Recurrent Epstein-Barr virus associated disease in a cardiac transplant patient: evolution from plasmacytic hyperplasia to diffuse large cell lymphoma

Clinico-pathologic conference L.M. Larratt M. Hamilton R. Coupland J.K. Preiksaitis

Key words: Epstein–Barr virus (EBV); large cell lymphoma; cardiac transplant; post-transplant lymphoproliferative disease; cytomegalovirus (CMV)

Recurrent Epstein–Barr virus associated disease in a cardiac transplant patient: evolution from plasmacytic hyperplasia to diffuse large cell lymphoma

Abstract: An Epstein–Barr virus (EBV)-seronegative 31-year-old male underwent cardiac transplantation in 1991 for congenital cardiomyopathy. He presented with a protracted course of waxing and waning lymphadenopathy beginning four years after transplantation with eventual progression to a fulminant EBV-positive large cell lymphoma eight years after transplantation. Risk factors for the development of post-transplant lymphoproliferative disease in this patient, the importance of a standardized approach to pathology in assessing therapeutic options, and the management strategies used are discussed.

Presentation of case

Authors’ affiliations: L.M. Larratt1, M. Hamilton1, R. Coupland2, J.K. Preiksaitis1 1 Department of Medicine, University of Alberta, Edmonton, Alberta, Canada, 2

Department of Pathology, Cross Cancer Institute, Edmonton, Alberta, Canada Correspondence to: L. M. Larratt 2E3.33 W.M.C. University of Alberta, Edmonton Alberta, T6G 2R7 Canada Tel: 1 780 407 7021 Fax: 1 780 407 2680 e-mail: llarratt/cha.ab.ca

The patient is a 31-year-old male who underwent cardiac transplantation in October 1991 for refractory congestive heart failure complicated by supraventricular and ventricular tachycardia secondary to a congenital cardiomyopathy. The cardiomyopathy was diagnosed at three years of age and was associated with other physical anomalies including short stature, webbed neck, narrow shoulder girth, Pierre Robin nuciograthia, and limitation of range of motion of the hands. These anomalies were not associated with a well-defined genetic syndrome. He was seronegative for cytomegalovirus (CMV) IgG, EBV viral capsid antigen (VCA) IgG, and Epstein–Barr nuclear antigen 1st (EBNA 1) IgG before transplant, and his donor was seropositive for both EBV VCA IgG and CMV IgG. The patient received acyclovir Received 11 September, accepted for publication 19 September 2000 Copyright c Munksgaard 2001 Transplant Infectious Disease . ISSN 1398-2273 Transpl Infect Dis 2001: 3: 119–123 Printed in Denmark . All rights reserved

800 mg po qid from day 7 to day 115 as CMV prophylaxis. Perioperative immunosuppression consisted of cyclosporin A, azathioprine, steroids, and induction therapy with anti-lymphocyte globulin (ALG) for eight days.

119

Larratt et al : Recurrent EBV in cardiac transplant

In the first two post-transplant months he seroconverted asymptomatically to EBV and developed a mild CMV syndrome that was successfully treated with IV ganciclovir. His postoperative course was otherwise uneventful.

immune globulin (IVIg) at 0.5 g/kg. The documented lymphadenopathy regressed rapidly. He was followed closely in the transplant clinic and was found to have significant fluctuation in his adenopathy but continued to

Monitoring of quantitative immunoglobulins and serum protein

feel quite well. He also had a significant flare of facial acne, which

electrophoresis (SPE) initially revealed a polyclonal gammopathy of

required both topical and systemic therapy with Minocycline. This

questionable significance. In December 1994 a faint monoclonal

flare was associated with worsening of his cervical adenopathy,

band appeared in the slow gamma region but was not identified on

which was noted to regress with control of the acne. In addition

follow-up SPE in April and July 1995. In June 1995 he was switched

intercurrent minor respiratory or other infections seemed also to be

from the Sandimmune to the Neoral cyclosporine preparation. A

associated with worsening adenopathy, often in areas unrelated to

monoclonal band was once again seen in the gamma region in Aug-

the site of infection, which resolved each time with resolution of the

ust 1995 and the patient was referred for hematological assessment.

infection.

