Progressive Multifocal Leukoencephalopathy

Journal Articles

JCV virology

Erickson KD, Bouchet-Marquis C, Heiser K, et al. Virion assembly factories in the nucleus of polyomavirus-infected cells. PLoS Pathog. 2012;8(4):e1002630.

Ferenczy MW, Johnson KR, Marshall LJ, et al. Differentiation of human fetal multipotential neural progenitor cells to astrocytes reveals susceptibility factors for JC virus. J Virol. 2013;87(11):6221-31.

Maginnis MS, Ströh LJ, Gee GV, et al. Progressive multifocal leukoencephalopathy-associated mutations in the JC polyomavirus capsid disrupt lactoseries tetrasaccharide c binding. MBio. 2013;4(3):e00247-13.

Nelson CD, Derdowski A, Maginnis MS, et al. The VP1 subunit of JC polyomavirus recapitulates early events in viral trafficking and is a novel tool to study polyomavirus entry. Virology. 2012;428(1):30-40.

Neu U, Maginnis MS, Palma AS, et al. Structure-function analysis of the human JC polyomavirus establishes the LSTc pentasaccharide as a functional receptor motif. Cell Host Microbe. 2010;8(4):309-19.

Saribas AS, Arachea BT, White MK, et al. Human polyomavirus JC small regulatory agnoprotein forms highly stable dimers and oligomers: implications for their roles in agnoprotein function. Virology. 2011;420(1):51-65.

Saribas AS, White MK, Safak M. JC virus agnoprotein enhances large T antigen binding to the origin of viral DNA replication: evidence for its involvement in viral DNA replication. Virology. 2012;433(1):12-26.

Shishido-Hara Y, Ichinose S, Uchihara T. JC virus intranuclear inclusions associated with PML-NBs: analysis by electron microscopy and structured illumination microscopy. Am J Pathol. 2012;180(3):1095-106.

Pathogenesis

Aly L, Yousef S, Schippling S, et al. Central role of JC virus-specific CD4+ lymphocytes in progressive multi-focal leucoencephalopathy-immune reconstitution inflammatory syndrome. Brain. 2011;134(Pt 9):2687-702.

Dalianis T, Hirsch HH. Human polyomaviruses in disease and cancer. Virology. 2013;437(2):63-72.

Gheuens S, Bord E, Kesari S, et al. Role of CD4+ and CD8+ T-cell responses against JC virus in the outcome of patients with progressive multifocal leukoencephalopathy (PML) and PML with immune reconstitution inflammatory syndrome. J Virol. 2011;85(14):7256-63.

Gorelik L, Reid C, Testa M, et al. Progressive multifocal leuoencephalopathy (PML) development is associated with mutations in JC virus capsid protein VP1 that change its receptor specificity. J Infect Dis. 2011;204(1):103-14.

Pastrana DV, Ray U, Magaldi TG, et al. BK polyomavirus genotypes represent distinct serotypes with distinct entry tropism. J Virol. 2013;87(18):10105-13.

Sunyaev SR, Lugovskoy A, Simon K, Gorelik L. Adaptive mutations in the JC virus protein capsid are associated with progressive multifocal leukoencephalopathy (PML). PLoS Genet. 2009;5(2):e1000368.

Tan CS, Broge TA Jr, Seung E, et al. Detection of JC virus-specific immune responses in a novel humanized mouse model. PLoS One. 2013;8(5):e64313.

Verschoor EJ, Groenewoud MJ, Fagrouch Z, et al. Molecular characterization of the first polyomavirus from a New World primate: squirrel monkey polyomavirus. J Gen Virol. 2008;89(Pt 1):130-7.

Wilson JJ, Pack CD, Lin E, et al. CD8 T cells recruited early in mouse polyomavirus infection undergo exhaustion. J Immunol. 2012;188(9):4340-8.

Risk stratification

Bloomgren G, Richman S, Hotermans C, et al. Risk of natalizumab-associated progressive multifocal leukoencephalopathy. N Engl J Med. 2012;366(20):1870-80.

Caraux A, Klein B, Paiva B, et al. Circulating human B and plasma cells. Age-associated changes in counts and detailed characterization of circulating normal CD138- and CD138+ plasma cells. Haematologica. 2010;95(6):1016-20.

Jelcic I, Aly L, Binder TM, et al. T cell epitope mapping of JC polyoma virus-encoded proteome reveals reduced T cell responses in HLA-DRB1*04:01+ donors. J Virol. 2013;87(6):3393-408.

Reid CE, Li H, Sur G, et al. Sequencing and analysis of JC virus DNA from natalizumab-treated PML patients. J Infect Dis. 2011;204(2):237-44.

