Contact Sensitization Pretransplantation PredictsAcute Hepatic Allograft Rejection

ANDREW BATHGATE,1 MATTHIAS DOLLINGER,2 JOHN PLEVRIS,1 CHRISTOPHER BELLAMY,2 AND PETER HAYES1

The loss of hepatic allografts to the rejection processes isnow relatively rare, and the reduction of adverse effectsrelated to immunosuppressive therapy is becoming moreimportant as patients survive longer after transplantation.We therefore investigated the response to a contact neoan-tigen before liver transplantation as a predictor of acuterejection after transplantation. Forty-one patients withchronic liver disease were sensitized with 0.1% diphenylcy-clopropenone while on the waiting list for orthotopic livertransplantation. Fourteen days later an elicitation reactionwas performed with 5 different concentrations of diphenyl-cyclopropenone. Nineteen responded to diphenylcyclopro-penone (score range, 1-9). Twenty-two patients had no re-sponse. Three patients died before transplantation (allnonresponders). Twelve (63%) of 19 responders had treat-ment for acute rejection compared with 1 of 19 nonre-sponders (P < .0001). In addition univariate analysis re-vealed recipient age, donor age, Child-Pugh class, andimmunosuppressive agent to be associated with acute rejec-tion. On multivariate analysis only skin test response was asignificant predictor of acute rejection (P 5 .02). All nonre-sponders had no or only mild rejection on biopsy, but 12 of19 responders had moderate or severe acute rejection onbiopsy. All patients requiring additional therapy to a singlecourse of corticosteroids for acute rejection had skin testscores greater than 1. We concluded that patients who donot respond to diphenylcyclopropenone sensitization be-fore transplantation develop at most mild acute rejectionand that skin test scores identify patients with troublesomerejection. Evaluation of skin test responses to a contact neo-antigen may facilitate tailoring of immunosuppressive ther-apy. (HEPATOLOGY 2001;33:1043-1046.)

Acute cellular rejection is common after orthotopic livertransplantation. A single episode of acute cellular rejection,unless it is severe, is not detrimental to graft survival1,2 al-though it may increase morbidity.3 At present the same im-

munosuppressive regimens are given to all patients. Identifi-cation of factors that could predict rejection may allowtailoring of immunosuppression to the individual permittingreduced immunosuppression in patients with low risk ofacute rejection and close monitoring of those with an in-creased risk.

A recent report of pretransplantation parameters of pa-tients, which predict subsequent development of acute rejec-tion, identified recipient age, donor age, HLA-DR mismatch,serum creatinine, etiology of liver disease, and cold ischemiatime as risk factors.2 Data published from our unit largelyagreed with these findings.4 The acute rejection of hepaticallografts is principally a T-cell–mediated response with acharacteristic histologic appearance of bile ductulitis, venousendotheliitis, and portal tract inflammation.5 The histologicseverity of acute rejection influences the decision to treat be-cause graft outcome is worse after an episode of severe rejec-tion.2

Contact hypersensitivity reactions are T-cell–mediated re-sponses, which are diminished in liver disease.6 The aim ofthis study was to investigate the relationship between contactsensitization to a neoantigen pretransplantation in patientswith chronic liver disease and acute cellular rejection post-transplantation with a view to individualization of immuno-suppressive regimens.

PATIENTS AND METHODS

Ethics. The local ethical committee approved the proposal for thisstudy and all patients gave written informed consent.