Immunosuppression at that time consisted of cyclosporine 100 mg

He remained clinically asymptomatic until June 1999 when the

bid, prednisone 5 mg qd and azathioprine 125 mg qd. There were

patient presented with rapidly growing extensive cervical lymph-

no B symptoms of fever, night sweats, or weight loss. The patient

adenopathy associated with a dry cough, weight loss, fever, and

had noticed transient minimal cervical lymphadenopathy but gener-

night sweats. He was pancytopenic, with a total white blood cell

ally felt well. A cervical lymph node aspirate was performed that

(WBC) count of 2.2, hemoglobin 68 and platelet count of 128. A

showed numerous benign lymphocytes with no malignant cells

cervical lymph node biopsy was performed. The lymph node archi-

identified. Staging with a chest x-ray and an abdominal ultrasound

tecture was effaced by a diffuse proliferation including large pleo-

was negative for any lymphadenopathy. Management consisted of

morphic malignant cells, some small lymphocytes, histiocytes, and

continued close clinical follow-up.

occasional plasma cells. The large cells were predominantly mono-

In November 1997 the patient presented with slowly growing

nuclear with large central nucleoli and moderately abundant am-

lymphadenopathy in the left axilla. There were no additional symp-

phophilic cytoplasm, with varying numbers of large anaplastic cells.

toms present. The patient’s immunosuppressive medications were

Mitoses were frequent and there was focal patchy necrosis. The

unchanged. Clinical examination revealed modest cervical and

malignant cells were positive for CD30, strongly positive for LCA,

inguinal nodes measuring less than 1 cm each. A computerized

and negative for all other markers including CD20, CD79a, CD22,

tomographic (CT) scan showed small (1–1.5 cm) anterior medias-

CD2, CD3, and CD15. This was felt to be consistent with a diffuse

tinal and para-aortic lymphadenopathy, mild splenomegaly at 15

large cell lymphoma, with large anaplastic cell features, although a

cm and suggestion of small bowel wall thickening. The bone mar-

Hodgkin-like PTLD could also be considered. In situ hybridization

row was negative for lymphoma. There was persistence of the poly-

for EBER was strongly positive. A CT scan showed massive lymph-

clonal gammopathy on SPE; the monoclonal peak previously seen

adenopathy in the cervical chains, submandibular area, left supra-

was not detected. A left axillary node was biopsied. The lymph

clavicular area, right axilla, mediastinum, and retroperitoneum. The

node architecture was variably effaced by a diffuse proliferation of

spleen was enlarged to 18 cm. The bone marrow was cellular with

lymphoid cells with morphology ranging from small lymphocytes

all hematopoietic cells present. The trephine biopsy revealed several

to plasmacytoid lymphocytes to plasma cells. Focally, both an in-

focal aggregates of lymphoid cells, some paratrabecular in location.

crease in large transformed cells and extension through the lymph

The cells were pleomorphic with predominantly large cells marking

node capsule were evident. Necrosis was not present. Marker studies

with CD30 and strongly for leucocyte common antigen (LCA), but

confirmed the B cell nature of most of the cells, but monoclonality

negative for other markers similar to those seen in the lymph node.

assessed by light chain restriction was not demonstrated. An in situ

This was consistent with bone marrow involvement by large cell

study for EBV RNA (EBER) was strongly positive in the majority

lymphoma. A peripheral blood EBV viral load was performed and

of the cells. The appearance was consistent with a post-transplant

showed 2500 EBV genomes per 100,000 lymphocytes (Dr. D. Rowe,

lymphoproliferative disorder (PTLD) with features intermediate be-

Pittsburgh, PA). A CMV antigenemia assay and buffy coat shell

tween plasmacytic hyperplasia and polymorphous B-cell prolifer-

vial culture for CMV were negative. A multigated analysis (MUGA)

ation. The patient was treated with reduction of immunosuppres-

scan showed that the patient’s cardiac ejection fraction was good at

sion, with his cyclosporin reduced to maintain levels below 100 mg/

62%.

