Liver transplantation with a strongly positive crossmatch: Case study and literature review

Authors

  • Garrett R. Leonard,

    Corresponding author
    1. Division of Transplantation, Department of Surgery, Penn State University College of Medicine, Hershey, PA
    • HLA Laboratory, Department of Pathology, Penn State University College of Medicine, Hershey, PA
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Hiroko Shike,

    Corresponding author
    • HLA Laboratory, Department of Pathology, Penn State University College of Medicine, Hershey, PA
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Tadahiro Uemura,

    1. Division of Transplantation, Department of Surgery, Penn State University College of Medicine, Hershey, PA
    Search for more papers by this author
  • Justine L. Gaspari,

    1. HLA Laboratory, Department of Pathology, Penn State University College of Medicine, Hershey, PA
    Search for more papers by this author
  • Francesca M. Ruggiero,

    1. Surgical Pathology, Department of Pathology, Penn State University College of Medicine, Hershey, PA
    Search for more papers by this author
  • Riaz A. Shah,

    1. Division of Transplantation, Department of Surgery, Penn State University College of Medicine, Hershey, PA
    Search for more papers by this author
  • Thomas R. Riley,

    1. Division of Gastroenterology and Hepatology, Department of Internal Medicine, Penn State Milton S. Hershey Medical Center, Penn State University College of Medicine, Hershey, PA
    Search for more papers by this author
  • Zakiyah Kadry

    1. Division of Transplantation, Department of Surgery, Penn State University College of Medicine, Hershey, PA
    Search for more papers by this author

  • The authors have no conflicts of interest to disclose.

Address reprint requests to Hiroko Shike, M.D., HLA Laboratory, Department of Pathology, Penn State Milton S. Hershey Medical Center, Penn State University College of Medicine, 500 University Drive, Mail Code H160, P.O. Box 850, Hershey, PA 17033. Telephone: 717-531-5117; FAX: 717-531-3803; E-mail: hshike@hmc.psu.edu

Address reprint requests to Zakiyah Kadry, M.D., Division of Transplantation, Department of Surgery, Penn State Milton S. Hershey Medical Center, Penn State University College of Medicine, 500 University Drive, MCH062, Hershey, PA 17033. Telephone: 717-531-5921; FAX: 717-531-5851; E-mail: zkadry@psu.edu

Abstract

A positive crossmatch has been associated with increased risk in liver transplantation. To study the clinical significance of preformed donor-specific human leukocyte antigen antibodies (DSAs) in liver transplantation, we reviewed patients who underwent liver transplantation with a strongly positive flow cytometry crossmatch. DSAs were evaluated with a Luminex solid phase assay. The complement-fixing ability of DSAs was tested with a complement component 1q (C1q) assay. Using an assay correlation between complement-dependent cytotoxicity crossmatch, flow cytometry crossmatch, and DSA results, we reviewed the effects of DSAs on the outcomes of our patients as well as reported cases in the literature. Five of 69 liver recipients had a strongly positive crossmatch: 4 had a positive T cell crossmatch [median channel shift (MCS) = 383.5 ± 38.9], and 5 had a positive B cell crossmatch (MCS = 408.8 ± 52.3). The DSAs were class I only in 1 patient, class I and II in 3 patients, and class II only in 1 patient. Cholestasis, acute rejection, or both were observed in 3 of the 4 patients with a positive T cell crossmatch with an MCS approximately greater than 300. The C1q assay was positive for 3 patients. Two had either persistent cholestasis or early acute rejection. One patient who was treated with preemptive intravenous immunoglobulin had an unremarkable outcome despite a positive C1q result. One of the 2 patients with a negative C1q assay experienced persistent cholestasis and early and recurrent acute rejection; the other had an unremarkable outcome. None of the patients died or lost a graft within the first year of transplantation. Our study suggests that human leukocyte antigen antibody screening, flow cytometry crossmatch MCS levels, DSA mean fluorescent intensity levels, and C1q assays may be useful in assessing the risk of antibody-mediated rejection and timely interventions in liver transplantation. Liver Transpl 19:1001–1010, 2013. © 2013 AASLD.