Schwab N, Schneider-Hohendorf T, Posevitz V, et al. L-selectin is a possible biomarker for individual PML risk in natalizumab treated MS patients. Neurology. 2013. [Epub ahead of print]

Warnke C, Pawlita M, Dehmel T, et al. An assay to quantify species-specific anti-JC virus antibody levels in MS patients. Mult Scler. 2013;19(9):1137-44.

Diagnosis and management

Clifford DB, Nath A, Cinque P, et al. A study of mefloquine treatment for progressive multifocal leukoencephalopathy: results and exploration of predictors of PML outcomes. J Neurovirol. 2013. [Epub ahead of print]

Martin-Blondel G, Cuzin L, Delobel P, et al. Is maraviroc beneficial in paradoxical progressive multifocal leukoencephalopathy-immune reconstitution inflammatory syndrome management? AIDS. 2009;23(18):2545-6.

Tan K, Roda R, Ostrow L, et al. PML-IRIS in patients with HIV infection: clinical manifestations and treatment with steroids. Neurology. 2009;72(17):1458-64.

Yousry TA, Pelletier D, Cadavid D, et al. Magnetic resonance imaging pattern in natalizumab-associated progressive multifocal leukoencephalopathy. Ann Neurol. 2012;72(5):779-87.

Websites

The Body — The Complete HIV/AIDS Resource: PML Fact Sheet
TheBody.com aims to use the Web to lower barriers between patients and clinicians, demystify HIV/AIDS and its treatment, improve the quality of life for all people living with HIV/AIDS, and foster community through human connection.

Deferno Trust
The Deferno Trust aims to facilitate information sharing, provide support to PML patients and their families, and raise funds for research.

National Institutes of Health. National Institute of Neurological Disorders and Stroke. Progressive Multifocal Leukoencephalopathy Information Page. 2011.

PML Consortium
The consortium is a joint effort between Biogen Idec, Bristol-Myers Squibb, Pfizer, MedImmune, and Roche that is intended to enhance the understanding of PML and identify more effective methods to predict, prevent, and treat PML.

Medscape: PML Overview
Medscape from WebMD offers specialists, primary care physicians, and other health professionals integrated medical information and educational tools.

NORD (National Organization of Rare Diseases): PML Abstract

Sponsors

Presented by

Keynote Speakers:
Alfred W. Sandrock Jr., Biogen Idec
Declan R. Walsh, Deferno Trust

PML and JCV

Progressive multifocal leukoencephalopathy (PML) is a rare but serious brain demyelinating disease. Symptoms typically include hemiparesis, ataxia, cognitive or behavioral changes, and visual disturbances. PML is associated with unifocal or multifocal lesions in the brain, which can be detected using magnetic resonance imaging (MRI). The lesions most commonly appear in the subcortical white matter, cerebellar hemispheres, peduncles, or brain stem. PML development, while dependent on the presence of JC virus (JCV), is the result of a confluence of viral and host risk factors. Many aspects of JCV virology and PML pathogenesis remain unknown.

JCV is a type of human polyomavirus and is genetically similar to BK virus and Simian virus 40 (SV40). It is believed that 50%–60% of the human population is seropositive for JCV, indicating exposure to the virus. JCV infection typically leads to a chronic, asymptomatic infection. The primary site of infection appears to be the kidneys: approximately 25% of adults shed JCV in their urine. JCV is rarely found in blood. While there have been reports of JCV in the brain, bone marrow, tonsils, and lymphocytes in peripheral blood, viral tropism (that is, the type(s) of cells and tissues that support JCV growth) and lifecycle outside the kidneys are not understood. An important unanswered question in JCV/PML research is how JCV accesses the central nervous system and ultimately infects oligodendrocytes in the brain. Recent data suggest that viral mutations in the VP1 capsid protein and/or the noncoding control region rearrangement might play an important role in the change in JCV tropism.

The emergence of PML is also dependent on impairment of host immune function. HIV/AIDS is the most common context in which PML is observed. PML has also been associated with hematologic malignancies, systemic lupus erythematosus, and certain therapeutics that result in significant immune suppression or immune modulation.

Criteria used by physicians to diagnose PML usually include clinical symptoms consistent with PML, MRI features characteristic of PML, and detection of JCV DNA in the cerebrospinal fluid (CSF) or evidence of JCV presence in a brain biopsy.

