|Home | About | Journals | Submit | Contact Us | Français|
Background: Recently novel Epstein–Barr virus (EBV) lymphoproliferative diseases (LPDs) have been identified in non-immunocompromised hosts, both in Asia and Western countries. These include aggressive T-cell and NK-cell LPDs often subsumed under the heading of chronic active Epstein–Barr virus (CAEBV) infection and EBV-driven B-cell LPDs mainly affecting the elderly.
Design: To better define the pathogenesis, classification, and treatment of these disorders, participants from Asia, The Americas, Europe, and Australia presented clinical and experimental data at an international meeting.
Results: The term systemic EBV-positive T-cell LPD, as adopted by the WHO classification, is preferred as a pathological classification over CAEBV (the favored clinical term) for those cases that are clonal. The disease has an aggressive clinical course, but may arise in the background of CAEBV. Hydroa vacciniforme (HV) and HV-like lymphoma represent a spectrum of clonal EBV-positive T-cell LPDs, which have a more protracted clinical course; spontaneous regression may occur in adult life. Severe mosquito bite allergy is a related syndrome usually of NK cell origin. Immune senescence in the elderly is associated with both reactive and neoplastic EBV-driven LPDs, including EBV-positive diffuse large B-cell lymphomas.
Conclusion: The participants proposed an international consortium to facilitate further clinical and biological studies of novel EBV-driven LPDs.
Over 90% of humans are infected with the Epstein–Barr virus (EBV) and the infection persists for life. Most persons have a chronic asymptomatic infection with EBV, but the virus has been associated with a number of malignancies and can infect B cells, T cells, NK cells, and epithelial cells. Patients with iatrogenic, congenital, or acquired immunodeficiency are at increased risk for EBV-associated lymphomas, which are in nearly all instances of B-cell lineage.
Chronic active Epstein–Barr virus (CAEBV) disease has been defined as a systemic EBV-positive lymphoproliferative disease (LPD) characterized by fever, lymphadenopathy, and splenomegaly developing after primary virus infection in patients without known immunodeficiency . Affected patients have high levels of EBV DNA in the blood, histological evidence of organ disease, and elevated levels of EBV RNA or viral proteins in affected tissues. While initially proposed as a progressive EBV infection of B cells as the primary target, the term as used in the recent literature refers to an aggressive EBV-positive T-cell, NK cell, or B-cell LPD, mainly affecting persons of Asian origin .
EBV-associated hemophagocytic syndrome (HPS), which can appear with CAEBV, or as a complication of other EBV-associated LPD, is due to excessive macrophage activation and hemophagocytosis [3, 4]. Patients present with fever, lymphadenopathy, pancytopenia, and hepatosplenomegaly, and have marked elevation of cytokines including tumor necrosis factor-α (TNF-α) and interferon (IFN)-γ [5, 6]. The disease is often fatal, despite therapy directed at the virus-infected T cells. In addition to corticosteroids or cyclosporine to inhibit cytokine production, bone marrow or hematopoietic stem-cell transplant may be effective in some cases.
In recent years, it has been appreciated that otherwise healthy adults of advanced age are at risk for EBV-associated B-cell lymphomas, initially reported as ‘senile EBV-associated B-cell LPD’ . Clinical series have identified a high risk of treatment failure [8, 9]. This process was incorporated in the fourth edition of the World Health Organization (WHO) classification of lymphomas as EBV-positive diffuse large B-cell lymphoma (DLBCL) of the elderly .
To better define the pathogenesis, classification, and treatment of EBV LPDs in non-immunocompromised hosts, an international meeting was organized at the National Institutes of Health in Bethesda, MD, on 8 and 9 September 2008. Virologists, immunologists, pathologists, infectious disease specialists, hematologists, and oncologists presented topics on a wide variety of EBV-associated diseases and discussed the pathogenesis, classification, and treatment of these diseases. This report presents highlights from that meeting and a consensus document regarding the classification of these diseases (Table 1).
