Immunoglobulin-G Subclass Antidonor Reactivity in Transplant Recipients Zu-hua Gao,1,2 Vivian C. McAlister,4 James R. Wright Jr.,2 Chloe C. McAlister,2 Kevork Peltekian,3 and Allan S. MacDonald1 Outcomes may differ after kidney transplantation compared to combined liver-kidney transplantation. In animal models, distinct patterns of antidonor immunoglobulin (Ig) G subclasses are associated with either rejection or transplant tolerance. Flow cytometry has increased the sensitivity of antidonor immunoglobulin detection. We compared antidonor IgG subclass responses in kidney transplant recipients to those in recipients of liver or multiorgan grafts. In this study of 19 organ (kidney, liver, pancreas) transplantations, recipient serum incubated with donor splenocytes was tested by flow cytometry for the presence of IgM, IgG, or IgG subclass 1 – 4. Sera before transplantation and 10 days and 100 days after transplantation were used. No differences were seen in antidonor IgM, IgG, or IgG subclass antibodies among recipients of kidney transplants and liver grafts or combination grafts, either before or after transplantation. IgG4 gradually but significantly increased after transplantation in all groups. High levels of antidonor IgG3 either before transplantation or produced after it were found in 3 kidney recipients who experienced acute rejection. No other patients experienced rejection, and no other increase in IgG3 was seen. In conclusion, antidonor IgG subclass profiles may be useful to distinguish populations at risk of rejection but they do not differentiate the immunological response after kidney transplantation from that after liver or combined transplantation. A late rise in antidonor IgG4 is consistent with decreased antidonor reactivity thought to occur late after transplantation. (Liver Transpl 2004;10: 1055–1059.)
S
ince the early reports of antidonor antibodies in the serum of the recipient before transplantation being a risk for hyperacute rejection, crossmatch-positive patients have been denied access to kidney transplantation.1,2 Unfortunately, a considerable number of candidates for kidney transplantation have alloantibodies and are unlikely to ever find a suitable (crossmatch negative) donor. Use of flow cytometry crossmatch (FCXM) to detect these antibodies, because it is more sensitive than conventional crossmatch based on complement-dependant cytotoxicity crossmatch (CDCXM), could increase the number of excluded candidates.3,4 Indeed, a recent series found a bleak outcome for CDC-XM-negative, FCXM-positive kidney transplants.5 In contrast, contemporaneous studies of liver recipients showed that donor-specific antibody (DSA) was not a barrier to transplantation, and it was found that the liver transplant protected a simultaneously transplanted crossmatch-positive kidney by either
absorbing DSA or by neutralizing them.6 Although the crossmatch-positive combined graft appeared to be at increased risk of acute cellular rejection, good outcomes continue to be recorded.7,8 Up to a quarter of kidney recipients develop FCXMdetectable antidonor immunoglobulin (Ig) G after transplantation, and this group is also at considerably increased risk of rejection.9,10,11 After living donor liver transplantation, 20% of recipients in one study developed post-transplantation DSA, all of whom experienced rejection.12 Antidonor Ig class type was not found to be related to rejection in this study, but many of the patients had received ABO incompatible grafts. In contrast, a study specifically looking at class switch found posttransplantation IgG DSA to be associated with rejection and IgM DSA with freedom from rejection after liver transplantation.13 We have noted the development of IgM DSA after crossmatch-positive liver-kidney transplantation, which failed to fix complement and appeared to protect the grafts.14 Protective IgM DSA has been reported in kidney transplantation in a study that also suggested that not all FCXM-detected IgG DSA was detrimental to the graft.15 We wondered if the variable responses to IgG DSA might be explained by IgG subclass identification. In the mouse, transplant recipients not receiving immunosuppression invariably make antidonor IgG, but the subclass (or isotype) profile determines the outcome: Antidonor IgG2a is associated with rejection, and IgG1 is associated with transplant tolerance.16 Humans and mice have 4 subclasses of IgG, although the nomenclature differs slightly. The light chain ( or Abbreviations: FCXM, flow cytometry crossmatch; CDC-XM, cytotoxicity crossmatch; DSA, donor specific antibody; Ig, immunoglobulin; MCF, mean channel of fluorescence. From the Departments of 1Surgery, 2Pathology,and 3Medicine, Dalhousie University, Halifax, Nova Scotia, and the 4Department of Surgery, University of Western Ontario, London, Ontario, Canada. Address reprint requests to Vivian McAlister, M.B., 4-TU44, University Hospital, London, Ontario, Canada N6A 5A5. FAX: 519-6633858; E-mail:
[email protected] Copyright © 2004 by the American Association for the Study of Liver Diseases Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/lt.20154
Liver Transplantation, Vol 10, No 8 (August), 2004: pp 1055 – 1059
1055
1056
Gao et al.