PML prognosis varies depending on comorbidities. If left unmanaged, the mortality rate is 30%–50% within the first three months of diagnosis. There are no known interventions that can reliably prevent or adequately treat PML. It is believed that the best prognosis for PML is afforded by reconstitution of the immune system leading to viral clearance. However, return of immune function also frequently leads to an immune reconstitution inflammatory syndrome (IRIS), leading to severe brain inflammation and significant clinical worsening (IRIS can cause severe brain injury or be fatal). Until an effective therapy is identified the only option available to maximize patient outcomes is early diagnosis via clinical vigilance combined with restoration of immune function, while managing inflammatory syndrome.

Lessons learned from PML in the natalizumab era

The first keynote speaker, Alfred W. Sandrock Jr. of Biogen Idec, provided a historical overview of natalizumab development, its association with PML, and current efforts to better understand and minimize the risk of PML. Natalizumab is a humanized monoclonal antibody against the cell adhesion molecule α4-integrin. It is used in the treatment of multiple sclerosis (MS) and Crohn's disease. It was approved in 2004 by the U.S. Food and Drug Administration (FDA) for relapsing forms of MS, but was voluntarily withdrawn from the market in 2005 after two PML cases associated with natalizumab treatment were reported. A comprehensive safety evaluation revealed no new PML cases, and the drug was returned to the U.S. market in 2006 under a special prescription program.

As of March 2013, more than 115 000 patients had been treated with natalizumab and 359 patients had developed confirmed cases of PML. Biogen Idec and its collaborators have conducted extensive research on PML and PML risk minimization, including the development of a PML risk stratification algorithm. Sandrock presented key lessons learned from natalizumab and PML, including the importance of focusing on patients and the value of timely communications among pharmaceutical companies, health care providers, and patients. Future research should include efforts to refine the anti-JCV assay, to understand the pathophysiology of PML, to develop additional risk stratification tools, and to better manage PML.

A patient and family perspective on PML

The second keynote speaker, Declan R. Walsh of Deferno Trust, shared his wife Natalie Murphy's story. Murphy is a pharmacologist who was diagnosed with PML in 2009 after 2-year natalizumab treatment for MS. Walsh established the Deferno Trust in 2010 to facilitate information sharing, provide support to other PML patients and families, and raise funds for research. Walsh noted that patients and their families often have similar questions regarding methods to manage PML and to prevent IRIS. He said that a list of doctors who have experience treating or studying PML would be very valuable to patients and health care providers. Like Sandrock, Walsh also emphasized the importance of a collaborative approach that involves all parties and places the focus on patients. Improved information sharing would provide patients with valuable knowledge and reduce the number of misconceptions about PML and its treatment.

The risk of developing PML increases after the first two years of natalizumab treatment and again after the third year of treatment. (Image courtesy of Alfred W. Sandrock Jr.)

Speakers:
Walter J. Atwood, Brown University
Mahmut Safak, Temple University School of Medicine
Robert L. Garcea, University of Colorado–Boulder

Highlights

• A compound known as Retro-2 was shown to reduce viral spread in established PML infections in vitro and to decrease the number of infectious virions produced in the cultures.
• Agnoprotein is a multifunctional regulatory protein of polyomaviruses, including JC virus. A Leu/Ile/Phe-rich domain of agnoprotein is critical for its structure and function.
• Foci of replicating mouse polyomavirus DNA co-localize with the viral T antigen and cellular DNA repair proteins, suggesting that polyomavirus replication utilizes the cellular DNA damage repair pathways for efficient viral DNA replication.

Infectious entry of JCV into host cells

Walter J. Atwood from Brown University discussed the JCV life cycle, which begins with a multistep attachment to virus receptors on the host cell. The receptors facilitate entry into the host cell by clathrin-dependent endocytosis, and then the virus is transported to the endoplasmic reticulum (ER), where host-cell chaperones begin the process of viral uncoating. With its protein coating (capsid) removed, the viral genome, along with some components of the capsid, exits the ER and is targeted to the nucleus by unknown mechanisms.

Atwood presented data suggesting that glycan lactoseries tetrasaccharide c (LSTc) is the major attachment receptor for JCV and that compounds targeting the LSTc binding site on the virus can reduce infection at high concentrations. He further showed that serotonin receptors can facilitate viral entry but do not affect the initial binding of the virus to cells.

Another compound, Retro-2, reduces viral spread in established in vitro infections and decreases the number of infectious virions produced in these cultures. Retro-2 does not interfere with binding or uptake of virions from the plasma membrane but instead inhibits retrograde trafficking to the ER, a necessary step for a productive infection. Further study of Retro-2 may increase our understanding of how polyomaviruses target the ER and could lead to therapeutic treatment options for diseases caused by human polyomaviruses, such as PML.