A variety of EBV viral proteins are expressed in EBV-associated LPD, and these influence the nature and effectiveness of the immune response and the potential risk for lymphomagenesis. As reviewed by E. Kieff (Boston), EBV nuclear antigen-1 (EBNA-1) maintains the EBV episome in these cells when the cells divide. EBNA-2 interacts with a variety of host cell proteins, including RBP-JK, p300/CBP, and p100 to upregulate expression of viral and cellular genes including c-myc, CD21, and CD23. EBNA-3A, B, and C also upregulate expression of cellular genes. EBV nuclear antigen-leader protein activates transcription in concert with EBNA-2 by reducing levels of histone deacetylase 4 in the nucleus. Latent membrane protein 1 (LMP1) is a functional homolog of CD40 and upregulates expression of a wide variety of cellular genes including nuclear factor-κB (NF-κB), c-Jun, AP1, AP2, and p38.
As discussed by R. Khanna (Brisbane), both the quantity and quality of the CD8+ T-cell response to EBV are critical to control infection. CD8+ T cells isolated from healthy seropositive individuals or from individuals who rapidly recover from infectious mononucleosis recognize a wide variety of EBV epitopes . In contrast, CD8+ T cells from persons with persistent infectious mononucleosis recognize only a few EBV epitopes (Figure 1). The diversity of the T-cell repertoire (as defined by Vβ T-cell receptors) is expanded in persons infected with EBV who are asymptomatic, while the repertoire is much narrower in persons with infectious mononucleosis . A similar reduction in diversity of the T-cell receptor repertoire was seen in some symptomatic transplant recipients, in contrast to a broader diversity observed in asymptomatic transplant recipients. No correlation was noted between the level of the EBV DNA load in the blood in transplant recipients and the diversity of the T-cell receptor repertoire.
A variety of potent cytokines and chemokines are induced by EBV infection of both B and T lymphocytes. These mediators are produced by the EBV-infected lymphoid cells directly, as well as bystander immune populations. As discussed by G. Tosato (Bethesda), vIL-10 (BCRF1) is structurally homologous to hIL-10 and is a paradigm for viral piracy of a cellular gene . Viral IL-10 stimulates B-cell growth, inhibits antigen presentation and T-cell growth, protects T cells from death, and suppresses IFN-γ secretion. EBV also promotes the expression of cellular cytokines in infected cells, including IL-6, IL-10, and EBI3. These lead to autocrine and paracrine stimulation of EBV-infected B-cell growth and likely promote the development of B-cell lymphomas.
EBV also stimulates secretion of a variety of cytokines and chemokines by noninfected cells, and these have an impact on the clinical features of acute and CAEBV infection. The LMP1-induced chemokines interferon-gamma-inducible protein-10 and the monokine induced by interferon-gamma mediate vascular damage, resulting in tissue necrosis . High levels of IFN-γ, soluble IL-2 receptor, TNF-α, and MIP-1α have been implicated in the pathogenesis of EBV-associated HPS, associated with CAEBV and T-cell and NK-cell lymphomas [5, 6, 15, 16].
CAEBV infection was first identified by Straus  as a disease related to chronic or persistent EBV infection of B cells. It was defined as a severe illness greater than 6-month duration that (i) begins as a primary EBV infection or is associated with markedly abnormal EBV antibody titers (e.g. anti-EBV viral capsid antigen IgG ≥ 5120, anti-EBV early-antigen IgG ≥ 640, or anti-EBNA < 2); (ii) histological evidence of major organ involvement such as interstitial pneumonia, hypoplasia of the bone marrow, uveitis, lymphadenitis, persistent hepatitis, or splenomegaly; and (iii) increased EBV RNA or proteins in affected tissues. Kimura et al.  revised the diagnostic criteria of CAEBV, so that patients could have either increased EBV RNA or proteins in infected tissues or increased levels of EBV in the peripheral blood, in addition to the other criteria. As reviewed by JIC (Bethesda), patients who have CAEBV of B cells usually develop a progressive cellular and humoral immunodeficiency and other complications including HPS and B-cell LPD. The etiology of B-cell CAEBV remains uncertain, although mutations in perforin were implicated in one patient, whose T cells were impaired for killing of Fas-deficient target cells .
