- Email: [email protected]
Blood Group Lewis Alloantibodies Cause Antibody-Mediated Rejection in Renal Transplant Recipients
OUTCOMES
Blood Group Lewis Alloantibodies Cause Antibody-Mediated Rejection in Renal Transplant Recipients M. Boratyn´ska, M. Banasik, A. Hałon´, D. Patrzałek, and M. Klinger ABSTRACT Three patients with negative Lewis phenotypes who displayed anti-Lewis antibodies suffered severe kidney allograft dysfunction. One woman and two men (22– 44 years) received ABO compatible kidney transplants with negative donor-recipient cross-match tests. Two patients had the phenotype Le(a⫺b⫺) with anti-Le(a) and anti-Le(b) complement binding antibodies. The third patient of phenotype Le(a⫹b⫺) developed anti-Lewis(b) antibody a few months after transplantation. One patient presented recurrence of worsened graft function from the day 6 to 4 months after transplantation; despite treatment there was not full recovery. The second patient had recurrences of acute graft dysfunction at 4 and 6 months after transplantation with nephrotic range proteinuria. The third patient showed progressive graft dysfunction at 7 months after transplantation. Biopsy specimens showed histological changes of antibodymediated rejection. In the third patient, we observed fibrinoid necrosis and thrombosis of arterioles and glomerular capillaries. Immunofluorescence studies showed immunoglobulin IgG and IgM in glomerular capillaries and C4d and C3 on endothelial cells of peritubular capillaries. Posttransplantation cross-match tests with donor lymphocytes were negative. Anti-Lewis antibodies were observed during follow-up. All patients were treated with methylprednisolone boluses. In addition, one subject received antithymocyte globulin (ATG) and 1 received plasmapheresis. Two patients had moderate renal dysfunction (creatinine levels 1.8 and 1.9 mg/dL) after 8 –17 months follow-up. The third patient lost her graft at 11 months after transplantation. Lewis antibodies may injure a renal allograft. C4d deposition and failure to show donor-specific anti-HLA antibodies suggested the participation of other antibodies.
T
HE LEWIS blood group system is a set of histocompatibility antigens.1 It is present on the surface of various cells, eg., renal cortex, pancreas, stomach mucosa,
small and large intestine, skeletal muscle, and adrenal glands. Soluble forms of the antigens are in saliva, urine, and all body fluids except cerebrospinal fluid. There are
From the Department of Nephrology and Transplantation Medicine (M. Bo., M. Ba., A.H., M.K.), and Department of General, Vascular and Transplantation Surgery (D.P.), Wroclaw Medical University, Wroclaw, Poland.
Address reprint requests to Maria Boratyn´ska, Department of Nephrology and Transplantation Medicine, Wroclaw Medical University, Traugutta 57/59, 50-417 Wroclaw, Poland. E-mail: [email protected]
© 2007 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/07/$–see front matter doi:10.1016/j.transproceed.2007.08.053
Transplantation Proceedings, 39, 2711–2714 (2007)
2711
BORATYN´SKA, BANASIK, HAŁON´ ET AL
2712
Lewis antigens on lymphocytes, monocytes, and platelets.2– 4 The synthesis of Lewis glycans occurs in exocrine epithelial cells, mostly of endodermal origin.5 The digestive system is probably a major site of Lewis plasma glycolipid synthesis. The expression of Lewis epitopes on erythrocytes is dependent on adsorption from plasma of Lewis-bearing glycolipids. The biosynthesis of Lewis antigens results from the interaction of 2 independent loci, LE and SE. Recent molecular studies have revealed several point mutations in the FUT3 and SE loci. People with Le (a⫺b⫺) have point mutations (Gly170¡Ser and Asp336¡Ala) in the coding region of FUT3 that inactivate its fucosyltransferase product. The phenotype Le (a⫺b⫺) is uncommon occurring in 6% of whites and 22% of blacks. The Lewis (a⫹b⫺) phenotype is found in 22%–23% of both races. The most common phenotype is Lewis (a⫺b⫹), namely, 72% of whites and 55% of blacks. Anti-Le (a) and anti-Le (b) are frequent, naturally occurring antibodies made by Le(a⫺b⫺) people. Individuals with the Lewis (a⫹b⫺) phenotype may produce anti-Lewis (b) alloantibodies. All of these antibodies can be cytotoxic to lymphocytes. Anti-Le alloantibodies are more frequently of the immunoglobulin IgM than the IgG class and may bind complement. Lewis antibodies may play a role in graft rejection because inferior renal graft survival has been observed among patients lacking Lewis antigens.6 – 8 Herein we have presented 3 patients with negative Lewis phenotypes who displayed anti-Lewis antibodies and suffered severe kidney allograft rejection. MATERIALS AND METHODS Among 339 patients who underwent transplantation from January 2004 to December 2006, 2 white men and 1 woman displayed Lewis antibodies. The patients were compatible with their donors (whites) for ABO blood groups. The donor Lewis phenotype was unknown in all cases. All recipients were not sensitized to HLA antigens. Their panel reactive antibodies values during the waiting time as well as immediately before transplantation were 0%. Donor-recipient cross-match tests before transplantation were negative.