Agnoprotein regulation of JCV

Agnoprotein is a multifunctional regulatory protein of polyomaviruses, including JCV. When this protein is not expressed, JCV is unable to sustain its productive life cycle. Mahmut Safak from Temple University School of Medicine showed that agnoprotein forms stable dimer/oligomers mediated by an amphipathic α-helix that spans amino acids (aa) 17–42. Deletion of the α-helix renders a replication-incompetent virus. Safak's lab characterized this region by a systematic deletion and substitution mutagenesis and demonstrated that a Leu/Ile/Phe-rich domain (spanning aa 28–39) within the α-helix is critical for the protein's structure and function, including dimer/oligomer formation, protein stability, and splicing of viral transcripts. These data suggest that this Leu/Ile/Phe-rich domain may represent a potential target for novel therapeutics for PML.

Virion assembly factories in polyoma-infected cells

Most DNA viruses replicate in the nucleus, where viral capsid proteins and viral genomes spatially intersect to assemble infectious progeny. Researchers in Robert L. Garcea's lab at the University of Colorado–Boulder observed filamentous structures in the nucleus of murine polyomavirus (MPyV)-infected mouse embryo fibroblasts via electron microscopy. Tomographic reconstructions suggest that these "bubble-like" structures travel the length of the filaments, and that virions are "shed" from their ends. The team hypothesize that the filaments represent "virus factories," where viral capsid proteins and viral genomes assemble to yield mature virions. These filamentous structures have also been seen in brain lesions from humans with PML. Garcea showed that foci of replicating MPyV DNA co-localize with the viral T antigen, as well as with a number of cellular DNA repair proteins. These observations suggest that polyomavirus replication utilizes the cellular DNA damage repair pathways for efficient viral DNA replication.

Short talk presentations

The session ended with a series of short talk presentations. Michael W. Ferenczy of the National Institute of Neurological Disorders and Stroke, NIH, discussed the effect cell-type differences have in determining the chromatin structure of JCV. The virus lytically infects human glial cells in PML patients, but can also nonproductively infect many cell types, including neurons. Ferenczy showed that JCV expresses less RNA in neurons than in astrocytes (a type of glial cell). This reduced expression is associated with repressive chromatin structure and repressive transcriptional factors.

JCV isolated from individuals with PML often contains mutations in the sialic acid-binding pocket of the VP1 capsid protein. Data presented by Melissa S. Maginnis from Brown University suggest that these mutations disrupt the binding of JCV to LSTc (the major receptor for JCV on cells and in tissues), and thus PML patient-associated viruses are, surprisingly, not infectious.

Yukiko Shishido-Hara of Kyorin University School of Medicine and others previously published data suggesting that promyelocytic leukemia nuclear bodies (PML-NBs)—dynamic subnuclear compartments that play a role in the transcription, replication, and repair of DNA and are implicated in tumor suppression—provide a scaffold for JCV replication. In this presentation, she presented data indicating that pathogenic JC virus progenies are efficiently reproduced at PML-NBs in the presence of agnoprotein and later disrupt PML-NB structures. These data advance our understanding of JCV replication.

Speakers:
Roland Martin, University of Zurich, Switzerland
Leonid Gorelik, Biogen Idec
Hans H. Hirsch, University of Basel, Switzerland
Igor J. Koralnik, Beth Israel Deaconess Medical Center; Harvard University

Highlights

• Increasing a protective CD4⁺ T-cell response may be particularly important for improving immune control of JCV in PML patients and may become an effective PML treatment approach.
• JCV-specific CD8⁺ cytotoxic T cells appear to play a crucial role in fighting JCV, and their presence has been associated with favorable clinical outcomes and improved survival of PML patients.
• PML-associated point mutations in the JCV capsid protein VP1 are located in or around the sialic acid binding site of the virus. These mutations may allow the virus to escape the recognition of neutralizing antibodies and/or may increase the likelihood that the virus will reach the central nervous system (CNS) from peripheral sites or be widely distributed throughout the CNS.
• Studies of BKV, a closely phylogenetically related polyomavirus to JCV, may shed light on JCV and PML.

Immunology of PML

Immune control of JCV involves all aspects of the adaptive immune system, including JCV-specific B cells and antibodies, CD4⁺ T cells, and CD8⁺ T cells. Antibodies are thought to be involved in neutralizing free JCV virions, and hence keeping the infection in the latent state and preventing the spread of free virus.

