Pancreas allograft rejection

INTRODUCTION — The outcomes of pancreas transplantation have steadily improved secondary to improvements in surgical techniques, donor and recipient selection and management, and diagnostic imaging, all of which have resulted in a reduction in early technical graft losses. In addition, advances in immunosuppression and immune monitoring have led to a commensurate decrease in immunological graft losses. However, pancreas allograft rejection remains a major clinical challenge and is the primary cause of death-censored pancreas allograft loss after three months posttransplant.

As of 2020, approximately 87 percent of pancreas transplants in the United States are performed as simultaneous pancreas-kidney (SPK) transplants, with the remainder performed as either sequential pancreas after kidney transplant (PAK; approximately 5 percent) or pancreas transplants alone (PTA; 7 to 10 percent) [1,2].

This topic reviews the epidemiology, clinical presentation, diagnosis, treatment, and prognosis of acute rejection of the pancreas allograft in the setting of SPK transplantation. However, this discussion is also applicable to pancreas allograft rejection in PAK and PTA recipients. Other aspects of pancreas transplantation are presented separately:

●(See "Pancreas and islet transplantation in diabetes mellitus".)

●(See "Pancreas-kidney transplantation in diabetes mellitus: Patient selection and pretransplant evaluation".)

●(See "Pancreas-kidney transplantation in diabetes mellitus: Benefits and complications".)

EPIDEMIOLOGY AND RISK FACTORS

●Incidence – The overall incidences of acute rejection among patients receiving a pancreas transplant are 12.5 percent for pancreas after kidney (PAK), 21.8 percent for pancreas transplant alone (PTA), and 10.6 percent for simultaneous pancreas-kidney (SPK) transplant [2].

●Risk factors – Risk factors for acute rejection include undergoing a nonprimary SPK transplant or a solitary pancreas transplant (PAK or PTA), race mismatch, human leukocyte antigen (HLA) mismatch, transplantation involving a male donor and female recipient, increasing donor age, and a panel reactive antibody (PRA) level of >40 percent [3-5]. Among recipients of a PTA, the number of HLA mismatches, particularly at the HLA-B and HLA-DR loci, also increases the risk of acute rejection [5].

Although earlier data were conflicting [6-10], several studies have found that the development of de novo donor-specific antibodies (DSAs) posttransplant is strongly associated with acute pancreas rejection as well as death-censored pancreas and kidney allograft loss [11-16]. As an example, a meta-analysis of eight studies (1434 patients) found that patients with a de novo DSA had a higher odds of graft failure (odds ratio [OR] 4.42, 95% CI 3.15-6.22) and rejection (OR 3.35, 95% CI 2.28-4.91) compared with those without DSA [14].

CLINICAL FEATURES

Clinical manifestations — Acute pancreas rejection occurs most commonly in the first three to six months following pancreas transplantation but may occur at any point in time thereafter. It may occur as early as in the first six to eight weeks posttransplant, especially if there was a clinical event leading to either a reduction in immunosuppression (eg, infection, malignancy, medication toxicity, surgical complication) or a decrease in drug levels because of changes in absorption or metabolism (eg, in the setting of gastroparesis, bowel dysmotility, or drug interaction).

Many patients with pancreas allograft rejection are asymptomatic although some occasionally present with mild graft tenderness and/or fever. Because the pancreas allograft is intra-abdominal in location, other clinical manifestations of acute rejection may include abdominal pain, distension (typically caused by ileus), or a change in bowel function (eg, constipation).

Laboratory findings — Elevated pancreatic enzyme levels (serum lipase and/or amylase) are the most common presentation of acute pancreas allograft rejection (>90 percent of cases). The sensitivities of an elevated serum lipase and elevated serum amylase for the diagnosis of rejection are approximately 70 and 50 percent, respectively [17]. However, these findings are not specific, as other factors can contribute to abnormal pancreatic enzyme elevation. In general, there is good correlation between the elevation in pancreatic enzymes and the degree of exocrine parenchymal injury or inflammation. (See 'Differential diagnosis' below.)

Other laboratory findings that can be observed in patients with pancreas allograft rejection include the following:

●Elevated serum creatinine – In simultaneous pancreas-kidney (SPK) transplant recipients, an elevated serum creatinine concentration may be a surrogate marker of pancreas allograft rejection since there is a 60 percent concordance of pancreas and kidney allograft rejection [18]. However, isolated pancreas rejection can occur in the setting of normal kidney allograft function and histopathology in 20 percent of cases, and, conversely, isolated kidney rejection can occur in the setting of normal pancreas allograft function and histopathology in 20 percent of cases [19]. Thus, a stable or normal serum creatinine level in SPK transplant recipients does not necessarily exclude the possibility of pancreas allograft rejection [20]. In addition, an elevated serum creatinine level may also reflect nonimmunologic causes of kidney function impairment.

●Decrease in urinary amylase level – In pancreas transplant recipients with bladder drainage, a decrease in urinary amylase level (measured as international units per hour) is suggestive of early rejection. In general, acute rejection should be suspected in patients who present with either a decrease in urinary amylase level of 25 percent or more from baseline on two consecutive measurements obtained at least 12 hours apart or a decrease of 50 percent or more from baseline on any single measurement. However, pancreas transplantation with bladder drainage is less commonly performed.

●De novo donor-specific antibody (DSA) or rising DSA levels – The development of de novo DSA or rising DSA titers in a patient with a preexisting DSA have been associated with antibody-mediated rejection and pancreas graft failure and should prompt an immediate evaluation for rejection [11-16].

