THE ROLE OF FLOW CYTOMETRY-DETECTED IgG AND IgM ANTI-DONOR ANTIBODIES IN CARDIAC ALLOGRAFT RECIPIENTS¹

Abstract

Background. 

At our transplant center, cardiac allograft recipients undergo transplantation following a negative IgG anti-human globulin (AHG) crossmatch (XM). Flow cytometry crossmatching (FCXM) is a more sensitive XM procedure than the AHG XM procedure, yet there is limited information regarding the clinical relevance of FCXM to cardiac allograft outcome.

Methods. 

FCXM was performed retrospectively using the pretransplant sera from 140 recipients of primary cardiac allografts who underwent transplantation after AHG-IgG-NEG XM. The FCXM results were correlated to posttransplant rejection and patient survival.

Results. 

All of the patients were auto-XM-NEG. Twenty-two of 140 patients (16%) displayed IgG (+) FCXM and had a significantly poorer 1-year survival rate than did 57 of the FCXM-NEG recipients (68% vs. 86%, P<0.02). Moreover, 50% of the IgG(+) FCXM recipients experienced early rejections (≤14 days postoperatively) compared with only 16% for the FCXM-NEG recipients (P<0.01). The survival rate of 92% for IgM(+) FCXM recipients (n=37) was significantly improved compared with the 86% survival rate for FCXM-NEG control recipients (P<0.05), suggesting a protective role for IgM. Consistent with this interpretation is that the 1-year survival rate of 79% for the IgG, IgM FCXM(+) recipients (n=24) was significantly better than the 68% survival rate for the IgG(+) FCXM recipients (P<0.02).

Conclusions. 

These data suggest that IgG(+) FCXM identifies a subset of AHG-IgG-NEG XM cardiac allograft recipients who are at risk for early rejections and poor survival. In contrast, the presence of IgM may be beneficial to survival.

*Abbreviations: AHG, anti-human globulin; CAD, coronary artery disease; CsA, cyclosporine; DTE, dithioerythritol; FCXM, flow cytometry crossmatch; Ig, immunoglobulin; NEG, negative; PBL, peripheral blood lymphocytes; PBS, phosphate-buffered saline; POS, positive; PRA, panel reactive antibody; UNOS, United Network of Organ Sharing; XM, crossmatch.

Before clinical allotransplantation is performed, a crossmatch (XM *) is performed to avoid hyperacute or accelerated rejection caused by recipient anti-donor antibodies ⁽¹⁾. Thus, the pretransplant XM may be the most important procedure performed to discriminate appropriate donor-recipient pairings for transplantation. Despite its importance, there has been a great deal of controversy regarding the interpretation and clinical significance of the various types of antibodies detected by the several different XM procedures ⁽²⁾. We have successfully used the anti-human globulin (AHG) XM procedure for renal, heart, and liver transplantation ^(3-6). Primary allograft recipients can undergo transplantation after a negative IgG AHG-XM or an IgM-positive AHG-XM when appropriate sera are tested against unseparated, donor peripheral blood lymphocytes (PBL). Flow cytometry crossmatching (FCXM) is a more sensitive assay than the AHG XM procedure and has the potential of detecting low levels of alloantibody sensitization in the absence of AHG reactivity. Moreover, it delineates IgM and IgG reactivity to either T or B cell targets ^(2,7). To date, most FCXM studies have been performed in renal allograft recipients ^(3,7,8). There are few studies reporting the significance of FCXM in association with cardiac transplantation ^(9,10). Therefore, we performed retrospective donor-specific FCXM using the pretransplant sera from recipients of primary cardiac allografts who underwent transplantation after AHG-IgG-NEG XM. We correlated the FCXM results to posttransplant rejection and patient survival.

MATERIALS AND METHODS

Patients. Between December 1991 and May 1997, 235 end-stage heart disease patients underwent transplantation at The Texas Heart Institute-University of Texas Medical School Transplant Center. Of these 235 recipients, there were 140 primary recipients who were crossmatched using both AHG and FCXM procedures and who were evaluated in this study. There were 112 (80%) men and 28 (20%) women, 112 (80%) of whom were Caucasian, 13 (9%) of whom were Mexican American, and 15 (11%) of whom were African American. The mean age of the patients was 48±1 years. Indications for transplantation included diagnoses of ischemic cardiomyopathy (n=70, 50%), idiopathic cardiomyopathy (n=42, 30%), and other pathological abnormalities (n=28, 20%), with 67 (48%) patients diagnosed as United Network of Organ Sharing (UNOS) Status I and 73 (52%) diagnosed as UNOS Status II. The mean ischemia time was 180±18 min. Forty percent of the recipients had received pretransplant blood transfusions, and the pretransplant panel reactive antibody (PRA) was 5±2%. The duration of follow-up patient evaluation ranged from 12 to 77 months.