CD8⁺ JCV-specific cytotoxic T cells appear to be ideally suited for viral elimination in PML, while CD4⁺ T cells probably orchestrate and/or support the generation and maintenance of efficient B-cell/antibody responses as well as the expansion of CD8⁺ T cells. When characterizing T-cell responses in the brains of patients with PML and PML immune reconstitution inflammatory syndrome (PML-IRIS), Roland Martin and his colleagues at the University of Zurich found a predominance of CD4⁺ T cells that are specific for JCV VP1 epitopes and frequently express a Th1-2 phenotype (i.e., secrete large amounts of interferon-γ and interleukin-4). The most prominently expressed T-cell clones in brain tissue use T-cell receptors that are capable of interacting with two or more different human leukocyte antigen (HLA)/peptide complexes.

A better understanding of JCV immune control in both a healthy state and PML is essential to the development of effective treatment and prophylaxis. Martin hypothesized that increasing a protective CD4⁺ T-cell response is particularly important for improving overall JCV immune control in PML patients and may lead to effective PML treatment. This hypothesis is supported by the successful treatment of two PML patients who had comorbidities such as idiopathic CD4⁺ lymphopenia and secondary immunocompromise with a combination of recombinant human interleukin-7 and vaccination with a JCV VP1 protein and an adjuvant, which together increase T-cell immune responses.

Viral determinants of PML pathogenesis

Researchers are also investigating how the form of the JC virus and viral point mutations contribute to the development of PML. Within individual hosts, JCV persists in at least two forms: a latent nonpathogenic form and a virulent neurotropic form. The neurotropic form contains a rearranged noncoding control region (NCCR) and is typically found in the cerebrospinal fluid (CSF) and brain in PML patients. The nonpathogenic form is most frequently detected in urine, and its NCCR is not rearranged.

Point mutations in the VP1 capsid protein have also been associated with PML. These PML-specific mutations are located in or around the sialic acid binding site of the virus. Leonid Gorelik and colleagues from Biogen Idec investigated the role(s) mutations play in viral pathogenesis. They found that these mutations do not increase the infectivity of many susceptible cell types but the mutant virions can infect key cell types in the CNS (such as astrocytes). These mutations also reduce binding to sialylated receptors that are abundant in the human body, and thus may increase the chance that the virus will reach the CNS from peripheral sites by allowing the virus to escape the recognition of neutralizing antibodies. Alternatively, the mutations might be acquired within the CNS and enable the virus to distribute throughout the CNS without binding to, and hence being stopped by, sialylated receptors.

PML-specific VP1 mutations are located in or around the sialic acid binding site of the virus. (Image courtesy of Leonid Gorelik)

Virus–host interactions: lessons from the BK polyomavirus

BKV, a phylogenetically closely related polyomavirus to JCV, is associated with hemorrhagic cystitis, a disease of the lower urinary tract, in allogeneic hematopoietic stem cell transplant patients and with nephropathy (kidney disease) in kidney transplant patients. Hans H. Hirsch from the University of Basel reviewed similarities and differences between BK polyomavirus and JC polyomavirus. JCV and BKV share several characteristics, including high seroprevalence in the general population, persistence in the renourinary tract, and asymptomatic shedding of archetype NCCR virus in the urine of healthy hosts. Interestingly, like JCV, more-virulent BKV variants have been found to bear rearranged NCCRs, which are associated with higher viral loads and more advanced histopathology. Hirsch discussed the reactivation and replication kinetics of BKV, viral markers for progression, and the role of immunosuppression in BKV-associated diseases. Because of the similarities between BKV and JCV, studies on BKV–host interactions may shed light on JCV and PML.

T cell-mediated cellular immune response against JC virus in PML patients

Igor J. Koralnik from Beth Israel Deaconess Medical Center and Harvard University focused on T cell-mediated cellular response against JCV in PML patients. He first introduced a few commonly used techniques to measure the presence and function of JCV-specific T cells. These techniques include chromium release assay, tetramer staining, Elispot, and intracellular cytokine staining assays.

Despite their low numbers, JCV-specific CD8⁺ cytotoxic T cells play a crucial role in PML, and their presence has been associated with favorable clinical outcomes and improved survival. T cells, however, have also been implicated in the pathogenesis of the immune reconstitution inflammatory syndrome (IRIS), a frequent manifestation, upon starting antiretroviral medications, in PML patients who also have AIDS. IRIS also occurs in most MS patients with natalizumab-associated PML after the medication is discontinued.

Short talk presentations

Although a high percentage of adults have serum antibodies for JCV, these antibodies somehow fail to protect a small subset of patients from PML. JCV isolates from the CSF of PML patients often contain distinctive mutations in the viral major capsid protein VP1. Christopher B. Buck and his team at the National Cancer Institute, NIH, tested the possibility that these JCV mutations can evade recognition by neutralizing antibodies. They found that while most healthy individuals have serum antibodies that can neutralize a broad range of wild type and PML-derived mutants, a small percentage of tested sera fail to neutralize a subset of PML-variant strains. These individuals might be at increased risk for PML.

