Post-transplantation lymphoproliferative disorders: advances in

Ž .Progress in Pediatric Cardiology 11 2000 145]157

Post-transplantation lymphoproliferative disorders:advances in diagnosis, prevention and management in

children

Steven A. Webber a,U, Michael Greenb

aDi ision of Cardiology, Uni ersity of Pittsburgh School of Medicine, Pittsburgh, PA, USAbDi ision of Allergy, Immunology & Infectious Diseases, Uni ersity of Pittsburgh School of Medicine, Pittsburgh, PA, USA

Abstract

With improving operative and postoperative survival after pediatric thoracic transplantation, attention has appropriatelybegun to focus on complications of long-term immunosuppression. One important complication is post-transplantationlymphoproliferative disorders. Much has been learnt about this spectrum of disorders over the last 15 years, including thepivotal role of primary Epstein]Barr virus infection in the etiology of most cases. Despite these advances, nomenclatureremains confusing for the clinician, prediction of outcome is imprecise and treatment strategies are poorly defined. Indeed, notreatments have been subjected to comparative prospective clinical trials. Only recently has attention focussed on strategiesfor prevention. This article will review the current state of knowledge of post-transplantation lymphoproliferative disorders,with emphasis on recent advances in diagnosis, prevention and management. Q 2000 Elsevier Science Ireland Ltd. All rightsreserved.

Keywords: Post-transplantation complications; Infancy and childhood; Lymphoproliferative disorders; Epstein]Barr virus; Immunosuppres-sion

1. Etiology: importance of primary Epstein–Barr virus( )EBV infection

Post-transplantation lymphoproliferative disordersŽ .PTLD are a spectrum of conditions that straddle theborders between infection and malignant neoplasia. Ithas long been recognized that normal immunesurveillance is essential for control of viral infectionand for prevention of development of neoplasms.

ŽLymphomas may arise in congenital e.g. X-linked. Žlymphoproliferative disease as well as acquired e.g.

.Acquired Immunodeficiency Syndrome immunodefi-

U Corresponding author. Present address: Division of Cardiology,Children’s Hospital of Pittsburgh, 3705 Fifth Avenue, Pittsburgh,PA 15213, USA. Tel.: q1-412-692-5541; fax: q1-412-692-6991.

Ž .E-mail address: [email protected] S.A. Webber

ciency syndromes. The association between therapeu-tic immunosuppression, as occurs in the transplantrecipient, and the development of lymphoid tumors

w xwas first recognized in 1968 1 . Recognition of anassociation between EBV infection and X-linked lym-phoproliferative disease stimulated a successful search

w xfor this virus within PTLD lesions 2 . Subsequently, alarge literature has accrued demonstrating the pivotalrole of Epstein]Barr virus in most cases of PTLD inpediatric solid organ transplant recipients. Evidence

w xcomes both from epidemiological observations 3 , aswell as from demonstration of EBV genome within

w xlesions 4 . We have found that approximately 80]90%of cases in pediatric thoracic recipients are EBVdriven, including all cases arising in the first 3 yearsafter transplantation. This is broadly comparable tofindings reported for other organ transplants and in

1058-9813r00r$ - see front matter Q 2000 Elsevier Science Ireland Ltd. All rights reserved.Ž .PII: S 1 0 5 8 - 9 8 1 3 0 0 0 0 0 4 4 - 8

( )S.A. Webber, M. Green r Progress in Pediatric Cardiology 11 2000 145]157146

adult recipients. In our experience, EBV negativecases usually arise late after transplantation, generallybeyond 5 years and their etiology remains an enigma.We have searched for other viruses that are known tocause lymphoma or other malignancies in AcquiredImmunodeficiency Syndrome, such as Human HerpesViruses 6 and 8, but have not found evidence forthese in EBV negative PTLD in the small number ofpatients that we have evaluated.

The source of EBV infection is not always clear.What is apparent is that most cases of PTLD occur inpatients who are seronegative for EBV at the time oftransplantation and who subsequently develop pri-

w xmary EBV infection 3 . Since the vast majority ofadults are seropositive at the time of transplantation,it is not surprising that PTLD occurs at a much lowerfrequency in adult recipients. Approximately 5]15%of all cases in pediatric thoracic transplantation havearisen in recipients who were seropositive at the timeof transplantation. In our own series, we found that18r22 cases of PTLD arose in patients who were

Ž . w xseronegative at transplantation Table 1 5 . Zangwilland colleagues noted that 12 of their 13 cases ofPTLD in pediatric heart recipients arose in patientswho were seronegative at the time of transplantation,and post-transplantation seroconversion was docu-

w xmented in all 12 cases 6 . The strong associationbetween the acquisition of primary EBV infectionafter transplantation and the risk of PTLD appears tobe true for all types of solid organ transplantationwith the exception of intestinal transplantation. In thelatter setting, EBV seropositive children frequently

w xdevelop PTLD 7 . This may reflect the intensity ofthe immunosuppressive regimens used, and the highload of lymphoid tissue that is contained within theintestinal allograft.

In the seronegative thoracic recipient, primary EBVinfection could come from several potential sources,

Ž .including the donor organ s , perioperative use ofblood products, or from subsequent community acqui-sition of the virus. In our own series, 23% of patients

Žin the primary mismatch group donor EBV seroposi-.tiverrecipient seronegative went on to develop

PTLD. When both recipient and donor were seroneg-ative, 16% of recipients developed PTLD. The demo-

graphic features of PTLD differed between these twogroups. In the ‘primary mismatch’ group, the mediantime to onset of PTLD was 4 months after transplan-tation, with all cases occurring within 1 year of trans-plantation. In the group with seronegative donors, a

Žmuch broader range of time to onset was observed 4.months]7 years . These observations provide indirect

evidence that the donor may be an important sourceof infection in many patients in the primary mismatchgroup. The low incidence of EBV seroconversion andPTLD in the early months after transplantation inseronegative recipients with seronegative donors sug-gests that perioperative blood transfusion is an un-likely source of EBV infection in pediatric thoracicrecipients. Interestingly, a small number of studieshave now used molecular techniques to unequivocally

w xdemonstrate that EBV was of donor origin 8,9 . Inone of these studies, a single donor provided kidneyand heart]lung block to two different recipients, onebeing EBV positive pre-transplantation and the otherbeing EBV negative. Both developed PTLD with EBV

w xof donor origin 8 . These observations suggest thatsome cases of PTLD associated with ‘reactivation’may in fact represent reinfection with a separatestrain of EBV

2. Frequency and risk factors

It is clear that acquisition of a primary EBV infec-tion post-transplantation is the most important riskfactor for the development of PTLD. It remains to beexplained, however, why the majority of patients whodevelop a primary EBV infection do not developPTLD. Clearly, there must be interplay with otherrisk factors. These factors have not yet to been fullyelucidated, but undoubtedly will include overall inten-sity of immunosuppression. In some studies use ofOKT3 as well as other lympholytic agents have beenassociated with increased risk of development of

w xPTLD 10,11 . Heart]lung and lung recipients receivehigher doses of immunosuppression than patients re-ceiving an isolated heart allograft. It is likely that thisincreased level of immunosuppression, and perhapsthe higher lymphoid load transmitted with the allo-

Table 1aOccurrence of EBV-associated PTLD based on donor and recipient EBV serologies at time of transplantation

EBV status DqrRy DyrRy Dq or yrRq

Number of patients 26 31 88Ž . Ž . Ž . Ž .Number developing PTLD % 6 23% 5 16% 4 4.5%

a Data from pediatric thoracic transplant program at Children’s Hospital of Pittsburgh, PA, USA. Notes: Seven additional cases of PTLDŽ .occurred among recipients with unknown donor serologies total 22 cases among 210 recipients . All seven were EBV seronegative at time of

transplantation. Abbre¨iations: D, donor; R, recipient; q, seropositive; y, seronegative.

