Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
The initial clinical descriptions of primary Epstein-Barr virus (EBV) infections are credited to Filatov and Pfeiffer at the end of the 19th century. Pfeiffer coined the term glandular fever , which described an illness consisting of fever, malaise, sore throat, and lymphadenopathy. In 1920, Sprunt and Evans introduced the term infectious mononucleosis (IM) to describe a series of patients with fatigue, fever, lymphadenopathy, and prominent mononuclear lymphocytosis ( Fig. 55.1 ). Serologic diagnosis of IM became available in the 1930s with the heterophile agglutination test developed by Paul and Bunnel and later modified by Davidson.
The identification of EBV as the causative agent of IM was impeded for many years by the inability to transmit the disease to animals or to grow the virus ex vivo. In 1958, Denis Burkitt described a lymphoma in Ugandan children, and investigators suspected an infectious etiology because the geographic distribution of disease coincided with the malaria belt of equatorial Africa. In 1964 Epstein, Achong, and Barr described herpesvirus-like particles in tumor biopsies from Ugandan children with Burkitt lymphoma (BL). Werner and Gertrude Henle developed an indirect immunofluorescent antibody assay to this new virus, now called Epstein-Barr virus , and showed that African children with BL, as well as 90% of American adults, had antibodies against EBV. In 1965 the Henles documented seroconversion to EBV of an individual who presented with clinical symptoms of IM. This initial observation was corroborated by larger studies confirming the association of EBV and IM.
EBV has subsequently been linked to a heterogeneous group of diseases. EBV was the first human virus implicated in oncogenesis, and the biology of the virus has been studied extensively on a cellular and molecular level. Because primary EBV infection is a self-limiting disease in almost all immune-competent individuals, clinical strategies have focused on the diagnosis and treatment of rare, potentially fatal EBV-associated diseases. Over the past two decades, successful immunotherapeutic approaches have been developed for EBV-associated malignancies, using either monoclonal antibodies or the adoptive transfer of EBV-specific T cells.
EBV belongs to the family of herpesviruses, which has almost 100 members. Membership is based on the architecture of the virion that is 120 to 300 nm in size and contains (1) a core of linear, double-stranded DNA, (2) an icosadeltahedral capsid with 162 capsomers, (3) an amorphous material between the capsid and envelope designated tegument, and (4) an envelope containing viral glycoproteins. Besides EBV, designated human herpesvirus 4 (HHV-4), seven other herpesviruses have been isolated from humans: herpes simplex viruses 1 and 2, cytomegalovirus (CMV), varicella-zoster virus, HHV-6, HHV-7, and the Kaposi sarcoma–associated herpesvirus (KSHV; HHV-8). Herpesviruses are further divided into subfamilies to reflect evolutionary relatedness and similar biologic properties. EBV and KSHV belong to the human gamma herpesvirus subgroup and have a limited tissue tropism to B and T lymphocytes and certain types of epithelial cells. Several variants of EBV have been identified by genomic polymorphisms. Initially, two EBV types were distinguished by sequence changes in EBV nuclear antigens 2 and 3 (EBNA2 and EBNA3). However, using polymorphisms in the latent membrane protein 1 (LMP1), further subtypes have been described. EBV strains vary by geography and have not been linked to a particular EBV-associated disease.
Primary EBV infection usually occurs through the oropharynx, where mucosal epithelial cells and/or B cells become infected ( Fig. 55.2 ). Infection of B cells by EBV is initiated by binding of the dominant viral glycoprotein gp350/220 to CD21, the C3d complement receptor; subsequent cell entry is mediated by a complex of three viral glycoproteins, gH, gL, and gp42. Gp42 binds to human leukocyte antigen (HLA) class II, which functions as a coreceptor, and gH is most likely involved in virus-cell fusion. The entry of EBV into epithelial cells may occur through multiple mechanisms because the majority of epithelial cells are CD21 negative. After viral entry, the capsid is dissolved and the EBV genome is transported into the nucleus, where it circularizes. Infection of epithelial cells results in lytic or abortive infection. In contrast, B-cell infection results predominantly in latency; however, lytic infection also occurs, resulting in the release of infectious virus into the saliva and other secretions. During primary infection, EBV establishes lifelong latency in B cells, and it is estimated that 1 to 50 cells per 1 × 10 6 B cells in the peripheral circulation are infected with EBV. The number of latently infected B cells within a person remains stable over years; however, intermittent reactivation of EBV in B cells into the lytic cycle at mucosal sites is probably responsible for the observed shedding of infectious virus into the saliva of asymptomatic carriers (see Fig. 55.2 ).