Abbreviations
AHG

anti-human globulin

ALT

alanine aminotransferase

AMR

antibody-mediated rejection

AST

aspartate aminotransferase

C1q

complement component 1q

C4d

complement component 4d

CDC

complement-dependent cytotoxicity

DSA

donor-specific human leukocyte antigen antibody

HLA

human leukocyte antigen

IgG

immunoglobulin G

IVIG

intravenous immunoglobulin

MCS

median channel shift

MELD

Model for End-Stage Liver Disease

MFI

mean fluorescent intensity

PRA

panel reactive antibody

SAB

single antigen bead

A positive lymphocyte crossmatch is not generally regarded as a contraindication to liver transplantation because liver transplantation is well known to be relatively resistant to preformed donor-specific human leukocyte antigen antibodies (DSAs) and antibody-mediated rejection (AMR). However, a positive crossmatch has been associated with lower patient survival,[1, 2] lower graft survival,[3-5] cholestasis,[6-8] thrombocytopenia,[7-9] increased rejection episodes,[10-12] and early rejection.[13-15] AMR has been reported in liver transplant recipients with a positive crossmatch, DSAs, and complement component 4d (C4d) staining.[10, 15-17] Adverse outcomes may be associated more significantly with strongly positive crossmatches versus weakly positive crossmatches when they are evaluated with complement-dependent cytotoxicity (CDC) titers or flow cytometry median channel shifts (MCSs). Reports regarding the levels of positive crossmatches and liver transplant outcomes remain limited to date. We present a case study of patients who underwent liver transplantation with strongly positive flow cytometry crossmatches and strong preformed DSAs. Preformed DSAs were evaluated for their complement-fixing ability with a complement component 1q (C1q) assay.[18] The C1q assay is a solid phase test that detects C1q binding of DSAs, which is the first step in the classic complement pathway that leads to the formation of the membrane attack complex and results in cell death. Thus, C1q-binding DSAs are expected to have the potential to exert cytotoxicity, and they have been associated with a greater risk of acute rejection and allograft loss in patients undergoing renal[19] and heart transplantation.[18, 20] We discuss levels of preformed DSAs, C1q assay results, transplant outcomes, posttransplant DSA monitoring, and treatment interventions in the setting of liver transplantation with a strongly positive crossmatch.

PATIENTS AND METHODS

Case Study

This study was approved by the institutional review board of the Pennsylvania State University College of Medicine (38601EM). We reviewed flow crossmatch results for all patients who underwent ABO-compatible liver transplantation at our center between October 2008 and June 2012. This case study was conducted with patients who had a positive flow crossmatch and were positive for DSAs. In this article, a crossmatch with an MCS > 200 is called strongly positive, whereas an MCS < 200 that generally results in a negative CDC crossmatch.

Liver transplantation was performed with a standard technique without a bypass. The posttransplant immunosuppression consisted of tacrolimus (target level = 8-10 ng/mL), mycophenolate mofetil (1000 mg twice/day), and prednisone. Prednisone (initial dose = 20 mg/day) was tapered off within the first 3 to 4 months. Patients were switched to cyclosporine A (target level = 200-300 ng/mL) if they developed neurological complications related to tacrolimus. C4d staining and histopathology were examined in formalin-fixed, paraffin-embedded tissues.