L, and his azathioprine was discontinued. He was started on acyclo-

The patient was initially managed with a further reduction in

vir 800 mg orally qid and given intermittent doses of intravenous

his cyclosporine and continued on acyclovir and IVIg. There was

120

Transplant Infectious Disease 2001: 3: 119–123

Larratt et al : Recurrent EBV in cardiac transplant

no clinical response over two weeks, and further intervention was

PTLD, the vital role of pathology in guiding therapy, and the man-

undertaken with CHOP chemotherapy (cyclophosphamide, adriamy-

agement strategies chosen.

cin, vincristine and prednisone) with an initial dose reduction of 33% of both cyclophosphamide and adriamycin due to marrow involvement. The patient experienced an episode of respiratory distress with hypoxia, sinus tachycardia, and fever 45 min into the

Risk factors for development of PTLD

cyclophosphamide infusion, requiring admission to the intensive care unit for monitoring, but he did not require intubation. This

Our patient had many of the major PTLD risk factors reviewed by

episode was not related to blood product administration. Apart from

Cockfield in this issue (1). He was both EBV and CMV mismatched

sinus tachycardia there were no documented arrhythmia and sub-

at the time of the transplant and developed primary infection with

sequent cardiac echocardiography was unremarkable. There was no

both viruses within the first few months after transplant. He also

evidence of pulmonary embolism on pulmonary angiography. Septic

experienced mild CMV disease. Although he received induction

work-up was also negative. Symptoms resolved quickly, and the

therapy with antilymphocyte globulin (ALG), the course was short.

patient was transferred back to the ward within 48 h of onset. Clin-

Receipt of a heart transplant placed him at higher risk of developing

ically, the patient improved following his first cycle of chemo-

PTLD than a comparable kidney transplant. Although our patient

therapy, with disappearance of all palpable lymphadenopathy and

developed EBV infection, presumably from the donor early after

reduction in spleen size. EBV viral load assays showed a reduction

transplant, he did not have clinical symptoms associated with EBV

to 400 genomes per 100,000 lymphocytes.

disease until four years later. This appeared coincident with a

The patient was given a second cycle of full-dose CHOP therapy

change in his cyclosporine formulation. Improved absorption of

three weeks later with cardiac monitoring. He tolerated it well with

cyclosporine when given as Neoral may have disrupted the pre-

no symptoms as seen in cycle 1. Subsequent chemotherapy was

viously established equilibrium between EBV and host immune sys-

equally uncomplicated. He received a total of five cycles of CHOP.

tem leading to the symptoms observed. Primary EBV and CMV

The sixth planned dose was not given due to a fall in his ejection

infection, CMV disease, use of ALG, and receipt of a non-renal allo-

fraction presumably due to anthracycline cardiotoxicity. His acyclo-

graft have all been clearly identified as risk factors for PTLD occur-

vir was switched to ganciclovir 1000 mg tid and he continued to

ring early after transplant. These factors have been less extensively

receive regular IVIg. Re-staging investigations were performed, with

studied in late PTLD. Our patient raises the question of whether

a repeat CT scan of the chest, abdomen and pelvis showing no

patients who are EBV seronegative but do not develop PTLD early

evidence of lymphadenopathy. His bone marrow was negative for

after transplant remain at increased risk of developing this compli-

disease as well. An EBV viral load assay at the completion of the

cation many years later.

chemotherapy demonstrated only 20 genomes per 100,000 lympho-

With the recognition of the risk factors in our patient, are there

cytes. His ganciclovir was stopped four months after completion of

strategies that could have been employed to decrease his cumulative

his chemotherapy. Within two months his EBV viral load climbed

risk? In this issue, Rowe et al. (2) have reviewed the role of EBV

to 400 genomes per 100,000 lymphocytes and the ganciclovir was

viral load monitoring in the management of patients with PTLD.

restarted. Follow-up EBV viral loads are now again low at 20 ge-

Studies have demonstrated that EBV viral loads in peripheral blood

nomes per 100,000 lymphocytes. The patient continues on oral gan-

are elevated at the time of PTLD diagnosis and rise prior to the

ciclovir and monthly IVIg and remains in ongoing clinical remission

development of symptomatic disease. There is evidence to suggest

13 months after initial chemotherapy administration.

that pre-emptive therapy in the form of reduction of immunosuppression, or antiviral or IVIg therapy given when viral loads are high may abort the subsequent development of PTLD (3, 4). EBV viral load assays were not available to us either early in our pa-