In recent years, it has been appreciated that defective immune surveillance for EBV may develop late in life and be associated with the development of EBV-positive B-cell LPD in individuals who otherwise have no apparent immune deficiency. As reviewed by SN (Nagoya), aggressive EBV-positive B-cell lymphomas that occur in older individuals are often extranodal, frequently involving the skin, gastrointestinal tract, or lung [7, 18, 19]. Termed EBV-positive DLBCL of the elderly in the 2008 WHO classification, the disease is characterized by proliferation of atypical large B cells including immunoblasts and Reed–Sternberg-like cells (Figure 2). Some cases have a more varied cytological composition and resemble the EBV-positive B-cell lymphomas that occur in iatrogenically immunosuppressed patients. Necrosis is prominent. Patients with EBV-positive DLBCL have a worse prognosis than patients with EBV-negative DLBCL or EBV-positive classical Hodgkin's lymphoma (CHL) [9, 20]. Most cases occur after the age of 60 with a median age of 70–79 years, and the incidence continues to increase with age.
ESJ (Bethesda) described the spectrum of EBV-associated B-cell LPD observed in a Western population without known immunodeficiency . As with the reports from Japan, most patients are of advanced age, generally >60 years; 116 cases were identified over a 7-year period and fell into five diagnostic categories: (i) lymph node-based reactive hyperplasia with increased EBV-positive B cells, (ii) EBV-positive nodal B-cell lymphoproliferations resembling post-transplant LPD (PTLD), (iii) EBV-positive extranodal B-cell lymphoproliferations resembling PTLD, (iv) EBV-positive diffuse DLBCLs, and (v) EBV-positive B-cell proliferations resembling CHL.
Twenty-eight patients had EBV-associated reactive lymphoid hyperplasia, with a median age of 67 years. The process was self-limited in most patients, with only one patient showing progression to a more aggressive lymphoproliferative process. All cases tested were polyclonal by IgH PCR. T-cell clonality or a restricted T-cell receptor gene rearrangement pattern was seen in three (11%) of the cases studied. This finding suggests a reduction in diversity of the T-cell receptor repertoire, as discussed by Khanna . Features included preserved architecture and a broad spectrum in cell size of the EBV-positive cells with frequent localization to germinal centers. Kojima et al.  reported similar EBV-associated reactive hyperplasias in middle-aged or elderly patients.
The median age was highest in patients with EBV-positive DLBCL (77 years), which included 11 nodal DLBCL and four nodal or extranodal plasmablastic lymphomas. There were 73 polymorphic B-cell lymphomas approximately equally divided between nodal and extranodal sites. Median ages were 73 and 76 years, respectively. Seven cases, median age 79, histologically and phenotypically resembled CHL (CD30+; CD15+) but presented in sites unusual for CHL such as the oral cavity (palate, gingiva, tongue, lips) and adrenal glands. Supporting the concept that a restricted T-cell response to EBV may be associated with defective immune response, ~20% of patients with either polymorphic B-cell lymphoma or DLBCL had evidence of a restricted clonal or oligoclonal T-cell response.
Lymphomatoid granulomatosis (LYG) is an EBV-related B-cell LPD that can affect patients with known immunodeficiency, but also occurs in adults without any known predisposing risk factors . As reviewed by K. Dunleavy (Bethesda) lung involvement is nearly constant (98% of patients); ~30% of patients have lesions in the kidneys, liver, skin, or central nervous system . The number of EBV-infected B cells is relatively low, in proportion to the number of T cells identified within the lesions. However, there is evidence of defective immune surveillance, as the mean CD4 and CD8 T-cell counts are below normal in most patients at diagnosis. The EBV viral load in the blood is usually not elevated. A clinical trial at the National Cancer Institute involving 40 patients used dose-adjusted IFN-α for patients with grade 1 or 2 LYG, and dose-adjusted EPOCH with rituximab for patients with grade 3 disease . Twenty-seven percent of the patients were previously untreated. Progression-free survival for patients receiving IFN-α was 56% with median follow-up of 5.1 years; nine patients progressed to grade 3 disease . The overall complete response (the disappearance of all signs of cancer in response to treatment) rate with dose-adjusted chemotherapy was 68%. The overall survival in patients with grades 1–3 disease was 69% with median follow-up of 4.3 years. These preliminary results represent an improvement over older series; however, due to the rarity of this condition, most reports are limited to few patients [24, 27, 28].
The spectrum of EBV-associated B-cell lymphomas and LPDs is extremely broad and includes acute infectious mononucleosis, benign reactivation as may be seen in the elderly; CAEBV infection involving B cells, LYG, post-transplant and other iatrogenically associated LPDs (e.g. methotrexate-associated LPD); pyothorax-associated lymphoma (now defined as EBV-positive DLBCL associated with chronic inflammation) ; EBV-positive DLBCL of the elderly; Burkitt lymphoma (EBV more often associated with endemic than sporadic); plasmablastic lymphomas (most cases associated with EBV), and CHL (EBV mainly in mixed cellularity and lymphocyte depletion subtypes). As discussed by the participants, the clinical syndrome and pathology are influenced by the virus and viral genes and the host, including both intrinsic and iatrogenic factors .
