Kim 3845 r1.qxd:layout

Kim 3845 r1.qxd:Layout 1 25/1/10 10:02 Page 124 Anti α-Gal Immune Response Following Porcine Bioprosthesis Implantation in ChildrenChun Soo Park1, Seong-Sik Park2, Sun Young Choi2, Sun Hee Yoon1, Woong-Han Kim1, Yong Jin Kim1,2 1Department of Thoracic and Cardiovascular Surgery, Seoul National University College of Medicine, 2Seoul NationalUniversity Hospital Clinical Research Institute, Xenotransplantation Research Center, Seoul, Republic of Korea Background and aim of the study: Porcine bioprosthe-
tively. All serum samples were analyzed using an
ses have been widely used in cardiac surgery in the
enzyme-linked immunosorbent assay to investigate
treatment of valvular heart disease. However, in
the α-Gal immune response.
younger patients, their use has been limited by early
Results: There were no operative deaths or complica-
failures known to be associated with an immune
tions. There was no statistically significant differ-
response and subsequent degeneration. The natural
ence between the titers of anti α-Gal antibodies of
antibodies directed at Galα1, 3-Galβ1-4GlcNAc-R(α-
the PVR and re-PVR groups. The titer of anti α-Gal
Gal), have been thought to initiate an immune
antibodies (IgM and IgG) was decreased on the first
response in humans transplanted with porcine organ
postoperative day, but increased in the first postop-
xenografts. The study aim was to determine the anti
erative week, regardless of the isotype. Whilst the
α-Gal immune response following commercial
titer of the anti α-Gal IgM antibody began to
porcine bioprosthesis implantation in children.
decrease after three weeks postoperatively, the titer
Methods: Between January 2008 and April 2008, 19
of anti α-Gal IgG antibody remained increased after
consecutive patients underwent pulmonary valve
two months.
replacement (PVR) with a commercially available
Conclusion: The implantation of a porcine biopros-
porcine bioprosthesis for an incompetent pulmonary
thesis elicits the increased formation of anti α-Gal
valve with congenital heart diseases. The median age
antibodies during the early postoperative period in
at surgery was 132 months (range: 14-330 months).
children, with different patterns between the two
Previous PVR with a porcine bioprosthesis had been
isotypes. The IgM antibody response was rapid and
performed in seven patients at a median of 44
transient, while the IgG antibody response was
months (range: 26-117 months) before surgery (re-
longer and more delayed.
PVR group). Sera were obtained sequentially five
times: immediately before surgery, and at one day,
one week, three weeks, and two months postopera-
The Journal of Heart Valve Disease 2010;19:124-130 Xenoreactive natural antibodies directed at the designated to the α-Gal epitope in humans, and Galα1, 3-Galβ1-4GlcNAc-R (α-Gal), epitope are known although anti α-Gal antibodies are produced through- to be a major barrier in xenotransplantation. The natu- out the patient’s lifetime due to exposure to the gut ral anti-pig antibodies in human serum react predomi- flora (3), the titers of natural antibodies show individ- ual differences (4). In xenotransplantation, anti α-Gal α-Gal-xenoreactive antibodies comprise at least 80- antibodies mediate the hyperacute rejection with com- 90% of anti-pig antibodies in humans (2).
plement activation, and delayed or chronic rejection Approximately 1% of the circulating antibodies are with an antibody-dependent cellular cytotoxicitymechanism (3,5-7).
The α-Gal epitope is expressed on a variety of tissues in pigs, including the cardiac valve endothelium (1,8- Yong Jin Kim MD, PhD, Department of Thoracic and Cardiovascular 10); recently, the presence of the α-Gal epitope was Surgery, Seoul National University College of Medicine, SeoulNational University Hospital Clinical Research Institute, documented in commercial bioprosthetic valves treat- Xenotransplantation Research Center, 28 Yongon-dong, Jongno-gu, Many patients who undergo procedures on the right Kim 3845 r1.qxd:Layout 1 25/1/10 10:02 Page 125 Anti α-Gal immune response 125 Table I: Patient preoperative characteristics. *Diagnosis: AORPA: Anomalous origin of right pulmonary artery; cc-TGA: Congenitally corrected TGA; PA: Pulmonaryatresia; PS: Pulmonary stenosis; PVR: Pulmonary valve replacement; TGA: Transposition of the great arteries; TOF: Tetralogy of Fallot; VSD: Ventricular septal defect.
+Preoperative medication: ADT: Aldactone; DCZ: Diclozide; DGX: Digoxin; ENL: Enalapril.