Case 1 A 22-year-old man of blood group phenotype A Rh⫹ DCw⫺ CCee, K⫺, Le (a⫺b⫺) displayed anti-Lewis (a) and anti-Lewis (b) antibodies that bound complement components C3c and C3d. His primary kidney disease was glomerulonephritis. He underwent kidney transplantation on November 27, 2005 from a 44-year old male deceased donor of blood group O Rh⫹ with 3 antigen mismatches (haplotype match). The patient was treated with tacrolimus (0.2 mg/kg/d), MMF (2-1 g/d), and prednisone (20 mg/d). The early posttransplantation follow-up was uneventful. The kidney graft functioned well initially; the patient was discharged with a serum creatinine level of 0.9 mg/dL. Four months later he was admitted to the hospital because of proteinuria (3– 4 g/d) and an increased creatinine level to 1.8 mg/dL. A transplant biopsy showed glomerulitis, neutrophilic interstitial infiltrate, and other features of antibody-mediated rejection (Table 1) but no cellular rejection. The patient was started on courses of plasmapheresis, initially daily, then every other day for 3 weeks. The serum creatinine level and proteinuria decreased. However, 6 months after transplantation the serum creatinine level increased to 3.2 mg/dL. Histological and immunohistochemical examinations of a second biopsy revealed mesangial glomerulonephritis with antibody-mediated rejection but no acute cellular rejection. He was treated with 3 500-mg doses of methylprednisolone and the prednisone dose was increased to 30 mg. His renal function improved. Currently, at 17 months after transplantation, the patient’s creatinine level is 1.4 –1.8 mg/dL.
Case 2 A 44-year-old man with blood group phenotype AB Rh⫹ DCw⫺ Ccee, K⫺, Le(a⫺b⫺) displayed anti-Lewis (a) and anti-Lewis (b) antibodies binding complement C3c and C3d. His primary kidney disease was polycystic kidney disease. He underwent transplantation on August 31, 2006. He received a kidney from a 53-year-old male deceased donor of blood group AB Rh⫹, with 3 mismatched HLA antigens (haplotype match). The initial immunosuppressive regimen consisted of CsA (6 mg/kg/d), MMF (2 g/d), and prednisone (30 mg/d). The kidney graft functioned immediately after transplantation, but from day 6 onward the serum creatinine level continued to increase reaching 3.9 mg/dL on day 8. A transplant biopsy showed features of antibody-mediated rejection with borderline cellular rejection according to Banff criteria. After treatment with boluses of methylprednisolone and conversion from CsA to tacrolimus, the creatinine level decreased to 1.9 mg/dL. Ten days
Table 1. The Outcomes of Histological and Immunohistological Examination of Renal Allograft Biopsy Specimens in Patients With Anti-Lewis Alloantibodies
Neutrophils in PTC Neutrophils in glomeruli Arterial fibrinoid necrosis Glomerulitis Thrombi in glomerular capillaries Thrombi in arterioles Acute tubular injury C4d in PTC C3 in glomeruli and vessels IgG in glomeruli and vessels IgM in glomeruli and vessels Abbreviation: PTC, peritubular capillaries.