Human Leukocyte Antigen (HLA) Test

HLA typing was performed with LABType (One Lambda, Inc., Canoga Park, CA), which is a reverse, sequence-specific oligonucleotide probe method. Flow cytometry crossmatching was performed on the day of transplantation with a current serum sample. Briefly, donor cells treated with pronase (1 mg/dL)[21] were incubated with a patient's serum, and HLA antibodies were detected with fluorescein isothiocyanate–conjugated goat [F(ab)′2] anti-human immunoglobulin G (IgG; Fc-specific) and acquired with FACSCalibur. Anti-CD3 peridinin chlorophyll protein and anti-CD19 R-phycoerythrin (BD Biosciences) were used to gate T and B cell populations, respectively. An MCS > 50 on a 1024-channel scale was considered positive. If the crossmatch was positive, DSAs were tested with a solid phase assay using pretransplant serum (from the day of transplantation). Panel reactive antibodies (PRAs) and DSAs (IgG class) were determined with LABScreen PRA and single antigen bead (SAB) solid phase microbead assays (One Lambda) using Luminex xMAP multiplex technology. Serum was pretreated with dithiothreitol in order to reduce interference by inhibitors.[22] DSAs were considered positive with a mean fluorescent intensity (MFI) > 1000. The highest MFI was used as an estimate of the DSA level if multiple DSAs were present. C1q-binding DSAs were detected with a C1q assay (One Lambda) using heat-inactivated pretransplant serum samples.

Assay correlations between flow cytometry crossmatches, CDC crossmatches, and DSA levels were evaluated on the basis of 50 T cell crossmatch results from 25 serum samples and 10 cells (provided by the American Society for Histocompatibility and Immunogenetics for proficiency testing). CDC crossmatching was performed with the standard National Institutes of Health technique with or without anti-human globulin (AHG) enhancement. Serum titration was prepared through a serial 2-fold dilution of pooled positive control serum with normal control serum (male and blood type AB). The crossmatch was performed with the serially diluted pooled positive control serum and 5 random donor cells.

RESULTS

HLA Test Results

Sixty-nine patients received ABO-compatible liver transplants between October 2008 and June 2012. The patient ages ranged from 24 to 70 years (mean age = 56.0 ± 10.9 years). The patients were followed for 6 months to 4.3 years (mean = 2.5 years). The crossmatch was positive in 9 patients (13.0%) and negative in 60 patients. A strongly positive crossmatch (MCS > 200) was detected in 5 patients (7.2%; patients 1-5 in Fig. 1) with DSAs (MFI > 9000). Four patients had a positive T cell crossmatch with an MCS of 383.5 ± 38.9 (range = 296-448). All 5 patients had a positive B cell crossmatch with an MCS of 408.8 ± 52.3 (range = 320-448).

Figure 1.

Flow cytometry crossmatch MCS. Liver transplant patients (n = 69) were tested with flow cytometry crossmatching. Patients 1 to 5 had a positive crossmatch with a T and/or B cell MCS > 200. Other patients (indicated by circles) had a negative or weakly positive crossmatch.

The correlation between flow cytometry crossmatch MCS, CDC crossmatch, and DSA MFI results is shown (Fig. 2B). Thirty-five negative CDC crossmatches with or without AHG enhancement showed an MCS < 200 according to flow cytometry crossmatching and DSAs with an MFI < 7500. Seven positive CDC crossmatches without AHG enhancement correlated with an MCS > 300 by flow cytometry crossmatching and with DSAs with an MFI > 10,000. A 1:4 titer from 7 flow cytometry crossmatches resulted in a mean MCS of 214 ± 46 (Fig. 2A). A flow cytometry titer of 1:16 from 5 crossmatches resulted in an MCS of 302 ± 61. A flow cytometry titer of 1:512 and a CDC titer of 1:32 corresponded to an MCS of approximately 400.

Figure 2.

Correlations between HLA antibody assays. (A) Correlation between the flow crossmatch MCS (1024 channels) and serum titration: (•) T cell crossmatches and (×) B cell crossmatches. (B) Correlation between the DSA MFI, flow crossmatch MCS, and CDC crossmatch results: (×) positive CDC crossmatch without AHG enhancement, (□) positive CDC crossmatch with AHG enhancement, and (•) negative CDC crossmatch with or without AHG enhancement.

