HLA match and other immunological parameters in relation to survival, rejection severity, and accelerated coronary artery disease after heart transplant
Marilyn S. Pollack, Christie M. Ballantyne, Charlotte Payton-Ross, Beth Cocanougher, Steven K. Rowe, Gloria J. Grant, John A. Farmer, George P. Noon, Donald G. Weilbaecher, Damien Flores and James B. Young
The Multi·Organ Transplant Center of The Methodist Hospital and Baylor College of Medicine, Houston, TX, U.S.A.
Pollack MS, Ballantyne CM, Payton-Ross C, Cocanougher B, Rowe SK, Grant OJ, Farmer JA, Noon OP, Weilbaecher DO, Flores D, Young JB. HLA match and other immunological parameters in relation to survival, rejection severity, and accelerated coronary artery disease after heart trans- plant. Clin Transplantation 1990: 4: 269-275.
Abstract: The relationship of donor/recipient HLA compatibility to survival, rejection severity and development of coronary arteriopathy after heart transplant is controversial. Furthermore, immunomodulating protocols in these patients have not been based on pretransplant knowledge of HLA match or other immunological parameters. In order to clarify this issue, immunological data from 113 consecutive cardiac transplant patients who were at least 3 months post-transplant were analyzed in relation to graft/patient survival and severity of rejection episodes. Ninety- two of these patients either reached the I-year post-transplant follow-up point and were studied with cardiac catheterization or died with autopsy study of coronary anatomy. These patients were also analyzed for pre- transplant immunological factors and post-transplant development of new lymphocytotoxic antibodies that might predict development of coronary artery disease. Results indicate that better donor/recipient HLA matches, lower pre-transplant T-cell panel reactive antibodies (PRA) and negative T-cell crossmatches are associated with improved graft/patient survival and/or less severe rejection episodes, especially in younger transplant patients. The development of new antilymphocyte antibodies after transplant was associated with the development of new coronary artery disease and with diminished survival. These data, collectively, suggest that patients, especially younger patients, with more poorly matched cardiac transplant donors or pretransplant cytotoxic antibodies should be candidates for more aggressive immunosuppression protocols because of their higher mortality and that all patients should be monitored for the development of new anti-lymphocyte antibodies.
Although, today, donor/recipient HLA matching and pretransplant crossmatch tests would rarely be practical for cardiac transplant recipients, analysis of the role of these and other immunological factors in cardiac graft outcome could be important for selecting high-risk patients for consideration of more aggressive post-operative immunosuppressive therepy. Several recent analyses of such parameters have generated confusion by reporting somewhat inconsistent results. For example, Yacoub et al. (1) and the Collaborative Heart Transplant Study organized by G. Opelz (2) found that good HLA-DR and/or B+DR matches were associated with better graft and patient survival while the Universities of Pittsburgh’s and Pennsylvania’s single-center studies found no relationship of matching parameters to survival (3, 4). A relationship to graft rejection episodes was observed, however, by the Pittsburgh group (4) and at least two other studies (6, 7). One additional study reported a relationship between donor/recipient incompatibility (in this case, development of anti-B cell antibodies) and post-transplant coronary artery disease (CAD) (8) while another study found none (9). Results reported from a third study (10), indicated an association between HLA antibodies and both mortality and CAD.
Since very few patients are well matched to their cadaver donors for HLA antigens, we thought that more relevant clinical data might be obtained if consideration were also given to selected major crossreactive antigen groups. That is, since particular pairs of mismatching HLA molecules are more homologous than other pairs, the role of HLA matching in cardiac transplantation might be clarified if donors mismatched for highly cross-reactive antigens with the recipients were considered to be less poorly mismatched than those mismatched for non-crossreactive antigens. Moreover, studies of sensitized platelet transfusion and renal transplant recipients have clearly shown that selected crossreactive mismatches are well-tolerated (11, 12). Selected crossreactive antigens were, therefore, counted as “half matched” in the analysis described in this report. In view of our own preliminary analyses indicating that older recipient age can be an independent positive factor in cardiac transplant success (13, 14), an analysis of the effect of HLA mismatch was also done separately for younger and older patients.