Although EBV can infect any B cell and express the full spectrum of latency proteins, typically only infection of naïve B cells results in persistent infection (see Fig. 55.2 ). EBV infection pushes the naïve B cell into a memory state independent of an antigen-dependent germinal center (GC) reaction by upregulation of cytosine deaminase, which induces both class switching and somatic hypermutation. The former also requires the expression of EBV-encoded LMP1, a constitutively activated CD40 molecule, or CD40 ligation, most likely provided by GC T helper (Th)3 cells that can provide T-cell help for B-cell differentiation by provision of CD40 ligand, interleukin (IL)-4, and IL-10 while preventing antigen-dependent effector T cell–mediated B-cell elimination by expression of transforming growth factor (TGF)-β. This reaction occurs within the lymph node and also involves downregulation of latency proteins and expression of latency type II. On exit from the lymph node, expression of latency proteins is completely inhibited. In this way, infected B cells can evade immune elimination. By contrast, primarily infected memory B cells enter and remain in latency type III and are rapidly eliminated by effector T cells and therefore do not contribute to virus persistence.
During latent infection, EBV persists episomally in resting memory B cells. Initially, it was thought that EBNA1 and LMP2 were expressed in memory B cells; however, more recent studies indicate that the majority of infected cells do not express viral proteins and of the almost 100 viral proteins, only EBNA1 is expressed during memory B-cell division (see Fig. 55.2 ). This extremely limited expression of viral proteins allows EBV to persist long term despite a robust cellular EBV-specific immune response.
Three other distinct types of EBV latency have been characterized in a heterogeneous group of malignancies ( Fig. 55.3 ). Latency type III, which can be readily produced by infecting B cells in vitro with EBV, is expressed in lymphoblastoid cell lines (LCLs). These cells express the entire array of nine EBV latency proteins: EBNA1, -2, -3A, -3B, and -3C, EBNA leader protein (LP), and the two viral membrane proteins LMP1 and LMP2. This pattern of EBV gene expression characterizes the EBV-associated lymphoproliferative disorders (EBV-LPDs) that occur in individuals severely immunocompromised by solid organ or hematopoietic stem cell transplantation (SOT, HSCT), congenital immunodeficiency, or human immunodeficiency virus (HIV) infection. Latency type II is the hallmark of EBV-positive Hodgkin lymphoma (HL), various EBV-associated non-Hodgkin lymphomas (NHLs), as well as nasopharyngeal carcinoma (NPC), which express EBNA1, LMP1, and LMP2. In addition, BARF1 is expressed in subsets of latency type II–associated malignancies. In latency type I, classically found in EBV-positive BL, only EBNA1 is expressed. However, variants in which all EBNAs are expressed in the absence of LMP1 have also been described. Latency type I or latency type II is found in EBV-associated gastric adenocarcinoma. Although grouping EBV-associated malignancies according to their dominant gene expression profile provided a useful framework for understanding EBV-driven oncogenesis, more recent studies using comprehensive gene expression array profiling have demonstrated expression of lytic cycle genes in BL or expression of lytic cycle and latency III genes in NPC.
The EBV proteins expressed during type III latency are involved in the transformation and growth of EBV-infected B cells. EBNA1 binds to the origin of replication of the latent viral genome and is responsible for the maintenance of the EBV episome in host B cells. EBNA2 upregulates the expression of the viral proteins LMP1 and LMP2 and cellular proteins that contribute to transformation. EBNA3A and EBNA3C are essential for EBV-induced B-cell transformation, and although EBNA3B is not essential for transformation, it is highly conserved and therefore must provide a survival function in vivo. EBNA-LP cooperates with EBNA2 in the induction of viral and cellular genes. LMP1, a viral oncogene, behaves like a constitutively activated CD40 molecule and is essential for EBV-mediated B-cell transformation. LMP2 mimics an activated B-cell receptor (BCR), allowing for long-term B-cell survival in the absence of antigen. In addition, it prevents the reactivation of EBV into the lytic phase of infection.