( )S.A. Webber, M. Green r Progress in Pediatric Cardiology 11 2000 145]157 147

Žgraft and thus higher EBV inoculum in the seroposi-.tive donor , contributes to the greater frequency of

PTLD in heart]lung and lung recipients.Currently, there is considerable controversy as to

whether tacrolimus based therapy is associated with agreater frequency of development of PTLD than cy-closporine-based regimens. Recent reports suggest this

w xmay be the case for pediatric liver 12 and for renalw x13 transplantation, though not all studies have con-

w xfirmed this observation 11 . It should be noted, how-ever, that no pediatric randomized trials comparingtacrolimus and cyclosporine therapy have been per-formed and these observations are based on historicalcontrols or registry data. Among pediatric heart recip-ients, we have noted comparable frequencies of PTLDbetween cyclosporine and tacrolimus treated patients

w xup to this time 14 , and our incidence in tacrolimus-treated patients is similar to that reported by the

w x w xColumbia 6 and Stanford University 15 groups us-ing cyclosporine based therapy.

It is currently unclear, whether age per se is a riskfactor for the development of PTLD or whether thismerely reflects donor and recipient EBV status.Younger age will be associated with lower risk ofpre-transplantation seropositivity, but also with lowerrisk of the donor being EBV seropositive sinceyounger recipients are more likely to receive organsfrom young donors to avoid excessive donorrrecipientsize mismatch. The Loma Linda group have observeda relatively low incidence of PTLD in infant heart

Žrecipients 11 of 255 recipients, 4.3%; R. Chinnock,.personal communication . Although this may reflect

the low frequency of donorrrecipient EBV mismatch-ing, it is of interest to speculate whether maternal]fetal transmission of anti-EBV antibodies might pro-tect the young infant during the period of greatestimmunosuppression early after transplantation. The

Žprecise role of antibody as opposed to cytotoxic T.cell responses in the protection of the patient against

the development of PTLD is unknown, though severalreports have suggested that lack of antibody response

Ž .to EBV nuclear antigens EBNA may increase thew xrisk of developing PTLD 16 . These observations are

important as prophylactic use of intravenous antibodypreparations containing high titers of anti-EBV anti-bodies has been suggested as a possible means of

Ž .preventing PTLD see below .Several problems in the definition and analysis of

PTLD cases have led to inconsistencies in the re-ported ‘incidence’ of this complication. Although therisk is highest in the first year after transplantation,the patient remains at risk indefinitely. Thus, thefrequency in any series will depend on post-operativesurvival and length of follow-up, as well as on casedefinition. Patients dying soon after transplant are ‘at

risk’ for only a very short period. Some reports have,therefore, quoted the frequency among 30-day sur-vivors. A more appropriate method of analysis for atime related event is to use Kaplan]Meier or similar‘survival analysis’ techniques. An overall ‘frequency ofoccurrence’ of PTLD of 4]10% for pediatric heart

w xrecipients 6,15,17,18 and approximately 10]20% forpediatric heart]lung and lung recipients have been

w xreported 18,19 . None of these studies used ‘survivalanalysis’ methodology.

A second, and perhaps more important, limitationis the method of case definition of PTLD used withinany given center. Most centers have not includedinfectious mononucleosis and other EBV-associatedviral syndromes as cases of PTLD, though they repre-sent part of the continuum of EBV driven pathology.

ŽWhen histological findings e.g. of enlarged lymph.nodes, tonsils or adenoidal tissues reveal only benign

lymphoid hyperplasia with architectural preservation,many centers do not consider the tissue to represent aPTLD. The importance of this variability in casedefinition is illustrated by the Columbia group, whofound that inclusion of such cases resulted in an

w xapproximate doubling of cases of PTLD 6 . Inclusionof these more ‘histologically benign’ cases will alsoresult in improved overall clinical outcomes since this

Ž .group of patients see below carry a very favorableprognosis. Thus, variation in the case definition willresult in problems in comparing both ‘incidence’ andoutcomes between centers

3. Clinical presentation and diagnostic evaluation

EBV is increasingly recognized to be associatedwith a wide range of disease manifestations in trans-plant recipients. This includes a non-specific viralsyndrome, mononucleosis, and PTLD including

ŽEBV-associated malignant lymphoma e.g. Burkitt’s.lymphoma . Although this classification is useful, it is

important to note that EBV presents as a continuousspectrum of illness, and benign manifestations canevolve to more serious syndromes within individualpatients. Furthermore, non-PTLD viral syndromes arenot always benign, and fatal viral sepsis may occur inthe absence of mass lesions.

The diagnosis of EBV disease in pediatric thoracicrecipients is based on clinical history and physicalexamination in combination with laboratory confir-mation. The most important factor in making thisdiagnosis is maintenance of a high index of suspicionat all times. In our experience, earlier diagnosis ap-pears to have correlated with more successful out-comes. A critically sick child with disseminated dis-ease involving multiple extranodal sites was seen on anumber of occasions in our early pediatric transplant


experience. These patients frequently died within daysor weeks of presentation, often from infection withsecondary pathogens. Such cases are now exceedinglyrare, suggesting that earlier diagnosis and improvedmanagement may be responsible for the changingpattern of presentation in this population.

In contrast to these severe manifestations, a historyof lethargy, malaise, weight loss and fever are nowcommon presentations of PTLD. An additional his-

Žtory of vomiting andror diarrhea which may be guiaic.positive is suggestive of gastrointestinal involvement,

which has been increasingly recognized to be a com-mon site of disease in pediatric thoracic recipients.Intestinal hemorrhage, obstruction and perforationmay also be seen in patients with intestinal PTLD.Surprisingly, the highest risk period for the later maybe during therapy when necrosis of transmural lesionsmay develop. Pulmonary disease is also very commonin cardiac recipients and is almost invariably present

Ž .in lung recipients Fig. 1 . Pulmonary presentationŽ .ranges from asymptomatic nodule s on routine chest

radiograph to life threatening pulmonary dysfunctionin the lung allograft. The latter may resemble lungrejection on chest radiograph with rather diffuse con-solidation without clearly defined mass lesions. Thismay lead to inadvertent augmentation of immunosup-

w xpression with severe consequences 19 . Other lesscommon presentations include persistent sore throat,adenopathy, and cutaneous nodules, as well asseizures, headaches and focal neurological lesions withCNS disease. We have even diagnosed PTLD fol-lowing elective removal of enlarged tonsils and ade-noids in a child developing airway obstruction duringconscious sedation for routine endomyocardial biopsy.Interestingly, the heart is the only organ transplant inwhich there is not a strong tendency for the disease to

involve the allograft. In most other solid organ trans-plants, allograft dysfunction may be a manifestationof PTLD and may mimic acute or chronic rejection.We have not seen cardiac dysfunction in a child dueto direct involvement with PTLD, though cardiac in-

w xvolvement has been reported 20 .Although physical examination may not reveal spe-

cific findings, there will frequently be evidence ofpallor, weight loss, peripheral adenopathy, and hep-atosplenomegaly. A full physical examination is essen-tial and should include thorough neurological exami-nation. Examination of the entire skin and sites of alllymph nodes is warranted. Careful examination of theoropharynx is required and should include evaluationby an ENT surgeon if there is clinical evidence oftonsillar or adenoidal hypertrophy.

