
cPRA (%) 2 y post-GF |
cPRA (%) 3 y post-GF |
cPRA (%) 4 y post-GF |
cPRA (%) 5 y post-GF |
cPRA (%) 6 y post-GF |
| #1 |
0 |
0 |
89 |
89 |
n/a |
n/a |
n/a |
n/a |
90 |
| #2 |
0 |
0 |
0 |
n/a |
n/a |
n/a |
n/a |
n/a |
n/a |
| #3 |
22 |
22 |
24 |
n/a |
n/a |
n/a |
n/a |
75 |
84 |
| #4 |
0 |
0 |
99 |
99 |
n/a |
n/a |
92 |
96 |
23 |
| #5 |
0 |
0 |
0 |
n/a |
n/a |
0 |
0 |
0 |
0 |
| #6 |
0 |
0 |
0 |
n/a |
n/a |
n/a |
n/a |
99 |
98 |
| #7 |
0 |
0 |
96 |
n/a |
n/a |
n/a |
n/a |
n/a |
100 |
| #8 |
0 |
0 |
0 |
n/a |
n/a |
n/a |
0 |
0 |
0 |
| #9 |
0 |
0 |
0 |
0 |
0 |
0 |
29 |
24 |
3 |
| #10 |
0 |
0 |
0 |
24 |
11 |
4 |
4 |
11 |
n/a |
| #11 |
0 |
0 |
22 |
n/a |
n/a |
n/a |
n/a |
n/a |
22 |
| #12 |
0 |
0 |
0 |
n/a |
n/a |
n/a |
n/a |
n/a |
n/a |
| #13 |
n/a |
0 |
0 |
n/a |
n/a |
70 |
24 |
24 |
0 |
|
Subject
|
cPRA (%) 7 y post-GF
|
cPRA (%) 8 y post-GF
|
cPRA (%) 9 y post-GF
|
cPRA (%) 10 y post-GF
|
cPRA (%) 11 y post-GF
|
cPRA (%) 12 y post-GF
|
cPRA (%) 13 y post-GF
|
cPRA (%) 14 y post-GF
|
cPRA (%) 15 y post-GF
|
| #1 |
78 |
27 |
n/a |
79 |
82 |
83 |
82 |
82 |
85 |
| #2 |
63 |
59 |
8 |
0a |
56 |
56 |
56 |
56 |
83 |
| #3 |
0 |
0 |
0 |
0 |
0 |
0 |
Enrolledb |
Enrolledb |
Enrolledb |
| #4 |
23 |
23 |
23 |
64 |
64 |
0 |
0 |
0 |
Enrolledb |
| #5 |
0 |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
| #6 |
81 |
81 |
82 |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LFTUP |
| #7 |
0a |
86 |
93 |
95 |
89 |
89 |
91 |
79 |
89 |
| #8 |
Enrolledb |
Enrolledb |
Enrolledb |
Enrolledb |
Enrolledb |
Enrolledb |
Enrolledb |
Enrolledb |
Enrolledb |
| #9 |
8 |
0 |
0 |
Enrolledb |
Enrolledb |
Enrolledb |
Enrolledb |
Enrolledb |
Enrolledb |
| #10 |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
| #11 |
30 |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
LTFUP |
| #12 |
0 |
0 |
0 |
0 |
n/a |
0 |
0 |
0 |
Enrolledb |
| #13 |
0 |
0 |
0 |
0 |
0 |
0 |
Enrolledb |
Enrolledb |
Enrolledb |
aThe presence of antigen-specific antibodies and values of mean fluorescence intensity below 3000, resulting in cPRA “0.”
bnrolled: patients have not reached this evaluation point (still enrolled in the protocol).
½ GS, half-life graft survival; cPRA, calculated panel-reactive antibody; GF, graft failure; IA, islet transplantation alone; ITx, islet transplant; LTFUP, lost to follow-up; n/a, data not available.