While CAEBV was first described as a persistent EBV infection targeting B cells, over the years the syndrome has been primarily associated with EBV infection of T cells and less often NK cells . The minimal diagnostic criteria for CAEBV are summarized above, all of which must be met. It has a strong racial predisposition, with most cases occurring in Japan and Korea and some cases in Native American populations in the Western Hemisphere from Mexico, Peru, and Central America. It is rare in Caucasians and African-Americans. The term T/NK cell CAEBV has been used in the literature to encompass a very broad spectrum of diseases, including a systemic form which may be polyclonal, fulminant and systemic EBV-positive T-cell LPDs that are clonal, hydroa vacciniforme (HV) of T-cell derivation, and severe mosquito bite allergy (usually of NK cell origin). The 2008 WHO classification has recognized the following disease entities that are considered neoplasms: systemic EBV-positive T-cell LPD of childhood (a clonal T-cell LPD) and HV-like T-cell lymphoma .
HK (Nagoya) carried out a nation-wide survey of T/NK–CAEBV in Japan in 2001 and identified 82 cases (42 males and 40 females). The mean age of onset of the disease was 11.3 years with a range of 9 months to 53 years and all patients had elevated levels of EBV DNA in the blood [29, 30]. The majority of patients had evidence of systemic disease, presenting with fever (93% of patients), hepatomegaly (79%), splenomegaly (73%), thrombocytopenia (45%), anemia (44%), and lymphadenopathy (40%). Cutaneous manifestations were common and included hypersensitivity to mosquito bites (33%), skin rash (26%), and HV (10%). Patients with only cutaneous disease had a better prognosis, although the criteria to distinguish HV, which may be clonal, from HV-like T-cell lymphoma are not well delineated [31, 32]. It has been controversial as to whether CAEBV is a type of T-cell or NK cell malignancy or a progressive infectious disease; of the patients in whom clonality could be analyzed, the proliferation was monoclonal in 76%, oligoclonal in 13%, and polyclonal in 11% . The EBV-infected cells were shown to be T cells in 46% of patients, NK cells in 33%, T/NK cells in 4%, B cells in 2%, and unclassified or not studied in the remaining 15%.
Patients with T-cell CAEBV often presented with high fever, lymphadenopathy, hepatosplenomegaly, high titer of EBV-specific antibodies, and had rapid progression of their disease. Patients with NK cell disease, in contrast, often had hypersensitivity to mosquito bites, rash, high levels of IgE, and did not necessarily have elevated EBV-specific antibody titers. The 5-year survival rate of patients with T-cell CAEBV was 59%, while that for NK cell disease was 87% . However, uncomplicated HV (of clonal T-cell derivation) had a better prognosis. Life-threatening complications of T/NK cell CAEBV included HPS (24% of patients), disseminated intravascular coagulation (16%), hepatic failure (15%), peptic ulcer disease/perforation (11%), coronary artery aneurysms (9%), central nervous system complications (9%), myocarditis (6%), and interstitial pneumonitis (5%).
The pathogenesis of T-cell and NK cell CAEBV is uncertain. EBV-positive T/NK cells have been identified in the tonsils and peripheral blood from patients with infectious mononucleosis , and the virus has been shown to infect NK cells in vitro . NK cells can acquire the EBV receptor, CD21, by synaptic transfer from B cells , allowing EBV binding to NK cells. T and NK cells from patients with CAEBV often have latency 2 phenotype with expression of EBV EBNA-1, LMP1, and LMP2 . There is evidence that defective T-cell and NK cell responses to EBV may play a role in the pathogenesis [29, 37].