Patients. Consecutive patients with chronic disease listed for or-thotopic liver transplantation between February 1997 and August1998 were sensitized with 100 mL of 0.1% diphenylcyclopropenone(University of Nijmengen, Nijmengen, The Netherlands) applied ona filter paper under a 2-cm Finn chamber (Epitest Ltd. Oy, Tuusula,Finland) to the upper arm for 48 hours. An elicitation test was ap-plied to the opposite forearm 12 days later with 15 mL of diphenyl-cyclopropenone at concentrations of 0.001%, 0.0025%, 0.005%,0.01%, and 0.025% placed on 8-mm filter paper discs on a strip ofFinn chambers (Epitest Ltd. Oy). This was removed by one person(A.B.), and the elicitation reaction was recorded. Each concentrationwas given a score: erythema and induration 5 1, vesicles 5 2, bulla 53. The scores for each concentration were summated giving a totalscore out of 15. This scoring system has been used by others usingdifferent contact sensitizing agents.7

All transplantations were performed at the Scottish Liver Trans-plant Unit between April 1997 and December 1998. The immuno-suppression regimen was triple therapy with either microemulsioncyclosporin 10 mg/kg/d or tacrolimus 0.1 mg/kg/d (target troughlevels were 175-200 nmol/L for cyclosporin and 10-15 ng/L for ta-crolimus), azathioprine (2 mg/kg), and prednisolone 20 mg/d. Ta-crolimus (Fujisawa, Munich, Germany) or cyclosporin (Neoral; No-

From the 1Scottish Liver Transplant Unit, Royal Infirmary of Edinburgh, LauristonPlace, Edinburgh, Scotland, UK; and the 2Department of Pathology, University of Edin-burgh, Teviot Place, Edinburgh, Scotland, UK.

Received September 25, 2000; accepted February 15, 2001.M.D. was supported by Fujisawa Ltd.Address reprint requests to: Andrew Bathgate, M.B., Ch.B., Department of Medicine,

Royal Infirmary of Edinburgh, Lauriston Place, Edinburgh, EH3 9YW, UK. E-mail:abathgate@ed.ac.uk; fax: (44) 131 229 2948.

Copyright © 2001 by the American Association for the Study of Liver Diseases.0270-9139/01/3305-0005$35.00/0doi:10.1053/jhep.2001.24030

1043

vartis, East Hanover, NJ) administration was randomly assigned topatients.

Acute rejection was defined as rejection requiring treatment withhigh dose corticosteroids. This decision was based on clinical andhistologic evidence and was made by the clinicians on the unit at thetime. These clinicians were blinded to the skin test score. Protocolbiopsies were performed at 7 days and further biopsies performed ifliver function tests deteriorated or were slow to improve after rejec-tion therapy. All biopsies in the first 30 days after transplantationwere reviewed. The histologic severity of acute rejection was gradedaccording to the Banff criteria.8 When more than one allograft biopsywas performed the most severe degree of rejection was used in anal-yses.

Details of the donor age and cold ischemia time were collected.Recipient age, etiology, severity of liver disease, and nutritional sta-tus (anthropometry assessed by a single dietician) were noted pre-transplantation. Lymphocytotoxicity testing and donor recipientHLA-DR mismatching were performed using standard techniques bythe tissue typing laboratory in our institution. Calcineurin inhibitorlevels were collected for each patient.

Chronic rejection was diagnosed using standard histologic criteriaafter appropriate clinical/radiologic exclusions.9 Graft loss was de-fined as patient death or retransplantation.

Follow-up and Statistics. All patients were followed-up for at least 1year. Univariate analysis was performed using Students t test forcontinuous variables or x2 testing with Fisher’s exact test, whereappropriate, for categorical variables. Logistic regression for univar-iate predictors was performed using SPSS statistical package (version9.0; SPSS Inc., Chicago, IL). P , .05 was taken as significant.

RESULTS

Skin Tests. Forty-one of 44 patients completed elicitationtests before transplantation. Nineteen of the 41 (46%) patientswere responders having a skin test score varying from 1 to 9.The mean age of the responders was 46.6 (SEM 8.9) comparedwith 55.8 (7.6) years (P 5 .001). There was no significantdifference between the responders and nonresponders withrespect to severity of liver disease as assessed by Child-Pughclassification or nutritional status as measured by anthropo-metric measurements. The skin test scores in the differentetiologies are shown in Table 1.