) of IgG is conserved across the subclasses. IgG subclasses are determined by heavy chain (␥) differences. For example, a single amino acid switch (proline to serine) at the core hinge reduces the complement fixing ability of human IgG1 to that of IgG4.17 Human IgG3, followed by IgG1, is the most potent complement binder of the subclasses. Human IgG4, the subclass present in the least concentration, is predominately expressed in association with chronic antigen stimulation, where it inhibits complement (C1q) binding.17 In mice, Th1 T cell clones have been shown to induce B cell memory and affinity maturation to produce predominantly IgG2a and IgG3, where as Th2 clones result in IgG1 and IgE production.18 We hypothesized that antidonor IgG subclass profiles in humans, either present before transplantation or developed as a response to it, might explain differences between kidney and liver recipients by their association with rejection or tolerance after transplantation.
Materials and Methods Before transplantation and after transplantation on day 10 and day 100, sera from isolated kidney recipients, isolated liver recipients, and recipients of multiple grafts (liver-kidney, pancreas-kidney) were tested for antidonor IgM, IgG, and subclass IgG 1 – 4. Serum was frozen and batch tested. The results were not available to the caregiving team, and no changes were made to the routine care of the patients, who were followed for rejection for 6 months. Immunosuppressants including cyclosporine, tacrolimus, mycophenlic mofetil, sirolimus, or prednisone were used in various combinations. DSA was measured using flow cytometry, following the technique described by Terasaki.3 The donor spleens were flushed with University of Wisconsin solution at organ retrieval and processed into single-cell suspensions within 24 hours of retrieval. This was done by fragmentation, wire-mesh straining, and splenocyte isolation using density-gradient sedimentation over Ficol-Hypaque (Sigma-Aldrich, Oakville, Ontario, Canada). Suspended in Roswell Park Memorial Institute – 1640 culture medium with 10% dimethyl sulfoxide (Sigma-Aldrich), the solution was divided into1 mL aliquots and frozen in liquid nitrogen. Because adequate storage space in liquid nitrogen was unavailable, the splenocytes were transferred to a freezer at ⫺80°C. Using this method of storage, we had lymphocyte survival rates of 95%, 80%, 60%, and 50% at 3, 6, 12, and 24 months, respectively, compared to 90% at 12 months in liquid nitrogen. Clinically surplus recipient serum collected at transplantation and on postoperative days 10 and 100 was frozen. At the time of testing, donor splenocytes were thawed in fetal calf serum, centrifuged at 233g and washed in phosphate buffered saline. Trypan blue stain was used to confirm the absence of dead splenocytes.
Samples with greater than 10% dead cells were discarded. The splenocytes were then counted. Heat-inactivated recipient serum (1 : 25 dilution) was first incubated with donor spleen cells for 30 minutes at 4°C and then stained with fluorescent conjugated goat antibody specific for human IgG Fc region on IgG 1 – 4 isotypes (Jackson ImmunoResearch Laboratories, Mountain View, CA). For each sample, 10,000 cells were analyzed in the lymphocyte gate of the flow cytometer. Intensity of staining expressed as mean channel of fluorescence (MCF) was recorded on a linear scale. Phosphate buffered saline was used instead of test serum in each study, and background staining was numerically subtracted from each test MCF. Actual MCF is reported for tests of recipient sera taken just before transplantation. Relative MCF of sera taken after transplantation is expressed as a ratio of MCF on the test day to the pretransplant MCF.