While a number of therapies have been tried for CAEBV including antiviral agents (acyclovir, ganciclovir), immunomodulators (IFN-α, IL-2), chemotherapy (etoposide, corticosteroids), cyclosporine, and EBV-specific cytotoxic T cells (CTLs), recently more promising results have been obtained with hematopoietic cell transplantation [38, 39]. Since the first report of successful allogeneic bone marrow transplantation for the disease , many successful cases have been reported using related or unrelated bone marrow transplants, with myeloablative or nonmyeloablative transplantation or with cord blood transplantation . Hematopoietic stem-cell transplantation can eliminate EBV-infected cells, reconstitute EBV-specific cellular immunity, and have a graft-versus-tumor effect. However, the procedure carries a high risk of transplantation-related complications and the 5-year survival rate was only 53% (Japanese Association for Research on EBV Study Group, unpublished data). Poor prognostic factors that argue for early intervention and transplant are (i) age at onset >8 years, (ii) platelet count <120000/μl, and (iii) T-cell- rather than NK cell-associated disease . Patients with HV have a better prognosis and may be followed conservatively, if there are no systemic symptoms. Measurement of the viral load after transplantation was helpful in determining the response to transplantation.
K. Oshima (Kurume) presented a proposed categorization of CAEBV from the CAEBV study group . They divided cases into four categories: A1 (polymorphic and polyclonal), A2 (polymorphic and generally monoclonal), A3 (monomorphic and monoclonal proliferation of T-cell or NK cell origin, and B (monomorphic and monoclonal T-cell LPD with fulminant clinical course). The clinical course in groups A1–A3 was generally protracted with most patients surviving for several years. Group B was defined as equivalent to fulminant EBV-positive LPD of childhood ; patients were under the age of five, had a fulminant clinical course that emerged soon after EBV infection, and morphology and phenotype that overlapped with group A3. Patients with the clinical syndromes of mosquito bite allergy and HV were distributed in groups A2 and A3. All patients had very high viral loads at presentation. Anti-EBV antibody titers were highest in A1 (VCA IgG 2560) and lowest in B (VCA IgG 160). Interestingly, antibody titers to EBV also were reported to be low in fulminant EBV-positive LPD of childhood  (now designated ‘systemic EBV-positive T-cell LPD of childhood’ in the WHO classification of 2008 ). It will be of interest to apply this classification system prospectively to CAEBV cases to determine its applicability as a diagnostic and prognostic system.
The perspective of EBV-related T-cell and NK cell disease in Korea was presented by Y-HK (Seoul) . Cases of systemic EBV-positive T-cell LPD and related entities were compared with more well-defined diseases such as extranodal NK/T-cell lymphoma and aggressive NK cell leukemia. Systemic EBV-positive T-cell LPD patients were mainly children and young adults and presented with acute illness with a fulminant clinical course, similar to aggressive NK cell leukemia, with death in a matter of weeks. These cases were comparable in behavior to those reported in the literature as ‘fatal infectious mononucleosis’ with HPS . There was a subset of children and young adults with CAEBV and a somewhat more protracted clinical course. Some of these patients had cutaneous manifestations, such as HV, but the median survival was still <1 year. Y-HK also identified a subset of patients presenting in adult life, who were often coinfected with hepatitis B or C virus, leading to reactivation of EBV .
The perspective of systemic T-cell LPD of childhood (CAEBV) in the Western hemisphere was presented by L. Quintanilla-Martinez (Tübingen). There is evidence of a strong racial predisposition, as nearly all patients were of Native American ethnic origin from Mexico or Central America . Previously healthy patients presented with acute onset of fever suggestive of an acute viral respiratory illness. Within a period of weeks patients developed hepatosplenomegaly and liver failure, sometimes accompanied by lymphadenopathy. Laboratory tests showed pancytopenia, abnormal liver function tests, and often an abnormal EBV serology with low or absent anti-VCA IgM antibodies. The disease was usually complicated by HPS, coagulopathy, multiorgan failure, and sepsis . The clinical course was aggressive, with a median survival of <1 year. The value of morphological subtyping was felt to be questionable, as in most cases the EBV-positive T cells lacked cytological atypia. The immunophenotype was of cytotoxic T-cell origin, CD8 > CD4. All cases studied were monoclonal for TCR gamma genes, and on this basis as well as the poor clinical outcome, the process has been considered to represent a form of mature T-cell malignancy in the 2008 WHO classification .