An overview of the laboratory evaluation for thediagnosis of PTLD is given in Table 2. Initial evalua-tion should include a complete blood count with whitecell differential and platelets. Leukopenia, often inassociation with atypical lymphocytosis, as well asthrombocytopenia are frequent findings. Anemia iscommon and may be normocytic and normochromicor may demonstrate findings of iron deficiency whenoccult gastrointestinal bleeding is present. Stools

Žshould be tested for blood. End organ function liver,.kidney should be evaluated. Elevations in uric acid

and lactate dehydrogenase are common and shouldalso be sought on blood chemistry testing. Immuno-globulin levels may be elevated. This is particularlytrue of IgE, perhaps due to a particular cytokine

Ženvironment present with PTLD so-called T helper.type 2 predominance . Serum protein electrophoresis

should also be performed. Although many patients donot demonstrate a monoclonal or oligoclonal gammo-

Fig. 1. Chest radiographs of a 5-year-old girl 3 years out from heart transplantation. She presented with low-grade fevers and mild respiratorysymptoms. The right-sided pulmonary infiltrates were treated as a pneumonia with oral antibiotics. Failure of clinical response ultimately ledto a diagnosis of EBV-positive polymorphic PTLD, which also involved the tonsils and adenoids. This case demonstrates that the diagnosis

Žmay be difficult unless high index of suspicion is maintained. There was rapid resolution with reduction of immunosuppression right-hand.panel .


Table 2Diagnostic evaluation of patient with suspected PTLD

Routine Selected patients

CBC, platelets, WCC with differential GI endoscopySerum electrolytes, calcium, BUN and Bone scan

creatinineLiver function tests Bone marrow biopsyUric acid Brain CTrMRILactate dehydrogenase Lumbar punctureQuantitative immunoglobulins Sepsis work-upSerum protein electrophoresis

Ž .EBV serologies anti-EBNA, VCA and EAEBV viral load by quantitative PCRStools for occult bleeding

Ž .Chest radiograph AP and lateralCT scan of chestrabdomenrpelvis

Ž .Core needle or excisional biopsy of lesion s

pathy, if present this provides an additional means offollowing the patient’s response to therapy.

A variety of imaging tests are also of help in theevaluation of the thoracic transplant recipient withsuspected PTLD. A chest radiograph often revealsevidence of pulmonary nodular disease andror evi-dence of mediastinal lymphadenopathy. The most in-formative diagnostic study is usually computed tomo-

Ž .graphic CT evaluation of the chest and abdomen.Evidence of nodal or extranodal disease will fre-quently be apparent on CT at one or more sites. Inthe chest, small pulmonary nodules or enlarged medi-astinal lymph nodes may be apparent even in thepresence of normal chest radiograph. In the abdo-men, disease may be found at normal lymph node

sites, within the GI tract, or at extranodal sites, in-Ž .cluding the liver and kidneys Fig. 2 . Some centers

routinely perform CT or magnetic resonance imagingof the brain. These studies should always be per-formed if there is any clinical suggestion of CNS

Ž .disease Fig. 3 .Other studies performed on selected patients as

directed by the clinical findings are shown in Table 2.Upper andror lower gastrointestinal endoscopyshould be performed when there are gastrointestinalsymptoms or evidence of occult gastrointestinal bleed-ing.

Evidence of EBV infection has traditionally beensought by serological studies. The presence of anti-

Ž .bodies to the viral capsid antigen VCA , nuclearŽ . Ž .antigens EBNA , and early antigens EA is fre-

quently assessed. These studies should be routinelyobtained when PTLD is suspected and certainly thepresence of anti EBV-IgM antibodies suggests anacute infection. Some patients, however, mount apoor, or even absent, antibody response and antibodytesting will also generally be non-contributory for thediagnosis of PTLD in the patients who are seroposi-tive pre-transplantation. Confusion may also arise dueto passive transmission of antibodies from bloodproducts, or of maternal origin in the infant recipient.

An important advance in the diagnosis of EBV-as-sociated PTLD, and other EBV-associated viral syn-dromes, is the measurement of EBV viral load inperipheral blood using quantitative or semi-quantita-

Ž .tive polymerase chain reaction PCR . We have re-cently reviewed the role of EBV PCR evaluation in

Fig. 2. Abdominal CT of a 6-year-old boy 5 years out from heart transplantation demonstrates the opposite end of the spectrum ofEBV-driven lymphoproliferation. He presented with weight loss and rapidly progressive jaw swelling. Histology confirmed Burkitt’s lymphoma.There was diffuse disease throughout the abdomen including the huge renal lesions shown here. Note also the large masses anterior to the leftkidney. The facial involvement extended into the floor of the orbit. No residual disease was present after the second course of chemotherapy.


Fig. 3. PTLD of the central nervous system is rare. This 6-year-oldboy with relapsed PTLD, re-presented with a sixth nerve palsy.MRI demonstrated five enhancing intracerebral masses with sur-rounding edema. Stereotactic biopsy showed monomorphic PTLD.These lesions resolved with aggressive reduction in immunosup-pression.

the diagnosis, management and possible prevention ofw xPTLD 21 . A growing experience suggests that a

markedly elevated EBV viral load is present in allpatients with EBV driven PTLD. Within our ownprogram, we use a quantitative competitive PCR tech-

Ž . w xnique QC-PCR developed by Rowe et al. 22 . Chil-dren with latent viral infection demonstrate very low

wcirculating viral loads generally -8 EBV genomecopies per 105 peripheral blood mononuclear cellsŽ .xPBMC . With PTLD, viral loads are usually ex-tremely high, being greater than 1000 genomecopiesr105 PBMC in most cases, and greater than200 in all cases that we have studied to date. How-ever, we have found that EBV viral loads are compa-rable among patients with EBV-associated viral syn-dromes and those with PTLD. Thus, while evaluationof the EBV viral load is a very useful rapid screeningprocedure for suspected EBV disease, it cannot re-place histologic examination of suspected sites of in-volvement when the diagnosis of PTLD is contem-plated.