Mechanisms underlying PML pathogenesis are poorly understood, partly because there is no animal model to study PML. Elizabeth L. Frost from Aron Lukacher's lab at Pennsylvania State University College of Medicine presented her work on developing a mouse model of mouse polyomavirus (MPyV)-induced CNS disease to elucidate the role that viral major capsid protein (VP1) mutations and immunosuppression play in JCV–PML pathogenesis. Her study shows that a PML-associated VP1 mutation, when incorporated into mouse MPyV, does not alter MPyV replication or T-cell responses in the CNS, and MPyV-specific T cells infiltrating the brain develop a tissue resident memory phenotype. As next steps, Frost will investigate whether this PML-associated VP1 mutation increases neurovirulence and whether MPyV-specific resident memory T cells contribute to CNS immune surveillance or pathology.

C. Sabrina Tan and her colleagues at Beth Israel Deaconess Medical Center and Harvard Medical School tackled the development of an animal model from a different angle, creating a humanized mouse by engrafting human hematopoietic cells and thymus into immunodeficient mice. Preliminary data suggested that, similar to human infections, primary JCV infection is asymptomatic in these mice. The virus is rarely detected in whole blood and plasma, but is occasionally excreted in the urine. These mice can develop both humoral and cellular immune responses to JCV.

John A. Vanchiere from Louisiana State University Health Sciences Center and colleagues are developing a nonhuman primate model of polyomavirus diseases. The Squirrel monkey polyomavirus (SMPyV) is commonly found in Bolivian squirrel monkeys. Immunosuppression induced by rituximab and anti-CD8 antibody treatments resulted in the production of SMPyV variants with rearrangements of the non-coding control regions in some of the animals. High viral loads of SMPyV were observed in kidney tissue in some animals and one animal developed multifocal encephalitis. These findings are consistent with an early-stage monkey model for polyomavirus-induced diseases.

Speakers:
Raphael P. Viscidi, Johns Hopkins University School of Medicine
Gary Bloomgren, Biogen Idec
Heinz Wiendl, University of Münster, Germany

Highlights

• Significant technical challenges have made it difficult to determine whether B cells are a reservoir for JCV, so the question is controversial.
• Positive anti-JCV antibody status, prior use of immunosuppressants, and increased duration of natalizumab treatment have all been identified as risk factors for PML in natalizumab-treated patients with multiple sclerosis (MS).
• The percentage of CD62L (a cell adhesion molecule)-expressing CD4⁺ T cells was significantly lower in natalizumab-treated patients who developed PML, compared with controls, suggesting that the measurement of CD62L on CD4⁺ T cells might serve as a risk stratification tool.

Are B cells a reservoir of JC virus?

Raphael P. Viscidi from Johns Hopkins University School of Medicine discussed studies published in the last 25 years that review whether B cells are a reservoir for JC virus; these studies generated very different results and the question remained unanswered. To answer it, Viscidi's lab prepared subpopulations of primary B cells (naïve and memory B cells, transitional B cells, and plasma cells) to determine their susceptibility to JCV transfection and infection. The team tested sorted subpopulations of cells from natalizumab-treated MS patients and drug-free MS patient controls for the presence of JCV DNA by quantitative real-time PCR. Their preliminary data suggest that it is technically challenging to culture primary B cells and to transfect and infect post-sorting or microbead-separated B-cell populations. Although they found that CD27-negative cells and CD10-positive cells can be infected with rearranged or archetype JC virus, it is still unknown whether the infected cells are B cells, as some non-B cells in peripheral blood mononuclear cells also express these surface markers.

What have we learned from natalizumab and PML risk stratification?

Positive anti-JCV antibody status, prior use of immunosuppressants, and increased duration of natalizumab treatment have been identified as risk factors for PML in natalizumab-treated paitents with MS. Gary Bloomgren from Biogen Idec presented a PML risk stratification algorithm based on these risk factors for patients treated or considering treatment with natalizumab. Recent studies show that PML patients have a significantly higher anti-JCV antibody index than antibody-positive non-PML patients. Therefore, anti-JCV antibody index might be another risk stratification tool.

Global-risk minimization and data/sample collection initiatives are informing the evolving understanding of PML and the factors associated with improved outcomes in MS patients treated with natalizumab. These factors may include shorter duration between symptom onset and PML diagnosis, localized PML on MRI at diagnosis, younger age, and lower pre-PML EDSS (Expanded Disability Status Scale) scores.