Patients and donors
All patients selected for cardiac transplantation had intractable congestive heart failure despite medical or surgical treatment, and were without irreversible damage to other vital organs, active infection or concurrent malignancy, pulmonary hypertension, recent pulmonary infarction, and insulin-dependent diabetes mellitus. The 113 consecutive patients included in this study had a mean age of 47.4 yr (range 13-68), a male/female ratio of 88: 12 and an ethnic distribution showing 89% Cau- casion, 7% Black, 3% Hispanic, and 1% Arabic.
Table 1. Relation of rejection episodes to donor recipient HLA mismatch A, Incidence of Severe Rejection
Table 2. Relationship of pretransplant PRA levels to severe rejection episodes in cardiac transplant patients
The donors for the patients in this study had a mean age of 26.9 yr (range 10-59), and all had normal clinical cardiac exams and electrocardiograms. They also were without preexisting heart % disease, cardiac trauma, significant coronary artery disease risk factors, active infection, systemic malignancy, prolonged hypotension or cardiac arrest. All donors were ABO-compatible with the selected recipient.
All patients received prophylactic rejection therapy with azathioprine, corticosteroids and OKT3 and/or cyclosporine (CYC) with early initiation of CYC treatment dependent on individual patient risk status for development of renal dysfunction (15). Individuals were monitored at 10 days, weekly intervals for 1-6 wk, biweekly intervals for 1-2 months, monthly intervals for 4-5 months, and as warranted thereafter with endomyocardial biopsy and supplementary, noninvasive echocardiographic studies and immunological techniques detailed below. Most patients (92) analyzed in this study were also evaluated for development of post-transplant coronary artery disease with angiograms and/ or autopsy evaluation. Angiograms were obtained with non-ionic contrast medium at the l-yr follow- up point. Vessels were classified as either angiograghically normal or as showing any evidence of CAD, whether minimally plaqued (defined as intra-luminal irregularity with or without significant stenosis) or significantly diseased (> 50% luminal obstruction or diffuse tapering). Patients whose post-transplant death was attributed to myocardial infarction were considered to have significant coronary artery disease.
HLA-A,B,C, and DR and DQ typing of patients and donors was performed using standard complement-dependent microcytotoxicity assays with nylon-column procedure-separated T and B-lymphocytes (16), respectively. For most cases, spleen or lymph node tissues were used as the source of donor lymphocytes. Crossmatch tests were performed retrospectively for all patient-donor pairs utilizing, for separated donor T cells, standard NIH, 37°C incubation, Amos wash, and anti-human-globulin (AHG) procedures and for donor B cells, a 37°C incubation procedure (17). Pretransplant panel-reactive antibody (PRA) for all patients was evaluated for T cells using the AHG and for B cells using the standard 37°C crossmatch procedures, respectively. Frozen/thawed T lymphocytes (from approximately 60 donors) and frozen/ thawed B-cell type CLL cells and fresh B lymphocytes (from approximately 30 donors) collectively representing all well defined HLA antigens were used as test cells. Fifty of the 92 patients for whom post-transplant CAD development data were available were also evaluated at approximately yearly intervals after transplant for the development of new anti-T and/or anti-B lymphocyte antibodies.