Case 1; Biopsy 1
Case 1; Biopsy 2
Case 2; Biopsy 1
Case 2; Biopsy 2
Case 3; Biopsy 1
⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫹/⫺ ⫹ ⫹ ⫹
⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹
⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹
⫹ ⫹ ⫺ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹
⫹ ⫹ ⫹ ⫹ ⫹ ⫹ Necrosis ⫹ only in glomeruli ⫹ ⫹
BLOOD GROUP LEWIS ALLOANTIBODIES later the patient’s creatinine level again increased to 2.8 mg/dL. Additional boluses of methylprednisolone were administered and the creatinine level decreased to 2.0 mg/dL only to increase to 2.4 mg/dL 4 weeks later. A second biopsy revealed features of antibody-mediated rejection with deposits of complement C3 and C4d as well as IgG and IgM antibodies, similar to the first biopsy (Table 1). He underwent 8 days of anti-thymocyte globulin treatment. The creatinine decreased to 2.1 mg/dL and is currently 1.9 mg/dL.
Case 3 A 37-year-old woman with blood group phenotype B Rh⫺ dccee, Kk, Le(a⫹b⫺) received a 4 HLA antigen mismatched cadaveric kidney transplant on May 31, 2004. The donor was a 55-year-old man of blood group B Rh⫹. Lewis antibodies were not detected in the recipient before transplantation. The patient was treated with tacrolimus (0.2 mg/kg/d), MMF (1 g/d), and prednisone (20 mg/d tapering to 5 mg/d). The early posttransplantation follow-up was uneventful. The kidney graft functioned well with a serum creatinine level of 1.0 mg/dL. Seven months after transplantation, the creatinine level increased to 1.8 mg/dL and then to 2.6 mg/dL by the ninth month. The patient refused a transplant biopsy or antirejection treatment. Eleven months after transplantation the creatinine level was 11 mg/dL. Despite 3 500-mg doses of methylprednisolone, she developed end-stage renal failure and returned to hemodialysis. Histological examination of the graft showed fibrinoid necrosis with thrombosis of the arterioles and glomerular capillaries. Immunofluorescence studies showed deposits of IgG, IgM, and C3 in the glomeruli and blood vessels. A posttransplantation cross-match test with donor lymphocytes was negative. Anti-Lewis (b) antibodies were detected in the patient’s serum. Currently, almost 2 years after graftectomy, anti-Lewis (b) antibodies are still present.
DISCUSSION
Our study revealed that the presence of Lewis alloantibodies in patients who lacked the Lewis blood group system antigens were the cause of acute antibody-mediated rejection. Lewis-negative recipients may mount an immune response to Lewis antigens present in the renal cortex. The role of antibodies was demonstrated by the identification of C4d deposits on the peritubular capillary endothelium as well as C3, IgG, and IgM deposits in the glomeruli and blood vessels. Morphological examination revealed changes specific for humoral rejection; whereas there were hardly any signs of cellular rejection. The major changes in the biopsy specimens were glomerulitis, changes in blood vessels, and presence of granulocytes in the glomerular and peritubular capillaries. In the most severe, case 3, anti-Lewis (b) antibody developed a few months after transplantation, causing fibrinoid necrosis and thrombosis of the arterioles and glomerular capillaries. No specific anti-donor alloantibodies were observed in the sera of the patients after transplantation. However, anti-Le (a) and anti-Le (b) antibodies that bound complement C3c and C3d were observed in 2 patients and anti-Lewis (b) in case 3. These antibodies were detected before transplantation in 2 patients and were still observed at 8 and 17 months after transplantation, respectively. In case 3, no anti-Lewis (b)
2713
antibodies were found before transplantation, but they were identified during the period of impaired renal function. It is assumed that they developed in response to exposure to Lewis (b⫹) epitopes in the graft. Although the Lewis phenotypes of the donors are unknown, they were most probably positive (a⫺b⫹), which is the most common one among whites. Previous studies showed that Lewis-incompatible renal allografts displayed poorer graft survival than those matched for the Lewis system.6 – 8 Spitalnick et al described 8 Lewis-negative patients who received renal grafts from Lewis-positive donors; all patients experienced acute graft rejection that required resuming hemodialysis. Anti-Lewis antibodies were detected in all patients.6 Similar studies by Fischer et al, indicating that Lewis antigens are important for transplant prognosis, suggested to consider Lewis compatibility in recipient selection.8 Opposite outcomes have been reported by other researchers.9,10 A retrospective study by Gratama et al, which included 1111 patients receiving a first cadaveric kidney graft, did not observe a difference among 1-year graft survival rates