Two demyelinating diseases in the brain of a single patient, PML and MS: how to minimize ‘one’ while treating the ‘other’

Keywords::

• demyelinating diseases
• JCV
• natalizumab
• patients

Although both progressive multifocal leukoencephalopathy (PML) and multiple sclerosis (MS) are demyelinating diseases in the human brain, the similarity between the two stops there. Neither the clinical course of these neurological diseases, their lesions and pathology or, most importantly, their causes have much in common. PML occurs in patients with underlying immune compromised conditions and is caused by JC virus lytic infection of oligodendrocytes, the myelin producing cell in the brain. JCV is a human polyomavirus named from the initials of the patient from whom the virus was first isolated in 1971 [1]. The correct taxonomic name is JCPyV and should be referenced that way; not referring to the actual name of the patient for ethical reasons, a practice that has crept into the literature.

There is no effective antiviral treatment for PML, although restoring a competent immune system does confer benefit by slowing the disease progression. In contrast, MS has complex causes that are still not well defined, although relapsing–remitting forms of MS (RRMS) involve inflammatory, cytolytic T cells directed against myelin and myelin-associated glycoproteins. RRMS can be treated with drugs that dampen or impede the immune system. Currently, immune modulatory biologicals, monoclonal antibodies, have been successful in reducing relapse episodes and appearance of new MS lesions.

One monoclonal antibody in that category is natalizumab, Tysabri^@ Biogen/Idec, which binds the alpha 4, beta 1 and 7 epitopes on integrins, VLA 4, consequently preventing extravasation of inflammatory mononuclear cells into the brain by blocking binding to vascular cell adhesion molecules [101]. What does PML then have to do with MS and patients treated with natalizumab? The answer is both clear and confounding. There is an overall 1/750 risk of developing PML in MS patients treated with natalizumab. That risk is even greater by 10-fold, 1/80, in MS patients treated with 24 or more monthly infusions who have previous treatment histories with immune suppressive drugs, and who are antibody positive to JCV. In fact, PML has only been reported in an MS patient if treated with natalizumab, even though many other MS patients have received immune suppressive drugs and exhibit prior exposure to JCV.

Moving forward, a critical issue is how best to determine which MS patients treated with natalizumab may be at greater versus lesser risk for PML. Important components of a risk mitigation strategy would potentially include: evidence of the virus in patients at times of greatest risk; ascertain the competency of T-cell immune responses that control infection; application of these surveillance assessments in order to preemptively reduce risk in patients who are candidates likely to benefit from therapies associated with PML risk.

Evidence of virus in patients

JCV, the cause of PML, is a well-characterized virus with a global distribution that can establish latency in kidney and lymphoid tissues [2]. Measurement of antibodies using serology, usually ELISA assays [3,4], is an indirect but routine method of determining exposure. However, lack of detection of antibodies may not correlate with other evidence of a prior JCV infection. Some patients may shed JCV in the urine, yet test negative for the antibody [5]. We have found that some individuals can be viremic, yet do not produce antibodies measureable in validated assays [6]. Whether these patients have a humoral immune defect or JCV is not presented adequately to mount a humoral response is not known. A direct measure of infection is the detection of the viral DNA genome in blood using quantitative PCR (qPCR) that is specific and sensitive. We have established a multiplex qPCR that satisfies both conditions; specificity and sensitivity to 10 copies/ml, as well as distinguishing between virulent and nonvirulent forms of JCV [7]. Consequently, using serology and qPCR for periodic monitoring of patients with underlying diseases, and/or utilizing therapies that place them at risk for PML, would represent an important advance in medical therapeutics.

T-cell responses to JCV antigens & vaccine approaches

As PML occurs most often in the setting of T-cell immunodeficiency, it is reasonable to assume that the mechanisms underlying its emergence in an individual involve reduced T-cell immune control of JCV replication. This view has been supported by a number of studies that have measured JCV-specific CD4 and CD8 T-cell responses in individuals with PML who had not received natalizumab therapy and which showed an association between PML and reduced frequency of such responses [8]. Although these studies focused predominantly on the magnitude of JCV-specific T-cell responses, the quality of the response in natalizumab-treated people appears to be critical.

We recently showed that when measuring T-cell secretion of multiple cytokines in response to stimulation with peptides covering the entire JC virus proteome, individuals with PML either made no detectable T-cell responses to JCV or had JCV-specific CD4 T-cell responses uniquely dominated by IL-10 production [9]. This is in marked contrast to PML-free individuals, whether treated or not with natalizumab, who mount robust IFNγ-dominated T-cell responses to multiple JCV proteins. Furthermore, individuals shortly after PML diagnosis had higher levels of IL-10, IL-5 and IL-15 in the CSF. Thus, longitudinal and comprehensive immune response monitoring of subjects prior to and during natalizumab treatment could be used to identify individuals who are at risk of developing PML. The potential of immune screening to identify such individuals should be considered as complementary to stratification strategies based on the antibody levels. Similarly, vaccine strategies could even be considered as a prerequisite ‘priming intervention’ that would promote expansion in the clonal frequency of JCV-specific T cells that produce cytokines associated with JCV control – IFNγ, TNF and IL-2 – the confirmation of which would be required prior to the initiation of PML-associated therapies such as natalizumab.

