SEARCH

SEARCH BY CITATION

Keywords:

  • Autoantibody;
  • autoimmunity;
  • recurrence;
  • simultaneous pancreas–kidney transplantation;
  • type 1 diabetes mellitus

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Discussion
  6. Disclosure
  7. References

We report herein the patterns of type 1 diabetes recurrence in a simultaneous pancreas–kidney transplant (SPK) recipient, in the absence of rejection. A 38-year-old female underwent SPK for end-stage nephropathy secondary to type 1 diabetes. Fasting blood glucose, HbA1c, fructosamine, C-peptide and autoantibodies (GAD-65, IA-2) were monitored throughout follow-up. At 3.5 years post-SPK, HbA1c and fructosamine increased sharply, indicating loss of perfect metabolic control, despite C-peptide levels in the normal-high range. Exogenous insulin was restarted 4 months later. C-peptide levels abruptly fell and became undetectable at 5.5 years. Autoantibody levels, which were undetectable at the time of SPK, never converted to positivity. Pancreas retranspantation was performed at 6 years. The failed pancreas graft had a normal macroscopic appearance. On histology, there were no signs of cellular or humoral rejection in the kidney or pancreas. A selective peri-islet lymphocytic infiltrate was observed, together with near-total destruction of β cells. At 2.5 years post retransplantation, pancreatic graft function is perfect. This observation indicates unequivocally that pancreas graft can be lost to recurrence of type 1 diabetes in the absence of rejection. GAD-65 and IA-2 autoantibodies are not reliable markers of autoimmunity recurrence.


Abbreviations: 
IHC, immunohistochemistry; ULN, upper limit of normal.

 

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Discussion
  6. Disclosure
  7. References

Simultaneous pancreas–kidney transplantation (SPK) has become the gold standard therapy for patients with type 1 diabetes and end-stage renal failure (1). The autoimmune nature of type 1 diabetes comprises both cellular and humoral components, represented by autoreactive CD4+ and CD8+ T lymphocytes and a variety of autoantibodies (2). The histologic hallmark of type 1 diabetes is a selective destruction of β cells by a lymphocytic cellular infiltrate, termed insulitis (3). Recurrence of type 1 diabetes on a pancreatic graft is usually well controlled and prevented by the immunosuppressive medication (4). Nonetheless, recurrence of autoimmunity, which may lead to loss of the pancreatic graft in the absence of rejection, has been occasionally reported (5–7).

Although the predominant role played by autoreactive T cells in the pathogenesis of type 1 diabetes is unquestioned, the pathogenicity of autoantibodies is debated, and may depend on a variety of host factors (2). Anyhow, it is firmly established that type 1 diabetes can occur in the absence of a functional B-cell compartment, and thus those autoantibodies are not a prerequisite for the inception of the disease (8).

In a recent in-depth immunological study of three cases of type 1 recurrence on a pancreatic graft, it has been proposed that the four cardinal features of type 1 diabetes recurrence were (i) hyperglycemia in the absence of rejection, (ii) insulitis and/or β-cell loss, (iii) persistence or reappearance of autoantibodies and (iv) presence of circulating autoreactive T cells (7). We report herein an unequivocal case of type 1 diabetes recurrence in a SPK recipient who did not develop autoantibodies, and discuss its metabolic and immunological features.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Discussion
  6. Disclosure
  7. References