of Lewis-negative (73%) versus Lewis-positive (73%) recipients.9 They suggested that Lewis-negative patients were at higher risk of graft failure only when receiving a poorly HLA-matched kidney. Data presented by Posner et al also showed no significant difference in graft survival between Lewis-matched and Lewis-mismatched donors and recipients. They did not recommend using Lewis compatibility for donor selection in renal transplantation.10 The disparate outcomes of these studies may have resulted from the latter group comparing Lewis phenotypes, whereas the former studied both the phenotype and the presence of anti-Lewis antibodies. The Lewis-negative phenotype may not occur uniquely as a result of an inactive allele, but may result from other causes, such as a failure of adsorption to the erythrocyte or the presence of an allele whose product is less active or may be selectively inactive toward glycolipid precursor substrates. Therefore, the Lewis-negative phenotype (usually assayed on erythrocytes) does not always identify cases producing anti-Lewis antibodies. Only individuals with an inactive allele can produce alloantibodies that are pathogenic factors in humoral rejection. Lewis epitopes are located particularly on epithelial cells of tubules, but may be protein-bound in other parts of kidney tissue, eg, in the glomeruli and blood vessels. Such an epitope distribution may explain the morphological and immunohistochemical findings intrinsic to antibody-mediated rejection in our recipients. Anti-Lewis antibodies present before and after transplantation seemed to be responsible for the recurrence and treatment resistance of renal allograft dysfunction. Improvement and stabilization of renal function occurred at 6 to 7 months after transplantation, suggesting the development of “accommodation,” which has been observed among transplantation, across incompatible ABO blood types.11 The development of anti-Lewis antibodies after transplan-
2714
tation led to terminal graft failure in a patient who did not undergo early treatment. Roy et al reported poor outcomes of retransplantation among Lewis(⫺) patients who had rejected Lewis(⫹) grafts.12 If such patients are not matched for Lewis phenotype, they will require elimination of antiLewis antibodies before and after retransplantation. In conclusion, Lewis antibodies may injure a renal allograft by a mechanism of antibody-mediated rejection. C4d deposition and failure to show donor-specific anti-HLA antibodies suggest the participation of other antibodies in allograft injuries. Compatibility for the Lewis system should be considered in the retransplantation of patients with anti-Lewis antibodies. ACKNOWLEDGMENTS We would like to thank Mrs. Barbara Wrzaszczyk for her help in assessing the Lewis phenotypes and antibodies.
REFERENCES 1. Oriol R, Cartron J, Yvart J, et al: The Lewis system: new histocompatibility in renal transplantation. Lancet 18:574, 1978 2. Cordon-Cardo C, Lloyd KO, Finstad CL, et al: Immunoanatomic distribution of blood group antigens in renal transplantation. Lab Invest 55:444, 1986
BORATYN´SKA, BANASIK, HAŁON´ ET AL 3. Henry S, Oriol R, Samuelsson B: Lewis histo-blood group system and associated secretory phenotypes. Vox Sang 69:166, 1995 4. Henry SM, Joval PA, Ghardashkhani ML, et al: Structural and immunochemical identification of Leb glycolipids in the plasma of a group O Le(a⫺b⫺) secretor. Glycoconj J 12:309, 1995 5. Hauser R: Le(a) and Le(b) tissue glycosphingolipids. Transfusion 35:577, 1995 6. Spitalnik S, Ptaff W, Cowles J, et al: Correlation of humoral immunity to Lewis blood group antigens with renal transplant rejection. Transplantation 37:265, 1984 7. Oriol R, Opelz G, Chun C, et al: The Lewis system and kidney transplantation. Transplantation 29:397, 1980 8. Fischer E, Lenhard V, Romer W, et al: Influence of Lewis blood group system on clinical kidney transplantation. Proc Eur Dial Transplant Assoc 16:377, 1979 9. Gratama JW, Hendriks GF, Persijn GG, et al: The interaction between Lewis blood group system and HLA-matching in renal transplantation. Transplantation 45:926, 1988 10. Posner MP, McGregore MB, Mendez-Picon G, et al: The importance of the Lewis system in cadaver renal transplantation. Transplantation 41:474, 1986 11. Takahashi K: A new concept of accommodation in ABOincompatible kidney transplantation. Clin Transplant 19(suppl 14):76, 2005 12. Roy R, Terasaki PI, Chia D, et al: Low kidney graft survival in Lewis negative patients after regrafting and newer matching schemes for Lewis. Transplant Proc 19:4498, 1987