The values of cPRA in IA group subjects (n = 13) before transplantation, during transplant follow-up and after graft failure are listed in Table 3 and are depicted in Figure 2. Before ITx and at the 1/2 GS time point only 1 recipient was cPRA-positive (Subject #3; cPRA: 22%). At graft failure, 5 subjects (subjects #1, #3, #4, #7, #11; cPRA: 89%, 24%, 99%, 96%, and 22%, respectively). In this group, immunosuppression was tapered and discontinued at 73.5 ± 69.4 d after graft failure. At 6 y postgraft failure, 55% of subjects (6/11 patients) had a cPRA>20% (cPRA: 22%–100%). Collectively, 10 of 13 subjects (77%) became allosensitized at any time point after graft failure, among which 6 of 13 subjects (46.2%) developed DSA positivity (DSA to 2 [1–3] donors). The degree of sensitization was variable (cPRA ranging between 22% and 100% during the follow-up after graft failure). Allosensitization persisted for up to 15 y in subject #2 (cPRA: 83%; Table 3).
FIGURE 2.: Values of cPRA (%) in IA group subjects (n = 13) before transplantation, during transplant follow-up and after graft failure. Box plot shows cPRA distribution mean, interquartile range, and minimum and maximum values. The lower bar shows the number of patients evaluated at each time point during the follow-up. 1/2 GS, half-life graft survival; cPRA, calculated panel-reactive antibody; GF, graft failure; IA, islet transplantation alone; ITx, islet transplantation.
Three subjects (#5, #8, and #13) were enrolled in a protocol to prevent allosensitization after discontinuation of immunosuppression following graft failure (ClinicalTrials.gov Identifier: NCT01999361; IRB Protocol Number: 20071058), which consisted of MMF monotherapy for 2 y. Two of these subjects (#5 and #8) were not allosensitized (cPRA-negative) after graft failure. The third subject (#13) was cPRA-positive 3 y postgraft failure (cPRA: 70%; DSA-negative), became cPRA-negative at 6 y postgraft failure (cPRA: 0%) and has remained negative since then (last follow-up 11 y postgraft failure) (Table 3). One subject (#12) required IVIG before graft failure for the treatment of parvovirus infection, at which time immunosuppression had to be discontinued. This subject had an initial cPRA negative before transplantation and subsequent cPRA measurements remained negative for the following 14 y (Table 3).
We compared the percentage values of cPRA before ITx, at the time of graft failure, and at 6 y postgraft failure to detect changes according to the time points. We found a positive direction change among groups (P = 0.01), indicating significantly higher levels of cPRA at 6 y postgraft failure as compared with those observed before ITx (Dunn–Bonferroni adjustment, P = 0.042) (Figure 3A). Due to our small sample size, we were unable to evaluate the association between cPRA values and numerous variables of interest, including age, transplanted IEQ/kg of recipient body weight, cytomegalovirus serologic status, immunosuppressive regimen, islet autoantibody-positivity, and T1D duration.
FIGURE 3.: Values of cPRA (%) in IA group subjects (n = 13) and ITX + BM subjects (n = 4). A, The change in cPRA (%) before transplantation, ½ time between transplant and graft failure, at graft failure, and 6 y after graft failure in the IA group (n = 13); *Dunn–Bonferroni Post hoc analysis between Pre-ITX and 6 y post GF, P = 0.042. B, The change in cPRA (%) before transplantation, at the time of graft failure, and 15 y after graft failure in the ITx + BM group (n = 4); α cPRA was performed in 2 of 4 patients before transplant. cPRA, calculated panel-reactive antibody; GF, graft failure; IA, islet transplantation alone; ITx + BM, islet transplantation plus bone marrow-derived CD34 + hematopoietic stem cells.