Results for HLA match and endomyocardial biopsy scores were evaluated in relation to clinical findings using the following scoring systems: The score for HLA-match (mismatch) was obtained by assigning 2 points for each completely mismatched A,B, or DR antigen, 1 point for tolerated cross-reactive antigens and 0 points for completely matched antigens (including “splits”). Decisions regarding tolerated crossreactive antigens were based on our own experience and that of others (e.g., Dahlke and Weiss (11) in evaluating mismatches that are usually non-reactive with sera from sensitized patients (e.g., recipient A2, donor A28 (but not patient A28, donor A2); recipient B7, donor B27 or recipient DR7, donor DRw9). The maximum score (worst match) was, therefore, 12 (2 points for each A,B and DR antigen). Biopsy scores were assigned using the McAllister 1-10 grading scale (15) with scores of 5-7 signifying moderate rejection and scores of 8-10 signifying acute severe rejection. Patients with more than 75% of biopsy scores of 3 or greater were considered to have evidence of chronic myocardial rejection.
All evaluations of immunological parameters, graft biopsies and angiographic assessment of CAD were done blindly without knowledge of other clinical parameters. Survival data were analyzed by Kaplan-Meier life table methods. Analyses of rejection and CAD data were done by the Wilcoxon rank sum method, the Student t-test or by Chi square analysis, as appropriate. Differences between groups were considered to be statistically significant when p ~ 0.05.
Results of these analyses indicated that there were significantly more severe rejection episodes (p= 0.02) and significantly more chronic myocardial rejection (p=0.03) with poorer HLA-A+B+DR matches (higher H L A scores), when the best and worst matched recipient/donor pairs were compared, omitting the middle group (Table 1). HLA mismatching also showed a trend toward a deleterious effect in relation to patient survival (Fig. 1, left), especially for the younger patients (Fig. 1, right, p = 0.07). The worst HLA-matched patients (HLA score 11-12) were also more likely than the best-matched (HLA score 0-7) to develop CAD (33% vs 17%), but this difference was also not statistically significant.
Table 3. Relationship of positive pretransplant crossmatch tests to severe rejection episodes in cardiac transplant patients
Table 4. A. Relationship of development of new anti-T or -B cell antibodies to posttrans- plant CAD
B. Relationship of development of new anti-T or -B cell antibodies to posttrans- plant CAD and/or death from severe rejection.
The analyses also indicate that patients with pretransplant T-cell PRA reactivies 10% or greater had significantly poorer survival (p = 0.02) when compared to those with lower P R A values (Fig. 2, left). Although B-cell PRA activity did not have a significant effect on survival (Fig. 2, right), patients with either T-cell (p=O.Ol) or B-cell (p= 0.0001) PRA reactivities 10% or greater were significantly more likely to have severe rejection episodes (Table 2). Patients with positive (retrospective) pre-transplant T-cell or B-cell crossmatch tests also had a trend towards more severe rejection episodes but this was not statistically significant (Table 3). In the limited number (50) of patients for whom both post-transplant CAD and new antibody data were available, we did not find that the development of new anti-lymphocyte antibodies after transplant was significantly associated with the development of post-transplant coronary artery disease (CAD) (Table 4A). However, when the end points of CAD and death from severe rejection were combined, the development of new anti-B cell antibodies was statistically significant (p = 0.04) as a poor prognostic sign (Table 4B). The development of new anti-T or anti-B lymphocyte antibodies was also significantly associated (p = 0.002) with poor survival per se (Fig. 3).
Two of the primary findings in this study, that patient/donor incompatibility, as assessed by HLA mismatching, and patient sensitization in general (pre-transplant anti T-lymphocyte PRA) are related to survival (Fig. 1) and the incidence of acute severe rejection episodes (Tables 1, 2; Figure 2), are not surprising and are in agreement with some aspects of other studies. In fact, the alternative finding by the Pittsburgh Group that there was no relationship of HLA match with survival (3), although there was with the incidence of graft rejection episodes (5), is confusing since it is expected that survival itself would be correlated with severe rejection episodes. Perhaps the inclusion of partially matched, cross-reactive antigens in our analysis has allowed for a clearer observation of an otherwise subtle effect and other centers should consider this alternative method of analysis.