●Elevated plasma dd-cfDNA level – Plasma donor-derived cell-free DNA (dd-cfDNA) levels are emerging as a biomarker of alloimmune activity following SPK transplantation, similar to the experience in kidney transplantation [21-23]. Although early one- to two-month "baseline" levels appear to be higher in SPK compared with kidney alone transplant recipients, fractional levels of dd-cfDNA >1 percent or serially increasing levels appear to be associated with episodes of rejection [22]. In addition, the absolute quantity of dd-cfDNA (>70 copies/mL) may have prognostic value as well [22]. Analogous to kidney transplantation, low fractional levels have a high negative predictive value whereas elevated levels >1 percent only have a sensitivity of 50 to 60 percent for predicting rejection. Stable intermediate fractional levels may be associated with allograft injury in the absence of rejection. (See "Kidney transplantation in adults: Clinical features and diagnosis of acute kidney allograft rejection", section on 'Laboratory findings'.)

●Elevated fasting glucose, elevated hemoglobin A1C level, or decrease in fasting C-peptide level – The presence of fasting hyperglycemia, an elevated hemoglobin A1C level (ie, >7 percent), or reduction in fasting C-peptide level is usually a late finding of acute rejection as early rejection usually affects exocrine tissue (ductitis and acinar inflammation) and the vasculature (venulitis and arteritis) before islet cells [24]. Hyperglycemia, when it coincides with elevated serum lipase and amylase levels, is more specific for acute rejection and may portend significant graft destruction [24]. However, hyperglycemia, an increase in hemoglobin A1C level, or a reduction in fasting C-peptide level without elevated pancreatic enzyme levels is nonspecific and can be caused by an increase in insulin resistance due to significant weight gain, recurrence of autoimmune type 1 diabetes, or posttransplant diabetes secondary to medications. In addition, measurement of hemoglobin A1C and C-peptide levels may be affected by a reduced glomerular filtration rate or anemia.

Imaging findings — Acute pancreas allograft rejection may be associated with elevated resistive indices on ultrasonography in either the kidney or pancreas allograft, but this is a nonspecific finding. Computed tomography (CT) imaging may show signs of allograft swelling, stranding, or "pancreatitis," but these are also nonspecific findings that are not unique to acute rejection.

MONITORING OF ASYMPTOMATIC PATIENTS — Since many pancreas transplant recipients with pancreas rejection are asymptomatic, the primary means of monitoring for rejection is laboratory testing. The frequency of laboratory monitoring depends upon the time since transplantation and may vary among transplant centers:

●Immediate posttransplant period – daily

●Time of hospital discharge until three months posttransplant – once or twice weekly

●Between three and six months posttransplant – every one to two weeks

●Between 6 and 12 months posttransplant – every two to four weeks

●After one year posttransplant – monthly

The following lab tests are obtained at every lab visit, except as noted (see 'Laboratory findings' above):

●Serum lipase and amylase

●Serum creatinine (in simultaneous pancreas-kidney [SPK] transplant recipients)

●Fasting plasma glucose

●Hemoglobin A1C and fasting serum C-peptide levels (performed quarterly)

●Amylase in a timed (8- to 12-hour) urine collection (only in pancreas recipients with bladder drainage)

In addition, we monitor donor-specific antibody (DSA) levels at specified time points (eg, monthly for the first three months posttransplant, then every three months thereafter), depending upon the patient's immunologic risk [25]. Some transplant centers, including ours, also monitor donor-derived cell-free DNA (dd-cfDNA) levels at these specified time points for certain high-risk patients. The frequency of monitoring DSA and dd-cfDNA levels may be reduced after the first posttransplant year. (See 'Epidemiology and risk factors' above and 'Laboratory findings' above.)

Some, but not all, transplant centers perform surveillance pancreas biopsies at specified time points (eg, 1, 3, 6, or 12 months posttransplant), given the difficulty in detecting pancreas allograft rejection by clinical features and laboratory monitoring. Centers performing surveillance biopsies may choose to biopsy the kidney allograft only in SPK transplant recipients as a surrogate marker for pancreas rejection although the concordance of rejection between these two organs is approximately 60 percent [18]. The role of pancreas allograft biopsy is discussed below. (See 'Biopsy for persistently elevated pancreatic enzyme levels' below.)

DIAGNOSIS

When to suspect rejection — The diagnosis of acute pancreas allograft rejection should be suspected in any pancreas transplant recipient who presents with a new elevation in pancreatic enzyme levels (ie, elevated serum lipase and/or amylase). In most cases, this is detected by routine laboratory surveillance of asymptomatic patients, but it may also be detected by laboratory testing in patients presenting with abdominal pain or discomfort. However, elevated pancreatic enzyme levels are not specific for acute rejection, as other factors in pancreas recipients can contribute to abnormal pancreatic enzyme elevation. (See 'Laboratory findings' above and 'Monitoring of asymptomatic patients' above.)