Case Study

Patient 1

The patient was a 30-year-old Hispanic female with end-stage liver disease secondary to primary biliary cirrhosis and a history of systemic lupus erythematosus with Sjogren's syndrome. Her Model for End-Stage Liver Disease (MELD) score was 40. She received a transplant from a 51-year-old Caucasian male donor with the same blood group (O). The donor's preprocurement laboratory tests showed an aspartate aminotransferase (AST) level of 35 IU/L, an alanine aminotransferase (ALT) level of 25 IU/L, and a total bilirubin level of 1.4 mg/dL. The cold ischemia time was 8.5 hours. T and B cell crossmatches were both strongly positive (Table 1). The PRA value was 100% for class I and 0% for class II. DSAs (A1, B8, and B39) were detected with an MFI > 10,000 with the LABScreen SAB assay, and complement fixing was shown with the C1q assay. After transplantation, DSAs in serum dramatically decreased (Fig. 3). The total bilirubin level initially decreased from 64.2 to 32 mg/dL but did not decrease after posttransplant day 2. A T-tube cholangiogram on day 4 showed normal findings. Vascular liver duplexes confirmed that all vessels were patent. The tacrolimus trough level was 7.6 ng/mL. Because of concerns about early acute rejection, a 250-mg bolus of Solu-Medrol was administered on day 4, and this was followed by another dose of Solu-Medrol (500 mg) on day 5. However, the bilirubin level remained significantly elevated. On the basis of the strongly positive crossmatch and high pretransplant DSA levels, we suspected AMR and administered intravenous immunoglobulin (IVIG; 2 g/kg) on day 6. By the next day (day 7), the bilirubin level started to improve significantly, and the DSA levels gradually declined. The patient continued to do well. The patient's most recent laboratory results (19 months after transplantation) indicated a stable and good graft outcome (ALT level = 68 IU/L, AST level = 45 IU/L, total bilirubin level = 0.5 mg/dL, and direct bilirubin level = 0.1 mg/dL).

Table 1. HLA Test Results
Patient NumberT/B MCSaDSAs (MFI)bC1q AssaycPosttransplant Coursed
  1. a

    T and B cell flow cytometry crossmatch MCS values are shown on a 1024-channel scale.

  2. b

    The specificity of DSAs (IgG class) detected with a solid phase assay (LABScreen) using anti-human IgG as a secondary antibody for detection is listed with the MFI levels.

  3. c

    The complement-fixing ability of DSAs was tested with a solid phase C1q assay (LABScreen) using an anti–complement 1 antibody for detection.

  4. d

    The posttransplant outcomes of the 5 cases are summarized.

1448/448A1, B8, B39 (>10,000)PositiveCholestasis and suspected AMR resolved after IVIG
2362/404B37, B65, DR13, DQ6 (>10,000)PositivePreemptive IVIG and uneventful outcome
3428/440A24, B35, B58, DR4 (>10,000)PositiveEarly rejection
4296/432A2, B3, B18, B58, DR14, DQ4, DQ5 (>10,000)NegativeCholestasis, early and recurrent rejection, hepatocellular carcinoma, and death (day 678)
540/320DR4, DR53, DQ8 (>9000)NegativeUneventful outcome
Figure 3.

Patients' laboratory results. Posttransplant ALT, total bilirubin, and DSA levels are depicted for the 5 patients. Patients 1 and 2 were treated with IVIG (shown by arrows).