‡Valve: CE: Carpentier-Edwards Bioprosthetic Valved Conduit Model 4300 (Edwards Lifesciences); H: Hancock® II Aortic,Medtronic™; SJM: Epic™ SUPRA valve Aortic, St. Jude Medical, Inc.
ventricular outflow tract require pulmonary valve increased formation of anti α-Gal antibodies; and (ii) replacement (PVR), with or without a conduit.
monitor perioperative changes in the titer of the anti α- Although, in most centers (including that of the pres- Gal antibodies (IgM and IgG) over time.
ent authors) the prosthesis of choice for PVR is ofporcine origin, such bioprostheses have a limited dura- Clinical material and methods
bility due to valve degeneration that occurs more fre-quently in younger patients (8,9,12-14). The Patients
degeneration of bioprostheses is multifactorial, and Between January 2008 and April 2008, 19 consecutive includes immunologic reactions, foreign body reac- patients underwent PVR with porcine bioprostheses tions, blood-surface interactions, chemical factors, produced by three manufacturers: the EpicTM SUPRA infection, mechanical factors, material fatigue, and sur- valve aortic (St. Jude Medical, Inc.; the HANCOCK® II gical factors (15). Although the mechanism leading to aortic (Medtronic); and the Carpentier-Edwards premature degeneration of the implanted bioprosthe- Bioprosthetic valved conduit Model 4300 (Edwards sis is not yet fully understood, the immune response Lifesciences). The patients’ diagnoses were tetralogy of has been considered to play an important role as an Fallot or variants thereof in 12 cases, truncus arteriosus initial trigger of the degeneration process (16), and α- in three, congenitally corrected transposition of the Gal has been noted as the major epitope.
great arteries with pulmonary stenosis in two com- The aim of the present study was to: (i) investigate plete transposition of the great arteries with pul- whether bioprosthesis implantation elicits an monary stenosis in one, and anomalous origin or right Kim 3845 r1.qxd:Layout 1 25/1/10 10:02 Page 126 126 Anti α-Gal immune response Figure 1: Changes in anti α-Gal activity in patient #14 (PVR group) and patient #17 (re-PVR group). PVR: Pulmonary valve replacement; OD: Optical density; Figure 2: Changes in the anti α-Gal antibody titer in the individual groups. A) Re-PVR IgM titer. B) Re-PVR IgG titer. C) PVR IgM titer. D) PVR IgG titer. PVR: pulmonary artery from the ascending aorta each in one Pulmonary valve replacement; POD: Postoperative day. case. A previous PVR with a porcine bioprosthesis hadbeen performed in seven patients (the re-PVR group)at a median of 44 months (range: 26 to 117 months) before surgery, while the remaining 12 patients under- An ELISA analysis was used to determine the went PVR for the first time (the PVR group). The pre- activity of IgM and IgG isotypes of the anti α-Gal anti- operative characteristics of the patients are listed in bodies. Bovine serum albumin (BSA) containing syn- Table I. The median age at surgery was 132 months thesized α-Gal (α-Gal-BSA), which was prepared by conjugation of α-Gal linker type 1 (Genkem, Postoperatively, all patients followed a similar med- Seoul, Korea) and bovine serum albumin (Armresco, ication protocol. Serum samples were obtained five Solon, OH, USA) with 1-ethyl-3-[3-dimethylamino- times from each patient: immediately before surgery, propyl]carbodiimide hydrochloride (Pierce, USA) in at postoperative day 1, during the first and third post- conjugation buffer (MES buffered saline pack; Pierce), operative weeks, and during the second postoperative was used as a solid-phase antigen. Microtiter plates month. All sera were stored in EDTA tubes and ana- were coated with 100 μl per well of α-Gal-BSA in PBS lyzed within one day of being obtained.
buffer (pH 7.4) (at a concentration of 1 μg/ml for the The study was approved by the institutional review IgM isotype, or 2.5 μg/ml for the IgG isotype), and board/ethical committee of the authors’ institution, incubated for 1 h at 37°C. The plates were then washed and informed consent was obtained from each of the with deionized water. Aliquots of the patient’s serum patients. The study was conducted in accordance with (100 μl per well) were added to the α-Gal-BSA-immo- bilized wells at a serial two-fold dilution, from 1:40 to1: 2,560 in BSA-Triton X-100 (pH 7.4, PBS, 3% BSA, Operative technique
0.01% Triton X-100). The plates were then incubated In all patients, PVR was performed using moderate for 1 h at 37°C. Donkey anti-human IgG and IgM anti- hypothermic cardiopulmonary bypass (CPB) and left bodies (Jackson, Human Research Laboratories Inc.) atrial venting through the right upper pulmonary vein, were used as a secondary antibody at a dilution of with or without myocardial ischemia, under ventricu- 1:15,000 for IgM and 1:10,000 for IgG in BSA-Triton X- lar fibrillation for isolated PVR (n = 10) and after 100. The optical density (OD) was measured at 450 nm release of the aortic clamp in patients with additional using the Thermo Electron-Lab Systems (Multiskan Kim 3845 r1.qxd:Layout 1 25/1/10 10:02 Page 127 Anti α-Gal immune response 127 Table II: Anti α-Gal titers in the two groups.* Values are mean ± SD.