Other laboratory findings that may be detected on routine surveillance and raise suspicion for pancreas rejection (especially in the setting of elevated pancreatic enzyme levels) include the following (see 'Laboratory findings' above):

●Elevated serum creatinine level (in simultaneous pancreas-kidney transplant [SPK] recipients)

●Elevated fasting plasma glucose or hemoglobin A1C level

●De novo donor-specific antibody (DSA) or an increase in DSA titer compared with pretransplant levels

●Plasma donor-derived cell-free DNA (dd-cfDNA) level >1 percent or a rising trend in serial dd-cfDNA measurements

●Decrease in urinary amylase level in an 8- to 12-hour timed urine collection (only in pancreas recipients with bladder drainage)

●Low blood tacrolimus or cyclosporine level

Initial evaluation and laboratory testing — In patients who present with a new elevation in serum lipase or amylase, our evaluation begins with an assessment for the presence of fever, abdominal symptoms (eg, pain or discomfort, distension, graft tenderness, nausea, vomiting, constipation), or dehydration (algorithm 1). We also assess the patient's adherence to their maintenance immunosuppression regimen since treatment nonadherence may be a cause of acute rejection.

●Patients with severe abdominal pain or distension should be referred to the emergency department for urgent evaluation in collaboration with the transplant team. This evaluation is discussed in more detail elsewhere. (See "Evaluation of the adult with nontraumatic abdominal or flank pain in the emergency department", section on 'Patient without abdominal catastrophe'.)

●In patients without severe abdominal pain or distension, we typically repeat serum lipase and amylase levels in two to three days to confirm that they are persistently elevated or rising. However, some experts may not repeat the enzyme levels if they are already markedly elevated (>2 to 3 times normal). Patients with recent constipation or dehydration, both of which may cause transient pancreatic enzyme elevations in pancreas recipients, should be treated as appropriate (eg, bowel regimen or increased oral hydration). If serum lipase and amylase levels return to normal, we resume routine monitoring of the patient. If serum lipase or amylase is persistently or markedly elevated, we obtain the following laboratory testing:

•Serum creatinine level (in SPK recipients)

•Fasting plasma glucose level

•DSA level (against the donor of the pancreas allograft)

•Plasma dd-cfDNA level

•Blood tacrolimus (or cyclosporine) concentration

•Blood cytomegalovirus (CMV), Epstein-Barr virus (EBV), and BK polyomavirus (BKPyV) viral loads by polymerase chain reaction

In pancreas recipients presenting with newly elevated pancreatic enzyme levels who have an elevated serum creatinine level (in SPK recipients), elevated fasting plasma glucose level, de novo DSA or increase in DSA titer compared with pretransplant levels, or elevated plasma dd-cfDNA level (>1 percent), a presumptive diagnosis of acute pancreas rejection can be made.

A subtherapeutic blood tacrolimus (or cyclosporine) concentration may suggest nonadherence or some other mitigating factor (such as malabsorption secondary to gastroparesis) and raise suspicion for the diagnosis of acute pancreas rejection.

The detection of CMV, EBV, or BKPyV viremia is important as the management of new viremia is different from that for acute pancreas rejection and involves antiviral therapy and a reduction, rather than augmentation, in immunosuppression. Some clinicians would treat new viremia first before considering a diagnosis or initiating treatment of rejection. Other clinicians would base decisions on the clinical setting and the extent of viremia. As an example, in a patient with elevated pancreatic enzyme levels who has recently discontinued CMV prophylaxis and is found to have low level CMV viremia, some authors would treat the viremia while proceeding to a pancreas allograft biopsy and treating for acute rejection if present. (See "Clinical manifestations, diagnosis, and management of cytomegalovirus disease in kidney transplant patients", section on 'Treatment'.)

Imaging to exclude other causes of pancreatic enzyme elevation — Following the initial evaluation (see 'Initial evaluation and laboratory testing' above), we obtain an abdominal imaging study to exclude other causes of pancreatic enzyme elevation, such as postsurgical complications (eg, ileus, fluid collection, leak) or other intra-abdominal processes (algorithm 1). Our choice of imaging study varies depending upon the timing of presentation as well as the presence or absence of abdominal symptoms:

●If the patient is within the early perioperative period (ie, first three months of transplant) or has abdominal symptoms or signs, we prefer computed tomography (CT) of the abdomen and pelvis with oral and intravenous (IV) contrast.

●If the patient is beyond the early perioperative period and does not have abdominal symptoms or signs, surgical causes of elevated pancreatic enzyme levels are less likely. In such patients, either a pancreas allograft ultrasound with Doppler study or CT of the abdomen and pelvis with oral and IV contrast is a reasonable choice.

If abdominal imaging reveals an intra-abdominal abnormality that could explain the elevation in pancreatic enzyme levels, then we treat with appropriate therapy (eg, percutaneous drain placement, bowel rest or catharsis, IV broad-spectrum antibiotics) and repeat measurements of the serum lipase and amylase levels. If the serum lipase and/or amylase levels return to normal, rejection is unlikely, and we resume routine monitoring. If the serum lipase and/or amylase levels remain persistently elevated, we proceed to a pancreas allograft biopsy (or, alternatively, a kidney allograft biopsy in SPK recipients), unless contraindicated. (See 'Biopsy for persistently elevated pancreatic enzyme levels' below.)

If abdominal imaging does not reveal a cause for the pancreatic enzyme elevation, we proceed to a pancreas allograft biopsy (or, alternatively, a kidney allograft biopsy in SPK recipients), unless contraindicated. (See 'Biopsy for persistently elevated pancreatic enzyme levels' below.)