Patient 2

The patient was a 53-year-old Caucasian male with end-stage liver disease secondary to alcoholic cirrhosis. His MELD score was 30. He received a liver transplant from a 41-year-old African American female donor with the same blood group (A). The donor's preprocurement test results were an AST level of 25 IU/L, an ALT level of 23 IU/L, and a total bilirubin level of 1.0 mg/dL. The cold ischemia time was 8 hours. T and B cell crossmatches were both strongly positive (Table 1). The PRA value was 100% for classes I and II. DSAs (B37, B65, DR13, and DQ6) were positive for complement fixing. After transplantation, his ALT, AST, and bilirubin levels and most DSA levels started to improve. The tacrolimus trough level was 7.8 ng/mL. However, on day 7, the bilirubin level increased again (Fig. 3), DSAs to B37 persisted, and a vascular duplex showed an increased resistance index and diminished hepatic artery flow. Therefore, IVIG (2 g/kg) was administered (1 g/kg/day for 2 days). His ALT, AST, total bilirubin, and DSA levels continued to improve. The patient was discharged on day 11. The patient continued to do well with no further episodes of rejection or graft dysfunction 18 months after transplantation. The patient's laboratory results were within the normal ranges (ALT level = 30 IU/L, AST level = 29 IU/L, total bilirubin level = 0.5 mg/dL, and direct bilirubin level = 0.1 mg/dL).

Patient 3

The patient was a 46-year-old Caucasian female with end-stage liver disease secondary to nonalcoholic steatohepatitis. Her MELD score was 32. She received a liver transplant from a 26-year-old Hispanic male donor with the same blood group (O) who before procurement had an AST level of 143 IU/L, an ALT level of 106 IU/L, and a total bilirubin level of 0.49 mg/dL. The cold ischemia time was 8 hours. T and B cell crossmatches were both strongly positive (Table 1). The PRA value was 100% for class I and 20% for class II. DSAs (A24, B35, B58, and DR4) were positive for complement fixing. The patient experienced increases in ALT, AST, and bilirubin on day 6; acute rejection was suspected and responded to steroid treatment (a 250-mg bolus of Solu-Medrol; Fig. 3). The tacrolimus trough level was 8.6 ng/mL at the time of suspected rejection. Biopsy was not performed. Her clinical course was unremarkable, and the patient was discharged on day 10. Three years and 8 months after transplantation, the patient had normal liver test results (ALT level = 13 IU/L, AST level = 30 IU/L, and total bilirubin level = 0.9 mg/dL).

Patient 4

The patient was a 25-year-old Caucasian male with end-stage liver disease secondary to hepatitis C virus and a history of cardiac surgery for congenital pulmonary atresia. His MELD score was 21. He received a liver transplant from a 20-year-old African American male with the same blood group (B). The donor's preprocurement tests showed an AST level of 106 IU/L, an ALT level of 33 IU/L, and a total bilirubin level of 1.4 mg/dL. The cold ischemia time was 10 hours. Hepatocellular carcinoma was not detected in the explanted liver. T and B cell crossmatches were both strongly positive (Table 1). The PRA values were 100% for classes I and II. DSAs (A2, B3, B18, B58, DR14, DQ4, and DQ5) were negative for complement fixation.

The patient's postoperative course was complicated. His immunosuppression was switched from tacrolimus to cyclosporine because of a mental status change on day 3 (suspected tacrolimus toxicity). His ALT, AST, and total bilirubin levels initially decreased until day 5. On day 6, these parameters started rising again (Fig. 3). A biopsy sample on day 7 demonstrated lymphocyte-predominant mixed inflammation in the portal, bile duct, and venous endothelium (rejection activity index = 4/9) that was compatible with mild early acute cellular rejection. The hepatitis C viral load was 240,000 IU/mL. The cyclosporine A level was 295 ng/mL. Biopsy findings could not exclude recurrent hepatitis or preservation injury. Focal C4d staining was detected only in the portal veins, portal capillaries, bile duct, and sinusoids and not in the central vein. The overall findings were inconclusive for a diagnosis of AMR. The ALT, AST, and total bilirubin levels as well as the C4d staining were resolved (biopsy on day 34) after the administration of Solu-Medrol (a 250-mg bolus) 3 times. DSAs were not tested after transplantation. Recurrent acute cellular rejection in conjunction with recurrent chronic hepatitis C was also diagnosed by biopsy (on days 131 and 250), which detected significant ductitis and venulitis with lymphocyte predominant infiltration. Although the hepatic biochemical parameters initially responded to the treatment with Solu-Medrol, bilirubinemia gradually became persistent. Biopsy also detected cholangiolar proliferation (day 12), cholestasis (day 154), and ductopenia (day 345), and these findings were consistent with chronic rejection and sclerosing cholestatic hepatitis. Hepatocellular carcinoma was diagnosed on day 640, and the patient died on day 678.