4. Pathological findings

The pathological description of PTLD remains con-fusing for many transplant physicians. This is regret-

table since an understanding of the histology andmolecular pathology of these disorders is importantfor determining therapy and prognosis. Fortunatelyseveral recent reviews have attempted to clarify this

w xtopic 4,23,24 . We have adopted a simple approach toŽ .the pathological description of PTLD Table 3 . First

and foremost, it should be noted that evaluationshould be done by a pathologist with extensive experi-ence in the evaluation of PTLD. The transplant physi-cian should forewarn the pathologist of the impendingbiopsy of suspected lesions and, ideally, the pathologistshould be present to receive the biopsy, which mustbe submitted fresh, rather than in formalin. When-ever possible, tissue from several involved sites shouldbe obtained since different morphologies may be pre-sent at different sites of disease. Core needle biopsies,under CT guidance, or excisional biopsies are gener-ally required for full diagnostic work up. Upperandror lower gastrointestinal endoscopy also offeropportunities to obtain multiple biopsy samples inpatients with involvement at these sites. Occasionally,laparoscopy or mediastinoscopy may be helpful tech-niques for obtaining tissue.

The diagnosis of PTLD is confirmed by histologyw x23,24 . The most benign end of the spectrum isrepresented by nonspecific reactive lymphoid hyper-plasia, including mononucleosis lesions. These lesionshave been variously labeled as benign lymphoid hy-perplasias, plasmacytic hyperplasias and even as ‘in-fectious mononucleosis-like PTLD’. The central char-acteristic of this group of lesions is the diffuse prolif-eration of mononuclear cells of various sizes, manywith plamacytoid features, but with preservation ofnormal tissue architecture. These lesions tend to oc-cur in lymph nodes, and in tonsils and adenoids ratherthan in other extranodal sites. These lesions representone end of the spectrum of PTLD, but it should benoted that many groups have not included them ascases of PTLD when reporting the frequency of thisproblem in their population of patients.

Polymorphic PTLD also demonstrate lymphoid in-filtrates of varying shapes and sizes. The entire rangeof lymphocyte differentiation may be seen. However,in these lesions there is effacement andror destruc-tion of normal tissue architecture. Areas of necrosisare frequently present. These lesions may occur inboth nodal tissue and at extranodal sites, and may

Žinvolve any organ. Large, bizarre cells atypical im-.munoblasts may be observed. In monomorphic PTLD,

the destructive lymphoid infiltrate has a much moremonotonous appearance with most cells appearing tobe transformed lymphocytes at one stage of differen-tiation. These lesions more closely resemble non-Hodgkin lymphomas in the non-transplant patient. Itis important to recognize that some degree of poly-morphism is often seen, although much less pro-


Table 3aPathological evaluation of suspected PTLD lesions

Technique Purpose

Routine histology Classify lesions:‘benign’ lymphoid hyperplasiapolymorphic PTLDmonomorphic PTLD

Ž .malignant lymphoma e.g. Burkitt’s

Ž .Immunophenotypingrflow cytometry Cell lineage and characterization e.g. T vs. B cell ;Ž .clonality Ig light or heavy chain expression

b Ž .Immunoglobulin gene rearrangement studies Host cell clonality monoclonal vs. polyclonal

ŽIn situ hybridization for EBER 1 or Presence of EBV genome.Southern blot or PCR analysis

Southern blot analysis of EBV terminal repeat region EBV clonality

Cytogenetic studies Structural chromosomal abnormalities

cMolecular studies of proto-oncogenes and tumor Define structural gene alterationssuppressor genes

a Full diagnostic work up requires core needle or excisional biopsy. Needle aspiration cytology is not adequate. Tissue must be submittedfresh, not in formalin.

b For clonal analysis of T cell PTLD, comparable analyses are performed for rearrangements of T cell receptor.cStudies of oncogenes remain a research tool at the present time.

nounced than in polymorphic lesions. Further confu-sion in nomenclature can arise when predominantlypolymorphic lesions contain localized areas that ap-pear monomorphic. Hence the need for description ofthe entire lesion, rather than placing undue relianceon short labels. Rarely, other specific histological vari-ants may also be noted including mature plasma celltype predominance and tumors that strongly resembleHodgkin’s lymphoma.

We attempt to distinguish monomorphic PTLD le-sions from specific overt malignant lymphomas suchas Burkitt’s lymphoma. In the Knowles’ classificationw x25 , the distinction between monomorphic PTLD andmalignant lymphomas is not made, and this group oflesions are all designated as ‘immunoblastic lym-phoma’. While we recognize that this distinction isoften arbitrary, we have continued to use the termmonomorphic PTLD, because a number of these le-sions may regress with reduction in immunosuppres-

Žsion, as do most cases of polymorphic PTLD see.below . By contrast, malignant lymphomas such as

Burkitt’s disease will invariably progress without ur-gent chemotherapy. If the term malignant lymphomais applied routinely to all monomorphic lesions, webelieve it may convey the wrong message to thetransplant physician, i.e. that chemotherapy is indi-cated as first line of therapy.

Further work-up of suspected PTLD includes im-munophenotyping by imunohistochemistry and flowcytometry. These procedures will demonstrate the cell

lineage of the lesions. Almost all PTLD are of B cellorigin, though rare cases of T cell PTLD are seen. It

Žshould be noted that variable numbers of T cells and.macrophages are generally interspersed between the

B cells populations in polymorphic lesions. Whetherthese may be ‘reactive’, and offer an improved prog-nosis if the T cell infiltrates are pronounced is underinvestigation. Immunophenotyping for immuno-globulin light and heavy chains may occasionally pro-vide information about clonality, when the lesionsexpress these products. Molecular studies offer themost definitive assessment of the clonality of thelesions. The concept of clonality is complex since it isnow recognized that a gradation of clonal alterationsmay be observed. Thus lesions cannot be simplylabeled as polyclonal or monoclonal. Sometimes, mi-nor clonal sub-populations may be present in a pre-dominantly polyclonal lesion. In other cases, individ-ual PTLD lesions within the same patient may bemonoclonal, yet each lesion is clonally distinct. Fur-thermore, the concept of clonality can be applied toboth the host cell and to the virus separately. Analysisof host cell clonality is based on the behavior ofimmunoglobulin genes, which rearrange uniquely inthe maturing B cell. Progeny of a B cell that hasalready arranged its immunoglobulin genes, will carry

w xthe same rearrangement 23]25 . EBV clonality isassessed by evaluation of the viral terminal repeatregion. When a single EBV fuses its terminal ends to

Ž .produce an episomal circular form, a fixed number


of terminal repeat segments are retained, and someare lost. With the use of specific molecular probes forthis region, it is possible to determine if the EBVvirus within PTLD lesions is itself clonal or polyclonalin nature. Monomorphic lesions are clonal in nature,and analysis of clonality is therefore of little rele-vance. Polymorphic lesions may be polyclonal, butfrequently represent clonal B cell proliferations. Somelesions are clonal by EBV, but not by immunoglobulingene analysis. Thus, analysis of clonality is most im-portant in polymorphic rather than monomorphic dis-

Ž .eases which are always clonal . However, these stud-ies will primarily remain research tools unless theycan be shown to influence therapy and outcome.There is some suggestion that regression of polymor-phic lesions in response to reduction in immunosup-pression may be influenced by their clonal pattern,

w xthough more work in this area is required 24 . An-other attempt to predict prognosis from the molecularpathological findings involves a search for abnormali-

Žties in oncogenes and tumor suppressor genes e.g..c-myc, N-ras and p53 . Polymorphic lesions, even if

clonal for EBV and immunoglobulin gene rearrange-ments, do not usually demonstrate mutations of onco-genes and tumor suppressor genes. By contrast, aproportion of monoclonal lesions will show evidence

w xof mutations of c-myc, N-ras or p53 4,25 . Suchstructural gene alterations may portend a worse prog-nosis in monomorphic PTLD.