ITx+BM Group (n = 4)
As per transplant protocol design, immunosuppression was discontinued in 3 patients 1 y after ITx.16 The remaining subject discontinued immunosuppression at the time of graft failure (178 d post-ITx). Subjects have been followed for up to 17 y postgraft failure. Subjects in this group were enrolled in the study protocol 8 (8–9) y after graft failure. Patient demographics and baseline characteristics are detailed in Table 1. Induction immunosuppression consisted of anti-IL-2 (daclizumab) and anti-TNF-alpha (infliximab) agents, while maintenance immunosuppression included sirolimus and tacrolimus. The remaining subject became positive (for IA-2 only) during follow-up. Median duration of graft function in this group was 1.25 (0.8–1.3) y (Table 1). Peripheral chimerism was maximal at 1 mo (5.92% ± 0.48%), highly reduced at 1 y (0.20% ± 0.08%) and was undetectable at graft failure.16
The values of cPRA (%) in ITx+BM group subjects before transplantation, during transplant follow-up and after graft failure are listed in Table S2 (SDC, https://links.lww.com/TP/C109). Before transplantation, subjects #1 and #2 were cPRA-negative, and remained negative while on immunosuppression. For subjects #3 and #4 the PRA tested by single antigen beds were not available for the pretransplant time. At the time of graft failure, subjects #1, #2, and #4 remained cPRA-negative, subject #3 was cPRA-positive (PRA: 97%, DSA-negative). After discontinuation of immunosuppression (as per protocol), subjects #1, #2, and #3 became DSA-positive and showed cPRA-positivity during the follow-up (Table S1, SDC, https://links.lww.com/TP/C109). Subject #3 presented cPRA 97% after transplant and remained cPRA positive for 16 y (DSA-negative). Subject #4 was negative at the time of graft failure and did not undergo further testing until 9 y postgraft failure, where results showed cPRA-positivity (cPRA: 51%) and the lack of DSA development. Repeat testing of subject #4 yearly thereafter was consistently negative until 17 y postgraft failure (cPRA: 41%; DSA-negative). Overall, allosensitization persisted for 15 y postgraft failure in 3 of 4 subjects (cPRA in subjects #1, #2, and #3: 98%, 88%, and 97%, respectively; see Table S1, SDC, https://links.lww.com/TP/C109). Percentage of cPRA before ITx, at the time of graft failure, and at 15 y postgraft failure is displayed in Figure 3B.
Recipients With Persistent Islet Graft Function
We included 18 subjects with persistent graft function for comparison in this analysis. Patient demographics and baseline characteristics are detailed in Table 1. ITx occurred between 2000 and 2017 and the majority were transplanted before 2011. Induction immunosuppression included daclizumab (50%) or T-cell depleting agents (50%) in combination with an anti-TNF-alpha agent (infliximab, n = 1; etanercept, n = 17). Sirolimus and tacrolimus were used for maintenance immunosuppression. Four of 18 subjects (#4 [IAK], #5 [IAK], #8[IA], and #10 [IA]) became allosensitized (cPRA: 71%, 23%, 54%, and 46%, respectively) at 10 y post-ITx while on immunosuppression, and 2 of them (#8 [IA] and #10 [IA]) developed DSA positivity.
DISCUSSION
After graft failure, immunosuppression is often discontinued in subjects receiving IA because there is no longer need for immunosuppressive therapy. PRA positivity subsequently develops in almost all subjects.8-11 In our study, allosensitization has persisted in the majority of subjects after discontinuation of immunosuppression for up to 15 y of follow-up. Three of 4 subjects who received ITx+BM developed allosensitization and DSA positivity after discontinuation of immunosuppression, with PRA positivity persisting for 17 y after graft failure in this group. The persistently high cPRA levels in these patients are most likely secondary to the BM cellular load. We hypothesize that DSA positivity was high likely due to the infusion of CD34-enriched cells, which contain a mixture of HSCs, hematopoietic progenitors, and various contaminants.16
At time of graft failure, there was minimal allosensitization. At 6 y postgraft failure, 55% of subjects (6/11; data not available in 2 subjects) in the IA group had cPRA > 20% (cPRA: 22%–100%), and 77% (10/13) became allosensitized at any time after graft failure. DSA positivity to multiple donors was observed in (6/13) 46% of subjects. Although the degree of sensitization was variable (cPRA ranging between 22% and 100%), it has persisted for up to 15 y of follow-up after discontinuation of immunosuppression. Among the 3 subjects in the IA group who experienced graft failure (18%) and did not develop allosensitization, 2 were treated with MMF for 2 y after discontinuation of protocol-specific immunosuppression in an attempt to specifically prevent the occurrence of allosensitization. Another subject received IVIG before graft failure due to parvovirus infection. This treatment is usually employed as part of desensitization protocols24 and may account for the prevention of allosensitization in this subject.