Results Sera from 19 recipients (10 males, 9 females), aged 29 – 69 years, of kidney (7), liver (6), kidney-pancreas (4) and liver-kidney (2) grafts were tested for antidonor IgM, IgG, and subclass IgG 1 – 4. Patient and graft survival were 100% and 95%, respectively. Table 1 reports the posttransplantation production of antidonor IgG subclasses in the individual recipients. Production was measured as the MCF relative to the baseline (fold increase from pretransplant MCF). The increase in IgG4 was statistically significantly greater than that in IgG1 or IgG2 at both 10 and 100 days. Pretransplantation IgM, IgG, and IgG subclass DSA and posttransplantation production were similar in kidney-, liver-, or combination-graft recipients (Table 2). Table 3 summarizes the antidonor IgG subclass responses of 3 recipients who experienced rejection compared to 16 recipients who had no rejection. One renal graft was lost to hyperacute rejection (patient 1). Two other kidney recipients, patient 2 and patient 3, were diagnosed with acute cellular rejection 28 and 103 days, respectively, after transplantation. No other rejection episodes were recorded during the study. Patient 1 had a high level of IgG3 DSA before transplantation. This was not known at the time of transplantation, which proceeded because the CDC-XM was negative. IgG3 MCF was 248 in this patient, compared to a mean (range) of 30 (1963) in nonrejecting patients. On day 10 after transplantation, IgG3 increased to 2.4 fold in patient 2, compared to a mean (range) of 0.9 (0.5-1.9) in nonrejectors. This patient had an episode of acute cellular rejection on day 28. On day 100, the increase in IgG3 remained at 2.5-fold in patient 2, compared to a mean (range) of 1.0 (0.7-1.53) in nonrejecting
1057
Immunoglobulin-G Subclass Antidonor Reactivity
Table 1. Relative Posttransplantation Antidonor IgG Subclasses With a Paired Comparison of IgG4 to the Other IgG Subclasses Posttransplant Day 10 Graft
Immunosuppression
kidney CMP kidney CMP kidney CMP kidney CMP kidney CMP kidney CMP liver TSP liver TSP liver TSP liver TSP liver TSP liver TSP P-K ATSP P-K ATSP P-K ATSP P-K ATSP L-K ATSP L-K TSP Paired Student t test with IgG4
Posttransplant Day 100
IgG1
IgG2
IgG3
IgG4
IgG1
IgG2
IgG3
IgG4
1.03 0.90 1.00 0.48 0.77 0.58 1.00 1.16 0.65 1.00 1.01 0.98 0.87 0.74
0.68 0.97 1.00 0.82 0.37 0.42 0.59 0.90 0.94 0.82 1.00 1.10 0.66 0.88
2.42 1.06 0.83 0.55 0.52 0.70 0.98 0.84 0.75 1.00 1.00 1.05 0.79 0.85
1.25 1.04 1.37 1.43 0.75 1.05 1.14 0.61 0.73 1.00 1.00 1.15 1.20 1.29
1.12 1.16 1.08 0.86 1.63 1.48 1.22 0.92 1.15 1.16 0.77 0.94 1.05 0.98 1.23
0.91 1.18 0.86 1.59 0.87 1.00 1.15 1.03 1.81 0.51 0.76 1.03 1.00 0.75 1.35
2.47 4.64 0.85 0.78 0.76 1.25 1.17 0.70 1.18 0.92 0.86 1.02 1.59 0.93 1.28
2.25 1.79 1.26 2.36 2.50 1.52 2.00 0.86 1.64 0.85 1.03 1.42 2.47 0.89 1.92
0.39 0.93 1.44 P ⴝ .02
0.73 0.97 0.44 P ⴝ .001
0.67 0.91 1.86 P ⴝ .41
0.94 1.22 1.25
0.98 0.86 P ⴝ .0008
0.90 0.60 P ⴝ .0003
1.02 1.36 P ⴝ .26
1.30 1.13
Abbreviations: C, cyclosporine; M, mycophenolate mofetil; P, prednisone; T, tacrolimus; S, sirolimus; P-K, pancreas-kidney; A, thymoglobulin; L-K, liver-kidney. NOTE. Posttransplant production of IgG is given as the mean channel fluorescence with serum of the day tested relative to before transplantation.
patients, but the day 28 episode of acute cellular rejection had resolved with steroid treatment and not recurred. On day 100, patient 3 had a large increase in IgG3 at 4.6-fold. This kidney recipient had an episode of steroid-sensitive acute cellular rejection on day 103. No other kidney recipient and no recipient of liver or
combination transplant experienced rejection. No other rises in IgG3 were seen. Two of 6 recipients who commenced with cyclosporine-mycophenolate-prednisone maintenance immunosuppression and none of 12 recipients on tacrolimus-sirolimus-prednisone developed IgG3 DSA or rejection.