Discussion

tient’s post-transplant course or at the onset of lymphadenopathy when pre-emptive therapy may have had the greatest effect. Monitoring of EBV viral loads may also be useful in monitoring response

This case highlights many of the major problems that the health

to therapy. During the period of several years when our patient

care team faces in caring for the transplant patient with PTLD. The

demonstrated waxing and waning of lymphadenopathy despite the

three main areas we would like to highlight for further discussion

reduction in immunosuppression and antiviral therapy, it is possible

are the risk factors present in this patient for the development of

that his EBV viral load remained high. Preliminary data suggest Transplant Infectious Disease 2001: 3: 119–123

121

Larratt et al : Recurrent EBV in cardiac transplant

that patients with high EBV viral loads may be at extremely low risk for rejection events (5). Withdrawal of immunosuppression is

Management

of significant concern in recipients of vital organ transplants. Had viral load monitoring been available during the polyclonal polymor-

Our patient’s long and protracted course exemplifies the precarious

phic phase of our patient’s illness, clinicians may have had signifi-

equilibrium that exists between immune control and EBV-driven

cantly more comfort in withdrawing immunosuppression totally in

lymphoproliferation in immunosuppressed transplant recipients (8).

this patient or more rationally titrating maintenance immunosup-

Molecular studies of oncogenes and tumor suppressor genes were

pression.

not performed on the tissues in this patient. However, his final pres-

Green et al. (6) have reviewed the controversies surrounding the

entation of a rapidly growing large cell lymphoma likely resulted

role of antiviral and immunoglobulin therapy in the prevention of

from genetic mutation in a rapidly proliferating EBV-infected cell

PTLD in this issue. Although presenting late after transplant, our

that made this clone resistant to control by the immune system. At

patient developed PTLD despite receiving high-dose acyclovir

this point in his clinical course, intervention with chemotherapy

prophylaxis for CMV prevention and intravenous ganciclovir for

was undertaken. Management with the monoclonal anti-CD20

mild CMV disease early in the transplant course. Our patient was

would have been attractive in view of its low toxicity profile (9), but

maintained on both high-dose acyclovir and IVIg from the time of

as our patient’s tumor was CD20 negative this option was not open

presentation of his documented PTLD until the onset of his rapidly

to us, illustrating the importance of examining PTLD tissue for this

progressive anaplastic disease, almost two years later. Whether the

marker. The choice of chemotherapy protocol was dictated by our

use of these agents slowed and altered the natural history of disease

institutional standard for treatment of intermediate-grade non-

in this patient remains speculative. It is hoped that randomized

Hodgkin’s lymphoma (10). Our patient did experience chemo-

prospective clinical trials currently underway will provide better

therapy-associated cardiotoxicity as demonstrated by a falling ejec-

data on the utility of these agents for the prevention of PTLD.

tion fraction. It has been suggested that the transplanted heart may be at greater risk than the native heart for the development of antracycline-associated cardiotoxicity (11). In retrospect, a choice of a protocol with a lower dose of anthracycline (e.g. PROMACE CytaBOM) may have been wise. The patient’s initial reaction to the

Pathology

chemotherapy was also unexpected. He had a high LDH and bulky adenopathy consistent with a high tumor load. Tumor lysis was

Our patient illustrates several of the issues raised by Nalesnik (7)

anticipated and appropriate preventative measures were taken. His

regarding the need for a standardized approach to the pathology in

symptoms, however, seemed more consistent with a cytokine storm

patients with PTLD. In 1995 the initial lymphadenopathy in our

with respiratory and circulatory compromise, which resolved rapid-

patient was evaluated by performing a lymph node biopsy inter-

ly with supportive measures only, unlike a more typical tumor lysis

preted as consisting of only benign-appearing lymphocytes. This

syndrome with electrolyte and renal abnormalities. This seems

gave us a false sense of security leading to no therapeutic change

more akin to a TRALI (transfusion reaction acute lung syndrome),

at this early stage in the patient’s course, when intervention may in

which is also felt likely to be a cytokine-driven event. A rebound in

fact have had the greatest impact. Excisional biopsy of the largest

viral load was observed in our patient in the absence of symptoms

accessible lymph node evaluated with a full range of diagnostic

and recurrent disease after completion of chemotherapy and with-

tests including histopathology, immuno-phenotyping, clonality

drawal of oral ganciclovir as described by Rowe et al. (2). Whether

studies, and EBV status would have been a more optimal diagnostic

the reintroduction of antiviral therapy resulted in the subsequent

pathway. Although presenting late after transplant, our patient’s

fall in EBV viral load is uncertain.