Biopsy for persistently elevated pancreatic enzyme levels — Pancreas recipients who have persistently elevated pancreatic enzyme levels after other causes of pancreatic enzyme elevation have been excluded should undergo a pancreas allograft biopsy, unless contraindicated (algorithm 1). A pancreas allograft biopsy is the gold standard for establishing the diagnosis of pancreas allograft rejection, differentiating between acute T cell-mediated rejection (TCMR) and antibody-mediated rejection (ABMR), and grading the severity of rejection; it can also reveal other causes of pancreatic inflammation and injury. (See 'Establishing the type and severity of rejection' below and 'Differential diagnosis' below.)

In SPK recipients, an alternative option is to first perform a kidney allograft biopsy since this is technically easier to perform and there is a 60 percent concordance of pancreas and kidney allograft rejection [18]. However, a pancreas allograft biopsy is warranted if the kidney allograft biopsy shows no evidence of rejection and pancreatic enzyme levels remain elevated [26,27]. Even when there is concordance of pancreas and kidney allograft rejection, there may be discordance in the type of rejection (TCMR, ABMR, or mixed) between the two organs [18]. In pancreas after kidney (PAK) recipients, kidney allograft biopsy will not mirror or predict pancreas allograft rejection, as the grafts come from different donors [28].

Ultrasound-guided pancreas allograft biopsy is the most commonly used approach and can be performed safely and effectively in either bladder-drained or enterically drained allografts [29,30]. In a study of 183 pancreas transplant recipients who underwent ultrasound-guided pancreas allograft biopsy, 88 percent of biopsies were considered adequate for diagnosis [30]. Complications occurred in 2.8 percent of biopsies and included bleeding and inadvertent biopsy of the kidney, liver, or small bowel. Other experienced transplant centers have similarly reported high rates of diagnostic adequacy and low complication rates with ultrasound-guided biopsy [26]. A CT-guided biopsy or, rarely, an open or laparoscopic surgical biopsy is an alternative option when a biopsy cannot be safely performed with ultrasound guidance.

Among patients who have a pancreas allograft with either enteric or bladder drainage, the duodenal segment of the allograft is an alternative site to obtain tissue [31]. However, the potential of discordant findings between pancreas and duodenal histology and the difficulty in accessing the allograft duodenum in patients with enteric drainage have limited its use [32]. Newer techniques of pancreas transplantation with either gastric or native duodenal (enteric) drainage may permit easier endoscopic access to the allograft duodenum for visualization and monitoring.

The management of patients who are unable to undergo a pancreas allograft biopsy is discussed below. (See 'Suspected (but not biopsy-proven) rejection' below.)

Establishing the type and severity of rejection

Acute T cell-mediated rejection — The diagnosis of acute TCMR is established by the presence of characteristic histologic features on pancreas allograft biopsy. These include inflammation of the ducts, acinar epithelium, and vascular endothelium (picture 1). Venulitis (subendothelial inflammatory cells in septal veins) is part of the histologic finding of acute TCMR in pancreas allografts but not included in the grading score in kidney transplantation. Septal inflammation by itself does not meet the criteria for grade I acute TCMR; thus, it is considered borderline for TCMR. Acute TCMR is classified into three grades according to the severity of inflammation and the degree of involvement of vascular endothelium (table 1) [33].

Chronic active TCMR may represent the transition phase from acute TCMR to chronic rejection. It includes the features of acute TCMR together with the presence of active transplant arteriopathy, which is defined as intimal proliferation leading to arterial luminal narrowing with the presence of intimal lymphocytes and macrophages.

Antibody-mediated rejection — The diagnosis of ABMR requires three components (table 2) [33]:

●The presence of DSA

●C4d positivity in the interacinar capillaries

●Morphologic evidence of pancreas tissue injury

Some patients have morphologic evidence of ABMR and a positive DSA without C4d positivity in the interacinar capillaries. Such patients should be suspected of having ABMR and treated in the same manner as patients with all three criteria.

The histologic grading of ABMR in the pancreas was recognized in the Banff criteria in 2011 and includes interacinar inflammation, interacinar capillaritis, vasculitis, and thrombosis [33]. Interacinar capillaritis is morphologically similar to peritubular capillaritis that occurs in cases of ABMR in kidney transplantation.

ABMR is classified according to the time of occurrence posttransplant. Hyperacute ABMR is due to preformed DSAs and occurs within seconds to minutes of vascular anastomosis, leading to hemorrhagic necrosis in acini, islets, and ducts as well as graft thrombosis [34]. Accelerated ABMR occurs within 24 to 48 hours after the anastomosis whereas acute ABMR occurs within days, weeks, or months after transplantation. Approximately 75 percent of acute ABMR occurs within the first six months of transplantation [35-40].

Mixed acute rejection — The diagnosis of mixed acute rejection is established by histologic findings of both acute TCMR and ABMR within the same pancreas allograft. (See 'Acute T cell-mediated rejection' above and 'Antibody-mediated rejection' above.)

In one study, the rates of acute TCMR, ABMR, and mixed acute rejection (both acute TCMR and ABMR) were 18, 10, and 7 percent at one year [3].

DIFFERENTIAL DIAGNOSIS — An elevation of serum pancreatic enzyme levels is the most common presentation of pancreas allograft rejection (>90 percent of cases). Other causes of elevated serum pancreatic enzyme levels in the pancreas recipient include the following [26]:

●Early posttransplant period (<6 weeks):

•Duodenal or parenchymal enzyme leak

•Infected fluid/abscess

•Graft thrombosis

•Ileus or constipation

•Bowel obstruction

•Vascular thrombosis (partial) or stenosis

●Mid-posttransplant period (>45 days to one year):

•Small bowel obstruction

•Pseudocyst

•Constipation

•Abscess

•Cytomegalovirus (CMV) pancreatitis

•Vascular thrombosis (partial) or stenosis

●Late posttransplant period (>1 year):

•Small bowel obstruction/ventral hernia

•Intrinsic pancreatic abnormality

•Native pancreatitis

•CMV pancreatitis

•Vascular thrombosis (partial) or stenosis

In most cases, these disorders can be distinguished from acute rejection by performing an abdominal imaging study, a pancreas allograft biopsy, or both. (See 'Diagnosis' above.)