Immunosuppression. Forty percent of the patients received either OKT3 or anti-thymocyte globulin induction therapy. This was done to avoid the nephrotoxic effects of cyclosporine in patients with severely impaired end-organ function before transplantation. All patients received either oral cyclosporine (6 mg/kg) or azathioprine (2 mg/kg) preoperatively and then methylprednisolone (500 mg administered intravenously) intraoperatively. Most patients received a triple-drug immunosuppressive regimen that consisted of orally administered cyclosporine (CsA) (6-10 mg/kg/day), azathioprine (2 mg/kg/day), and prednisone (2 mg/kg/day), with a standard taper to a level of 0.4 mg/kg/day at 2 weeks posttransplant. CsA trough levels were monitored daily using a whole-blood monoclonal antibody assay (TDx; Abbott Diagnostics, North Chicago, IL). During the first 30 days after transplant, target levels were decreased to between 200 and 400 ng/ml; thereafter, levels were further reduced to below 300 ng/ml. For patients with pretransplant renal or hepatic dysfunction, either antithymocyte globulin or OKT3 was substituted for CsA during induction therapy ⁽¹¹⁾.

Endomyocardial biopsies were routinely performed. Treatment for rejection was based on biopsy findings and clinical signs of rejection (arrhythmias, or univentricular or biventricular failure). A biopsy score of IIIA (International Society for Heart, Lung Transplantation) or greater indicated rejection. Specific rejection treatment options have previously been described ⁽¹²⁾. The presence of allograft coronary artery disease (CAD) was determined by serial left heart catherization and coronary arteriography.

Histocompatibility testing. All donors and recipients were ABO-compatible. HLA A, B, C, and DR tissue typing was performed using the microlymphocytotoxicity technique and commercially available tissue typing trays ⁽¹³⁾. Patient sera were tested for panel reactive antibodies (%PRA) against a 60-member donor panel representing HLA specificities ⁽⁴⁾. All recipients had AHG and dithioerythritol (DTE)-AHG XM performed using the highest PRA and pretransplant sera ⁽⁴⁾.

When the donor heart was obtained locally, the XM was performed preoperatively; when the donor heart was obtained from an area geographically distant from the Texas Medical Center, the XM was performed retrospectively.

FCXM was performed with a FACSCAN Flow Cytometer (256 channels, log scale) from Becton Dickinson (San Jose, CA), according to a modification of the protocol presented by Garovoy ⁽⁷⁾ and Garovoy et al. ⁽¹⁴⁾. Donor PBL were isolated using the Ficoll-Hypaque density centrifugation technique, washed three times with phosphate-buffered saline (PBS) heat-inactivated 10% fetal bovine serum (FBS), and brought to a concentration of 2-4×10⁶ PBL·10% FBS/ml. Donor target PBL (2-4×10⁶ in a 0.2-ml volume) were added to 0.1 ml of recipient test serum, and the mixture was incubated for 30 min at room temperature. Thereafter, the cells were washed twice with PBS·10% FBS, the supernatant was discarded, and the cell pellet was resuspended in 0.05 ml (of a 1:50 dilution) of fluoresceinated goat F(ab)′2 anti-human IgG or IgM, Fc fragment-specific Ig (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA.). Anti-CD3 Per CP and anti-CD19 PE (Becton Dickinson) were added to allow discrimination of T and B cells. The mixture was then incubated for 20 min at 5°C. After the incubation, 1.0 ml of lysing reagent was added for 4 min to remove red blood cell contamination. The mixture was then centrifuged at 200×g for 5 min and the supernatant discarded; the cells were then resuspended in 0.3 ml of PBS and were ready for the FACSCAN analysis. Background controls consisted of cells with PBS·10% FBS and the goat anti-human Ig, cells with AB+ sera and the goat anti-human Ig, and cells with a high PRA serum (sera collected from at least five patients with 95-100% PRA). Positive FCXM was defined as a shift in the mean channel of fluorescence of ≥20 channels to the right of the T-cell peak or ≥30 channels for the B-cell peak in test sera compared with AB+ control sera.

The decision to transplant a donor organ to a recipient was based on the AHG, DTE/AHG XM result. FCXM was performed retrospectively (usually the day after transplantation) using fresh donor target cells.