Although guidelines to prevent or decrease PRA positivity after graft failure in ITx have not yet been established, preliminary data from an ongoing study (ClinicalTrials.gov Identifier: NCT01999361) suggest that administration of MMF for 2 y after graft failure may be a valid option to prevent, decrease or reverse PRA positivity. Similarly, Campbell et al9 recommend avoiding discontinuation of immunosuppression until patients are evaluated for the need for subsequent transplant; during this evaluation process, subjects are maintained on mycophenolic acid therapy for a 2-y period. A single-center study conducted on kidney transplantation recipients reported that prolonging immunosuppression for >3 mo after graft failure resulted in decreased allosensitization and lower mean PRA levels, as compared with early discontinuation of immunosuppression.25 Moreover, gradual tapering of immunosuppression after kidney graft failure has been shown to decrease allosensitization compared with abrupt cessation of the immunosuppressive regimen.25
ITx has evolved over time and most current protocols use T-cell depleting agents (alemtuzumab and antithymocyte globulin) for induction. In our study, 85% of subjects in the graft failure group (transplanted between 2000 and 2006) (11/13) and 50% in the islet graft function group (transplanted between 2001 and 2017) did not receive T-cell depleting agents for induction. Therefore, our findings in the graft failure group are only applicable to subjects who received Edmonton-like immunosuppression. Brooks et al26 evaluated whether alemtuzumab induction affected DSA positivity (de novo) at 12 mo post-ITx and found that in subjects who did not develop DSA, 95% received alemtuzumab, whereas in subjects who developed DSA, 60% received alemtuzumab. Furthermore, the Edmonton group showed that recipients who received alemtuzumab did not developed allosensitization up to 36 mo post-ITx.26 Of note, 2 of 3 subjects (66%) in our study cohort who remained cPRA negative following cessation of immunosuppression underwent induction with T-cell depleting agents; however, 1 of them was also kept on MMF for 2 y postgraft failure to prevent allosensitization. Thus, it is unclear whether the prevention of allosensitization on this subject was attributed to T-cell depleting induction, MMF therapy postgraft failure or a combination of both. In the ITX+BM group 3 of 4 subjects islet graft failure developed months after discontinuation of immunosuppression (as per protocol); therefore, graft failure is the result of immunosuppression discontinuation.
The number of ITx recipients with persistent graft function who became allosensitized while on immunosuppression in our study is comparable to data reported for kidney transplant recipients with stable serum creatinine during maintenance immunosuppression.27 The development of DSA positivity is generally associated with rapid loss of islet allograft function, although results are controversial across studies.11,26,28 However, the development of DSA positivity within 4 wks after ITx has been associated with decreased islet allograft function at 3 mo and can predict graft failure at 12 mo.26 It still remains controversial whether antibodies detected exclusively by solid-phase techniques influence long-term graft outcomes. Unlike kidney transplantation, the clinical impact of anti-HLA antibodies on ITx remains unclear. Recently, Chen et al29 showed that allogeneic islets are resistant to DSA-mediated rejection, with a similar rate of islet graft attrition between patients with or without DSA (n = 49, patients who were successfully grafted with allogeneic islet).