Blood Group Lewis Alloantibodies Cause Antibody-Mediated Rejection in Renal Transplant Recipients M. Boratyn´ska, M. Banasik, A. Hałon´, D. Patrzałek, and M. Klinger ABSTRACT Three patients with negative Lewis phenotypes who displayed anti-Lewis antibodies suffered severe kidney allograft dysfunction. One woman and two men (22– 44 years) received ABO compatible kidney transplants with negative donor-recipient cross-match tests. Two patients had the phenotype Le(a⫺b⫺) with anti-Le(a) and anti-Le(b) complement binding antibodies. The third patient of phenotype Le(a⫹b⫺) developed anti-Lewis(b) antibody a few months after transplantation. One patient presented recurrence of worsened graft function from the day 6 to 4 months after transplantation; despite treatment there was not full recovery. The second patient had recurrences of acute graft dysfunction at 4 and 6 months after transplantation with nephrotic range proteinuria. The third patient showed progressive graft dysfunction at 7 months after transplantation. Biopsy specimens showed histological changes of antibodymediated rejection. In the third patient, we observed fibrinoid necrosis and thrombosis of arterioles and glomerular capillaries. Immunofluorescence studies showed immunoglobulin IgG and IgM in glomerular capillaries and C4d and C3 on endothelial cells of peritubular capillaries. Posttransplantation cross-match tests with donor lymphocytes were negative. Anti-Lewis antibodies were observed during follow-up. All patients were treated with methylprednisolone boluses. In addition, one subject received antithymocyte globulin (ATG) and 1 received plasmapheresis. Two patients had moderate renal dysfunction (creatinine levels 1.8 and 1.9 mg/dL) after 8 –17 months follow-up. The third patient lost her graft at 11 months after transplantation. Lewis antibodies may injure a renal allograft. C4d deposition and failure to show donor-specific anti-HLA antibodies suggested the participation of other antibodies.
T
HE LEWIS blood group system is a set of histocompatibility antigens.1 It is present on the surface of various cells, eg., renal cortex, pancreas, stomach mucosa,
small and large intestine, skeletal muscle, and adrenal glands. Soluble forms of the antigens are in saliva, urine, and all body fluids except cerebrospinal fluid. There are
From the Department of Nephrology and Transplantation Medicine (M. Bo., M. Ba., A.H., M.K.), and Department of General, Vascular and Transplantation Surgery (D.P.), Wroclaw Medical University, Wroclaw, Poland.
Address reprint requests to Maria Boratyn´ska, Department of Nephrology and Transplantation Medicine, Wroclaw Medical University, Traugutta 57/59, 50-417 Wroclaw, Poland. E-mail: [email protected]
© 2007 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/07/$–see front matter doi:10.1016/j.transproceed.2007.08.053
Transplantation Proceedings, 39, 2711–2714 (2007)
2711
BORATYN´SKA, BANASIK, HAŁON´ ET AL
2712
Lewis antigens on lymphocytes, monocytes, and platelets.2– 4 The synthesis of Lewis glycans occurs in exocrine epithelial cells, mostly of endodermal origin.5 The digestive system is probably a major site of Lewis plasma glycolipid synthesis. The expression of Lewis epitopes on erythrocytes is dependent on adsorption from plasma of Lewis-bearing glycolipids. The biosynthesis of Lewis antigens results from the interaction of 2 independent loci, LE and SE. Recent molecular studies have revealed several point mutations in the FUT3 and SE loci. People with Le (a⫺b⫺) have point mutations (Gly170¡Ser and Asp336¡Ala) in the coding region of FUT3 that inactivate its fucosyltransferase product. The phenotype Le (a⫺b⫺) is uncommon occurring in 6% of whites and 22% of blacks. The Lewis (a⫹b⫺) phenotype is found in 22%–23% of both races. The most common phenotype is Lewis (a⫺b⫹), namely, 72% of whites and 55% of blacks. Anti-Le (a) and anti-Le (b) are frequent, naturally occurring antibodies made by Le(a⫺b⫺) people. Individuals with the Lewis (a⫹b⫺) phenotype may produce anti-Lewis (b) alloantibodies. All of these antibodies can be cytotoxic to lymphocytes. Anti-Le alloantibodies are more frequently of the immunoglobulin IgM than the IgG class and may bind complement. Lewis antibodies may play a role in graft rejection because inferior renal graft survival has been observed among patients lacking Lewis antigens.6 – 8 Herein we have presented 3 patients with negative Lewis phenotypes who displayed anti-Lewis antibodies and suffered severe kidney allograft rejection. MATERIALS AND METHODS Among 339 patients who underwent transplantation from January 2004 to December 2006, 2 white men and 1 woman displayed Lewis antibodies. The patients were compatible with their donors (whites) for ABO blood groups. The donor Lewis phenotype was unknown in all cases. All recipients were not sensitized to HLA antigens. Their panel reactive antibodies values during the waiting time as well as immediately before transplantation were 0%. Donor-recipient cross-match tests before transplantation were negative.