Transfusion history. Transfusion information for all patients was obtained from the patient or from hospital records. Forty percent of all patients had positive pretransplant blood transfusion histories.

Date analysis. Patient survivals were calculated using standard actuarial life table techniques ⁽¹⁵⁾. The significance of differences between groups was determined by a Wilcoxon rank sum test ⁽¹⁶⁾. Finally, when appropriate, chi-square analysis was used for statistical correlation.

RESULTS

Between December 1991 and May 1997, 235 end-stage heart disease patients underwent transplantation at The Texas Heart Institute-University of Texas Medical School Transplant Center. The patients underwent transplantation after an AHG-IgG-NEG XM, and the 1-year patient survival rate was 84%. Of these 235 patients, there were 140 primary recipients who were also crossmatched using flow cytometry procedures. We report the correlation of the FCXM results of these 140 recipients to posttransplant rejection and patient survival. The 1-year survival rate of these 140 AHG and FCXM recipients was also 84%.

Table 1 shows the FCXM and graft survival results. All 140 recipients were donor AHG-IgG-XM-NEG and also flow autologous XM-NEG. Fifty-seven of the 140 patients (41%) were FCXM-NEG to donor targets and had an 86% 1-year graft survival rate. However, 22 of the 140 (16%) patients displayed donor IgG(+) FCXM results and had a significantly poorer 1-year survival rate than did the 57 FCXM-NEG recipients (68% vs. 86%, P<0.02). There were 11 IgG(+) FCXM sera reactive to T cells only, 3 sera reactive to B cells only, and 8 sera reactive to both B and T targets. The 1-year survival rate of 92% for IgM(+) FCXM recipients (n=37) was significantly better than the 86% survival rate for FCXM-NEG control recipients (P<0.05). There were 5 IgM(+) FCXM sera reactive to T cells only, 22 sera reactive to B cells only, and 10 sera reactive to both B and T targets. Of great interest was the 1-year survival rate of 79% for the IgG, IgM(+) FCXM recipients (n=24) was intermediate between the 68% survival rate for the IgG(+) only FCXM recipients (P<0.02) and the 92% survival rate for IgM(+) only FCXM recipients (P<0.02). There were 2 IgG, IgM(+) FCXM sera reactive with T cells only, 10 sera reactive with B cells only, and 12 sera reactive with both B and T target cells.

Table 1: Flow cytometry crossmatch results and graft survival^a

Table 2 shows the data for early rejections occurring within 14 days posttransplantation. Only 16% (9 of 57 patients) of the FCXM-NEG recipients experienced early rejections compared with 50% (11 of 22 patients) for the IgG(+) FCXM recipients (P<0.01) and 38% (9 of 24 patients) for the IgG, IgM(+) FCXM recipients (P<0.01). Only 24% (9 of 37 patients) of the IgM(+) FCXM recipients experienced early rejections compared with 50% (11 of 22 patients) for the IgG(+) FCXM recipients (P<0.05). There were no significant differences in frequency of early rejections when comparing FCXM-NEG with IgM(+) recipients and IgG(+) with IgG, IgM(+) FCXM recipients.

Table 2: Early rejections (≤14 days after transplantation)

There were no demographic differences among FCXM-NEG, IgM(+), IgG, IgM(+), or IgG(+) recipients when considering original disease diagnosis, age, gender, race, UNOS status, induction therapy, pretransplant recipient-donor HLA matching, or PRA, or ischemia time. The only difference observed was that IgG, IgM(+) group had a significantly greater percentage of patients treated with pretransplant blood transfusions than did the other groups combined (71% vs. 34%, P<0.001).

Table 3 shows the survival rates to 3 years and the occurrence of CAD. The survival differences observed at 1 year hold up through 3 years of follow-up. The 3-year survival rate of 64% for IgG(+) FCXM recipients is significantly poorer than the 79% survival rate for FCXM-NEG recipients (P<0.02) and the 86% survival rate for the IgM(+) FCXM recipients (P<0.02). Similarly, the 3-year survival rate of 86% for IgM(+) FCXM recipients is still better than the 79% survival rate for FCXM-NEG recipients (P<0.05). Of interest was the observation that development of posttransplant CAD was higher in the IgM(+) FCXM recipient group compared with the FCXM-NEG group (27% vs. 9%, P<0.02).