initial PTLD lesion was EBV positive and had histologic features

In summary, this case represents an atypical late presentation of

more commonly seen in the early post-transplant period. This sug-

EBV-positive PTLD in a cardiac transplant patient who was EBV

gested that it would be responsive to a reduction in immunosuppres-

seronegative in the pre-transplant period. PTLD progression was

sion as was observed. The final ‘‘recurrent’’ PTLD in our patient

slow and occurred over several years. Our initial treatment ap-

differed in histology, cell phenotype, and clonality when compared

proach of reduction in immunosuppression, acyclovir, and IVIg ap-

to the pathologic findings two years earlier. This reinforces the need

peared to have slowed progression of his PTLD initially, but there

for repeat biopsy of tissue and pathologic examination in ‘‘recur-

was final breakthrough with rapid clonal proliferation and trans-

rent’’ disease (7) in order to rationally plan therapeutic intervention.

formation to large cell lymphoma. CHOP chemotherapy sequentially

122

Transplant Infectious Disease 2001: 3: 119–123

Larratt et al : Recurrent EBV in cardiac transplant

eradicated all clinically detectable disease and the patient remains

ment strategies to reduce the incidence of this life-threatening com-

well. Our challenge for the future is to further refine our manage-

plication of solid organ transplantation.

References 1. COCKFIELD SM. Identifying the patient at risk for post-transplant lymphoproliferative disorder. Transpl Infect Dis 2001: 3: 70–78. 2. ROWE DT, WEBBER S, SCHAUER EM, REYES J, GREEN M. Epstein–Barr virus load monitoring: its role in the prevention and management of PTLD. Transpl Infect Dis 2001: 3: 79–87. 3. MCDIARMID SV, JORDAN S, LEE GS, et al. Prevention and preemptive therapy of posttransplant lymphoproliferative disease in pediatric liver recipients. Transplantation 1998: 66: 1604–1611. 4. GREEN M, BUENO J, ROWE DT, et al. Predictive negative value of persistent low EBV viral load after intestinal transplantation in children. Transplantation 2000: 70: in press.

5. WEBBER SA, ROWE E, BOYLE GJ, et al. Natural history of EBV load in pediatric thoracic transplant recipients with posttransplant lymphoproliferative disorders and other primary EBV infections. Transplantation 1999: 67: 215. 6. GREEN M, REYES J, WEBBER S, ROWE D. The role of antiviral and immunoglobulin therapy in the prevention of Epstein–Barr virus infection and post-transplant lymphoproliferative disease following solid organ transplantation. Transpl Infect Dis 2001: 3: 97–103. 7. NALESNIK M. The diverse pathology of posttransplant lymphoproliferative disorders: importance of a standardized approach. Transpl Infect Dis 2001: 3: 88–96.

8. TANNER JE, ALFIERI C. The Epstein–Barr virus and post-transplant lymphoproliferative disease: interplay of immunosuppression, EBV and the immune system in disease pathogenesis. Transpl Infect Dis 2001: 3: 60–69. 9. DURANDY A. Anti-B cell and anti-cytokine therapy for the treatment of posttransplant lymphoproliferative disease; past, present and future. Transpl Infect Dis 2001: 3: 104–107. 10. DAVIS CL. Interferon and cytotoxic chemotherapy for the treatment of posttransplant lymphoproliferative disorder. Transpl Infect Dis 2001: 3: 108–118. 11. SWINNEN LJ, MULLEN GM, CARR TJ, COSTANZO MR, FISHER RI. Aggressive treatment for post-cardiac transplant lymphoproliferation. Blood 1995: 86: 3333– 3340.

Transplant Infectious Disease 2001: 3: 119–123

123