*Titer defined as the dilution at 1.0 of the OD.
PVR: Pulmonary valve replacement; POD: Postoperative day.
Statistical analysis
immediately postoperatively (at one day) and then Background OD-values were subtracted from the increased after one week, when compared to the pre- test OD-values. The titer was defined as the dilution at operative OD, regardless of the isotype. The ODs of the 1.0 of the OD. All data were presented as mean ± SD, IgG isotype remained elevated at the end of the study or median plus range, using Microsoft Excel 2007.
(at 60 days after surgery), but those of the IgM isotype Comparisons were performed using either Student’s t- began to decrease at three weeks postoperatively.
test or a paired t-test. A p-value <0.05 was considered Although the postoperative titer seemed to be high- er in the re-PVR group (Table II; Fig. 2), there was nosignificant difference between the two groups (PVR and re-PVR) with regards to the anti α-Gal IgM andIgG titers (Table II). The change in titer of the anti α- Clinical results
Gal antibodies with time was similar in each patient No early deaths occurred among the patients. The (see Fig. 1). The immediate postoperative (one-day) median CPB time was 167 min (range: 94 to 392 min), titer appeared to decrease, though not statistically sig- while in nine patients aortic cross-clamping was nificantly (Table III; Fig. 3). At one week after surgery required for additional intracardiac procedures. The the titer increased significantly, up to seven-fold in median myocardial ischemic time was 108 min (range: IgM and 32-fold in IgG (Table III; Fig. 3). The IgM titer 55 to 140 min). In all patients, the preoperative prepa- began to decrease at three weeks after surgery, but that ration, anesthetic management and operative strategy of IgG increased from three weeks and was maintained were similar and the hospital course was uneventful, until the end of the study (at two months after surgery) and all were discharged within 10 days after surgery, without complications. Treatment with an angiotensin-converting enzyme inhibitor was required during the Discussion
early postoperative period in eight patients, and anti-coagulation for six months in all patients. The median As α-Gal is a major barrier for xenotransplantation, duration of follow up was 11 months (range: 8 to 14 much effort has been expended to avoid the immune months). All patients were in NYHA functional class 1, response directed against it, including immunoglobu- and none of the patients was administered any cardiac lin or enzymatic treatment and genetic manipulation medication (except warfarin) at the last follow up (α1,3-galactosyltransferase knock-out) (17-19).
Unfortunately, the α1,3- galactosyltransferase knock-out pig is not yet available within a clinical setting, and ELISA: Anti α-Gal antibody titer
other methods such as enzymatic removal of the α-Gal Results from random samples (patient #14 of the epitope have caused only a delay in the rejection. The PVR group and patient #17 of the re-PVR group) are major antibodies that are thought to initiate the hyper- shown in Figure 1. In all cases, the OD decreased acute rejection of porcine xeno-organs include IgM Kim 3845 r1.qxd:Layout 1 25/1/10 10:02 Page 128 128 Anti α-Gal immune response Table III: Anti α-Gal titer in all patients (versus Figure 3: Changes in the anti α-Gal antibody titer in all patients. A) Changes in IgM titer. B) Changes in IgG titer. turnover would not be expected) these patients still *Titer defined as the dilution at 1.0 of the OD.
PVR: Pulmonary valve replacement; develop calcification soon after transplantation along with valve destruction. This suggests that anotherprocess is involved, which may be an immune reaction.
xenoantibodies; however, IgG xenoantibodies can also Manji et al. (24) reported that significant inflamma- mediate complement activation as well as antibody- tion of the xenografts, as well as a significant humoral dependent cellular cytotoxicity mechanisms (6).