Cytogenetic abnormalities are mostly found in ma-lignant lymphomas, such as Burkitt’s disease. Fewdata exist on the presence and significance of cytoge-netic abnormalities in polymorphic or monomorphicPTLD lesions. Where present, it is likely they mayalso portend a poor prognosis if managed by reduc-tion in immunosuppression alone.

The pathologist must demonstrate whether PTLDlesions contain evidence of EBV. A number of tech-niques are available including Southern blot analysisŽ .generally combined with analysis of EBV clonality ,PCR and in situ hybridization. We, and many pro-grams, use in situ hybridization with the EBER-1probe which labels EBV-encoded early RNA tran-scripts in infected cells. This technique is reliable,rapid and is performed on routinely processed paraf-fin sections. Overall, approximately 80]90% of pedi-atric thoracic PLTD cases are EBV positive, includingall cases arising in the first year or two after trans-plantation. In our experience, EBV negative casesdevelop late after transplantation. The spectrum ofhistopathological findings is similar in EBV positiveand negative lesions, though Burkitt’s lymphoma hasalways been EBV driven.

Finally, it should be noted that non-lymphoid Ep-stein]Barr induced neoplasia may very rarely developafter solid organ transplantation, especially in chil-

w xdren 26 . These tumors are composed of spindle cellswith smooth muscle features. We have observed threesuch cases among over 1300 pediatric solid organtransplants at Children’s Hospital of Pittsburgh from1982]1995. All showed evidence of clonal EBV infec-tion. Previous PTLD was present in one of our cases.

5. Prevention of PTLD

The ability to prevent PTLD would be of enormousbenefit to pediatric transplant recipients. The mostlogical approach would be to immunize all seronega-tive recipients prior to transplantation. Several vac-cine preparations are under evaluation but progresshas been slow, despite early optimism from primate

w xwork 27 . This may reflect the low interest fromindustry, given the low morbidity and mortality associ-ated with EBV infection in the immunocompetenthost in developed countries. An alternate strategywould be to avoid transplanting seronegative recipi-ents, especially with a seropositive donor. This ap-proach would effectively exclude many pediatricthoracic recipients from receiving organs, and entailstremendous logistical problems since donor EBVserologies are generally not available at the time of

w xacceptance of the donor organ 5 . Various otherpreventive strategies have been suggested. Chemopro-phylaxis, with short term intravenous ganciclovir, fol-lowed by long-term oral acyclovir therapy was nothelpful in preventing PTLD in a randomized trial in

w xpediatric liver transplantation 28 . Several studieshave also demonstrated rise in EBV viral load byPCR while patients are receiving therapy with ganci-clovir or acyclovir.

Ž .Use of intravenous gammaglobulin IVIG prepara-tions containing high levels of anti-EBV antibodytiters has also been suggested. IVIG has been shownto prevent the development of EBV-associated lym-phomas in a severe combined immunodeficiency

w xmouse model 29 . A multicenter, randomized trial ofŽprophylactic CytoGam which contains high anti-EBV

.titers vs. a placebo infusion in seronegative liverrecipients is currently in progress.

Another appealing preventive strategy is to targetŽpatients with evidence of early EBV infection prior

.to development of symptoms for preemptive therapy.This targeted approach avoids treatment of allseronegative recipients. Published experience suggeststhat EBV PCR levels rise during primary infection,prior to both antibody development and the onset ofsymptoms. A quantitative or semi-quantitative EBVPCR would, therefore, seem to be an appealingscreening test. We currently perform monthly EBVQC-PCR assays on all seronegative recipients for thefirst 6 months after transplantation, and then at less


frequent intervals. When the PCR becomes positive,increased clinical surveillance is instituted and cau-tious temporary reduction in immunosuppression isperformed for those children whose recent biopsiesrevealed no rejection. Other potential preemptivestrategies include institution of ganciclovir or Cyto-Gam at the time of evidence of primary infection byPCR. All these strategies require investigation byrandomized clinical trials.

6. Management of established PTLD

We have recently reviewed the various treatmentw xstrategies for histologically confirmed PTLD 28,30 .

Treatment options are also summarized in Table 4. In1984, Starzl et al. reported the reversibility of PTLDby reduction in immunosuppression in cyclosporine-

w xtreated patients 31 . This strategy remains the main-stay of therapy for most patients. As stated earlier, webelieve this should be the initial approach to patientswith both polymorphic and monomorphic disease.However, patients with overtly malignant disease, such

as Burkitt’s lymphoma, should be managed with im-mediate chemotherapy. Some authorities believe thatmonomorphic PTLD should also be initially managedwith chemotherapy, although resolution of such le-sions by reduction in immunosuppression has beenwell documented. As discussed earlier in this review,there is an urgent need for more research to definethose clinical, histological and molecular pathologicalfindings that most accurately predict response to re-duction in immunosuppression.

The goal of reduced immunosuppression is to allowthe host to recover natural immune surveillance andsubsequently to regain control over the proliferationof EBV-infected cells. Wide variation in response tothis approach has been noted between centers. Thisvariation may be explained by a number of reasonsincluding differences in patient populations, defini-tions of PTLD, and strategies for reducing immuno-suppression. In our experience with pediatric heart

Ž .transplant recipients, the vast majority 80]90% ofnon-malignant lesions will respond to reduction inimmunosuppression combined with anti-viral therapy,with most children showing evidence of clinical re-

Table 4Treatment strategies for PTLD

Therapy Comments

First line therapies

Reduced immunosuppression Effective in most pediatric heart recipients,especially in polymorphic PTLD

Ž .Antiviral therapy e.g. ganciclovir Widely used but unproven

Chemotherapy First line for overt malignancy, e.g. Burkitt’s lymphoma

Second line therapies

Interferon-alpha May cause severe rejection

ŽIntravenous immunoglobulins containing Unproven. Trials in PTLD prevention underway..anti-EBV antibodies

Anti-B cell monoclonal antibodies Limited experience in PTLD. Norandomized trials. Pediatric trial in preparation

Cellular immunotherapy e.g. autologous Promising techniques. Limited experienceor recipient HLA matched EBV-specific to date. No controlled trials.cytotoxic T lymphocytes; autologous

Ž .lymphokine activated killer LAK cells

Chemotherapy Established second line therapy forrefractory and relapsed PTLD; more oftenrequired in monomorphic PTLD and lateonset EBV-negative disease

Surgeryrradiation therapy Reserved for treatment of localcompression of critical structures, bowelobstruction, etc. Also for excisional biopsyof localized, easily accessible lesions at presentation


sponse within 2]4 weeks of reduction of immunesuppression. Results in heart]lung and lung recipi-ents have been less satisfactory with approximatelyhalf showing response to reduced immunosuppres-sion, perhaps due the less aggressive reduction inimmunosuppression that we have used in thesepatients.

