EBV is a DNA virus of the herpesvirus family with a tropism for B-lymphocytes. The life cycle consists of two forms—the latent and lytic phases. The latent phase results in activation, proliferation, and somatic hypermutation of infected B-cells. The phase is characterized by viral expression of proteins EBER and LMP-1. LMP-1 binds to the CD40 receptor on B-cells and prevents cellular apoptosis. These immortalized B-cells establish a viral cache. In the lytic phase, the virus replicates, destroys host cells, and infects other B-cells 1. Remission is achieved by the control of latent-infected B-cells, primarily by the response of cytotoxic T-cells. Compromise of T-cell mediated immunity results in the loss of the ability to control dormant, immortalized, infected B-cells, causing a neoplastic proliferation in PTLD. This is analogous to pathogenesis of primary CNS lymphoma in AIDS.1
In contrast to PTLD, EBV encephalitis and myelitis are manifestations of the lytic phase of the viral cycle. It is hypothesized that the neuronal damage is due to the infiltrative process of infected B-cells and the reactive immunologic cascade rather than a primary infection of neural cells1.
Epidemiology and risk factors
PTLD was initially described in 1970 and is defined by the World Health Organization as lymphoid proliferation or lymphoma occurring in patients who have undergone organ transplantation and are maintained on immunosuppressive regimens3. PTLD is the second most common malignancy in the post-transplant population, with approximately 15% of cases involving CNS.3,4. Risk factors include type of organ transplanted, greater immunosuppression, EBV status of host and donor, and younger host age. Intestine and lung transplantations have the highest rate of PTLD (20%), whereas kidney and liver transplants have a lower rates (1–3%)3,5−8. Seronegative hosts and EBV positive donors have increased risk likely due to the introduction of an EBV-infected graft into a naïve host 3.
CNS involvement in PTLD is commonly accompanied by involvement of other organ systems; however, PCNS-PTLD may occur without evidence of systemic involvement. Incidence is highest at six months to 1-year post-transplant, but PCNS-PTLD may develop several years after the transplant (median 4.4 years)3. Presentations include increased intracranial pressure, headache, seizure, and focal neurologic deficits 3–6. Intraocular spread may occur as well, and a slit-lamp exam is recommended.
Brain and spinal cord MRI with contrast is the imaging modality of choice. Radiographically, lesions are commonly multifocal with preferential involvement of periventricular structures with or without meningeal enhancement or hemorrhage. Lesions are typically contrast enhancing with variable patterns (homogenous, ring, or heterogeneous)3,9.
The CSF profile is non-specific with minimal pleocytosis and increased protein. Cytology may show malignant cells, and flow cytometry can sometimes detect a monoclonal population10. CSF testing alone is usually insufficient for diagnosis; tissue biopsy is needed in most cases.
Pathologic diagnosis requires the detection of latent viral proteins EBER and LMP-1 utilizing in-situ hybridization or immunohistochemistry. As in our case, it may be diagnostically challenging to differentiate EBV encephalitis from PCNS-PTLD. The pathogenies of CNS-PTLD is not binary and the disease itself exists as a continuum from early lesions which resemble reactive lymphoplasmacytic proliferation to the polymorphic subtype with mixed lymphoid and plasma cells to monomorphic PTLD with malignant lymphoid cells of single clonality8,11. Furthermore, it is recognized that PTLD lesions demonstrate a mixed molecular pattern of latent and lytic EBV gene expression12. Thus, early CNS-PTLD presents a true diagnostic dilemma since these lesions have a histologic appearance that is difficult to distinguish from encephalitis and detection of latent EBV proteins may be scant as in our case.
Treatment and Prognosis
The goal of therapy requires balancing cure of the malignancy and preservation of graft function. It is typically recommended that the initial step be either reduction or complete withdrawal of immunosuppression3,8,10. Antiviral therapies that target thymidine kinase such as ganciclovir are ineffective since the enzymatic expression does not occur in the latent phase of the viral cycle. Immune reconstitution is typically insufficient and concurrent anti-neoplastic interventions are required. Rituximab is a preferred first-line agent given tolerability, limited toxicity, and high response rate. Other therapies include whole-brain radiation, methotrexate, and autologous or allogenic EBV-specific cytotoxic T-lymphocyte infusions3,4, 6–8,10. With treatment, median survival ranges from 26–47 months3,4.
Non-PTLD Manifestation of EBV in the CNS
Presentation of EBV encephalitis, myelitis, and encephalomyelitis bears semblance to other viral infections of the CNS. Infection may be heralded by prodromal systemic manifestations of infectious mononucleosis or neurologic manifestation may be the initial symptomatology1. A diverse array of neurologic deficits have been described with EBV encephalitis2. Moreover, EBV encephalitis may trigger para-infectious acute disseminated encephalomyelitis or, as in our case, vasculitis resulting in cerebral infarcts1,2 .
CSF studies demonstrate a variable degree of lymphocytic pleocytosis with increased protein and normal glucose. While PCR detection of the virus is utilized for diagnosis of EBV encephalitis and myelitis a positive EBV PCR may reflect incidental reactivation13. Conversely, there may be false negative CSF EBV PCR14. Corresponding viral serologies to the viral capsid (IgG and IgM), EBNA, and EA may be helpful in supporting the presence of an acute or subacute EBV infection.
Brain MRI with contrast in EBV encephalitis may range from completely normal to multifocal areas of T2 hyperintensity in the parenchyma, diffusion restriction, or contrast-enhancement of the meninges and cranial nerves1,2,15,16. In organ transplant recipients, EBV encephalitis has been reported to present as a tumor-like lesion15.
For EBV myelitis, a complete spine MRI with contrast is the imaging modality of choice. There is no known pathognomonic finding, although there are reports of longitudinally extensive lesions with or without contrast enhancement14.
The histologic findings in EBV encephalitis consist of predominantly perivascular lymphocytic infiltrates of microglia, macrophages, T-cells, and infected B-cells. The infected B-cells may show some clonality, lymphoblastoid-appearance, and increased mitotic figures, making the distinction between malignancy and infection difficult1,17.
Treatment and Prognosis
The most crucial step is reduction of immunosuppression to restore the T-cell mediated immune response. There is limited evidence for the use of antiviral agents; however, their use should be strongly considered, especially in severe presentations. Intravenous ganciclovir or parenteral valganciclovir are preferred given their known activity against EBV replication in-vitro 2,18,19. In organ transplant patients, valganciclovir and immunoglobulin therapy have been reported to be effective in EBV encephalitis, but valganciclovir may cause myelosuppression with prolonged use19. The optimal duration of treatment is unknown and is usually made based upon clinical response. Steroids are not routinely administered, and the decision should be made at the discretion of the physician.
In immunocompetent patients with EBV encephalitis, the mortality rate is approximately 10%, with most patients having good outcomes without long-term deficits16,20. However, in post-transplant populations outcomes are unknown.