X-linked lymphoproliferative disease (XLPD), which is caused by mutations in the SAP gene, and CAEBV share many clinical features in common. Acute infection usually results in a fulminant disease with infiltration of multiple organs by EBV-infected B cells and activated T cells with HPS and tissue necrosis. Survivors often have hypogammaglobulemia and may develop B-cell lymphomas. Other complications include aplastic anemia, necrotizing vasculitis, or LYG. Based on data derived from a SAP knockout mouse model, J. Sullivan (Worcester) suggested that patients with XLPD may have defective apoptosis of CD8 T cells that predisposes them to the HPS and fatal EBV infections .
The clinical spectrum of HV in Asia and the Western hemisphere was presented by X. Zhou (Beijing) and C. Barrionuevo (Lima). The median age of patients from China was 7 years (range 3–15 years), with an increased male-to-female ratio. All the patients presented with a papulovesicular rash, with ulceration and crusting, primarily affecting sun-exposed areas of the skin. Twenty-five percent (4 of 16) of the patients also reported hypersensitivity to mosquito bites. EBV-positive cells were abundant in the lesions during periods of active disease (spring, summer); lesions often regressed during the autumn and winter. Most of the HV patients also had evidence of systemic disease. About 80% (13 of 16) of patients presented with high fever 38°C–40°C and 38% of patients had hepatosplenomegaly and/or lymphadenopathy. Follow-up data (mean 22 months; range 4–46 months) was available for 44% (7 of 16) of cases. Two patients died of liver or multiple organ failure, and five were still alive with a stable or smoldering disease.
C. Barrionuevo (Lima) presented cases with HV-like lesions from Peru, which were categorized as HV-like T-cell lymphoma in their series based on infiltrative growth pattern, often aberrant T-cell phenotype, clonal rearrangement of TCR genes, and poor clinical outcome . The clinical and pathological features are very similar to those observed in Japan and Korea (Figure 3). The mean age of patients in Peru was 11 years (range 5–17 years). Lesions most commonly involved sun-exposed areas (face and upper limbs). Lesions often showed edema, papules, blisters, crusts, ulcers, and healed as vacciniforme scars. Some patients had hypersensitivity to insect bites. Systemic symptoms were common and lymphadenopathy was present in 30% of cases and hepatosplenomegaly in 10%. Less frequent were intercurrent infections, HPS, or visceral involvement. The 2-year survival rate was 43%. Patients receiving chemotherapy or chemotherapy and radiation therapy had partial response (a decrease in the size of a tumor, or in the extent of cancer in the body in response to treatment) rates of 30%. Deaths were due to sepsis, liver failure, malignancy, or HPS.
The criteria for the distinction of HV from HV-like T-cell lymphoma have not been clearly delineated in the literature. Based on the published experience and reports presented at the meeting, EBV and T-cell clonality were found in both types of cases. Some patients with HV have eventual resolution of their disease in adult life, whereas other patients develop progressive disease with worsening of cutaneous symptoms and eventual systemic dissemination [31, 32, 48, 49]. In addition, some patients with HV-like symptomatology have severe CAEBV early in the course of the disease. Cases of HV lacking clonal rearrangement of TCR genes appear to have a more benign clinical course . The entity of HV-like lymphoma as included in the WHO classification stipulates an EBV-positive clonal proliferation . However, it is not clear that T-cell clonality is always predictive of a progressive clinical course, as discussed by HK. A related issue is severe mosquito bite allergy, which is usually of NK cell derivation, but shows overlap with HV [52, 53]. Both HV and severe mosquito bite allergy are considered part of the spectrum of CAEBV, with a broad spectrum of clinical aggressiveness.
C. Rooney (Houston) discussed EBV-specific T-cell therapy for patients with chronic and persistent EBV infection. In Western populations (e.g. United States, Europe) these patients usually have expansion of EBV-infected B cells, i.e. the B-cell type of CAEBV. Historical therapeutic approaches, before the use of bone marrow or stem-cell transplantation, have included high-dose immunoglobulin, IL-2, antiviral agents, IFN-α or IFN-γ, corticosteroids, and rituximab. Other than isolated case reports, these therapies generally have not been successful and relapses were common. More recently autologous EBV-specific T cells have been used in the therapy of persistent active EBV infection . This therapy also has been successful in the post-transplant setting . Incubation of the patient's peripheral blood mononuclear cells with their irradiated EBV-transformed B cells results in activation of EBV-specific T cells in all seropositive donors, including both healthy persons and those with EBV-associated LPDs. In some cases, however, generation of EBV-transformed B cells may be difficult or impossible if the patient received ritxumab within the prior 6 months or high doses of chemotherapy. The induced activated EBV-specific T cells recognize EBV EBNA-3A, 3B, and 3C and early-lytic antigens to a greater extent than T cells that recognize LMP1, LMP2, or EBNA-1.