A 38-year-old Caucasian woman (blood group B+; HLA A3,24; B8,47; DR4,17) with a 20-year history of type 1 diabetes and end-stage nephropathy underwent SPK. Autoantibodies were negative before transplant. The donor was a 29-year-old woman (blood group O+; HLA A1,28; B15,37; DR4) with brain death secondary to a head trauma. The pancreas transplantation procedure was performed using systemic–enteric drainage. Her immunosuppressive treatment consisted of thymoglobulin induction, a tacrolimus–sirolimus association and steroids. Target trough levels were 8–15 ng/mL for tacrolimus and 5–10 ng/mL for sirolimus. The early postoperative course was characterized by an acute episode of kidney function impairment 1 month posttransplant. Kidney biopsy revealed thrombotic microangiopathy and borderline kidney rejection. This was managed by a switch from tacrolimus to mycophenolate mofetil (MMF; 500 mg bid), and was followed by a rapid return to normal kidney function. Because she could only tolerate reduced doses of MMF, sirolimus trough levels were raised to 10–15 ng/mL for the first 2 years and lowered to 8–10 ng/mL thereafter. Targets were consistently achieved in the first 2 years. Regarding endocrine pancreatic function, insulin was withdrawn immediately after transplantation. The first 6 months were characterized by high levels of basal C-peptide (in the range of 3–4.5 ng/mL), normal fasting blood glucose (FBG) and normalization of HbA1c by the third month. Between 6 months and 1 year posttransplant, in parallel with steroid tapering, C-peptide levels dropped to normal values and remained in the range of 1.8–2.5 ng/mL until 4 years posttransplant. However, abnormally elevated FBG values were recorded from 2.5 years posttransplant. Results of an oral glucose tolerance test (OGTT) at 3 years were in the diabetic range (glucose at 2 h postload: 16.4 mmol/L), and HbA1c reached values >6.0% for the first time at 3.5 years posttransplant. Because of the rather high C-peptide values and a weight gain of 6 kg (BMI increase from 20.7 to 24.2 kg/m2), posttransplant diabetes mellitus (PTDM) was suspected, rosiglitazone treatment was briefly started and insulinotherapy was resumed shortly thereafter. From then on, C-peptide steadily declined to undetectable levels by 5.5 years posttransplant, at which time she had returned to her pretransplant insulin requirements. Metabolic data are summarized in Figure 1. Except for a transient and modest elevation of GAD-65 antibody at 3 months posttransplant, no significant and prolonged elevation of autoantibodies (ICA, insulin, GAD-65, IA-2) was observed. Autoantibody follow-up is summarized in Figure 2. Throughout follow-up, anti-HLA antibodies remained negative, serum amylase and lipase remained within normal values. Kidney function was assessed yearly by an isotopic method and remained stable throughout follow-up, varying between 66 mL/min and 77 mL/min, with a value of 70 mL/min at 8 years.

image

Figure 1. Long-term endocrine function after SPK. On the left axis, HbA1c (%) is represented as black histograms and fasting blood glucose (mmol/L) is represented as grey histograms. On the right axis, fasting C-peptide (pmol/L) is represented as open squares and C-peptide/glucose ratio is represented as open triangles. Daily insulin requirements, daily prednisone dose and blood level of sirolimus trough levels are indicated below the graph.

Download figure to PowerPoint

image

Figure 2. Follow-up of autoantibodies after SPK. ICAs (UJDS) are represented as white histograms, anti-insulin Abs (E/mL) are represented as black histograms, anti-GAD-65 Abs (IE/mL) are represented as dark gray histograms, anti-IA-2 (U/mL) are represented as light gray histograms and autoantibody levels are represented as ratio over upper limit of normal, to which a value of 1 is attributed (dotted line).

Download figure to PowerPoint

She was listed for a second pancreas transplant, which she received 6 months later, i.e. 6 years after SPK. The explanted graft was macroscopically normal in size and appearance. Histological examination (Figure 3) revealed normal exocrine pancreatic tissue with preserved lobulation and no sign of acinar damage. Pancreatic islets were normal in terms of size, number and distribution. However β-cell numbers were markedly decreased. Immunohistochemistry studies revealed that only 10.7% of chromogranine A positive cells also stained for insulin, as compared to 86.3% staining for glucagon. Insulitis was observed in the form of a T-cell infiltrate. CD3+/CD8+ T cells were seen surrounding the islets, whereas a few CD3+/CD4+ T cells were visible within the islets. CD20 and CD138 staining confirmed the absence of B lymphocytes and plasmocytes. C4d staining was negative. Congo red staining was negative, showing the absence of amyloid deposits. Importantly, a kidney biopsy taken during surgery was essentially unremarkable, with no sign of cellular or humoral rejection.

image

Figure 3. Macroscopy and histology of the pancreas specimen at the time of re-Tx. (A) Whole pancreas specimen showing a normal size pancreas. (B) H&E staining showing essentially intact pancreatic parenchyma (×100). (C) IHC staining for insulin and (D) for glucagon showing near-total disappearance of β cells in the islets. (E) IHC staining for CD8 showing moderate insulitis (×200).