In our own program, all patients are initially takenoff tacrolimus or cyclosporine, as well as azathioprine

Žor mycophenalate mofetil if they were receiving these.medications . If a patient is receiving corticosteroid

therapy, it is usually continued at maintenance levels.The time at which calcineurin inhibitors are reintro-duced varies depending on severity of disease, priorrejection history, time from transplantation and timerequired for tacrolimus or cyclosporine levels to fall.We have frequently observed that immunosuppres-sant levels are high at time of diagnosis and it maytake many days for drug levels to become immeasur-able. This presumably reflects impaired drugmetabolism, often associated with mild liver dysfunc-tion. Careful monitoring is carried out for evidence ofrejection. Echocardiograms are performed two tothree times per week and an endomyocardial biopsy isroutinely obtained 1 week into therapy, and subse-quently every 1]2 weeks while a major reduction inimmune suppressive therapy is continued. Forheart]lung and lung transplant recipients, the chestradiograph, pulse oximetry and pulmonary functiontests of the patient are followed carefully and a trans-bronchial biopsy is performed approximately 7]10days into therapy.

In addition to the above strategies, we now recom-mend following the EBV viral load in the peripheralblood by PCR, initially on a weekly basis, followingthe diagnosis of PTLD. This has proved very helpful

w xin patient management 21,30 . A fall in the EBV viralload is consistent with the development of an immuneresponse against the EBV-infected B-cells and has

Žpredicted a good clinical response often before re-.duction in size of lesions is observed . Among eightŽpediatric heart recipients with PTLD all with very

high viral loads at presentation; range 200])50005 .genome copiesr10 PBMC , seven cleared their viral

loads to F200 at a median of 2 months following theŽ .diagnosis of PTLD range: 7 days]9 months . The

drop in the viral load appeared to be concomitantwith resolution of their PTLD. Two of the childrenwho initially cleared their viral loads showed a tran-sient rebound to values greater than 200 withoutclinical symptoms. Rebound rejection occurred in fiveof eight cases at a time when the viral load was 8]100genome copiesr105 PBL. No rejection was seen athigher viral load values. In one 6-year-old boy, 1 yearout from cardiac re-transplantation, we held all im-munosuppression for 2 months after he presented

with viral ‘septic shock’ and multi-system failure. Per-sistent elevations in QC EBV-PCR to )5000 genomecopiesr10 5 PBMC for approximately 8 weeksprompted us to withhold immunosuppression for thisvery prolonged period. A precipitous fall in PCR to-200 occurred at this time and low-dose tacrolimustherapy was successfully introduced without develop-ment of rebound rejection. Shortly, thereafter, resolu-tion of multi-system failure occurred and the patientwas successfully discharged home. These observa-tions, supported by similar results in our liver recipi-

w xents 21 , suggest that viral load monitoring by PCRmay be very helpful in monitoring response to therapyand determining the timing of reintroduction of im-munosuppression.

Prior to PCR monitoring, we had no clear indica-tors of when immunosuppression should be reintro-duced, other than when acute rejection developed.Ideally, one wishes to reintroduce immunosuppres-sion before important rejection evolves. It is impor-tant to note, however, that acute rebound rejection, ifit develops, needs to be treated in a conventional

Žmanner usually with bolus intravenous solumedrol.for cardiac 3A rejection . Acute rejection often coin-

cides with resolution of PTLD and PTLD usually doesnot recur after treatment of acute rejection. We sus-pect that reports of failure of polymorphic lesions torespond to reduced immunosuppression may fre-quently be due to overly cautious reduction in im-munosuppression, because of fear of the conse-quences of rebound rejection. For patients showinggood response to therapy, we typically re-institutemaintenance therapy with low doses of their primary

Žimmunosuppressant typically at half the previous.dose . We attempt to use the minimum dose neces-

sary to prevent acute rejection and we accept anincrease in surveillance biopsies as the price neces-sary to achieve this aim. Azathioprine or my-cophenalate mofetil are frequently not reintroduced.

In addition to reduction of immune suppression,most centers also use antiviral therapy. Both acyclovirand ganciclovir have been shown to inhibit lytic EBVDNA replication in vitro and would seem to be ofvalue in treating the lytic phase of EBV infections.Pathologic analysis has shown that most EBV in-fected cells within PTLD lesions are transformedB-cells and are not undergoing lytic infection. Neitheracyclovir nor ganciclovir suppress EBV-driven prolif-eration of transformed B cells in vitro, nor are theyactive against B-cells that are latently infected withEBV. A few studies have demonstrated, however,evidence of some lytic phase EBV within PTLD le-sions. It, therefore, seems reasonable to continue

Žtherapy until there is clinical and virologic e.g. falling.EBV viral load evidence of resolution of

EBVrPTLD. We do acknowledge that there is no


clear proof of efficacy of ganciclovir or acyclovir inthe management of PTLD.

Beyond reduction of immune suppression andanti-viral therapy, the optimal management of PTLDin solid-organ transplant recipients is controversial.The use of interferon has been described in anecdotalreports as a therapeutic option in the management of

w xPTLD 30,32,33 . There is a lack of prospective, con-trolled trials to establish the therapeutic role of this

Žagent. We have treated six children three liver and.three thoracic transplant recipients with interferon.

Five of these patients received alpha-interferon 2bwhile one was treated with gamma-interferon. Three

Žof the six patients one liver and two thoracic trans-.plant recipients appeared to have responded to the

interferon therapy without significant side effects.However, each of these patients was receiving either areduced dose or no immune suppression at the timeof initiation of interferon. One lung transplant recipi-ent had to prematurely discontinue the use of inter-feron because of the development of severe rejection.In addition, each of the three pediatric liver trans-plant recipients treated with interferon developed re-jection while receiving this therapy. These observa-tions, and those from the literature, suggest that therole of this agent in the treatment of PTLD is farfrom established and that rejection will frequentlyoccur in children receiving this therapy.

A potential role for the use of intravenous immuneŽ .globulin IVIG for the treatment of PTLD has also

w xbeen suggested 32,33 . Several reports have docu-mented an association between loss, or absence, ofantibody against at least one of the Epstein]Barr

Ž .nuclear antigens EBNA in EBV infected organ re-cipients and the subsequent development of PTLDw x16 . In addition, Riddler et al. demonstrated a corre-lation between an increasing level of anti-EBNA anti-

Žbodies including those introduced through transfu-. w xsions with a decrease in EBV viral load 16 . IVIG

has been used in combination with interferon-alphaas treatment for PTLD at the University of Minne-

w xsota and other centers 32,33 . As with the use ofantiviral agents and interferon, there are no compara-tive trials evaluating the role of intravenous immuno-globulin preparations in the management of PTLD.