A phase I study was reported using autologous EBV-specific T cells in patients with mild CAEBV defined as >6 months of symptoms (most often fever and fatigue) and either elevated peripheral blood EBV load or free EBV DNA in serum/cerebrospinal fluid or EBV VCA antibody titer >1:640 . The study showed improvement or resolution of symptoms and the follow-up on these patients is now 2–6 years. Many of these patients have had normalization or reduction of their EBV VCA antibody titers, a decrease in the EBV DNA load, and an increase in circulating EBV-specific CTLs after therapy. Additional patients have now been enrolled with severe CAEBV involving B or T cells; while some of these patients have shown clear responses to therapy, other have required hematopoietic stem-cell transplantation. Newer techniques are being developed to target additional EBV proteins, particularly LMP1, LMP2, and EBNA-1. Adenovirus vectors have been used to infect both dendritic cells and EBV-transformed B cells to enrich for the frequency of these antigens as targets for the patient's peripheral blood mononuclear cells [56, 57] (Figure 4). Adenovirus expressing EBV LMP2 has been used to generate LMP2-specific autologous CTLs in a clinical trial in patients with EBV-positive lymphomas . Most patients with relapsed disease had a complete response to the CTLs.
H. Heslop (Houston) discussed the use of hematopoietic stem-cell transplantation for treatment of CAEBV. This approach has been used in children with hemophagocytic lymphohistiocytosis (HLH), a hereditary disorder that shares many features with CAEBV . As with CAEBV, most of the deaths are related to early complications of transplantation . More recent studies in HLH using reduced intensity conditioning with fludarabine, melphalan, and alemtuzumab have resulted in improved survival rates of 75%–93% at 1 year , as compared with fully ablative transplantation.
Transplantation for CAEBV has resulted in survival rates of 50%–64%. Younger patients, those with lower viral loads and those with less intense conditioning regimens have had improved survival . A number of strategies are being considered to reduce the risk of relapse including the use of additional boosts of EBV-specific T cells after transplant, infusions of T cells that recognize antigens on lymphoma cells (e.g. CD70), or enhancing alloreactivity through the use of antibodies that can link tumor cells and CTLs by binding both tumor cell antigens and antigen receptors on CTLs.
W. Wilson (Bethesda) discussed the use of novel therapies for EBV LPD. Antiviral therapy may have some activity, particularly at an early stage when virus replication is more prominent. A study of valganciclovir in 47 children with EBV infection after liver transplantation showed a reduction in EBV viral load and no new cases of post-transplant LPD . Histone deacetylase inhibitors, such as sodium butyrate induce viral gene expression with lytic EBV replication. These agents induce expression of the viral thymidine kinase which can phosphorylate ganciclovir which is toxic to cells . A recent clinical trial using arginine butyrate and ganciclovir resulted in improved survival . High-dose sodium butyrate and ganciclovir can block the phosphotidylinositol-3-kinase kinase/Akt pathway to kill virus-infected cells. Valproic acid is a potent histone deacetylase inhibitor and induces lytic EBV gene expression. Combination therapy of valproic acid with gemcitabine kills EBV-transformed B cells in vitro and in SCID mice more effectively than chemotherapy alone [64, 65]. EBV-transformed B cells show activation of NF-κB, and bortezomib, a proteasome inhibitor, leads to increased levels of IκB kinase and inhibits activation of NF-κB. EBV LMP1 inhibits apoptosis to prevent death of transformed B cells. HA14-1 is a small molecule inhibitor of Bcl-2 that kills EBV-transformed B cells . The combination of bortezomib and HA14-1 was synergistic for killing EBV-transformed B cells in vitro. Another NF-κB inhibitor, dehydroxymethylepoxyquinomicin induced apoptosis of EBV-transformed B cells . The mammalian target of rapamycin pathway is another target for killing EBV-transformed B cells. Inhibition of this pathway affects multiple downstream signaling molecules required for protein synthesis and cell cycle progression. Rapamycin kills EBV-transformed B cells in vitro and in SCID mice . Thus, the study of molecular pathways activated in EBV-transformed B cells has led to candidates for the treatment of EBV LPD.