Besides EBV proteins, small nonpolyadenylated viral RNAs termed EBERs 1 and 2 and the BamHI-A rightward transcripts microRNAs (BART-miRs) are expressed in all forms of latency. In addition, the expression of at least 17 distinct EBV-derived microRNAs has been reported. The EBERs are the most abundant viral RNAs in latently infected cells. They enhance the oncogenic phenotype of EBV-transformed cells but are nonessential for EBV-mediated transformation. The expression pattern of the microRNAs depends on the latency type, and it is therefore likely that they play an important role during the life cycle of the virus.
Healthy individuals mount vigorous humoral and cellular immune responses to primary EBV infection. Although antibodies to the viral membrane proteins neutralize virus infectivity, the cellular immune response is essential for controlling virus-infected cells during both lytic and latent phases.
Heterophile antibodies, originally described by Paul and Bunnell, are present in 90% to 95% of EBV infections at some point during the illness. However, in infants and children younger than the age of 4 years with primary EBV infection, heterophile antibody responses may not be detected. Heterophile antibodies are immunoglobulin M (IgM) antibodies, which agglutinate erythrocytes from different species including bovine, camel, horse, goat, and sheep. EBV-induced heterophile antibodies have no reactivity against guinea pig kidney cells in contrast to naturally occurring antibodies (Forssman antibodies) or antibodies present in patients with serum sickness and other conditions.
In addition to heterophile antibodies, cold agglutinins directed preferentially against the anti-I antigen on red cell membranes are frequently detected in the sera of patients with IM; however, hemolytic anemia is rare. Other antibodies (including anti-I, anti-N, Donath-Landsteiner antibodies, platelet antibodies, and anti–smooth muscle antibodies) have been described in association with EBV infection
EBV-specific antibody responses are detected with immunofluorescence assays developed in the first decades of EBV research. EBV antibodies are directed against (1) EBNA, (2) early antigen (EA), (3) the membrane antigen expressed on the surface of cells late in the lytic cycle, and (4) the viral capsid antigen (VCA) expressed within cells late in the lytic cycle. Each antigen is a composite of several distinct viral proteins, and attempts have been made to replace the aforementioned assays with tests using specific viral proteins; however, no single test has attracted widespread use.
VCA-IgM and IgG antibodies are usually present at the onset of clinical symptoms because of the prolonged viral incubation period ( Table 55.1 ). VCA-IgM antibodies are a good marker for an acute infection because they rapidly disappear within 4 to 8 weeks. VCA-IgG antibodies persist for life and are commonly used to document prior EBV infection. IgG antibodies against EA are present at the onset of the clinical illness in approximately 70% of patients. EA antibodies are divided into methanol-sensitive (anti-D) and methanol-resistant (anti-R) antibodies, and the majority of EA antibodies detected are anti-D antibodies. The presence of anti-D antibodies is consistent with recent infection, because titers disappear after recovery. IgG antibodies to EBNA appear late in the course of almost all cases of EBV infection and persist throughout life; their presence early in a suspected case of primary EBV infection excludes the diagnosis. Aberrations in this pattern of serum reactivity are observed in many EBV-associated diseases and will be discussed under the specific disease sections. For example, the absence of EBNA antibodies despite previous EBV infection is one of the serologic markers suggestive for chronic active EBV (CAEBV) infection.