Correlation With Acute Rejection. Thirty-eight of 41 patientssurvived to transplantation at a median interval of 34 days(range, 1-301) from elicitation to transplantation. The 3 pa-tients who died while waiting transplantation were nonre-sponders. Acute rejection requiring treatment occurred in 19(50%) of 38 patients. Five of these 19 patients, all responders,required multiple allograft biopsies (2-4) in the first 30 days.Three of the 5 had additional therapy for acute rejection.Seven of 38 patients did not undergo biopsies at 7 days at the

discretion of the clinical team at the time. None of these 7patient required allograft biopsy in the first 30 days, and theywere regarded as negative for acute rejection.

Table 2 shows the occurrence of acute rejection requiringtreatment according to skin test response, donor and recipientage (median values used as cutoff), immunosuppression lev-els, and severity of liver disease. No difference in rejection wasseen according to etiology of liver disease, cold ischemia time,HLA-DR mismatch, or lymphocytotoxic crossmatch (data notshown).

On univariate analysis skin test response (P , .001), donorage (P 5 .05), recipient age (P 5 .05), and primary immuno-suppression (P 5 .003) were significantly associated withacute rejection requiring treatment. On multivariate analysisskin test response was the only independent factor associatedwith acute rejection (P 5 .02).

The original histologic assessments were performed by 1 of3 pathologists. A fourth pathologist (C.B.) rescored all thebiopsies to eliminate interobserver variation in scoring. Theseresults are shown in Table 3. There was no increase in thescore of severity of rejection as assessed by the single pathol-ogist, but 6 biopsies were scored as mild compared with mod-erate and 6 biopsies were scored as having no rejection com-pared with mild rejection. Table 4 shows the occurrence of

TABLE 1. Skin Test Score According to Etiology of Liver Disease (n 5 41)

NoResponse

TestScore 1-3

TestScore 4-6

TestScore 7-9

Primary biliary cirrhosis(n 5 17) 9 4 3 1

Alcohol-induced liver disease(n 5 10) 7 3 0 0

Primary sclerosing cholangitis(n 5 5) 2 0 1 2

Chronic viral hepatitis (n 5 4) 0 2 2 0Autoimmune hepatitis (n 5 2) 2 0 0 0Cryptogenic cirrhosis (n 5 3) 2 1 0 0

TABLE 2. The Occurrence of Acute RejectionRequiring Treatment (n 5 38)

NoRejection

AcuteRejection Significance

Elicitation responseNo 18 1 P , .001Yes 5 14

Age (yr),53 8 10 P 5 .05$53 15 5

Donor age,45 8 10 P 5 .05$45 15 5

ImmunosuppressionTacrolimus 16 3 P 5 .003Cyclosporin 7 12

Mean level first 7 daysTacrolimus 9.5 13 P 5 NSCyclosporin 185 218

Child-Pugh classA 1 1 P 5 .03B 6 10C 16 4

Abbreviation: NS, not significant.

TABLE 3. Skin Test Score and Histologic Severity of Acute Rejection asScored by a Single Pathologist (n 5 31)

Skin TestScore

NoRejection

MildRejection

ModerateRejection

SevereRejection

0 7 5 0 01-2 2 3 2 13-4 0 2 2 15-6 0 0 3 07-8 0 0 1 2

1044 BATHGATE ET AL. HEPATOLOGY May 2001

moderate/severe acute rejection compared with no/mild rejec-tion in responders and nonresponders.

All 5 patients who required more than a single biopsy in thefirst 30 days had a skin test score greater than 1 compared withno patients with a score less than or equal to 1 (P , .001).

Follow-up. Two patients, both nonresponders, died withinthe first 30 days from sepsis and multiorgan failure. One allo-graft, in a patient with a skin test score of 9, was lost to chronicrejection at 8 weeks. The other graft loss was to hepatic arterythrombosis at 6 weeks in a patient with no skin test response.All other grafts were functioning well at 12 months.