The finding that younger patients show a more pronounced deleterious effect of HLA mismatching than older patients (Fig. 1, right) is also not surprising in view of the known decreases in immune function with age (18, 19). Indeed, we previously reported (13) that when, during a 24-consecutive month period, 20 of 42 cardiac transplants were performed in patients with ages greater than 50, multivariate analysis of maximal biopsy rejection grades in relation to age and cyclosporine levels indicated that the maximal biopsy score within the first 2 months post-transplant was negatively associated with age even after controlling for cyclosporine use (the regression of rejection grade on age was statistically significant at p = 0.03). The fact that younger patients do not, over all, show poorer survival than older patients undoubtedly reflects a balance between their better general health and their more active immunological responses.
Although HLA mismatch and presensitization is shown in this and has been shown in most other studies (1-2, 5-7) to influence the development of acute rejection and/or survival, the role of these or any other immunologic parameters in the development of accelerated post-transplant CAD remains controversial. In our study, there is a trend toward a role of HLA mismatch in the development of CAD (33% vs 17% in the worst vs best matched groups). Of more statistical significance was the finding that the development of new anti-B cell antibodies is associated with both death from rejection and the development of post-transplant coronary artery disease (Table 4B). One group that had reported (9) that HLA mismatch had no relationship to CAD (the association of CAD being primarily with cytomegalovirus infection) did, in fact, demonstrate a striking trend with 7 of 16 CAD patients (62%) having no matched HLA antigens vs only 15 of 86 patients (17%) without CAD having no matched HLA antigens (9), Table 1). Cytomegalovirus infection (or other unknown factors) could lead to local production of lymphokines like gamma interferon that stimulate expression of HLA antigens in general and induce de novo endothelial cell class II antigen expression. These antigens, in turn, could stimulate immunologically-mediated damage as part of the etiological basis for the CAD. Our finding (Table 4B) – also reported by others (8, 10) – of increased anti-lymphocyte antibodies in post-transplant CAD patients is consistent with this hypothesis. In addition, we find that new anti-lymphocyte antibodies is a poor prognostic indicator in general (Fig. 3), and would suggest that patients be monitored for this at regular intervals.
Although, as noted above, little can be done to avoid mismatching in relation to donor selection, these findings collectively suggest that immunological parameters do affect post-transplant clinical course and that patients with the worst HLA matches and/or pretransplant T-cell PRA ~ 10% should be considered candidates for closer monitoring in relation to both acute and possible chronic rejection (and in relation to post-transplant development of CAD). In fact, because of the very poor prognosis of patients with pre-transplant PRA ~ 10% (Fig. 2, left), our own and several other cardiac transplant centers have endeavored to require pre-transplant T-cell crossmatch tests for this potential recipient group. Although, in our own experience, only 2 of 8 patients with pretransplant T-cell PRA ~ 10% (including Ire-transplant patient not included in the above data analyses) had positive (retrospective) T-cell crossmatch tests, this was a much higher incidence (25%) than in the low P R A group (7 positive crossmatches for 105 patients) (6.7%) (p = 0.06). Further analysis of the inter-relationship of immunological parameters, post-transplant blood lipid levels, the incidence of cytomegalovirus infection and other clinical factors in relation to the development of CAD in our patient population is currently in progress.
Supported in part by the Cullen Trust for Health Care. Computational assistence was provided by the CLINFO project of the National Institutes of Health and the General Clincial Research Clinic of Baylor College of Medicine. The authors also thank Lucy Stewart for expert typing and editorial assistance.
Marilyn S. Pollack, Christie M. Ballantyne, Charlotte Payton-Ross, Beth Cocanougher, Steven K. Rowe, Gloria J. Grant, John A. Farmer, George P. Noon, Donald G. Weilbaecher, Damien Flores and James B. Young
1. YACOUB M, FESTENSTEIN H, DOYLE P, et al. The influence of HLA matching in cardiac allograft recipients receiving cyclosporine and azathioprine. Transplant Proc 1987: 19: 2487.