Antibody Specificity | Positive in IM (%) | Time of Appearance in IM | Persistence | Comments |
---|---|---|---|---|
Viral Capsid Antigen (VCA) | ||||
VCA-IgM | 100 | At clinical presentation | 4–8 weeks | Highly sensitive and specific; of major diagnostic utility |
VCA-IgG | 100 | At clinical presentation | Lifelong | Useful for documentation of past EBV infection |
Early Antigen (EA) | ||||
Anti-D | 70 | Peaks 3–4 weeks after onset | 3–6 months | Correlates with disease severity; seen in patients with NPC |
Anti-R | Low | 2 weeks to several months after onset | 2 months to >3 years | Occasionally seen with unusually severe cases; seen in patients with African Burkitt lymphoma |
EBNA | 100 | 3–4 weeks after onset | Lifelong | Presence excludes primary EBV infection |
In normal individuals, primary EBV infection often results in a massive expansion of activated, antigen-specific T cells. Using tetramer technology to enumerate antigen-specific T cells, it has been documented that the CD8 + T-cell response may be dominated by T cells specific for a limited number of epitopes, as seen with T-cell responses against other herpesviruses. T cells specific for epitopes derived from immediate early and several early EBV proteins of the lytic cycle are dominant during the acute phase of IM, and long-term persistence of EBV-specific, CD8 + T cells has been documented after primary EBV infection. As many as 5.5% of the circulating CD8 + T cells in a healthy virus carrier may be positive for a single EBV epitope, illustrating how persistent EBV infection can influence the composition of the host's T-cell pool. Besides EBV-specific CD8 + T cells, EBV-specific CD4 + T cells play an important role in the control of EBV infections, and EBNA1-specific CD4 + T cells have been implicated in the control of newly infected B cells. As for CD8 + T-cell responses, there is a marked hierarchy of immunodominance, with the majority of CD4 + T cells being specific for EBNA1 and to less extent EBNA3C.
At present there is only limited experience with human EBV vaccines. Vaccine studies have been conducted in healthy donors in the prophylactic setting using recombinant vaccinia virus encoding the major viral glycoprotein gp350 (1), recombinant gp350 protein (2), or an EBNA3A peptide. While vaccination with gp350 induced neutralizing antibodies and in one study prevented the clinical picture of IM, these vaccines did not significantly reduce the incidence of EBV infection. Also, no reduction in the incidence of EBV infection was observed with the EBNA3A peptide vaccine. In addition, a recombinant gp350 protein was evaluated in patients with chronic kidney disease prior to kidney transplantation. Although the vaccine-induced transient neutralizing antibodies, it did not reduce the incidence of EBV-LPD post transplant. Newer vaccines consisting of gp350-ferritin complexes that induce significantly higher titers of neutralizing antibodies in preclinical models have been developed but require testing in humans.
Vaccine strategies for EBV-associated malignancies should seek to elicit or boost the EBV-specific cellular immune response against EBV latency. Individuals likely to benefit from this approach are EBV-seronegative patients scheduled to undergo SOT or patients who have an EBV-associated malignancy with a low tumor burden or are in remission. A variety of vaccine studies have been conducted in patients with EBV + NPC. In two studies patients with advanced disease were vaccinated with either dendritic cells (DCs) loaded with peptides derived from LMP2 or DC transduced with an adenoviral vector encoding full-length LMP2 and an inactive form of LMP1. Administration of DC vaccines was safe, and a transient increase in the frequency of LMP2-specific T cells was observed on the DC/peptide vaccine trial. However, the clinical benefit in both studies was limited. Thus future studies will need to explore DC vaccines with greater potency administered to subjects with less tumor burden. In another phase I clinical study, after frontline therapy, patients with NPC were vaccinated with a modified vaccinia virus Ankara encoding the c-terminal portion of EBNA1 and LMP2 (MVA-EL). Induction of CD4 + and CD8 + antigen-specific T-cell responses was observed in the majority of patients after vaccination, and a phase II clinical study is in progress. Lastly, a recombinant adenovirus vaccine encoding LMP2 was evaluated in patients with NPC. Although vaccine administration was safe, no data are available if the vaccine boosted LMP2-specific T-cell responses. Ultimately, although development of an effective EBV vaccine has remained an elusive achievement, numerous studies have been undertaken in pursuit of this goal.
Although the aggregate number of patients around the world with EBV-associated lymphomas and lymphoproliferative disorders (LPDs) is modest, when EBV-associated gastric carcinoma and NPC are considered, global estimates of EBV-associated malignancies approach 200,000 cases per year. Acknowledging the burden of these malignant complications from EBV, in addition to the toll of IM in adolescents and young adults infected by the virus, a case for vaccine development gains traction. As precedent for effective vaccination against other oncogenic viruses such as human papillomavirus and hepatitis B has been established, EBV should be considered a target for wide-scale vaccine development.