Patient 5

The patient was a 48-year-old Caucasian female with end-stage liver disease secondary to autoimmune-induced cirrhosis. Her MELD score was 29. She received a transplant from a 27-year-old donor with the same blood group (A). The donor's preprocurement laboratory tests showed an AST level of 129 IU/L, an ALT level of 723 IU/L, and a total bilirubin level of 0.6 mg/dL. The cold ischemia time was 8 hours. The patient had a strongly positive B cell crossmatch and PRA values of 2% and 80% for classes I and II, respectively (Table 1). DSAs (DR4, DR53, and DQ8) were negative for complement. Her postoperative course after liver transplantation was uncomplicated (Fig. 3). DSAs were not detected on day 4. She was well 12 months after transplantation, and her liver function tests were stable with an AST level of 12 IU/L, an ALT level of 21 IU/L, and a total bilirubin level of 0.7 mg/dL.

DISCUSSION

Despite strong DSAs and strongly positive crossmatches, none of the 5 recipients died or lost his or her liver graft within the first year after transplantation. However, cholestasis and/or steroid-responsive early rejection episodes were observed in 3 of the 5 patients. It is intriguing that 2 of the 3 patients with a strongly positive T cell crossmatch and C1q-binding DSAs developed either cholestasis or early acute rejection (patients 1 and 3); the exception (patient 2) was treated with preemptive IVIG. The limitations of this retrospective case study include the small number of patients and the lack of a biopsy diagnosis. Biopsy either was not performed or was inconclusive. However, our cases highlight the potential usefulness of preformed DSA testing. We outline our DSA testing algorithm for liver transplant recipients and the difficulty of diagnosis and treatment in the clinical setting.

Preformed DSAs can cause acute and chronic liver allograft injuries.[23] The mechanism has been described as antibody-mediated microvascular injury, which leads to hepatocyte necrosis, cholangiolar proliferation, and cholestasis.[16] For HLA antibody–mediated rejection, a demonstration of DSAs is required. In addition to DSAs, a comprehensive evaluation of the evidence for graft dysfunction and biopsy findings including positive C4d staining are necessary.[16]

The clinical presentation of liver allograft AMR is nonspecific. Increases in AST, ALT, and cholestasis can also be caused by many different etiologies such as ischemic injury, initial graft dysfunction, acute cellular rejection, infections, medications (including azathioprine, cyclosporine, and tacrolimus), hepatitis, hepatic artery thrombosis, biliary complications, and disease recurrence.[24, 25] AMR should be considered as part of the differential diagnosis if DSAs are present.[6, 10, 15, 26] A biopsy finding of C4d staining can aid in the diagnosis of AMR and rule out other causes of graft dysfunction. In patient 1, a strong suspicion of AMR was triggered by the strongly positive crossmatch, the high levels of pretransplant DSAs (which were complement-fixing), the persistence of DSAs after transplantation, and bilirubinemia resistant to standard steroid boluses. The improvement in bilirubinemia after IVIG treatment was so rapid that liver biopsy was not performed, although there was a clinical suspicion of AMR. The DSAs also paralleled the clinical improvement. Retrospectively, a biopsy might have established the diagnosis of AMR. However, if the biopsy results had been inconclusive, we doubt that we would have withheld treatment from patient 1 in light of our strong clinical suspicion.