Case 1 A 22-year-old man of blood group phenotype A Rh⫹ DCw⫺ CCee, K⫺, Le (a⫺b⫺) displayed anti-Lewis (a) and anti-Lewis (b) antibodies that bound complement components C3c and C3d. His primary kidney disease was glomerulonephritis. He underwent kidney transplantation on November 27, 2005 from a 44-year old male deceased donor of blood group O Rh⫹ with 3 antigen mismatches (haplotype match). The patient was treated with tacrolimus (0.2 mg/kg/d), MMF (2-1 g/d), and prednisone (20 mg/d). The early posttransplantation follow-up was uneventful. The kidney graft functioned well initially; the patient was discharged with a serum creatinine level of 0.9 mg/dL. Four months later he was admitted to the hospital because of proteinuria (3– 4 g/d) and an increased creatinine level to 1.8 mg/dL. A transplant biopsy showed glomerulitis, neutrophilic interstitial infiltrate, and other features of antibody-mediated rejection (Table 1) but no cellular rejection. The patient was started on courses of plasmapheresis, initially daily, then every other day for 3 weeks. The serum creatinine level and proteinuria decreased. However, 6 months after transplantation the serum creatinine level increased to 3.2 mg/dL. Histological and immunohistochemical examinations of a second biopsy revealed mesangial glomerulonephritis with antibody-mediated rejection but no acute cellular rejection. He was treated with 3 500-mg doses of methylprednisolone and the prednisone dose was increased to 30 mg. His renal function improved. Currently, at 17 months after transplantation, the patient’s creatinine level is 1.4 –1.8 mg/dL.
Case 2 A 44-year-old man with blood group phenotype AB Rh⫹ DCw⫺ Ccee, K⫺, Le(a⫺b⫺) displayed anti-Lewis (a) and anti-Lewis (b) antibodies binding complement C3c and C3d. His primary kidney disease was polycystic kidney disease. He underwent transplantation on August 31, 2006. He received a kidney from a 53-year-old male deceased donor of blood group AB Rh⫹, with 3 mismatched HLA antigens (haplotype match). The initial immunosuppressive regimen consisted of CsA (6 mg/kg/d), MMF (2 g/d), and prednisone (30 mg/d). The kidney graft functioned immediately after transplantation, but from day 6 onward the serum creatinine level continued to increase reaching 3.9 mg/dL on day 8. A transplant biopsy showed features of antibody-mediated rejection with borderline cellular rejection according to Banff criteria. After treatment with boluses of methylprednisolone and conversion from CsA to tacrolimus, the creatinine level decreased to 1.9 mg/dL. Ten days
Table 1. The Outcomes of Histological and Immunohistological Examination of Renal Allograft Biopsy Specimens in Patients With Anti-Lewis Alloantibodies
Neutrophils in PTC Neutrophils in glomeruli Arterial fibrinoid necrosis Glomerulitis Thrombi in glomerular capillaries Thrombi in arterioles Acute tubular injury C4d in PTC C3 in glomeruli and vessels IgG in glomeruli and vessels IgM in glomeruli and vessels Abbreviation: PTC, peritubular capillaries.