Table 3: Graft survival and frequency of coronary artery disease^a

DISCUSSION

Preoperative crossmatching for renal transplantation has routinely been performed to obviate the occurrence of hyperacute and/or accelerated rejection. Detection of recipient anti-donor humoral immunity often precludes the transplant. During the last 10 years, XM procedures have changed dramatically, and the clinical relevance of XM results must be correctly interpreted to avoid paring inappropriate donors and recipients and incorrectly denying a recipient a donor organ ⁽¹⁷⁾. For heart transplantation, the XM cannot always be performed before the transplant operation and therefore, in these instances, is not helping to select an appropriate donor. However, a positive XM result after a concurrent or retrospective XM may alert clinicians for rapid intervention in cases of suspected early rejection episodes.

We have used the AHG-NEG and the IgM(+)-AHG XM procedure for renal transplantation since 1989 ^(3,4). At that same time, we began using these XM for heart recipients ⁽⁵⁾. The FCXM procedure is a more sensitive assay to detect the presence of anti-donor antibody reactivity than the AHG-XM procedure ^(7,14). In this study, the patients were all AHG-IgG-XM-NEG. None presented with an IgM(+)-AHG XM. In addition, all patients were found to be flow cytometry auto-XM-NEG. Therefore, the positive flow XM were detecting anti-donor reactivity in the absence of AHG-detected antibody or flow autoantibody. Sixteen percent of our study population displayed IgG(+) FCXM and experienced a significantly poorer survival rate (P<0.02) and more early rejection episodes (P<0.01) than did the FCXM-NEG recipients.

This finding is similar to those of two previous reports on FCXM in heart recipients ^(9,10). In 1991, Shenton et al. ⁽⁹⁾ reported that positive FCXM results predicted those cardiac allograft recipients who were at risk for moderate or severe rejection. More recently, Aziz et al. ⁽¹⁰⁾ reported that FCXM-detected anti-donor antibodies in cardiac allograft recipients identified patients with increased cellular and vascular rejections within the first 6 months postoperatively. The majority of early graft dysfunction and early deaths occurred in FCXM(+) recipients ⁽¹⁰⁾. Therefore, IgG(+) FCXM identifies a subset of AHG-XM-NEG recipients who are at significant risk for early rejection episodes and graft loss.

Risk factors for sensitization, such as transfusion history, recipient-donor HLA mismatches, gender, or %PRA, did not help to predict those recipients displaying IgG(+) FCXM ⁽¹⁰⁾. Moreover, it was not clear whether the IgG reactivity was anti-donor MHC class I or II because the IgG reactivity was directed against both B and T cell targets.

The most provocative finding of our study was that IgM(+) FCXM recipients enjoyed significantly improved survival rates and fewer early rejection episodes. The data suggest a protective role for IgM. Consistent with this hypothesis is that patients displaying both IgG and IgM had a survival rate of 79%, which was between the 68% survival rate for IgG(+) and the 86% survival rate for FCXM-NEG recipients. That IgM could be thought of as an immunoregulatory Ig is not a new idea. In 1978, Iwaki et al. ⁽¹⁸⁾ reported improved renal allograft survival in the presence of IgM anti-B cell antibodies detected in the cold. Subsequently, Iwaki et al. ⁽¹⁹⁾ reported successful transplants across T-warm-positive XM because of IgM. Patients with preformed IgM HLA class II antibodies have been reported to have significantly better graft survival rates than those with anti-HLA class I antibodies ⁽²⁰⁾. In our previous publication, we also noted better survival rates for IgM(+)-AHG recipients than for AHG-XM-NEG recipients ⁽⁵⁾. Recently, McCalmon et al. ⁽²¹⁾ reported that IgM-specific anti-donor HLA antibody did not result in hyperacute renal allograft rejection and was consistent with long-term allograft survival. Finally, some investigators have reported that the presence of IgG anti-HLA antibodies is associated with activated, cyclosporine-resistant cytotoxic T cells, whereas in the case of IgM antibodies, it is associated with naïve cytotoxic T cells that are cyclosporine-sensitive ⁽²²⁾. This observation may explain the different survival outcomes in the presence of IgG compared with IgM.

Finally, the finding of increased incidence of CAD in the IgM(+) FCXM group is provocative. Whether this is directly correlated to CAD needs to be determined in a large patient population study.

In conclusion, the presence of pretransplant, flow cytometry-identified anti-donor IgG antibody correlates with early rejections and poor survival rates for cardiac allograft recipients. IgM may play a protective role against rejection and graft loss. Flow cytometry-identified anti-donor Ig may alert the clinician to those allograft recipients who are at clinical risk and may allow a more intelligent therapeutic intervention. The results of this study need further confirmation in a multicenter study.