The lack of understanding of uncommon EBV LPDs affecting B cells, T cells, and NK cells is aggravated by confusion in the literature regarding terminology and diagnostic criteria for individual disease entities and clinical syndromes. Meeting participants concluded that the term CAEBV should be applied to systemic LPDs that are not frank lymphomas and that arise during primary infection and persist for over 6 months. In addition, the nature of the EBV-infected cell, B, T, or NK, should be specified in all instances. CAEBV of B-cell origin also has been referred to as chronic (or persistent) infectious mononucleosis, which is a distinct entity from the chronic fatigue syndrome. The term ‘systemic EBV-positive T-cell LPD’, as adopted by the WHO classification, is the preferred pathologic designation over CAEBV for those cases that are clearly clonal, as they are generally associated with an aggressive clinical course and require aggressive treatment. These cases may, however, arise in the background of CAEBV. HV, which may be clonal, has a chronic and protracted clinical course and may regress spontaneously in adult life. As it has distinctive clinical and pathological features, the term HV is preferred over CAEBV for patients with these features. Criteria for the distinction of HV and HV-like lymphoma remain to be defined. Severe mosquito bite allergy appears to be a related syndrome, but is usually of NK cell origin.
Participants proposed the establishment of an international consortium to collect further clinical outcome data on CAEBV and related disorders. In addition, a proposed biobank will allow studies of cellular immunity, gene expression profiling, and sequencing of candidate genes. It is hoped that these studies will identify new pathways involved in the pathogenesis of these diseases and lead to multicenter international clinical trials to evaluate novel therapies for these diseases.
Intramural Research Programs of the Center for Cancer Research: National Cancer Institute; National Institute of Allergy and Infectious Diseases; Office of Rare Diseases at the National Institutes of Health.
We thank the participants of the meeting: Jeffrey I. Cohen, National Institute of Allergy and Infectious Diseases, Bethesda, MD; Elaine S. Jaffe, National Cancer Institute, Bethesda, MD (organizers); Carlos Barrionuevo, Instituto Nacional de Enfermedades Neoplasticas, Lima, Peru; Kishor Bhatia, National Cancer Institute, Bethesda, MD; John K. C. Chan, Queen Elizabeth Hospital, Kowloon, Hong Kong; Janet Dale, National Institute of Allergy and Infectious Diseases, Bethesda, MD; Kieron Dunleavy, National Cancer Institute, Bethesda, MD; Nancy Lee Harris, Massachusetts General Hospital and Harvard Medical School, Boston, MA; Helen Heslop, Baylor College of Medicine, Houston, TX; James Jones, Centers for Disease Control and Prevention, Atlanta, GA; Shannon Kenney, University of Wisconsin, Madison, WI; Rajiv Khanna, Queensland Institute of Medical Research, Brisbane, Australia; Elliot Kieff, Harvard University, Boston, MA; Hiroshi Kimura, Nagoya University, Nagoya, Japan; Y.-H. Ko, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea; Richard Little, National Cancer Institute, Bethesda, MD; Shigeo Nakamura, Nagoya University, Nagoya, Japan; Koichi Ohshima, Kurume University, Kurume, Japan; Miguel A. Piris, Centro Nacional de Investigaciones Oncologica, Madrid, Spain; Stefania Pittaluga, National Cancer Institute, Bethesda, MD; Leticia Quintanilla-Martinez, University of Tubingen, Tubingen, Germany; Koneti Rao, National Institute of Allergy and Infectious Diseases, Bethesda, MD; Alan Rickinson, University of Birmingham, Birmingham, England; Cliona Rooney, Baylor College of Medicine, Houston, TX; John Sixbey, Louisiana State University, Shreveport, Louisiana; John Sullivan, University of Massachusetts, Worcester, MA; Steven Swerdlow, University of Pittsburgh, Pittsburgh, PA; Giovanna Tosato, National Cancer Institute, Bethesda, MD; Larry Weiss, City of Hope National Medical Center, Duarte, CA; Denise Whitby, National Cancer Institute; Wyndham Wilson, National Cancer Institute, Bethesda, MD; Zhou Xiaoge Beijing, Friendship Hospital, Beijing, China; Robert Yarchoan, National Cancer Institute, Bethesda, MD; Tadashi Yoshinino, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Science, Okayama, Japan.