Table 2. Antidonor IgM, IgG, and IgG Isotype (Median and Range) in Recipients of Liver, Kidney, or Combination Grafts Graft Type Day 0, Pre Tx MCF
Day 10 Post Tx MCF relative to day 0 (ratio)
Day 100 Post Tx MCF relative to day 0 (ratio)
Kidney (n ⫽ 7) Liver (n ⫽ 6) Combination (n ⫽ 6)
IgM
IgG
IgG1
IgG2
IgG3
IgG4
35 (27 – 50) 45 (28 – 61)
71 (59 – 363) 76 (64 – 81)
48 (30 – 85) 45 (36 – 61)
33 (17 – 76) 31 (27 – 55)
50 (19 – 248) 49 (41 – 63)
21 (14 – 28) 29 (22 – 44)
42 (34 – 58)
81 (51 – 134)
45 (36 – 67)
30 (20 – 51)
41 (22 – 58)
26 (16 – 35)
Kidney (n ⫽ 6) Liver (n ⫽ 6) Combination (n ⫽ 5)
0.99 (0.78 – 1.43) 0.88 (0.75 – 1.28)
1.01 (0.88 – 1.17) 1.00 (0.86 – 1.07)
0.83 (0.48 – 1.03) 1.00 (0.65 – 1.16)
0.75 (0.37 – 1.00) 0.92 (0.59 – 1.10)
0.77 (0.52 – 2.42) 0.99 (0.75 – 1.05)
1.15 (0.75 – 1.43) 1.00 (0.61 – 1.15)
1.00 (0.53 – 1.12)
0.97 (0.68 – 1.16)
0.90 (0.39 – 1.44)
0.80 (0.44 – 1.00)
0.88 (0.67 – 1.86)
1.21 (0.94 – 1.29)
Kidney (n ⫽ 5) Liver (n ⫽ 6) Combination (n ⫽ 5)
1.15 (0.69 – 1.57) 0.97 (0.67 – 1.43)
1.17 (0.91 – 1.25) 1.00 (0.90 – 1.08)
1.14 (0.86 – 1.63) 1.04 (0.77 – 1.22)
0.96 (0.86 – 1.59) 1.03 (0.51 – 1.15)
1.05 (0.87 – 1.59) 0.97 (0.86 – 1.18)
2.02 (1.26 – 2.50) 1.23 (0.86 – 2.00)
0.95 (0.72 – 1.29)
0.92 (0.68 – 1.35)
0.98 (0.39 – 1.23)
0.83 (0.60 – 1.35)
1.15 (0.67 – 1.36)
1.21 (0.89 – 2.47)
Abbreviation: MCF, mean channel of fluorescence. NOTE. Two kidney-liver and 4 kidney-pancreas patients are included in the combination group; 3 kidney recipients who experienced rejection are included; 1 kidney graft was removed on day 5, and the recipient was not tested on day 10 or 100.
1058
Gao et al.
Table 3. Flow Cytometry of Donor Splenocytes for IgG and IgG Isotype (Median and Range) Binding From Serum From 16 Organ Recipients Without Rejection Compared to 3 Recipients With Rejection Day MCF
Rejection
IgG
IgG1
IgG2
IgG3
IgG4
0 (Pre Tx)
No Yes No Yes No Yes
74.5 (51 – 134) 363,* 71,† 82‡ 1.0 (0.7 – 1.2) 1.03,† 0.88‡ 1.04 (0.7 – 1.3) 1.14,† 2.23‡
44.5 (30 – 67) 85,* 33,† 49‡ 0.96 (0.4 – 1.4) 1.03,† 0.9‡ 1.07 (0.4 – 1.6) 1.12,† 1.16‡
30.0 (16 – 51) 76,* 34,† 33‡ 0.85 (0.4 – 1.0) 0.69,† 0.97‡ 0.96 (0.5 – 1.5) 0.91,† 1.18‡
30.0 (19 – 63) 248,* 19,† 47‡ 0.85 (0.5 – 1.9) 2.42,† 1.06‡ 1.02 (0.7 – 1.53) 2.47,† 4.64‡
47.5 (14 – 44) 23,* 16,† 28‡ 1.1 (0.7 – 1.4) 1.25,† 1.04‡ 1.47 (0.9 – 2.4) 2.25,† 1.79‡
10 (Post Tx)§ 100 (Post Tx)§
*Patient 1, kidney recipient diagnosed with hyperacute rejection 6 hours after transplant (not tested on day 10 or day 100). †Patient 2, acute cellular rejection, day 28. ‡Patient 3, acute cellular rejection, day 103. §MCF relative to day 0 ratio.