EBV infections occur worldwide, and, in most populations, 90% to 95% of adults have antibodies against EBV. Depending on geographic and socioeconomic factors, there is a wide variation in the age of primary EBV infections. Early, asymptomatic primary EBV infection occurs more often in individuals from lower socioeconomic groups, particularly those living in low- and middle-income countries. In higher socioeconomic groups in high-income countries, the age of primary infection is more often delayed until the second decade of life and clinically apparent IM is more prevalent.
Humans are the only source of EBV, which is present in the saliva of patients with IM. A majority of EBV-positive adults shed virus into their saliva, and this percentage is increased in immunocompromised patients such as SOT recipients. EBV is viable outside the body for 2 weeks at 4°C but is susceptible to drying; the virus has not been recovered from environmental sources, suggesting that close contact is needed for viral spread. The incubation period of EBV is estimated to be 30 to 50 days.
Primary EBV infection in infants and young children is either asymptomatic or accompanied by mild, nonspecific symptoms and signs such as fever, upper respiratory tract infection, pharyngitis with or without tonsillitis, and cervical lymphadenopathy. In contrast, approximately 30% to 50% of adolescents and young adults present with the clinical picture of IM. Frequently, a prodrome consisting of fatigue, malaise, and low-grade fever is present for 1 to 2 weeks. Prominent pharyngitis with exudative tonsillitis is often the cardinal sign of IM; other signs and symptoms are listed in Table 55.2 . The adenopathy in IM most commonly affects the posterior cervical lymph nodes, although diffuse adenopathy can occur. The enlarged lymph nodes are not fixed, may be tender to palpation, and lack overlying skin erythema. Hepatomegaly is uncommon; however, splenomegaly develops in 15% to 65% of patients and is more prominent in the second to fourth week of the illness. Skin manifestations include a faint, morbilliform rash reminiscent of rubella and less commonly erythema multiforme and erythema nodosum. Most patients with primary EBV infection have symptoms for 2 to 4 weeks and recover without significant complications or sequelae.
Manifestation | Percentage (Range) |
---|---|
Symptoms | |
Sore throat | 82 (70––88) |
Malaise | 57 (43–76) |
Headache | 51 (37–55) |
Anorexia | 21 (10–27) |
Myalgias | 20 (12–22) |
Chills | 16 (9–18) |
Abdominal discomfort | 9 (2–14) |
Signs | |
Lymphadenopathy | 94 (93–100) |
Pharyngitis | 84 (69–91) |
Fever | 76 (63–100) |
Splenomegaly | 52 (50–63) |
Hepatomegaly | 12 (6–14) |
Palatal enanthem | 11 (5–13) |
Rash | 10 (0–15) |
Jaundice | 9 (4–10) |
Although the majority of patients with IM fully recover without severe complications, any organ system can be affected in this clinical syndrome of primary EBV infection.
Patients with IM may present with a wide range of hematologic findings besides the atypical lymphocytosis ( Fig. 55.4 ). These include anemia, neutropenia, thrombocytopenia, and rare cases of aplastic anemia.
Autoimmune hemolytic anemia occurs in approximately 3% of patients with IM. It presents in the first 2 weeks of the illness, and the majority of patients recover within 1 to 2 months. Patients usually have a positive direct Coombs test. Most common anti-I antibodies are present; however, anti-I, anti-N, and Donath-Landsteiner antibodies have also been reported. In addition to hemolysis, IM-associated anemia can be caused by transient erythroblastopenia.
Mild, self-limiting neutropenia is a common finding during the first 4 weeks of the disease. However, severe neutropenia associated with fatal bacterial infections has been reported.
Mild thrombocytopenia (50,000 to 150,000/mm 3 ) is a common finding in patients with IM. It usually occurs within the first 2 weeks of presentation and resolves within 2 months. Severe thrombocytopenia with overt bleeding is rare; however, death from intracranial hemorrhage has been described. The etiology of the thrombocytopenia is not completely understood, and a variety of explanations have been suggested. Because bone marrow examination shows normal or increased numbers of megakaryocytes, peripheral platelet consumption is most likely due to the presence of antiplatelet antibodies or platelet pooling and destruction within an enlarged spleen.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here