PML risk stratification algorithm based on three risk factors: positive anti-JCV antibody status, prior use of immunosuppressants, and increased duration of natalizumab treatment. (Image courtesy of Gary Bloomgren)

Risk stratification in the MS patient population

As Bloomgren discussed, one risk factor for PML in natalizumab-treated patients with MS is increased duration of natalizumab treatment. Heinz Wiendl and colleagues from University of Münster explored how the immunological environment is influenced after long-term natalizumab therapy. They found that natalizumab treatment in MS patients induced T cells to switch from central-memory phenotype to effector-memory phenotype in CSF. They also observed that the percentage of CD62L-expressing CD4⁺ T cells was significantly lower in natalizumab-treated patients who developed PML compared to natalizumab-treated patients who did not develop PML Thus, CD62L-expressing CD4⁺ T cells might serve as a risk stratification tool and biomarker for PML.

Short talk presentations

JCV and BKV are closely related polyomaviruses. Ortwin Adams and colleagues from the Institute of Virology at the University Hospital of Düsseldorf tested a hypothesis that co-infection, or infection with BKV, decreases the risk of PML development, possibly by cross-reactive adaptive immune responses that protect from a symptomatic dissemination of JCV to the brain. They observed that the anti-BKV antibody level and reactivity was significantly lower in patients with PML compared with controls. Adams concluded his talk by discussing the use of anti-BKV antibody status and reactivity as a tool for PML risk assessment.

Host genetic factors may contribute to PML development since the disease is rare although the virus is common. John P. Carulli and colleagues from Biogen Idec tested this hypothesis by performing a genome-wide association scan of 109 AIDS–PML patients with 1200 healthy volunteers as controls. They also conducted whole-genome sequencing of 26 MS patients who developed PML while being treated with natalizumab, as well as 28 MS controls who did not develop PML after natalizumab treatment for two years or more. Thus far no genes have been found to be clearly associated with PML, although some SNPs and rare mutations identified in the studies are under further investigation.

Kory R. Johnson from the National Institute of Neurological Disorders and Stroke, NIH, presented a collaborative effort to standardize anti-JCV serology assays. Anti-JCV antibody status can be used as a risk stratification tool for PML development; to improve sero-status measures, 10 laboratories in the U.S. and Europe individually tested a panel of 50 sera and compared results. Of the 50 sera, 46% were concordantly classified as seropositive or seronegative by all labs; this low concordance rate was due to differences in data scale and to the use of lab-specific "cut off" criteria of JCV antibody level. The labs proposed a probability-based reporting scheme for JCV sero-status derived from the antibody assay output (i.e., level of JCV antibody detected), which is more informative than simply reporting a seropositive or seronegative category. There are also efforts underway to define an "international unit" for serology assays that could be distributed by the World Health Organization.

Clemens Warnke from the University of Düsseldorf presented a high-throughput screening/association study that looked at whether human leukocyte antigen (HLA) class I and II genotypes regulate anti-JCV antibody status and levels. They observed a strong negative association between anti-JCV antibody positivity and the DRB1*15 haplotype and a strong positive association between anti-JCV antibody positivity and the DQB1*06:03 haplotype. They concluded that HLA class II alleles are associated with anti-JCV antibody status and speculated that HLA class II-restricted immune responses may be important to anti-JCV host defense.

Speakers:
Avindra Nath, National Institute of Neurological Disorders and Stroke, NIH
Nancy D. Richert, Biogen Idec
David B. Clifford, Washington University School of Medicine
Imke Metz, University Medical Center, University of Göttingen, Germany

Highlights

• An ideal treatment for immune reconstitution inflammatory syndrome would both prevent non-specific immune activation and enhance JCV-specific cytotoxic immune responses.
• PML may be detected by MRI three to four months before symptoms develop.
• Clinical trials are difficult but provide valuable information.
• Small-molecule CCR5 inhibitors may be effective drugs for IRIS treatment.

Immune reconstitution inflammatory syndrome (IRIS)

It is believed that the best prognosis for PML is afforded by reconstitution of the immune system leading to viral clearance. However, return of immune function also frequently leads to an immune reconstitution inflammatory syndrome (IRIS), causing severe brain inflammation and significant clinical worsening (IRIS can cause severe brain injury or be fatal).

Avindra Nath of the National Institute of Neurological Disorders and Stroke, NIH, noted that IRIS may occur either at the time of PML presentation or after successful restoration of the immune system. The symptoms differ based on the location of the lesions. Risk factors include the degree and duration of immunosuppression, the rapidity of correction, and genetic susceptibility. An ideal treatment would both prevent non-specific immune activation and enhance cytotoxic immune responses specifically to JCV, but corticosteroids are the best available treatment option. Nath emphasized that the current goal is to treat, not to prevent, IRIS in PML patients, as there is no known way to prevent it. Signs of inflammation should be observed before steroid treatment is initiated to ensure that T cells have accessed the area to combat PML.