Download figure to PowerPoint

The second donor was a 25-year-old man (blood type B+; HLA A1,32; B39,61; DR11,52) with brain death secondary to intracerebral hemorrhage. The same immunosuppressive induction and maintenance were used. Again, insulin was stopped immediately after transplantation. Two years after the second pancreas transplant, pancreas function remained normal, with excellent metabolic control and normal OGTT (glucose at 2 h postload: 4.7 mmol/L).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Discussion
  6. Disclosure
  7. References

We report herein an unequivocal case of recurrence of autoimmunity on a pancreatic graft in a patient with type 1 diabetes. The diagnosis of type 1 diabetes recurrence was based on: (i) loss of insulin secretory function, (ii) lymphocytic insulitis on the pancreatic explant, (iii) selective destruction of β cells in the islets of Langerhans and (iv) absence of evidence of exocrine pancreas or kidney acute cellular rejection (9).

With respect to alternate diagnoses, humoral rejection can be ruled out because of absence of anti-HLA antibodies and C4d staining in the kidney, chronic rejection because of the normal macroscopic and histologic aspect of the exocrine pancreas (10) and type 2 diabetes or PTDM given the rapid β-cell destruction observed, in a lean patient (BMI 21 kg/m2) who was not on calcineurin inhibitor immunosuppression.

Recurrence type 1 diabetes in a pancreatic graft is a rarely reported event. It may be a rare occurrence because of the fact that autoimmunity is generally well controlled by conventional immunosuppression. It was initially reported in recipients of segmental pancreatic grafts from living-related HLA-identical twins or two-haplotype-matched siblings who received either no or reduced immunosuppression (4). The difficulties in establishing the diagnosis may also have led to underreporting, as suggested by the unexpectedly high occurrence observed in a large histopathologic series (11).

Two of the cardinal features of type 1 diabetes recurrence proposed in the recent study from the University of Miami are lacking in this case. We did not demonstrate the presence of autoreactive T cells, because it requires tetramer technology, which is not yet available in our institution. Interestingly, our patient never increased her autoantibody levels above the reference interval in a significant and more than transient fashion. Subjects in the Miami study increased at least two autoantibodies in a prolonged fashion, up to 70× upper limit of normal (ULN), and heralded diabetes recurrence 3 months to 3 years in advance (7). GAD-65 antibody titers increased only very transiently in our patient to 3× ULN at 3 months posttransplant and remained undetectable for the next 4 years. They reappeared at very low levels, fluctuating around ULN from 6 months after recurrence of diabetes until 1 year after retransplant and became undetectable again thereafter. All other autoantibodies measured were always in the normal range. In this case, autoantibodies have been poor indicators of diabetes recurrence, let alone actors in the pathogenesis of autoimmunity.

To briefly touch upon islet of Langerhans transplantation, autoimmunity is considered by some authors to be a major contributor to islet graft decay and loss. The presence of autoreactive T cells at the time of transplantation was uniformly associated to graft failure in one study, whereas autoantibodies had no predictive value (12). In other reports, the presence of both anti-IA-2 and GAD-65 autoantibodies at the time of transplant, or the rise in GAD-65 titers was associated with poorer islet graft outcome (13,14).

The rather high C-peptide levels that were maintained in this patient even as HbA1c had started to rise were wrongly interpreted as peripheral insulin resistance. The observed pattern of sustained high C-peptide levels before an abrupt drop could possibly represent unregulated C-peptide release because of ongoing β-cell destruction. The progressive decrease in C-peptide/glucose ratio observed from 1 year posttransplant until ultimate graft loss may support this view.

In summary, this case shows that unequivocal recurrence of autoimmunity can occur on a whole pancreas graft and lead to β-cell destruction even in the absence of a sizeable increase in GAD and/or IA-2 autoantibody titers. This suggests that autoantibodies are not necessarily a cardinal feature of this condition and that the pathogenesis of type 1 diabetes recurrence on a pancreatic graft is not a homogeneous phenomenon.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Discussion
  6. Disclosure
  7. References

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Discussion
  6. Disclosure
  7. References