The reason for the persistence of alloantibodies after discontinuation of immunosuppression following allograft failure in subjects receiving ITx is unclear. The mechanism for the variable alloantibody response and persistence in certain subjects is also undetermined. No islet alloantigens should theoretically persist after allograft rejection and discontinuation of immunosuppression, and thus, the stimuli for persistent alloantibody production are still unclear. However, factors accounting for persistent alloantibody production may include the persistence of cells of donor origin (microchimerism) or the persistence of alloantigens in dendritic cells30 and the presence of long-lived plasma cells generated during the alloimmune response.31-33
Allosensitization to allografts utilized for repair of heart defects has also been described.34-36 In a study of adults receiving homograft aortic graft implantation, anti-HLA antibodies were shown to persist for up to 15 y.34 Pediatric patients receiving implantation of cryopreserved valved and nonvalved allografts showed persistence of anti-HLA antibodies for up to 8 y after implantation, although antibody levels decreased with time.35 Similarly, another study of pediatric surgical patients observed long-term sensitization for up to 18 y.36 Our study is the first to highlight the persistence of allosensitization for such duration specifically in ITx recipients.
Allosensitization after ITx is of concern given that T1D patients may require pancreas, kidney, or other organ transplantation in the future. Therefore, a high cPRA will increase wait time to identify a potential HLA compatible donor. Sensitization may impact outcomes although careful matching based upon detailed HLA antibody analysis and current immunosuppression strategies may reduce this risk.
Several studies suggest that pancreas after islet transplantation can be a successful strategy to achieve long-term insulin independence in previously failed islet allograft recipients.12,37 However, allosensitization remains of concern given the prolonged wait time for subsequent transplantation.37,38
The main limitations of our study include the small sample size, retrospective design, inclusion of participants from different study protocols, and patients lost to follow up. In addition, we may have underestimated allosensitization with flow cytometry since the cutoff of mean fluorescence intensity was set at 3000.
The high frequency of allosensitization development and persistence after islet allograft failure is a significant finding, which argues in favor of minimizing the number of islet infusions (donors) per each recipient. Naziruddin et al11 showed that the presence of low levels of PRA at the first islet infusion significantly increased the development of PRA >20% after transplantation. Mohanakumar et al28 showed that preexisting HLA sensitization may be one of the factors that lead to loss of transplanted islets. Accurate alloantibody characterization using the same methodology, single-antigen solid-phase HLA antibody assays is mandatory before transplantation. In many recent trials of islet allograft transplantation, the presence of any level of PRA (using single-antigen solid-phase assays) above background is listed as exclusion criteria for the first infusion. For subsequent infusions, alloantibodies are detected before those unacceptable antigens are avoided. T-cell depleting agents are now more commonly used in islet allograft transplantation and there is evidence5,39 that DSA formation is much less commonly seen. Therefore, our results need to be interpreted in this context and may not be applicable to the people that have been more recently transplanted on T-cell depleting agents. In the case of sensitization, it is possible that combining solid-phase HLA antibody identification methodologies with virtual crossmatching will increase the allocation of deceased donor organs to highly sensitized patients for islet transplantation similar to kidney transplantation.40,41 This approach needs to be evaluated in islet allograft transplant trials.
Monitoring for allosensitization should also be considered in study protocols using allogeneic cell products (eg, stem cells). Last, strategies to prevent allosensitization before discontinuation of immunosuppression after graft failure should be included as part of transplant protocols.
In conclusion, allosensitization is a highly prevalent and persistent feature following islet allograft failure with potential negative implications regarding the need for future organ transplantation (ie, increased wait list times). Treatment with MMF for up to 2 y to prevent allosensitization after islet graft failure may be an effective therapeutic option, although further large-scale studies are required to confirm our findings.
ACKNOWLEDGMENTS
The authors are grateful to the members of the cGMP Human Cell Processing Facility, the preclinical Human Immunology and Immunogenetics Program, the Clinical Cell Transplant Program at the Diabetes Research Institute, and the University of Miami CTSI for their support of this work.