Application of PML risk-predilection markers: clinical utility

The development of selective adhesion molecule inhibitors represented an important advance in the neurotherapeutics of MS; from both the perspective of elucidating mechanisms of immune dysregulation, which are germane to tissue injury in MS, and with respect to the formulation of novel treatment interventions that precisely and specifically target the transmigration of mononuclear cells from the circulation into the CNS [10].

Those of us in neurologic practice recall the paucity of effective treatment for MS and its destructive impact upon the patient’s personal, family and professional life. The arrival of parenteral therapies such as interferon beta and glatiramer acetate represented an important transformation in the capabilities of the neurologist caring for the patient with MS. The more recent development of orally administered disease-modifying therapy for MS further illustrates the progress that has been achieved in translating scientific principles at the bench and bedside into effective and more convenient treatments for our patients [11].

A significant proportion of treated patients, however, continue to exhibit disease activity and the accumulation of sustained disability. However, natalizumab, more than any other MS disease-modifying agent, has been demonstrated to significantly reduce both clinical and radiographic measures of disease activity, while correspondingly producing added benefits on physical, mental and emotional facets of quality of life. The robust efficacy observations reported from the randomized controlled clinical trials have been corroborated in clinical practice, the magnitude of which appears to eclipse the benefits derived from the other FDA-approved disease-modifying therapies.

A validated risk mitigation strategy would facilitate evidence-based counseling of our patients on the crucial subject of treatment-related benefits versus the real and now quantifiable, risks of contracting PML. An additional dividend of a risk mitigation strategy that takes into account the various factors previously characterized, concerns the ability of the treating neurologist to modify the natalizumab treatment regimen to a longer time period between infusions. Originally, this strategy was conceived after considering observations reported in the Phase II clinical trials, where cessation of treatment was not associated with recrudescence of clinical or radiographic measures of disease activity until about 12–16 weeks of treatment interruption. Based on these highly salient observations, our group at UT Southwestern Medical Center began to offer our MS patients an 8-week treatment regimen (dating back to the fall of 2006, shortly after natalizumab returned to the market, following its withdrawal on 28 February 2005), when frequent infections appeared to be a derivative of natalizumab therapy.

Employing the modified 8-week dosing scheme was associated with a marked resolution of infection risk (principally urinary tract infections, shingles, pneumonia and fungal vaginitis), and yet, without any apparent compromise in treatment efficacy (e.g., we observed no evidence of significant breakthrough; either clinically or on 3-month serial surveillance imaging studies over a year following the transition from ever 4 to every 8 week dosing (hypothetically producing recurrent oscillations of ‘mini-immune’ reconstitution). When the STRATIFY study results became available, designating that JCV IgG⁺ patients were at higher risk of developing PML compared to those with negative serologies, we began to counsel JCV IgG⁺ patients who were stable on natalizumb, concerning the option of remaining on therapy (rather than having to transition to an alternative agent), albeit at every 8 week intervals. To date, we are unaware of PML being associated with the modified 8-week natalizumab dosing regimen.

The integration of our new immune response profiling assay system (if validated and CLIA certified) into the rubric of a risk mitigation strategy, it would appear to further stratify our patients into markedly distinctive risk groups; with patently clear ramifications for offering modified dosing schedules. Ultimately, much remains to be done in order to reconcile the formidable challenges associated with the application of disease-modifying treatment of unprecedented efficacy, that are at the same time associated with serious, life altering and in some potentially life-threatening complications. Without equivocation, we believe that future progress in the arena of risk mitigation in neurotherapeutics will no doubt be the principal dividend of forging collaborative alliances among patients, care providers, regulatory agencies, investigators and the biopharmaceutical industry.

Impact on patients

The diagnosis of drug-related PML has been devastating to patients and their families, particularly since it usually follows the distressing diagnosis of MS. Although relatively rare, but increasing in occurrence, there is surprisingly insufficient information on PML that is widely available to the patient communities. More active visible and continuous communication with the current and future patients and their families in the areas of risk stratification, diagnosis, worldwide treatments, therapies and care are urgently required. A clear goal should be a continuing collaborative approach among the medical, regulatory, research, clinical and, most importantly, the patient communities.

Financial & competing interests disclosure

EM Frohman received consultant and lecture fees from Biogen Idec, TEVA, Genzyme, Abbott Laboratories, Acorda and Novartis. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.