DISCUSSION

The occurrence of acute cellular rejection in patients with-out hepatitis C infection does not detrimentally affect graftoutcome unless it is severe.2 It may therefore be possible toreduce immunosuppression in patients with a decreased riskof acute rejection in the knowledge that, should they developacute rejection as a consequence, there is no evidence to sug-gest that this will adversely affect outcome. On the other hand,patients with a high risk of acute rejection should be moni-tored carefully to maintain immunosuppressive levels be-cause severe acute rejection and recurrent rejection can ad-versely affect graft outcome.

The occurrence of early acute rejection requiring treatmentin this study was similar to that in previous years in our center(50%-55%) for patients transplanted for chronic liver diseaseundergoing immunosuppression with triple therapy. Thissuggests that the process of contact sensitization itself did notfacilitate the occurrence of acute rejection. This is furthersupported by the fact that no patient developed a reaction atthe site of sensitization at the time of acute rejection.

The pretransplantation parameters investigated that havebeen shown to affect posttransplantation rejection are age,etiology of liver disease with alcohol-induced liver diseasehaving reduced acute rejection, preoperative creatinine, se-verity of liver disease, and nutritional status.2,4,10,11 Therehave been some reports suggesting that cold-ischemia timeand donor age may also influence acute rejection, althoughthis is not every center’s experience.12 All these parameterswere investigated in our population and none was found to beas useful as skin testing in predicting those less likely to sufferacute rejection. The threshold for treatment variesfrom unit to unit with mild acute rejection often not beingtreated.13 In patients without a response to diphenylcyclopro-penone, the most severe rejection seen was mild rejectionsuggesting that there is scope to diminish immunosuppres-sion in these patients.

In this study we also found increased severity of acute re-jection in patients with high skin test scores. Because there arereports suggesting that severe acute rejection does lead to adecrease in graft survival, it was interesting that in this studythe patient who developed chronic rejection had the highest

skin test score. In addition the only other patients who re-quired additional therapy to a single course of methylpred-nisolone had skin test scores greater than 1.

The immunosuppressive agents used also affected acuterejection in our study. There is already evidence to suggestthat tacrolimus is more effective in preventing acute rejec-tion,14,15 although our study involved the new microemulsionpreparation of cyclosporin of which there is less comparativedata.

There are studies suggesting an increased susceptibility toinfection in patients with impaired responses to skin testing.Two of the 3 patients who died while on the waiting list diedfrom sepsis and the 2 patients who died in the first 2 monthsalso had infection and multiorgan failure. The potential ben-efit of reducing immunosuppression in patients with no re-sponse on skin testing may be to reduce the early problemsassociated with infection in the postoperative period. Anotherpotential benefit of reducing immunosuppression may be toprevent renal toxicity because there is evidence to suggestthat the long-term renal problems seen with cyclosporin areinfluenced by levels in the first 30 days. This study, however,investigated patients given the older preparation of cyclo-sporin.16

Our results do suggest a relationship between pretransplan-tation reaction to a contact neoantigen, and the logical nextstep would be to study a reduction in immunosuppression inthe patients with no elicitation response in the early trans-plantation period to determine any reduction in adverse ef-fects and also graft outcome. There is also potential for inves-tigating patients in whom a reduction in immunosuppressionis desired because of established adverse effects, such as renalimpairment or malignancy, to determine if there is any rela-tionship to skin test scores while on immunosuppression andpropensity to develop acute rejection after a reduction in im-munosuppression.

In conclusion, we present data showing that the inability tomount a response to a contact neoantigen pretransplantationpredicts a reduced likelihood of developing acute cellular re-jection after transplantation. We have shown a relationshipbetween skin test scores and severity of acute rejection. Thesefindings may provide a basis for tailoring immunosuppressiveregimens to the individual.

REFERENCES

1. Dousset B, Conti F, Cherruau B, Louvel A, Soubrane O, Houssin D,Calmus Y. Is acute rejection deleterious to long-term liver allograft func-tion? J Hepatol 1998;29:660-668.