REFERENCES
1. Hering BJ, Clarke WR, Bridges ND, et al.; Clinical Islet Transplantation Consortium. Phase 3 trial of transplantation of human islets in type 1 diabetes complicated by severe hypoglycemia. Diabetes Care. 2016;39:1230–1240.
2. Cure P, Pileggi A, Froud T, et al. Improved metabolic control and quality of life in seven patients with type 1 diabetes following islet after kidney transplantation. Transplantation. 2008;85:801–812.
3. Foster ED, Bridges ND, Feurer ID, et al.; Clinical Islet Transplantation Consortium. Improved health-related quality of life in a phase 3 islet transplantation trial in type 1 diabetes complicated by severe hypoglycemia. Diabetes Care. 2018;41:1001–1008.
4. Vantyghem MC, Kerr-Conte J, Arnalsteen L, et al. Primary graft function, metabolic control, and graft survival after islet transplantation. Diabetes Care. 2009;32:1473–1478.
5. Shapiro AM, Pokrywczynska M, Ricordi C. Clinical pancreatic islet transplantation. Nat Rev Endocrinol. 2017;13:268–277.
6. Ryan EA, Paty BW, Senior PA, et al. Five-year follow-up after clinical islet transplantation. Diabetes. 2005;54:2060–2069.
7. Collaborative Islet Transplant Registry (CITR): 2016 (Ninth) Annual Report. Collaborative Islet Transplant Registry Website. Available at
https://citregistry.org/content/citr-9th-annual-report. Accessed January 10, 2020.
8. Cardani R, Pileggi A, Ricordi C, et al. Allosensitization of islet allograft recipients. Transplantation. 2007;84:1413–1427.
9. Campbell PM, Senior PA, Salam A, et al. High risk of sensitization after failed islet transplantation. Am J Transplant. 2007;7:2311–2317.
10. Pouliquen E, Baltzinger P, Lemle A, et al.; GRAGIL Network. Anti-donor HLA antibody response after pancreatic islet grafting: characteristics, risk factors, and impact on graft function. Am J Transplant. 2017;17:462–473.
11. Naziruddin B, Wease S, Stablein D, et al. HLA class I sensitization in islet transplant recipients: report from the collaborative islet transplant registry. Cell Transplant. 2012;21:901–908.
12. Wisel SA, Gardner JM, Roll GR, et al. Pancreas-after-islet transplantation in nonuremic type 1 diabetes: a strategy for restoring durable insulin independence. Am J Transplant. 2017;17:2444–2450.
13. cPRA Calculator. U.S. Department of Health & Human Services. 2019.
https://optn.transplant.hrsa.gov/resources/allocation-calculators/cpra-calculator/. Acessed October 12, 2020.
14. Cecka JM. Calculated PRA (CPRA): the new measure of sensitization for transplant candidates. Am J Transplant. 2010;10:26–29.
15. Froud T, Ricordi C, Baidal DA, et al. Islet transplantation in type 1 diabetes mellitus using cultured islets and steroid-free immunosuppression: Miami experience. Am J Transplant. 2005;5:2037–2046.
16. Mineo D, Ricordi C, Xu X, et al. Combined islet and hematopoietic stem cell allotransplantation: a clinical pilot trial to induce chimerism and graft tolerance. Am J Transplant. 2008;8:1262–1274.
17. Shapiro AM, Lakey JR, 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.
18. Baidal DA, Ricordi C, Berman DM, et al. Bioengineering of an intraabdominal endocrine pancreas. N Engl J Med. 2017;376:1887–1889.
19. Froud T, Baidal DA, Faradji R, et al. Islet transplantation with alemtuzumab induction and calcineurin-free maintenance immunosuppression results in improved short- and long-term outcomes. Transplantation. 2008;86:1695–1701.
20. Peixoto E, Vendrame F, Arnau A, et al. Ten years of preserved kidney function after islet transplant graft failure. Diabetes Care. 2016;39:e209–e211.