Case 1; Biopsy 1
Case 1; Biopsy 2
Case 2; Biopsy 1
Case 2; Biopsy 2
Case 3; Biopsy 1
⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫹/⫺ ⫹ ⫹ ⫹
⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹
⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹
⫹ ⫹ ⫺ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹
⫹ ⫹ ⫹ ⫹ ⫹ ⫹ Necrosis ⫹ only in glomeruli ⫹ ⫹
BLOOD GROUP LEWIS ALLOANTIBODIES later the patient’s creatinine level again increased to 2.8 mg/dL. Additional boluses of methylprednisolone were administered and the creatinine level decreased to 2.0 mg/dL only to increase to 2.4 mg/dL 4 weeks later. A second biopsy revealed features of antibody-mediated rejection with deposits of complement C3 and C4d as well as IgG and IgM antibodies, similar to the first biopsy (Table 1). He underwent 8 days of anti-thymocyte globulin treatment. The creatinine decreased to 2.1 mg/dL and is currently 1.9 mg/dL.
Case 3 A 37-year-old woman with blood group phenotype B Rh⫺ dccee, Kk, Le(a⫹b⫺) received a 4 HLA antigen mismatched cadaveric kidney transplant on May 31, 2004. The donor was a 55-year-old man of blood group B Rh⫹. Lewis antibodies were not detected in the recipient before transplantation. The patient was treated with tacrolimus (0.2 mg/kg/d), MMF (1 g/d), and prednisone (20 mg/d tapering to 5 mg/d). The early posttransplantation follow-up was uneventful. The kidney graft functioned well with a serum creatinine level of 1.0 mg/dL. Seven months after transplantation, the creatinine level increased to 1.8 mg/dL and then to 2.6 mg/dL by the ninth month. The patient refused a transplant biopsy or antirejection treatment. Eleven months after transplantation the creatinine level was 11 mg/dL. Despite 3 500-mg doses of methylprednisolone, she developed end-stage renal failure and returned to hemodialysis. Histological examination of the graft showed fibrinoid necrosis with thrombosis of the arterioles and glomerular capillaries. Immunofluorescence studies showed deposits of IgG, IgM, and C3 in the glomeruli and blood vessels. A posttransplantation cross-match test with donor lymphocytes was negative. Anti-Lewis (b) antibodies were detected in the patient’s serum. Currently, almost 2 years after graftectomy, anti-Lewis (b) antibodies are still present.
DISCUSSION
Our study revealed that the presence of Lewis alloantibodies in patients who lacked the Lewis blood group system antigens were the cause of acute antibody-mediated rejection. Lewis-negative recipients may mount an immune response to Lewis antigens present in the renal cortex. The role of antibodies was demonstrated by the identification of C4d deposits on the peritubular capillary endothelium as well as C3, IgG, and IgM deposits in the glomeruli and blood vessels. Morphological examination revealed changes specific for humoral rejection; whereas there were hardly any signs of cellular rejection. The major changes in the biopsy specimens were glomerulitis, changes in blood vessels, and presence of granulocytes in the glomerular and peritubular capillaries. In the most severe, case 3, anti-Lewis (b) antibody developed a few months after transplantation, causing fibrinoid necrosis and thrombosis of the arterioles and glomerular capillaries. No specific anti-donor alloantibodies were observed in the sera of the patients after transplantation. However, anti-Le (a) and anti-Le (b) antibodies that bound complement C3c and C3d were observed in 2 patients and anti-Lewis (b) in case 3. These antibodies were detected before transplantation in 2 patients and were still observed at 8 and 17 months after transplantation, respectively. In case 3, no anti-Lewis (b)
2713
antibodies were found before transplantation, but they were identified during the period of impaired renal function. It is assumed that they developed in response to exposure to Lewis (b⫹) epitopes in the graft. Although the Lewis phenotypes of the donors are unknown, they were most probably positive (a⫺b⫹), which is the most common one among whites. Previous studies showed that Lewis-incompatible renal allografts displayed poorer graft survival than those matched for the Lewis system.6 – 8 Spitalnick et al described 8 Lewis-negative patients who received renal grafts from Lewis-positive donors; all patients experienced acute graft rejection that required resuming hemodialysis. Anti-Lewis antibodies were detected in all patients.6 Similar studies by Fischer et al, indicating that Lewis antigens are important for transplant prognosis, suggested to consider Lewis compatibility in recipient selection.8 Opposite outcomes have been reported by other researchers.9,10 A retrospective study by Gratama et al, which included 