Acute cellular rejection often accompanies or follows AMR in biopsy findings.[10, 23] An increased risk for early and recurrent rejection has been reported for liver transplant recipients with a positive crossmatch.[13, 14, 27] Thus, there appears to be some shared immunological connection between AMR and cellular rejection. Patients 3 and 4 experienced acute rejection that responded to steroid treatment. In patient 4, there were repeated episodes of acute rejection, and cholestasis gradually became persistent within 3 months after transplantation. C4d staining in repeated biopsy samples was inconclusive, and the diagnosis of AMR was not made. Retrospectively, posttransplant DSA monitoring could have been helpful in evaluating the possibility of AMR. If DSAs had been persistent, it is possible that repeated cellular rejection, underlying AMR, and repeated steroid treatment may have contributed to a progression of fibrosis in a patient with recurrent hepatitis C virus.

Biopsy findings for AMR have been described as platelet aggregates in portal and/or central veins, neutrophilic exudation, patchy hepatocyte necrosis and centrilobular swelling, cholangiolar proliferation, and cholestasis.[16, 23] However, a biopsy diagnosis of liver allograft AMR is difficult. It resembles preservation injury, sepsis, and biliary/vascular complications.[23] Acute cellular rejection is frequently present and obscures the findings of AMR.[10] AMR can present as a phenotype of chronic rejection.[28] Thus, the application of C4d staining has been shown to aid in the diagnosis of AMR.[10, 15-17] C4d patterns have been reported to be linear/granular sinusoidal[15, 17] or diffuse portal.[10] However, a consensus interpretation of C4d staining is less established for liver allograft rejection versus kidney allograft rejection.[16, 29] C4d staining results can be inconsistent and depending on the staining methods.[17] Currently, there are no guidelines regarding the interpretation of negative/minimal/focal C4d staining in liver allograft biopsy findings. For kidney allograft AMR, the limitations of C4d staining as a diagnostic marker have been recognized, and a new concept of C4d-negative AMR has recently been under discussion.[30]

If we know that a preformed DSA level in a patient is considered clinically significant, the posttransplant management strategy may be adjusted to minimize the risk. However, a clinically significant level for liver transplantation has not yet been defined. We were interested in exploring such a level, and we studied our cases as well as cases reported in the literature. We focused on a T cell crossmatch that reflects the class I DSA level because B cell crossmatch reports are even more limited and complex to interpret.

CDC titers reported in association with graft loss include titers greater than 1:32,768,[31] titers greater than 1:64,[32] titers greater than 1:32,[9] a median titer of 1:16 (from 1:8 to 1:64),[33] and a median titer of 1:8[34] (Fig. 4). Although these cases are mostly from decades ago, such high CDC titers appear to have shown effects on allograft outcomes at least under the immunosuppression of that time. Flow cytometry crossmatch MCS values reported for AMR/graft dysfunction have included 366 ± 219 for 3 patients with AMR,[15] 343 for a case with refractory AMR,[35] and 383.5 ± 38.9 (patients 1-4). Overall, preformed DSAs with a CDC titer > 1:8 (approximately equivalent to an MCS of 300 in our laboratory) or with an MCS > 300 by flow cytometry crossmatching may indicate a significant risk. We are aware that the clinically meaningful threshold that predicts an increased risk in liver transplantation needs further investigation. Meanwhile, we have decided to keep our initial MCS threshold of 200 for a strongly positive crossmatch, and we have used it to screen patients at higher risk (Fig. 5). We have also applied the C1q assay to evaluate the complement-fixing ability of DSAs. The use of the C1q assay to evaluate the risk of rejection has been reported for renal[19] and heart patients[20] before transplantation.

Figure 4.

Crossmatch results and liver transplant outcomes. Crossmatch results [CDC titers and flow cytometry MCSs (1024-channel scale)] for pretransplant sera from reported cases and patients 1 to 5 are summarized. The approximate correlations between CDC crossmatches and flow crossmatches are based on our results.