TREATMENT

Pretreatment assessment — Prior to the treatment of acute rejection, we assess both pancreatic function and the chronicity of the injury to the pancreas allograft. In addition, the benefits of treatment (eg, restoration of allograft function) must be weighed against the risks of antirejection therapy (eg, worsening of hyperglycemia, infection, malignancy).

●Assessment of pancreatic function is based upon evaluation of the patient's fasting plasma glucose concentration, serum C-peptide, and hemoglobin A1C levels. New abnormalities in these indices are suggestive of abnormal islet allograft function.

●The chronicity of allograft injury is determined by the pancreas allograft biopsy and imaging of the pancreas allograft. Evidence of significant fibrosis or a reduction in pancreatic islets on biopsy, or the finding of a small sized pancreas allograft by imaging, is consistent with a chronic component to the rejection.

In general, we treat nearly all patients who have evidence of acute rejection on biopsy irrespective of the perceived chronicity of allograft injury or degree of graft dysfunction. Because exocrine immunological injury usually occurs before islet injury, extensive parenchymal fibrosis without endocrine dysfunction is not a contraindication to treatment in the setting of biopsy-proven acute rejection. In addition, failure to treat acute rejection may lead to an intra-abdominal catastrophe or further sensitization. However, in patients with clinical evidence of allograft endocrine dysfunction (eg, elevated fasting serum glucose, low serum C-peptide, or hemoglobin A1C level ≥7 percent) and histologic evidence of a reduction in pancreatic islets, some experts would not treat for acute rejection.

Initial therapy

Biopsy-proven rejection — There is no high-quality evidence to guide the optimal initial treatment of biopsy-proven pancreas allograft rejection. Our approach to therapy is primarily based upon the clinical experience of the authors and editors and data relevant to both kidney (alone) and pancreas transplantation.

Patients with acute TCMR — In patients with biopsy-proven acute T cell-mediated rejection (TCMR), our approach to initial therapy is guided by the severity of rejection per the Banff classification (table 1):

●Borderline rejection – For patients with biopsy-proven borderline TCMR, we suggest treatment with glucocorticoids rather than observation or other immunosuppressive therapies.

•We typically administer pulse intravenous (IV) methylprednisolone 3 to 5 mg/kg daily for three doses (maximum daily dose of 500 mg) or dexamethasone 100 mg daily for three doses, followed by a tapered dose of oral prednisone as follows: 60 mg daily for two days, then 40 mg daily for two days, then 20 mg daily for two weeks, then 10 my daily for two weeks, then 5 mg daily indefinitely depending upon the clinical response.

•In addition, we augment maintenance immunosuppression by targeting higher levels of tacrolimus (ie, 8 to 12 mcg/L) and administering mycophenolate mofetil (1000 mg twice daily) or mycophenolate sodium (720 mg twice daily) for the ensuing two to three months. After two to three months, maintenance immunosuppression can be returned to levels similar to or slightly higher than they were prior to the episode of rejection.

●Grade I, II, or III rejection – For patients with biopsy-proven grade I, II, or III acute TCMR, we suggest treatment with rabbit antithymocyte globulin (rATG)-Thymoglobulin plus glucocorticoids rather than glucocorticoids alone.

•We administer rATG-Thymoglobulin at 6 to 10 mg/kg (1.5 to 2 mg/kg per dose, divided into five to seven daily or alternate-day doses as tolerated).

•We administer IV methylprednisolone 125 to 250 mg as premedication prior to the first two to three doses of rATG-Thymoglobulin, followed by a tapered dose of oral prednisone as detailed above for borderline rejection. We do not administer additional pulse high-dose glucocorticoids in patients who receive rATG-Thymoglobulin.

•In addition, we augment maintenance immunosuppression by targeting a tacrolimus level of 8 to 12 mcg/L and administering full doses of mycophenolate. Depending upon the white blood cell and platelet counts, reduced doses of mycophenolate may be required during rATG therapy.

•In all patients who are treated with rATG-Thymoglobulin and high-dose glucocorticoids, we recommence antimicrobial and antiviral prophylaxis for at least three months with a regimen that is identical to that administered in the immediate posttransplant period. This includes prophylaxis against Pneumocystis jirovecii pneumonia (PCP), cytomegalovirus (CMV) infection and disease, and herpes simplex infection (in patients who are at low CMV risk). A detailed discussion of the different prophylactic regimens is presented separately. (See "Prophylaxis of infections in solid organ transplantation", section on 'Pneumocystis pneumonia' and "Prophylaxis of infections in solid organ transplantation", section on 'Cytomegalovirus'.)

Some centers treat grade I acute TCMR with pulse high-dose glucocorticoids and reserve the use of rATG-Thymoglobulin for patients who do not respond to initial treatment. In a single-center study of 158 pancreas transplant recipients with biopsy-proven rejection, the addition of rATG-Thymoglobulin to glucocorticoids, compared with glucocorticoids alone, was associated with an improvement in treatment response rates and long-term graft survival among patients with grade II or III rejection but not those with grade I rejection [41]. If high-dose glucocorticoids (rather than rATG-Thymoglobulin) are used, a follow-up pancreas biopsy in approximately four to eight weeks should be considered to confirm the resolution of rejection.