response to the xenografts, had occurred within a short In cardiac surgery, non-viable porcine bioprosthetic period of time, and that the calcification correlated valve implantation has been performed for four with the amount of inflammation in a young animal decades, and today the main concern of cardiac sur- model. Konakci and colleagues (16) noted that degen- geons following the implantation of a bioprosthesis is eration of the bioprosthetic valve begins with the pen- not a catastrophic hyperacute rejection but rather a late etration of immunoglobulins into the valve-matrix, degeneration of the implanted valve, which often macrophages onto the valve surface; the process is Although the currently available porcine biopros- then completed with collagen breakdown and calcifi- thetic valves include a variety of design, manufactur- cation. The same group (16) reported that the α-Gal ing, and treatment options, the vast majority are epitope still existed on the commercial porcine bio- glutaraldehyde-treated. It is generally accepted that prosthetic valve, thus confirming the findings of bioprosthetic valve degeneration is due to calcification Kasimir et al. (11), and that the bioprosthesis implanta- of the valve and tissue disruption or tearing.
tion elicited a specific humorally-mediated immune Glutaraldehyde fixation has been introduced to reduce response directed against α-Gal. Several in vivo stud- valve antigenicity and to improve its mechanical ies have reported that the titer of anti α-Gal IgM or IgG strength (20). However, the failure of glutaraldehyde- antibodies was increased, as measured using ELISA treated porcine xenografts in clinical series has been reported, while others have shown that glutaralde- In the present study, it was observed that the titer of hyde treatment does not remove xenograft antigenici- anti α-Gal antibodies changed with different patterns ty. Thus, one of the major causes of bioprosthesis between the two isotypes. At one week after surgery, degeneration is an immune reaction (21-24).
the anti α-Gal antibody titer increased, regardless of The immune mechanism has been considered to play the isotype. Whilst the anti α-Gal IgM antibody titer an important role as an initial trigger of the degenera- began to decrease after three weeks, that of IgG was tion process, and the α-Gal epitope is the major antigen.
maintained over two months after surgery. The imme- In the same context, the younger the patient’s age, the diate postoperative (day 1) titers showed a tendency to more frequently will valve degeneration occur, which decrease, which may be ascribed to the binding of cir- suggests a more active and robust immunologic culating anti α-Gal antibody to the α-Gal isotope on response to xenoantigens (8,9,14). Simionescu (25) noted the valve and/or a dilutional effect of CPB or intravas- that excessive wear and tear, and a higher calcium turnover secondary to growth, is responsible for early There was no statistically significant difference in the calcification in children; however, even after the adoles- anti α-Gal antibody titer between the PVR and re-PVR cent period (young adulthood, when a high calcium groups. Before surgery, the anti α-Gal antibody titer Kim 3845 r1.qxd:Layout 1 25/1/10 10:02 Page 129 Anti α-Gal immune response 129 was similar between the two groups, which suggested that the antigenicity of the α-Gal epitope on a previ- 4. Buonomano R, Tinguely C, Rieben R, Mohacsi PJ, ously implanted valve may have decreased and almost Nydegger UE. Quantitation and characterization of disappeared with time. However, the mean postopera- anti-Galα1-3Gal antibodies in sera of 200 healthy tive titer (on days 21 and 60) seemed to be higher in the persons. Xenotransplantation 1999;5:173-180 re-PVR group (though not statistically significantly), 5. Galili U, LaTemple DC, Walgenbach AW, Stone KR.
such that the possibility of an influence due to previ- Porcine and bovine cartilage transplants in ous exposure on the immune response to the second cynomolgus monkey. Transplantation 1997;63:646- exposure, perhaps by host sensitization, may not be 6. Cooke SP, Pearson JD, Savage COS. Human IgG xenoreactive antibodies mediate damage to porcine Study limitations
endothelial cells in vitro by both humoral and cel- The primary limitation of the study was the small lular mechanisms. Transpl Immunol 1997;5:39-44 patient numbers, and the wide range of age and body 7. Lin SS, Kooyman DL, Daniels LJ, et al. The role of weight. The study was also limited by the use of dif- natural anti-Galα1-3Gal antibodies in hyperacute ferent types of bioprosthesis with different anticalcifi- rejection of pig-to-baboon cardiac xenotransplants.