Discussion This study is in agreement with previous reports that the presence of DSA before transplantation or its development afterward is associated with an increased risk of hyperacute or acute cellular rejection.1 – 7, 9 – 14 Our novel observation is that IgG3 may be the responsible subclass. This is not surprising, at least with respect to hyperacute rejection, as IgG3 is the strongest complement activator among the IgG subclasses. However, the histological features of the rejections suffered by patients 2 and 3 in our series were those of acute cellular rather than humoral rejection. A large prospective study of FCXM-positive kidney transplantations showed an increased incidence, compared to FCXM-negative recipients, of acute cellular rejection but not of vascular lesions associated with humoral rejection.11 Similarly the production of IgG3 DSA appeared to increase the risk of acute cellular rejection. The mechanism for this is not known. It is possible that T cell activation helps B cells mature with resulting antidonor antibody production. Th1 and Th2 clones induce different IgG subclasses, most likely mediated by different responses of the B cells to the cytokines, which differentiate the T helper cell clones.18 If this is so, the production of IgG3 DSA may be a bystander phenomenon. However, it was detected before evidence of graft dysfunction, and therefore it could become useful in surveillance after transplantation. IgG4 DSA increased in the recipients over time. This subclass is normally present in the least concentration in humans, and increases are associated with chronic antigenic stimulation.17,19 A reversal in IgG3 / IgG4 ratio has been observed with acquired immunodeficiency.20 It is well known that less immunosuppression is required to maintain a stable transplant than is
required soon after transplantation. No mechanism has been demonstrated. The production of IgG4 is consistent with the balance between persistent antigenic stimulation and continuous immunosuppression that is achieved in stable transplant recipients, but again it may represent a bystander phenomenon as a role for complement in long-term graft loss is unknown. IgG subclass profile differences between recipients of kidney, liver, or combination grafts are not apparent. Even though the sample size was small, the variation among individuals in a group was far bigger that the variation among the groups, indicating that very large studies indeed will be required to see differences, should there be any. This study is also too small and heterogenous to conclude anything regarding association between DSA production after transplantation and different immunosuppressive protocols. However, differences in outcome among immunosuppressive protocols in FCXM-positive renal transplantation have been reported.21 It may therefore be useful to include examination of IgG subclass DSA responses in randomized control trials of different immunosuppressive regimens. Despite the knowledge in experimental transplantation and in nontransplant human immunology that IgG subclass profiles are associated with different immune responses, this is the first study of IgG subclasses in clinical transplantation. Donor splenocytes were used as a “catch-all” target. Although donor splenocytes are plentiful at retrieval, prolonged viable storage is difficult. While the previous studies cited refer to the risk of antidonor T cell antibodies, B cell FCXMpositivity has also been shown to increase the risk of rejection.22 We did not distinguish between class I and class II or individual human leukocyte antigen (HLA) specificities. The value of using newer synthetic targets
Immunoglobulin-G Subclass Antidonor Reactivity
of FCXM is being established.23 IgG subclass identification is compatible with these technological improvements. If these findings are confirmed in other centers and in larger populations, antidonor IgG subclass surveillance before and after transplantation may be more sensitive and more specific than total IgG in the determination of patients at risk.