There are no high-quality data supporting the use of high-dose glucocorticoids or rATG-Thymoglobulin in pancreas recipients with acute TCMR, and rationale for these therapies is based primarily upon studies showing their efficacy for the treatment of acute TCMR in kidney transplant recipients. In a retrospective single-center analysis of 102 pancreas transplant recipients with biopsy-proven TCMR, 47 (43 with grade I, 4 with grade II, and 0 with grade III) were treated with glucocorticoids alone, and 55 (16 with grade I, 27 with grade II, and 12 with grade III) were treated with glucocorticoids plus rATG-Thymoglobulin [41]. Among those who received glucocorticoids alone, response rates were 84 percent for grade I and 75 percent for grade II. Among those who received glucocorticoids plus rATG-Thymoglobulin, response rates were 75 percent for grade I, 82 percent for grade II, and 75 percent for grade III. There was no significant difference in the incidence of infections at one year after rejection treatment among patients treated with glucocorticoids alone and those treated with glucocorticoids plus rATG-Thymoglobulin. (See "Kidney transplantation in adults: Treatment of acute T cell-mediated (cellular) rejection".)

Patients with ABMR — The primary goal of treatment of antibody-mediated rejection (ABMR) is to eradicate the clonal population of B cells or plasma cells that is responsible for production of the donor-specific antibody (DSA). Our approach is similar to that for the treatment of ABMR in kidney alone transplant recipients, for which there is more supportive evidence.

For patients with biopsy-proven ABMR diagnosed within the first year posttransplant (early onset), we suggest treatment with the combination of plasmapheresis, intravenous immune globulin (IVIG), and rituximab, rather than other therapies. For patients with biopsy-proven ABMR diagnosed after the first year posttransplant (late onset), we suggest treatment with IVIG and rituximab, rather than plasmapheresis, IVIG, and rituximab. We do not perform plasmapheresis in patients with late-onset ABMR, based on the lack of evidence supporting the safety and efficacy of plasmapheresis in later-onset ABMR among kidney transplant alone recipients since there are no data in pancreas transplant recipients. However, some transplant centers continue to administer plasmapheresis for ABMR diagnosed after one year posttransplant. (See "Kidney transplantation in adults: Prevention and treatment of antibody-mediated rejection", section on 'Approach to initial therapy'.)

●We administer plasmapheresis daily or every other day for a minimum of three to five sessions and maximum of 10 to 12 sessions as tolerated by the patient. There is no high-quality evidence to support this or any other quantity of treatments. Exchanges are usually performed with albumin to avoid the risk of sensitization; however, if coagulation parameters become abnormal, exchanges with half albumin and half fresh frozen plasma would be reasonable. (See "Therapeutic apheresis (plasma exchange or cytapheresis): Indications and technology".)

●We administer IVIG at a dose of 100 mg/kg after each session of plasmapheresis, except after the final session, when we administer a larger dose (500 to 1000 mg/kg, rounded to the nearest 5 g). If plasmapheresis is not performed (eg, in patients with late-onset ABMR), the dose and frequency of IVIG vary among different centers. Some centers administer a single dose of IVIG 1000 mg/kg (maximum dose 100 g) and repeat this dose at two to four week intervals as needed depending upon the severity of ABMR and response to therapy. Other centers administer IVIG weekly for five to six weeks at a dose of 250 to 500 mg/kg. There are no data to support one approach over another.

●After the final session of plasmapheresis and IVIG, rituximab can be administered as a single IV dose of either 200 to 500 mg or 375 mg/m², which can be repeated at two-week intervals for a total of one to four doses. We usually administer two doses. However, data in support of rituximab are very limited.

●In addition, we augment maintenance immunosuppression as we do in patients with acute TCMR and also administer oral prednisone as follows: 20 mg daily for two to four weeks, then 15 mg daily for two weeks, then 10 mg daily for two weeks. Depending upon the severity of ABMR, response to therapy, and patient adherence, prednisone is either maintained at 10 mg daily indefinitely or reduced to 5 mg daily indefinitely. (See 'Patients with acute TCMR' above.)

●In all patients who are treated for ABMR, we recommence antimicrobial and antiviral prophylaxis for at least three months with a regimen that is identical to that administered in the immediate posttransplant period. This includes prophylaxis against PCP, CMV infection and disease, and herpes simplex infection (in patients who are at low CMV risk). A detailed discussion of the different prophylactic regimens is presented separately. (See "Prophylaxis of infections in solid organ transplantation", section on 'Pneumocystis pneumonia' and "Prophylaxis of infections in solid organ transplantation", section on 'Cytomegalovirus'.)

Our rationale for the use of plasmapheresis and IVIG, with or without rituximab, is based primarily upon studies of these therapies for the treatment of acute ABMR in kidney transplant recipients. These data are presented elsewhere. (See "Kidney transplantation in adults: Prevention and treatment of antibody-mediated rejection".)

Patients with mixed acute rejection — Patients with mixed acute rejection on biopsy should be treated for both TCMR and ABMR. There is no uniformly accepted approach; some clinicians treat such patients concomitantly for TCMR and ABMR, while others treat first for ABMR and then treat for TCMR (or vice versa). The individual approaches to TCMR and ABMR are discussed above. (See 'Patients with acute TCMR' above and 'Patients with ABMR' above.)