cation treatments. Despite these limitations, the implantation of a porcine bioprosthesis was found to elicit the anti α-Gal immune response in each patient McKenzie IFC. Studies on human naturally occur- ring antibodies to pig xenografts. Transplant Proc1993;25:2917-2918 In conclusion, porcine bioprosthesis implantation 9. Stone KR, Walgenbach AW, Abrams JT, Nelson J, elicits an increased production of anti α-Gal antibodies Gillett N, Galili U. Porcine and bovine cartilage in humans, albeit with different patterns between the transplants in cynomolgus monkey: I. A model for two isotypes (IgM and IgG). Notably, the IgM antibody chronic xenograft rejection. Transplantation response was more rapid and transient, while that of IgG was longer and more delayed. The most important 10. Farivar RS, Filsoufi F, Adams DH. Mechanisms of requisites for the ideal valve substitute in cardiac sur- Gal(alpha)1-3Gal(beta)1-4GlcNAc-R (alphaGal) gery are reliability and durability. Although an expression on porcine valve endothelial cells. J immunologic response to the xenoantigen other than α-Gal (non-gal antigen) has been reported (27-29), the 11. Kasimir MT, Rieder E, Seebacher G, Wolner E, α-Gal epitope is a major antigen that must be over- Weigel G, Simon P. Presence and elimination of the come in pig-to-human transplantation, including bio- xenoantigen gal (alpha1,3) gal in tissue-engineered prosthesis implantation. For this reason, further heart valves. Tissue Eng 2005;11:1274-1280 studies are required to either avoid or minimize the 12. Silver MM, Pollock J, Silver MD, Williams WG, immune response to the α-Gal epitope, including Trusler GA. Calcification in porcine xenograft decellularization as well as enzymatic or genetic valves in children. Am J Cardiol 1980;45:685-689 13. Jamieson WRE, Rosado LJ, Munro AI, et al.
Carpentier-Edwards standard porcine bioprosthe- Acknowledgements
sis: Primary tissue failure (structural valve deterio- These studies were supported by grants from the Korea Health 21 Research & Development Project, Korean Ministry of Health, Welfare & Family (A04004- 14. Fann JI, Miller DC, Moore KA, et al. Twenty-year clinical experience with porcine bioprostheses. AnnThorac Surg 1996;62:1301-1312 References
15. Gabbay S, Kadam P, Factor S, Cheung TK. Do heart valve bioprostheses degenerate for metabolic or McKenzie IFC. Anti-pig IgM antibodies in human mechanical reasons? J Thorac Cardiovasc Surg serum react predominantly with Gal(α1-3)Gal epi- topes. Proc Natl Acad Sci USA 1993;90:11391-11395 16. Konakci KZ, Gohle B, Blumer R, et al. Alpha-Gal on 2. Bracy JL, Cretin N, Cooper DKC, Iacomini J.
bioprostheses: Xenograft immune response in car- Xenoreactive natural antibodies. Cell Mol Life Sci diac surgery. Eur J Clin Invest 2005;35:17-23 17. LaVecchio JA, Dunne AD, Edge ASB. Enzymatic 3. Galili U. The α-Gal epitope (Galα1,3-Galβ1- removal of alphagalactosyl epitopes from porcine 4GlcNAc-R) in xenotransplantation. Biochimie endothelial cells diminishes the cytotoxic effect of Kim 3845 r1.qxd:Layout 1 25/1/10 10:02 Page 130 130 Anti α-Gal immune response natural antibodies. Transplantation 1995;60: 24. Manji RA, Zhu LF, Nijjar NK, et al. Glutaraldehyde- fixed bioprosthetic heart valve conduits calcify and 18. Phelps CJ, Koike C, Vaught TD, et al. Production of fail from xenograft rejection. Circulation α1,3-galactosyltransferase-deficient pigs. Science 25. Simionescu DT. Prevention of calcification in bio- 19. Schussler O, Shen M, Shen L, Carpentier SM, prosthetic heart valves: Challenges and perspec- tives. Expert Opin Biol Ther 2004;4:1971-1985 26. Kozlowski T, Monroy R, Xu Y, et al. Anti- Gal(alpha)1-3Gal antibody response to porcine bone marrow in unmodified baboons and baboons 20. Carpentier A, Lemaigre G, Robert L, Carpentier S, conditioned for tolerance induction.
Dubost C. Biological factors affecting long-term results of valvular heterografts. J Thorac 27. Stone KR, Abdel-Motal UM, Walgenbach AW, Turek TJ, Galili U. Replacement of human anterior 21. Villa ML, DeBiasi S, Pilotto F. Residual heteroanti- cruciate ligaments with pig ligaments: A model for genicity of glutaraldehyde-treated porcine cardiac anti-non-Gal antibody response in long-term xeno- transplantation. Transplantation 2007;83:211-219 22. Schoen FJ. Are immune mechanisms important in 28. Baumann BC, Stussi G, Huggel K, Rieben R, tissue heart valve failure? A debate. J Heart Valve Seebach JD. Reactivity of human natural antibodies to endothelial cells from Gala(1,3)Gal-deficient 23. Vyavahare N, Hirsch D, Lerner E, et al. Prevention of bioprosthetic heart valve calcification by ethanol 29. Zhu A, Hurst R. Anti-N-glycolylneuraminic acid preincubation: Efficacy and mechanisms.
antibodies identified in healthy human serum.

Source: http://www.drheart.net/korean/pdf/2010_02r.pdf