The use of anti-B-cell monoclonal antibodies hasrecently been suggested as a therapeutic option forpatients with PTLD not responding to conventionaltherapy. Limited experience with the use of anti-CD21and anti-CD24 monoclonal antibodies in combina-tion, as well as use of an anti-CD22 immunotoxin

w xhave been reported 34,35 . The results were mostw xpromising for polyclonal disease 34 . These anti-

bodies are not currently available, though an anti-CD20 humanrmouse chimeric monoclonal antibodyŽRituximab; Genentech Inc. & IDEC pharmaceuti-

.cals is currently commercially available for treatmentof certain CD20-positive B-cell non-Hodgkin lympho-mas in adult non-transplant recipients. A multi-center,clinical trial is currently being planned to evaluatethis product in the treatment of pediatric PTLD.

Another interesting strategy for the treatment ofrefractory PTLD is cellular immunotherapy. Nalesnik

Ž .et al. have used interleukin-2 IL2 ex vivo-stimulatedŽ .lymphokine-activated killer LAK cells of recipient

w xorigin to treat seven patients 36 . Autologous PBMCwere obtained by leukapheresis and depleted ofmonocytes, and then cultured for 10]11 days in thepresence of IL2. The cells were then returned to thepatient intravenously. The use of this strategy ap-peared to successfully treat four patients with EBV-associated refractory PTLD, although the patientsalso were treated with decreased immunosuppression.Three patients with EBV negative disease did notrespond. A concern with this therapy is that it mayalso stimulate anti-donor lymphocytes leading to re-jection.

A more logical approach to cellular therapy is togive the patient an infusion of cytotoxic T-lympho-

Ž .cytes CTLs directed against EBV-specific antigens.This should result in control of the proliferation ofEBV infected B cells, but without the risk of rejec-tion. The use of EBV-specific CTL therapy has al-ready been developed and applied for the manage-

Ž .ment of PTLD in bone marrow transplant BMTw xrecipients 37 . In this situation, the donor is often

available to provide the CTL, is not receiving im-munosuppression and will hopefully be EBV seroposi-tive. Application of this technique is more proble-matic in solid organ transplantation since the PTLDlesions are of recipient cell origin and the recipientwill need to be the source of CTL, unless an HLAidentical source of CTL is available. Since EBV-asso-ciated PTLD occur much more frequently in patientswho are EBV seronegative prior to transplantation,pre-existing immunity specific to EBV does not exist.To obtain functional EBV-specific CTLs, one wouldneed to ‘immunize’ and stimulate recipient T-cellsagainst EBV ex vivo. Efforts to do just this are under-way in several centers. There is one report in theliterature of a pediatric lung recipient with centralnervous system PTLD, who received mononuclear cellinfusions from an HLA-identical, EBV positive, sib-ling. The patient did develop significant rejection, butdramatic response of the monoclonal PTLD was seen.The response correlated with in vivo reconstitution of

w xnormal EBV-specific cytotoxic activity 38 .Other therapeutic options available for the treat-

ment of PTLD include the use of chemotherapeuticagents, radiation and surgery. The very high viralloads documented in the peripheral circulation ofpatients with PTLD, emphasizes the systemic nature


of this disease, even when only a solitary lesion isidentified. These observations suggest that surgeryand radiation are only appropriate for the manage-

Žment of local complications e.g. gastrointestinalhemorrhage, intestinal obstruction, or local compres-

.sion of critical structures . Some centers strongly sup-port the use of chemotherapy, especially for mono-morphic PTLD. These agents are immunosuppressivein nature and interfere with the recovery of the host’snatural immune surveillance mechanisms. They do,however, offer protection to the allograft from rejec-

w xtion while simultaneously treating the PTLD 39 .Thus, they may have a role for treating patients whorelapse during reintroduction of immunosuppressionnecessitated by rejection, or in the rare case of activePTLD with concomitant rejection. Their role inpatients with overt malignancy, e.g. Burkitt’s lym-phoma, has already been discussed. The optimalchemotherapy regimen has not been determined,though most are based on protocols used for treat-ment of non-Hodgkin’s lymphoma. Guidelines onwhich children should receive chemotherapy, andwhich chemotherapeutic regimen should be used, arenot available at this time.

7. Conclusions

Over the last decade, much has been learnt aboutthe nature of PTLD. The pivotal role of EBV infec-tion in the majority of cases has been established andthe pathological description of lesions has entered themolecular era. It seems likely that further understand-ing of the molecular pathology may lead to greaterability to define optimal treatment regimens andprognosis. Quantitative PCR techniques for EBV holdgreat promise for enhancing the diagnosis and pre-vention of PTLD, and for monitoring response totherapy. In particular, this technique appears to helppredict when re-introduction of immunosuppressionshould be instituted. A number of exiting new thera-pies are on the horizon, including use of monoclonalantibodies against B cell surface antigens and thedevelopment of cellular therapies, such as use ofinfusions of autologous or HLA matched EBV speci-fic CTLs. Such strategies offer the promise of con-trolling abnormal B cell proliferation without the riskof allograft rejection, since the host alloresponse isnot enhanced by these therapies. The place ofchemotherapy, and the optimal regimens required,remain to be defined. Our understanding of the eti-ology, behavior and optimal treatment for EBV nega-tive PTLD remains limited, in part due to the rarity ofthese lesions. Encouragingly, there is an increasinglevel of interest in PTLD among clinical and basicinvestigators, as well as recognition of the need for

multi-center trials to define optimal prevention andtreatment strategies. A degree of optimism seemswarranted!

Acknowledgements

The authors wish to thank Dr Paul Dickman for hishelpful review of the manuscript.

References

w x1 Starzl TE. Discussion of Murry JE, Wilson RE, Tilney NL etal. Five years’ experience in renal transplantation with im-munosuppressive drugs: survival, function, complications andthe role of lymphocyte depletion by thoracic duct fistula. AnnSurg 1968;168:416.

w x2 Briggs JD, Hamilton DNH, Macsween RNM et al. Infectiousmononucleosis, herpes simplex infection and diffuse lym-phoma in a renal transplant patient. Transplantation 1978;25:227.

w x3 Ho M. Risk factors and pathogenesis of posttransplant lym-phoproliferative disorders. Transplant Proc 1995;27:38]40.

w x4 Chadburn A, Cesarman E, Knowles DM. Molecular pathologyof posttransplant lymphoproliferative disorders. Sem DiagnPathol 1997;14:15]26.

w x5 Webber SA. Post-transplant lymphoproliferative disorders: apreventable complication of solid organ transplantation?w xEditorial . Pediatr Transplant 1999;3:95]99.

w x6 Zangwill SD, Hsu DT, Kichuk MR et al. Incidence andoutcome of primary Epstein]Barr virus infection and lym-phoproliferative disease in pediatric heart transplant recipi-ents. J Heart Lung Transplant 1998;17:1161]1166.

w x7 Finn L, Reyes J, Bueno J, Yunis E. Epstein]Barr virusinfections in children after transplantation of the small intes-tine. Am J Surg Pathol 1998;22:299]309.

w x8 Cen H, Breinig MC, Atchison RW, Ho M, McKnight JLC.Epstein]Barr virus transmission via the donor organ in solidorgan transplantation: polymerase chain reaction and restric-tion fragment length polymorphism analysis of IR2, IR3 andIR4. J Virol 1991;65:976]980.