2. Wiesner RH, Demetris AJ, Belle SH, Seaberg EC, Lake JR, Zetterman RK,Everhart J, et al. Acute hepatic allograft rejection: incidence, risk factors,and impact on outcome. HEPATOLOGY 1998;28:638-645.

3. Fisher LR, Henley KS, Lucey MR. Acute cellular rejection after livertransplantation: variability, morbidity and mortality. Liver Transpl Surg1995;1:10-15.

4. Bathgate AJ, Hynd P, Sommerville D, Hayes PC. The prediction of acutecellular rejection in orthotopic liver transplantation. Liver Transpl Surg1999;5:475-479.

5. Snover DC, Sibley RK, Freese DK, Sharp HL, Bloomer JR, Najarian JS,Ascher NL. Orthotopic liver transplantation: a pathological study of 63serial liver biopsies from 17 patients with special reference to the diag-nostic features and natural history of rejection. HEPATOLOGY 1984;4:1212-1222.

6. Pirisi M, Vitulli D, Falleti E, Fabris C, Soardo G, Del Forno M, Bardus P,et al. Increased soluble ICAM-1 concentration and impaired delayed-typehypersensitivity skin tests in patients with chronic liver disease. J ClinPathol 1997;50:50-53.

TABLE 4. Severity of Rejection in Respondersand Nonresponders, n 5 31 (P < .001)

No/MildRejection

Moderate/SevereRejection

Nonresponder 12 0Responder 7 12

HEPATOLOGY Vol. 33, No. 5, 2001 BATHGATE ET AL. 1045

7. Watson MA, Briggs JD, Diamandopoulos AA, Hamilton DN, Dick HM.Endogenous cell-mediated immunity, blood transfusion, and outcome ofrenal transplantation. Lancet 1979;2:1323-1326.

8. Demetris AJ, Batts K, Dhillon AP, Ferrell L, Fung J, Geller S, Hart J. Banffschema for grading liver allograft rejection: an international consensusdocument. HEPATOLOGY 1997;25:658-663.

9. International Panel. Update of the international Banff schema for liverallograft rejection: working recommendations for the histopathologicstaging and reporting of chronic rejection. HEPATOLOGY 2000;31:792-799.

10. Weisner RH, Lombadero M, Lake J, Everheat J, Detre KM. Liver trans-plantation for end stage alcoholic liver disease: an assessment of out-come. Liver Transpl Surg 1997;3:231-239.

11. Farges O, Saliba F, Farhamant H, Samuel D, Bismuth A, Reynes M, Bis-muth H. Incidence of rejection and infection after liver transplantation asa function of the primary disease: possible influence of alcohol and poly-clonal immunoglobulins. HEPATOLOGY 1996;23:240-248.

12. Shackleton CR, Martin P, Melinek J, Stothers L, Millis JM, Olthoff KM,Imagawa DK, et al. Lack of correlation between the magnitude of pres-ervation injury and the incidence of acute rejection, need for OKT3, andconversion to FK506 in cyclosporine-treated primary liver allograft re-cipients. Transplantation 1995;60:554-558.

13. Dousset B, Hubscher SG, Padbury RT, Gunson BK, Buckels JA, MayerAD, Elias E, et al. Acute liver allograft rejection—is treatment alwaysnecessary? Transplantation 1993;55:529-534.

14. The U.S. Multicenter FK506 Liver Study Group. A comparison of tacroli-mus (FK 506) and cyclosporine for immunosuppression in liver trans-plantation. N Engl J Med 1994;331:1110-1115.

15. European FK506 Multicentre Liver Study Group. Randomised trial com-paring tacrolimus (FK506) and cyclosporin in prevention of liver allo-graft rejection. Lancet 1994;344:423-428.

16. Fisher NC, Nightingale PG, Gunson B, Lipkin G, Neuberger JM. Chronicrenal failure in liver transplant recipients treated with cyclosporin A: aretrospective analysis. Transplantation 1998;66:59-66.

1046 BATHGATE ET AL. HEPATOLOGY May 2001