References 1. Starzl TE, Marchioro TL, Terasaki PI, Porter KA, Faris TD, Herrmann TJ, et al. Chronic survival after human renal homotransplantation. Lymphocyte-antigen matching, pathology and influence of thymectomy. Ann Surg 1965;162:749 – 787. 2. Kissmeyer-Nielsen F, Olsen S, Petersen VP, Fjeldborg O. Hyperacute rejection of kidney allografts, associated pre-existing humoral antibodies against donor cells. Lancet 1966;2:662 – 665. 3. Iwaki Y, Lau M, Cook DJ, Takemoto S, Terasaki PI. Crossmatching with B and T cells and flow cytometry. Clin Transpl 1986:277 – 284. 4. Ogura K, Terasaki PI, Johnson C, Mendez R, Rosenthal JT, Ettenger R, et al. The significance of a positive flow cytometry crossmatch test in primary kidney transplantation. Transplantation 1993;56:294 – 298. 5. Karpinski M, Rush D, Jeffery J, Exner M, Regele H, Dancea S, et al. Flow cytometric crossmatching in primary renal transplant recipients with a negative anti-human globulin enhanced cytotoxicity crossmatch. J Am Soc Nephrol 2001;12:2807 – 2814. 6. Gordon RD, Fung JJ, Markus B, Fox I, Iwatsuki S, Esquivel CO. The antibody crossmatch in liver transplantation. Surgery 1986; 100:705 – 715. 7. Saidman SL, Duquesnoy RJ, Demetris AJ, McCauley J, Ramos H, Mazariegos G, et al. Combined liver-kidney transplantation and the effect of preformed lymphocytotoxic antibodies. Transpl Immunol 1994;2:61 – 67. 8. Neumann UP, Lang M, Moldenhauer A, Langrehr JM, Glanemann M, Kahl A, et al. Significance of a T-lymphocytotoxic crossmatch in liver and combined liver-kidney transplantation. Transplantation 2001;71:1163 – 1168. 9. Piazza A, Poggi E, Borrelli L, Servetti S, Monaco PI, Buonomo O, et al. Impact of donor-specific antibodies on chronic rejection occurrence and graft loss in renal transplantation: posttransplant analysis using flow cytometric techniques. Transplantation 2001;71:1106 – 1112. 10. Pelletier RP, Hennessy PK, Adams PW, VanBuskirk AM, Ferguson RM, Orosz CG. Clinical significance of MHC-reactive alloantibodies that develop after kidney or kidney-pancreas transplantation. Am J Transplant 2002;2:134 – 141. 11. Scornik JC, Clapp W, Patton PR Van der Werf WJ, Hemming
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
1059
AW, Reed AI, Howard RJ. Outcome of kidney transplants in patients known to be flow cytometry crossmatch positive. Transplantation 2001;71:1098 – 1102. Kasahara M, Kiuchi T, Takakura K, Uryuhara K, Egawa H, Asonuma K, et al. Postoperative flow cytometry crossmatch in living donor liver transplantation: Clinical significance of humoral immunity in acute rejection. Transplantation 1999;67: 568 – 575. Farges O, Buffello D, Shi YM, Berth A, Bismuth H. Anti-donor antibody class switching after liver transplantation. Transplantation 1995;60:296 – 300. McAlister CC, Gao Z, McAlister VC, Gupta R, Wright JR Jr., MacDonald AS, Peltekian K. Protective anti-donor IgM production after crossmatch positive liver-kidney transplantation. Liver Transpl 2004;10:315 – 319. Kerman RH, Susskind B, Buyse I, Pryzbylowski P, Ruth J, Warnell S, et al. Flow cytometry-detected IgG is not a contraindication to renal transplantation: IgM may be beneficial to outcome. Transplantation 1999; 68:1855 – 1858. Gao ZH, Zhong R, Jiang J Garcia B, Xing JJ, White MJ, Lazarovits AI. Adoptive transferable tolerance induced by CD45RB monoclonal antibody. J Am Soc Nephrol 1999;10:374 – 381. Xu Y, Oomen R, Klein MH. Residue at position 331 in the IgG1 and IgG4 CH2 domains contributes to their differential ability to bind and activate complement. J Biol Chem 1994;269:3469 – 3474. Rizzo LV, DeKruyff RH, Umetsu DT. Generation of B cell memory and affinity maturation. Induction with Th1 and Th2 T cell clones. J Immunol 1992;148:3733 – 3739. Jenmalm MC, Bjorksten B, Macaubas C, Holt BJ, Smallacombe TB, Holt PG. Allergen-induced cytokine secretion in relation to atopic symptoms and immunoglobulin E and immunoglobulin G subclass antibody responses. Pediatr Allergy Immunol 1999; 10:168 – 177. de Martino M, Rossi ME, Azzari C, Chiarelli F, Galli L, Vierucci A. Low IgG3 and high IgG4 subclass levels in children with advanced human immunodeficiency virus-type 1 infection and elevated IgE levels. Ann Allergy Asthma Immunol 1999;83: 160 – 164. Kimball P, Wagner B, King A, Fisher RA, Dawson S, Cotterell A, Posner M. Comparison of two drug regimens upon clinical outcome among renal transplant recipients with positive flow cytometric crossmatches. Clin Transplant 2002;16:290 – 294. Kotb M, Russell WC, Hathaway DK, Gaber LW, Gaber AO. The use of positive B cell flow cytometry crossmatch in predicting rejection among renal transplant recipients. Clin Transplant 1999;13:83 – 89. Muller-Steinhardt M, Fricke L, Kirchner H, Hoyer J, Kluter H. Monitoring of anti-HLA class I and II antibodies by flow cytometry in patients after first cadaveric kidney transplantation. Clin Transplant 2000;14:85 – 89.