2. OPELZ G. Collaborative Heart Transplant Study, NeWSletter 1, March 1, 1988.
3. COLSON YL, KORMOS RL, HARDESTY RL, et al. Histocom_ patibility in cardiac transplantation. In: Tenth International Histocompatibility Conference, New Y ork, November 18-23, 1987.
4. SHAKIN-EsHLEMAN SH, CAVAROCCHI NC, ZMIJEWSKI CM. HLA compatibility and clinical outcome among cardiac transplant recipients. Clin Transplantation 1990: 4: 98- 101.
5. ZERBE T, ARENA V, KORMOS R, COLSON Y, GRIFFITH B, HARDESTY R. Role of major histocompatibility complex (HLA) matching in cardiac allograft rejection. Transplant Proc 1988: 20: 1, Supplement 1,74–75.
6. PFEFFER PF, FOERSTER A, FROYSAKER T, THORSBY E. Correlation between HLA-DR mismatch and rejection episodes in cardiac transplantation. Transplant Proc 1987: 19:’ 691.
7. KERMAN RH, VAN BUREN CT, LEWIS RM, FRAZIER OH, COOLEY D, KAHAN BD. The impact of HLA A, B, and D,R blood transfusions and immune responder status on cardiac’ allograft recipients treated with cyclosporine. Transplantation 1988: 45: 333.
8. HESS ML, HASTILLO A, MOHANAKUMAR T, et al. Accelerated atherosclerosis in cardiac transplantation: Role of cytotoxic B-cell antibodies and hyperlipidemia. Circulation ‘ 1983: 68 (Supp II): 94,
9. McDONALD K, RECTOR TS, BRAUNLIN EA, KUBO SH, OLlVARI MT. Association of coronary artery disease in cardiac transplant recipients with cytomegalovirus infection, Am J Cardiol 1989: 64: 359.
10. PETROSSIAN GA, HERNADI S, NICHOLS AB, REED E, POWERS ER. Increased mortality and coronary artery disease in cardiac transplant recipients with lymphocytotoxic anti-HLA antibodies. Circulation 1988: 78: II-251.
11. DAHLKE MB, WEISS KL. Platelet transfusion from donors mismatched for crossreactive HLA antigens. Transfusion 1984: 24: 299.
12. SCORNIK JC, LEFOR WM. Antibodies to crossreactive HLA antigens. Transplantation 1987: 43: 235.
13. POLLACK MS, SHORT HD, YOUNG JB, PIWINSKI SE, CALLAWAY C, DEBAKEY ME. Graft stability in a heart transplant recipient whose immunosuppressive therapy was discontinued for 8 months. Transplantation 1988: 45: 242.
14. BALLANTYNE CM, ROWE SK, POLLACK MS, et al. Relationship of histocompatibility matching and recipient age to rejection in heart transplant patients. Abstracts, Annual Meeting, American Society of Transplant Physicians, 1989.
15. YOUNG JB, LEON CA, LAWRENCE EC, WHISENNAND HH, NOON GP, DEBAKEY ME. Heart replacement for terminal cardiac disease: Cardiac transplantation and mechanical sustenance of the cardiovascular system. Baylor Cardiology Series 1989: 12(2): 4-27 and 12(3): 4-27.
16. DANlLOVS JA, AYOUB G, TERASAKI PI. B lymphocyte isolation by thrombin-nylon wood. In: TERASAKI PI, ed. Histo-combatibility Testing 1980. Los Angeles: UCLA pp. 287-288 .
17. ZACHARY AA, TERESI GA (eds.) ASHJ Laboratory Manual, American Society for Histocompatibility and Immunogenetics, 1990.
18. GREENBERG LJ, YUNIS EJ. Histocompatibility determinants, immune responsiveness and aging in man. Fed Proc 1978: 37: 1258.
19. WALFORD RL. Immunology and aging. Am J C1in Pathol 1980: 74: 247.