w x9 Haque T, Thomas JA, Falk KI et al. Transmission of Ep-Ž .stein]Barr virus EBV in transplanted organs causes lym-

phoproliferative disease in EBV-seronegative recipients. JGen Virol 1996;77:1169]1172.

w x10 Swinnen LJ, Costanzo-Nordin MR, Fisher SG et al. Increasedincidence of lymphoproliferative disorders after immunosup-pression with the monoclonal antibody OKT3 in cardiactransplant recipients. New Engl J Med 1990;323:1723]1728.

w x11 Newell KA, Alonso EM, Whitington PF et al. Posttransplantlymphoproliferative disease in pediatric liver transplantation.Transplantation 1996;62:370]375.

w x12 Cox KL, Lawrence-Miyasaki LS, Garcia-Kennedy R et al. Anincreased incidence of Epstein]Barr virus infection and lym-phoproliferative disease in young children on FK506 afterliver transplantation. Transplantation 1995;59:524]529.

w x13 Ciancio G, Siquijor AP, Burke GW et al. Post-transplantlymphoproliferative disease in kidney transplant recipients inthe new immunosuppressive era. Clin Transplant 1997;11:243]249.

w x14 Webber SA. Fifteen years of pediatric heart transplantationat the University of Pittsburgh: lessons learned and futureprospects. Pediatr Transplant 1997;1:8]21.

w x15 Bernstein D, Baum D, Berry G et al. Neoplastic disorders


Žafter pediatric heart transplantation. Circulation 1993;88 part.2 :230]237.

w x16 Riddler SA, Breinig MC, McKnight JLC. Increased levels ofcirculating Epstein]Barr virus-infected lymphocytes and de-creased EBV nuclear antigen antibody responses are associ-ated with the development of posttransplant lymphoprolifera-tive disease in solid-organ transplant recipients. Blood1994;84:972]984.

w x17 Harwood JS, Gould FK, McMaster A et al. Significance ofEBV status and post-transplant lymphoproliferative diseasein pediatric thoracic transplantation. Pediatr Transplant1999;3:100]103.

w x18 Boyle GJ, Michaels MG, Webber SA et al. Posttransplanta-tion lymphoproliferative disorders in pediatric thoracic organrecipients. J Pediatr 1997;131:309]313.

w x19 Sweet SC, Spray TL, Huddleston CB et al. Pediatric lungtransplantation at St. Louis Children’s Hospital, 1990]1995.Am J Resp Crit Care Med 1997;155:1027]1035.

w x20 Eisen HJ, Hicks D, Kant JA et al. Diagnosis of posttransplantlymphoproliferative disorder by endomyocardial biopsy in acardiac allograft recipient. J Heart Lung Transplant 1994;13:241]245.

w x21 Green M, Reyes J, Webber S, Michaels MG, Rowe D. Therole of viral load in the diagnosis, management and possibleprevention of EBV-associated post-transplant lymphoprolif-erative disease following solid organ transplantation. CurrOpin Organ Transplant 1999;4:292]296.

w x22 Rowe DT, Qu L, Reyes J et al. Use of quantitative competi-tive PCR to measure Epstein]Barr virus genome load in theperipheral blood of pediatric transplant recipients with lym-phoproliferative disorders. J Clin Microbiol 1997;35:1612]1615.

w x23 Harris NL, Ferry JA, Swerdlow SH. Post-transplant lym-phoproliferative disorders: summary of Society for Hemato-pathology work shop. Sem Diagn Pathol 1997;14:8]14.

w x24 Nalesnik MA, Starzl TE. Epstein]Barr virus, infectiousmononucleosis, and post-transplant lymphoproliferative dis-orders. Transplant Sci 1994;4:61]79.

w x25 Knowles DM, Cesarman E, Chadburn A et al. Correlativemorphologic and molecular genetic analysis demonstratesthree distinct categories of posttransplant lymphoproliferativedisorders. Blood 1995;85:552]565.

w x26 Lee ES, Locker J, Nalesnik M et al. The association ofEpstein]Barr virus with smooth muscle tumors occurringafter organ transplantation. New Engl J Med 1995;332:19]25.

w x27 Epstein MA, Morgan AJ, Finerty S et al. Protection ofcottontop tamarins against Epstein]Barr virus-induced ma-lignant lymphoma by a prototype subunit vaccine. Nature1985;318:287]289.

w x28 Green M, Reyes J, Rowe D. New strategies in the preventionand management of Epstein]Barr virus infection and post-transplant lymphoproliferative disease following solid organtransplantation. Curr Opin Organ Transplant 1998;3:143]147.

w x29 Abedi MR, Linde A, Christensson B, Mackett M, Ham-marstrom L, Smith C. Preventive effect of IgG from EBV-seropositive donors on the development of human lym-phoproliferative disease in SCID mice. Int J Cancer 1997;71:624]629.

w x30 Green M, Michaels MG, Webber SA, Rowe D, Reyes J. Themanagement of Epstein]Barr virus associated post-transplantlymphoproliferative disorders in pediatric solid organ trans-plant recipients. Pediatr Transplant 1999;3:271]281.

w x31 Starzl TE, Nalesnik MA, Porter KA et al. Reversibility oflymphomas and lymphoproliferative lesions developing undercyclosporin-steroid therapy. Lancet 1984;1:583]587.

w x32 Shapiro RS, Chauvenet A, McGuire W et al. Treatment ofB-cell lymphoproliferative disorders with interferon alfa and

w xintravenous gamma globulin letter . New Engl J Med1988;318:1334.

w x33 Ippoliti, Martinelli L, Loernzutti F et al. Post-transplantlymphoproliferative disease after heart transplantation onSandimmune therapy: treatment with interferon-alpha 2b andintravenous immunoglobulin. Trans Proc 1995;26:2660]2661.

w x34 Fisher A, Blanche S, Le Bidois J et al. Anti-B-cell monoclo-nal antibodies in the treatment of severe B-cell lymphoprolif-erative syndrome following bone marrow and organ trans-plantation. New Engl J Med 1991;324:1451]1456.

w x35 Senderowicz AM, Vitetta E, Headlee D et al. Completesustained response of a refractory, post-transplantation, largeB-cell lymphoma to an anti-CD22 immunotoxin. Ann InternMed 1997;126:882]885.

w x36 Nalesnik MA, Rao AS, Furukawa H et al. Autologous lym-phokine-activated killer cell therapy of Epstein]Barr viruspositive and negative lymphoproliferative disorders arising inorgan transplant recipients. Transplantation 1997;63:1200]1205.

w x37 Rooney CM, Smith CA, Ng CYC et al. Use of gene-modifiedvirus-specific T lymphocytes to control Epstein]Barr virusrelated lymphoproliferation. Lancet 1995;345:9]13.

w x38 Emanuel DJ, Lucas KG, Mallory GB et al. Treatment ofposttransplant lymphoproliferative disease in the central ner-vous system of a lung transplant recipient using allogeneicleukocytes. Transplantation 1997;63:1691]1694.

w x39 Swinnen LJ, Mullen GM, Carr TJ, Costanzo MR, Fisher RI.Aggressive treatment for postcardiac transplant lymphoprolif-eration. Blood 1995;86:3333]3340.

Documents

Post-transplantation lymphoproliferative disorders: advances in