Suspected (but not biopsy-proven) rejection — Confirming the diagnosis of pancreas allograft rejection requires a pancreas allograft biopsy (see 'Biopsy for persistently elevated pancreatic enzyme levels' above). However, a pancreas biopsy may not be possible in patients who are on chronic anticoagulation or antiplatelet therapy that cannot be temporarily discontinued or in patients who cannot undergo a pancreas biopsy for technical reasons. In such patients, the decision to empirically treat for rejection is made on a case-by-case basis:

●For patients who have a high likelihood of acute rejection, we empirically treat for acute TCMR. Such patients include those with risk factors such as medication nonadherence, low calcineurin inhibitor levels, reduced immunosuppression in the setting of previous infection (eg, BK polyomavirus or cytomegalovirus [CMV] infection), presence of de novo DSA or markedly elevated donor-derived cell-free DNA (dd-cfDNA) levels (arbitrarily defined as >1 percent), high panel reactive antibody (PRA) level pretransplant, pancreas retransplant, and history of prior rejection episodes.

Our regimen consists of pulse IV glucocorticoids, using an approach similar to that used to treat biopsy-proven TCMR. Patients who do not have a clinical response can be treated with rATG-Thymoglobulin. In general, however, we prefer not to administer rATG-Thymoglobulin in the absence of biopsy corroboration of the diagnosis of acute rejection. (See 'Patients with acute TCMR' above.)

●For patients who have a de novo DSA or an increase in DSA compared with pretransplant levels in the setting of graft dysfunction, we also empirically treat for ABMR. Some experts would also empirically treat ABMR in the setting of a rising and markedly elevated dd-cfDNA levels (arbitrarily defined as >1 percent). The treatment of ABMR is described above. (See 'Patients with ABMR' above.)

Monitoring response — In all patients being treated for acute rejection, serum lipase and amylase levels should be monitored daily during the first five to seven days of treatment, followed by two to three times per week during the second week, and then at least weekly thereafter for a minimum of one month. We consider a positive response to therapy to be a decrease in serum lipase and/or amylase levels of at least 50 percent (and ideally to normal) within three to five days of treatment. In patients who respond to therapy, we maintain augmented maintenance immunosuppression and continue low-dose oral prednisone indefinitely. The duration of augmented immunosuppression depends upon the initial severity of rejection as well as the timing and response to therapy. Patients without a response to therapy are considered to have ongoing or refractory rejection and may require modification of therapy. (See 'Refractory rejection' below.)

Patients with de novo DSA or an increase in DSA compared with pretransplant levels should have DSA levels monitored every two to four weeks. However, DSA will persist even in patients who respond to therapy and should not be viewed as a specific target for treatment.

In patients who are treated with rATG-Thymoglobulin, we also monitor the blood absolute lymphocyte count and CD3 lymphocyte counts (usually after the third dose of rATG) to assess the response to therapy. The absolute lymphocyte count should decrease to <50 cells/microL and CD3 lymphocyte count to <20 cells/microL.

In addition, some transplant centers perform a repeat pancreas biopsy one month following the initial treatment course to evaluate the histologic response to treatment.

Refractory rejection — Patients with acute pancreas rejection (and no other cause of allograft pancreatitis) whose serum lipase and/or amylase levels decrease by less than 50 percent with treatment, or who have an increase in lipase and/or amylase level after an initial decrease, are considered to have ongoing or refractory rejection and may require modification of therapy. Our approach to such patients is as follows:

●Refractory TCMR – In patients with acute TCMR who do not respond to initial treatment with glucocorticoids only, we administer rATG-Thymoglobulin. Dose and administration are discussed elsewhere in this topic. (See 'Acute T cell-mediated rejection' above.)

In patients with acute TCMR who do not respond to initial treatment with rATG-Thymoglobulin and glucocorticoids, we repeat a pancreas allograft biopsy (or a kidney allograft biopsy if the pancreas allograft is not accessible) to assess for ongoing rejection. If the repeat biopsy shows evidence of ongoing acute rejection, dose escalation with rATG-Thymoglobulin (or longer duration of therapy) or treatment with horse antithymocyte globulin (10 to 15 mg/kg for two doses) or alemtuzumab (30 mg, given subcutaneously, for one dose) can be considered on a case-by-case basis. If the repeat biopsy does not show evidence of ongoing acute rejection, the patient should be evaluated for other causes of pancreatitis. (See 'Patients with acute TCMR' above.)

●Refractory ABMR – In patients with ABMR who do not respond to initial treatment, we treat with another three to five sessions of plasmapheresis followed by IVIG, as described elsewhere in this topic (see 'Patients with ABMR' above). After the last session of plasmapheresis, rATG-Thymoglobulin (1.5 to 2 mg/kg for two to three doses) or rituximab (375 mg/m²) can be given.

The decision of when to stop treating patients with rejection is difficult and is determined by factors such as patient age, frailty, tolerability of treatment, cumulative immunosuppression, and history of infection. We generally do not administer more than 10 total doses (15 mg/kg) of rATG-Thymoglobulin and typically stop therapy after giving a second round of anti-T cell therapy (rATG-Thymoglobulin or alemtuzumab).

Data on the treatment of refractory pancreas allograft rejection are very limited, and the above approach is based primarily upon the clinical experience of the authors and editors of this topic.