1111 patients receiving a first cadaveric kidney graft, did not observe a difference among 1-year graft survival rates of Lewis-negative (73%) versus Lewis-positive (73%) recipients.9 They suggested that Lewis-negative patients were at higher risk of graft failure only when receiving a poorly HLA-matched kidney. Data presented by Posner et al also showed no significant difference in graft survival between Lewis-matched and Lewis-mismatched donors and recipients. They did not recommend using Lewis compatibility for donor selection in renal transplantation.10 The disparate outcomes of these studies may have resulted from the latter group comparing Lewis phenotypes, whereas the former studied both the phenotype and the presence of anti-Lewis antibodies. The Lewis-negative phenotype may not occur uniquely as a result of an inactive allele, but may result from other causes, such as a failure of adsorption to the erythrocyte or the presence of an allele whose product is less active or may be selectively inactive toward glycolipid precursor substrates. Therefore, the Lewis-negative phenotype (usually assayed on erythrocytes) does not always identify cases producing anti-Lewis antibodies. Only individuals with an inactive allele can produce alloantibodies that are pathogenic factors in humoral rejection. Lewis epitopes are located particularly on epithelial cells of tubules, but may be protein-bound in other parts of kidney tissue, eg, in the glomeruli and blood vessels. Such an epitope distribution may explain the morphological and immunohistochemical findings intrinsic to antibody-mediated rejection in our recipients. Anti-Lewis antibodies present before and after transplantation seemed to be responsible for the recurrence and treatment resistance of renal allograft dysfunction. Improvement and stabilization of renal function occurred at 6 to 7 months after transplantation, suggesting the development of “accommodation,” which has been observed among transplantation, across incompatible ABO blood types.11 The development of anti-Lewis antibodies after transplan-
2714
tation led to terminal graft failure in a patient who did not undergo early treatment. Roy et al reported poor outcomes of retransplantation among Lewis(⫺) patients who had rejected Lewis(⫹) grafts.12 If such patients are not matched for Lewis phenotype, they will require elimination of antiLewis antibodies before and after retransplantation. In conclusion, Lewis antibodies may injure a renal allograft by a mechanism of antibody-mediated rejection. C4d deposition and failure to show donor-specific anti-HLA antibodies suggest the participation of other antibodies in allograft injuries. Compatibility for the Lewis system should be considered in the retransplantation of patients with anti-Lewis antibodies. ACKNOWLEDGMENTS We would like to thank Mrs. Barbara Wrzaszczyk for her help in assessing the Lewis phenotypes and antibodies.
REFERENCES 1. Oriol R, Cartron J, Yvart J, et al: The Lewis system: new histocompatibility in renal transplantation. Lancet 18:574, 1978 2. Cordon-Cardo C, Lloyd KO, Finstad CL, et al: Immunoanatomic distribution of blood group antigens in renal transplantation. Lab Invest 55:444, 1986
BORATYN´SKA, BANASIK, HAŁON´ ET AL 3. Henry S, Oriol R, Samuelsson B: Lewis histo-blood group system and associated secretory phenotypes. Vox Sang 69:166, 1995 4. Henry SM, Joval PA, Ghardashkhani ML, et al: Structural and immunochemical identification of Leb glycolipids in the plasma of a group O Le(a⫺b⫺) secretor. Glycoconj J 12:309, 1995 5. Hauser R: Le(a) and Le(b) tissue glycosphingolipids. Transfusion 35:577, 1995 6. Spitalnik S, Ptaff W, Cowles J, et al: Correlation of humoral immunity to Lewis blood group antigens with renal transplant rejection. Transplantation 37:265, 1984 7. Oriol R, Opelz G, Chun C, et al: The Lewis system and kidney transplantation. Transplantation 29:397, 1980 8. Fischer E, Lenhard V, Romer W, et al: Influence of Lewis blood group system on clinical kidney transplantation. Proc Eur Dial Transplant Assoc 16:377, 1979 9. Gratama JW, Hendriks GF, Persijn GG, et al: The interaction between Lewis blood group system and HLA-matching in renal transplantation. Transplantation 45:926, 1988 10. Posner MP, McGregore MB, Mendez-Picon G, et al: The importance of the Lewis system in cadaver renal transplantation. Transplantation 41:474, 1986 11. Takahashi K: A new concept of accommodation in ABOincompatible kidney transplantation. Clin Transplant 19(suppl 14):76, 2005 12. Roy R, Terasaki PI, Chia D, et al: Low kidney graft survival in Lewis negative patients after regrafting and newer matching schemes for Lewis. Transplant Proc 19:4498, 1987