Solid phase assay antibody testing is convenient because the assay, unlike crossmatch methods, does not depend on donor cell availability. Sensitization status screening and DSA detection in pretransplant and posttransplant settings are performed with relative ease without invasive procedures. However, we have to remember that the assays have not been validated or approved for quantitative assessment.[36] In particular, very high HLA antibody levels may result in a falsely low or negative result known as the prozone, which is attributable to the interference of immunoglobulin M,[22] complement component 1,[37] and antigen saturation. Despite the report of dithiothreitol serum treatment being used to reduce assay interference,[22] the DSA MFI does not necessarily correlate with titration, especially at high concentrations in our hands. Therefore, the evaluation of DSA levels with pretransplant DSA MFIs should be verified by comparisons to flow crossmatch MCS values, and posttransplant DSA MFIs may need to be evaluated with serum dilution. For this reason, we confirmed a decrease in the posttransplant DSA MFIs for patients 1, 2, and 5 by repeating the SAB assay after the dilution of both pretransplant and posttransplant serum samples (data not shown).

The disappearance of DSAs has been described as an important indicator of good outcomes.[14, 15, 38] This disappearance has been reported to occur as early as immediately[39] or 20 minutes after liver implantation.[32] Pretransplant DSAs (CDC titers ≤ 1:16) are more likely to become undetectable after transplantation.[9] Persistent or recurrent posttransplant DSAs are associated with adverse outcomes such as increases in acute rejection[14, 15, 39] and cellular rejection,[23] a higher rate of graft loss,[4] and lower patient survival.[1] Therefore, we repeated the testing of DSA levels within the first week after transplantation for patients 1, 2, and 5 (Fig. 3). If they persisted, the levels of DSAs were monitored weekly until their disappearance or a declining trend was confirmed. Rejection and graft loss in a setting of persistent DSAs have usually been reported from a few days after transplantation up to the first month.[34, 40-43] If persistent DSAs or graft dysfunction remains, biopsy should be considered.

Treatments with IVIG, rituximab, plasmapheresis, and bortezomib have been reported for AMR.[36] However, at this time, there are no Food and Drug Administration–approved treatments for AMR for liver transplant recipients, and no controlled trial has been conducted to evaluate the effectiveness of treatments. Thus, all the aforementioned treatments are considered options for each case of AMR. In addition to patient 1, patient 2 was also treated with IVIG because of persistent C1q-fixing DSAs. We cannot determine whether the IVIG treatment truly helped these patients. However, both patients 1 and 2 had good outcomes and did not suffer adverse effects from IVIG.

On the basis of this review of our own cases and other cases that have been published, we have outlined an HLA antibody test algorithm for liver transplant recipients (Fig. 5). Briefly, we screen pretransplant patients for HLA antibodies with solid phase testing. If HLA antibodies are present, we think that the crossmatch at the time of transplantation is important. If a strongly positive crossmatch (MCS > 200 by flow cytometry) is detected, we evaluate the DSA level with both the flow cytometry MCS and the Luminex MFI. If the crossmatch is strongly positive because of strong DSAs, we test the DSAs for complement fixation (with the C1q assay) and monitor the DSAs after transplantation. Strongly positive DSAs, C1q-positive DSAs, and persistent posttransplant DSAs indicate an increased risk for AMR, so biopsy should be considered in the presence of graft dysfunction.

Figure 5.

Test algorithm for a liver recipient with a strongly positive crossmatch. A strongly positive crossmatch with an MCS > 200 is tested for DSAs with a solid phase assay. A patient with strong DSAs and C1q-positive DSAs is considered at higher risk and is monitored for posttransplant DSAs. If persistent DSAs are detected, the patient is monitored as being at higher risk for AMR.

In conclusion, the evaluation of DSAs for their levels and complement-fixing ability in a pretransplant patient is potentially useful. If a patient has an increased risk for AMR, we should be vigilant, follow graft function and DSAs, perform biopsy in a timely fashion, and start treatment modalities if they are indicated.

Ancillary