21. Zachary AA, Braun WE. Calculation of a predictive value for transplantation. Transplantation. 1985;39:316–318.
22. Bingley PJ, Bonifacio E, Mueller PW. Diabetes antibody standardization program: first assay proficiency evaluation. Diabetes. 2003;52:1128–1136.
23. Vendrame F, Pileggi A, Laughlin E, et al. Recurrence of type 1 diabetes after simultaneous pancreas-kidney transplantation, despite immunosuppression, is associated with autoantibodies and pathogenic autoreactive CD4 T-cells. Diabetes. 2010;59:947–957.
24. Marfo K, Lu A, Ling M, et al. Desensitization protocols and their outcome. Clin J Am Soc Nephrol. 2011;6:922–936.
25. Casey MJ, Wen X, Kayler LK, et al. Prolonged immunosuppression preserves nonsensitization status after kidney transplant failure. Transplantation. 2014;98:306–311.
26. Brooks AM, Carter V, Liew A, et al. De novo donor-specific HLA antibodies are associated with rapid loss of graft function following islet transplantation in type 1 diabetes. Am J Transplant. 2015;15:3239–3246.
27. Banasik M, Boratyńska M, Kościelska-Kasprzak K, et al. The impact of de novo donor-specific anti-human leukocyte antigen antibodies on 5-year renal transplant outcome. Transplant Proc. 2013;45:1449–1452.
28. Mohanakumar T, Narayanan K, Desai N, et al. A significant role for histocompatibility in human islet transplantation. Transplantation. 2006;82:180–187.
29. Chen CC, Pouliquen E, Broisat A, et al. Endothelial chimerism and vascular sequestration protect pancreatic islet grafts from antibody-mediated rejection. J Clin Invest. 2018;128:219–232.
30. Thomson AW, Lu L, Murase N, et al. Microchimerism, dendritic cell progenitors and transplantation tolerance. Stem Cells. 1995;13:622–639.
31. Ionescu L, Urschel S. Memory B cells and long-lived plasma cells. Transplantation. 2019;103:890–898.
32. Woodle ES, Tremblay S, Rossi A, et al. Plasma cell targeting to prevent antibody-mediated rejection. Am J Transplant. 2020;20(Suppl 4):33–41.
33. Zhang Z, Schuster SJ, Lacey SF, et al. Stable HLA antibodies following sustained CD19+ cell depletion implicate a long-lived plasma cell source. Blood Adv. 2020;4:4292–4295.
34. Smith JD, Ogino H, Hunt D, et al. Humoral immune response to human aortic valve homografts. Ann Thorac Surg. 1995;60(2 Suppl):S127–S130.
35. Hooper DK, Hawkins JA, Fuller TC, et al. Panel-reactive antibodies late after allograft implantation in children. Ann Thorac Surg. 2005;79:641–644. Discussion 645.
36. O’Connor MJ, Lind C, Tang X, et al. Persistence of anti-human leukocyte antibodies in congenital heart disease late after surgery using allografts and whole blood. J Heart Lung Transplant. 2013;32:390–397.
37. Andres A, Livingstone S, Kin T, et al. Islet-after-failed-pancreas and pancreas-after-failed islet transplantation: Two complementary rescue strategies to control diabetes. Islets. 2015;7:e1126036.
38. Gruessner AC, Sutherland DE, Gruessner RW. Matching in pancreas transplantation–a registry analysis. Transplant Proc. 2001;33:1665–1666.
39. Shapiro AM, Toso C, Koh A, et al. Alemtuzumab + Tac/MMF substantially improves long-term insulin-independence, and strongly suppresses auto and alloreactivity after clinical islet transplantation. Transplantation. 2010;90:134.
40. Bray RA, Gebel HM. Strategies for human leukocyte antigen antibody detection. Curr Opin Organ Transplant. 2009;14:392–397.
41. Schinstock CA, Gandhi MJ, Stegall MD. Interpreting Anti-HLA antibody testing data: a practical guide for physicians. Transplantation. 2016;100:1619–1628.