PROGNOSIS — In simultaneous pancreas-kidney (SPK) recipients, acute rejection is associated with decreased graft survival [42,43]. In one study of 159 pancreas transplant recipients, the one-year pancreas graft survival rate after treated rejection was 80 percent [3]. In another study, rejection in both the kidney and pancreas allograft, but not in either kidney or pancreas alone, was associated with an increased risk of pancreas graft failure (relative risk 1.54, 95% CI 1.25-1.88) [42].

Outcomes depend upon the severity and timing of rejection and the timeliness and intensity of therapy. Early diagnosis and treatment of pancreas allograft rejection results in preservation of graft endocrine function [44,45]. Following successful treatment of rejection, pancreatic enzyme levels frequently recover although not always to the previous baseline level [46,47].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Kidney transplantation".)

SUMMARY AND RECOMMENDATIONS

●General principles – Approximately 85 percent of pancreas transplants in the United States are performed as simultaneous pancreas-kidney (SPK) transplants, with the remainder performed as either sequential pancreas after kidney (PAK) or pancreas transplants alone (PTA). Pancreas allograft rejection is the primary cause of death-censored pancreas allograft loss after three months posttransplant. (See 'Introduction' above and 'Epidemiology and risk factors' above.)

●Clinical features – Many patients with pancreas allograft rejection are asymptomatic although some occasionally present with mild graft tenderness and/or fever. Other clinical manifestations may include abdominal pain, distension (caused by ileus), or a change in bowel function. Elevated serum enzyme (lipase and/or amylase) levels are the most common presentation of acute rejection (>90 percent of cases). (See 'Clinical manifestations' above and 'Laboratory findings' above.)

●Monitoring for rejection – The primary means of monitoring pancreas transplant recipients for pancreas rejection is laboratory testing. The frequency of monitoring depends upon the time since transplantation and may vary among different centers. (See 'Monitoring of asymptomatic patients' above.)

●Diagnosis – Pancreas allograft rejection should be suspected in patients with either clinical manifestations or laboratory abnormalities (most commonly an elevation in serum lipase and/or amylase levels). Our approach to the evaluation of such patients is detailed above (algorithm 1). The gold standard for the diagnosis of rejection is a pancreas allograft biopsy, which is the only reliable method for grading the severity of rejection and differentiating between acute T cell-mediated rejection (TCMR), antibody-mediated rejection (ABMR), and other causes of pancreatic inflammation and injury. In SPK recipients, an alternative option is to first perform a kidney allograft biopsy since this is technically easier to perform and there is a 60 percent concordance of pancreas and kidney allograft rejection. (See 'Diagnosis' above.)

●Treatment – In general, we treat nearly all patients who have evidence of acute rejection on biopsy irrespective of the perceived chronicity of allograft injury or degree of graft dysfunction. There is no high-quality evidence to guide the optimal initial treatment of biopsy-proven pancreas allograft rejection. Our approach to therapy is primarily based upon the clinical experience of the authors and editors and data relevant to both kidney (alone) and pancreas transplantation:

•Biopsy-proven TCMR – For patients with biopsy-proven borderline TCMR, we suggest treatment with glucocorticoids rather than observation or other immunosuppressive therapies (Grade 2C). For patients with biopsy-proven grade I, II, or III acute TCMR, we suggest treatment with rabbit antithymocyte globulin (rATG)-Thymoglobulin plus glucocorticoids rather than glucocorticoids alone (Grade 2C). In addition, we augment maintenance immunosuppression. For patients receiving rATG-Thymoglobulin and high-dose glucocorticoids, we recommence antimicrobial and antiviral prophylaxis. (See 'Acute T cell-mediated rejection' above.)

•Biopsy-proven ABMR – For patients with biopsy-proven ABMR diagnosed within the first year posttransplant (early onset), we suggest treatment with the combination of plasmapheresis, intravenous immune globulin (IVIG), and rituximab, rather than other therapies (Grade 2C). For patients with biopsy-proven ABMR diagnosed after the first year posttransplant (late onset), we suggest treatment with IVIG and rituximab, rather than plasmapheresis, IVIG, and rituximab (Grade 2C). We do not perform plasmapheresis in patients with late-onset ABMR based on the lack of evidence supporting the safety and efficacy of plasmapheresis in later-onset ABMR among kidney transplant alone recipients. However, some transplant centers continue to administer plasmapheresis for ABMR diagnosed after one year posttransplant. In addition, we augment maintenance immunosuppression and recommence antimicrobial and antiviral prophylaxis. (See 'Antibody-mediated rejection' above.)

•Mixed acute rejection – Patients with biopsy-proven mixed acute rejection should be treated for both TCMR and ABMR. There is no uniformly accepted approach; some clinicians treat such patients concomitantly for TCMR and ABMR, while others treat first for ABMR and then treat for TCMR (or vice versa). (See 'Patients with mixed acute rejection' above.)

•Suspected (but not biopsy-proven) rejection – In patients who are unable to undergo a pancreas biopsy (eg, due to chronic anticoagulation or antiplatelet therapy that cannot be discontinued or for technical reasons), the decision to empirically treat for rejection is made on a case-by-case basis. We empirically treat patients for acute TCMR if the likelihood of acute rejection is judged to be high. In addition, we also treat for ABMR if there is a detection of de novo donor-specific antibodies (DSA) or an increase in DSA compared with pretransplant levels in the setting of graft dysfunction. (See 'Suspected (but not biopsy-proven) rejection' above.)