PML MRIs in natalizumab-treated MS patients

Nancy D. Richert of Biogen Idec presented the findings from an evaluation of the MRI scans from 302 natalizumab-treated patients with confirmed cases of PML. Lesions are most commonly located within the frontal lobe of natalizumab-treated patients. PML lesions can be classified as unilobar, multilobar, or widespread, and it is believed that all cases start as unilobar and progress to multilobar or widespread. PML can affect both the white and the gray matter of natalizumab-treated patients.

Richert noted that deep gray matter lesions can be mistaken for multiple sclerosis (MS). A delayed PML diagnosis could prevent the timely management of PML. MS lesions are most often periventricular, while PML lesions are peripheral and subcortical. PML lesions are often extensive and U-shaped; MS lesions can be U-shaped as well but are smaller. FLAIR and DWI are sensitive MRI methods for detecting PML.

Richert also noted the qualitative difference between IRIS and inflammatory PML; they appear similar on MRIs and it is necessary to watch the progression to distinguish the two. A few cases demonstrated that PML may be detected through MRIs three to four months before PML diagnosis. MRIs are thus a very important screening tool, especially for asymptomatic patients, who have improved survival and less functional disability than symptomatic patients.

PML clinical trials

David B. Clifford of Washington University School of Medicine reviewed the three PML clinical trials, testing cytosine arabinoside (1998), cidofovir (2002), and mefloquine (2013), with a focus on the mefloquine trial. Although the mefloquine trial failed, it illustrated the current obstacles to testing interventions for PML, which include the rarity of PML, the challenges to diagnosis, the rapid course of infection, and confounding/predisposing immune deficiency conditions.

Clifford elaborated on six design issues faced by the mefloquine trial: diagnostic criteria, patient population, patient numbers, trial duration, therapy, and organization. The variable course of PML in the most common clinical settings, including HIV-, malignancy-, or natalizumab-associated PML, together with poor patient compliance with a treatment plan, continue to be major challenges to PML clinical trials. Clifford suggested that optimal endpoints for PML clinical trials should center on biomarkers (such as quantitative JCV DNA in the cerebrospinal fluid), imaging, and clinical outcomes, rather than on survival. He concluded that clinical trials for PML are difficult to conduct and substantial multicenter support is required.

Pathology of IRIS in MS with natalizumab-associated PML

Imke Metz of University of Göttingen gave an overview of IRIS in MS patients with natalizumab-associated PML and emphasized the need for better management of IRIS. His lab analyzed brain biopsy tissues from five MS patients with natalizumab-associated PML and found a pronounced cytotoxic T cell-dominated infiltrate and high numbers of macrophages and plasma cells.

Non-demyelinated white matter and grey matter were also inflamed and few or no viral infected cells were found, indicating IRIS. Furthermore, chemokine receptor 5 (CCR5) was expressed on monocytes and T cells in autopsy CNS tissue from an MS patient with natalizumab-associated PML and IRIS, suggesting that CCR5-positive immune cells are potential effectors in pathophysiology of IRIS. Treatment of an MS patient with natalizumab-associated PML with the CCR5-antagonist maraviroc (an antiretroviral) resulted in marked clinical improvement. Maraviroc could potentially be an effective drug to treat IRIS.

Chemokine receptor 5 (CCR5) was expressed on T cells (stained with CD3) in autopsy CNS tissue from an MS patient with natalizumab-associated PML–IRIS. (Image courtesy of Imke Metz)

Short talk presentations

Benoit Combaluzier of Neurimmune Holding AG hypothesized that human-derived antibodies generated during a naturally occurring immune response against JCV strains in PML patients are attractive candidates for PML treatment. The team screened JCV-specific antibodies from memory B-cell repertoires of healthy donors and patients with full recovery from PML/PML–IRIS and successful control of JCV infection. They then selected antibodies that specifically target conformational JCV capsid epitopes and efficiently block cell infection and viral spread in vitro. These antibodies are promising candidates for the development of PML therapies, with high safety and efficacy profiles.

Roumen Balabanov of Rush University Multiple Sclerosis Center presented the favorable results of treatment/disease management of eleven patients with natalizumab-associated PML/IRIS. Their results suggested that early and frequent MRIs, immunostimulation with granulocyte colony stimulating factor (G-CSF), and anti-inflammatory treatment with maraviroc might be an effective treatment/management protocol for natalizumab-associated PML.

© 2013 The New York Academy of Sciences.