Background
Multiple sclerosis (MS) is considered a T cell-mediated autoimmune disease of the central nervous system (CNS) with a complex genetic background [
1]. It is accepted that blood-brain barrier (BBB) breakdown and T cells migration across BBB initiate an immune response against CNS myelin antigens and contribute to disease pathogenesis [
2,
3]. In addition, degeneration including loss of axons, diffuse damage to normal appearing white matter and involvement of deep and cortical gray matter contribute substantially to the disability progression [
1]. Clinically, the focal myelin and neuronal destruction leads to a variety of relapsing-remitting symptoms, which later in the course may become persistent or progressive [
4].
Seizures can occur in MS patients and the risk of epilepsy seems to be three-times higher in patients with MS than in the general population [
5]. Seizures can be the presenting symptom of MS but have been observed in relapsing-remitting as well as in secondary or primary progressive MS. β-interferons, which are often used for the treatment of MS, may have pro-convulsant effects [
6]. Moreover, MS symptoms can be aggravated by several antiepileptic drugs (AEDs), which can mimic disease activity [
5]. Up to now, no clinical trials for the treatment of epilepsy in MS patients have been performed and, therefore, no clear recommendations can be given.
Recent evidence suggests that inflammation mechanisms play a role in the pathogenesis of epilepsy [
7‐
12]. Moreover, recent studies performed in an experimental mouse model of epilepsy suggested that leukocyte trafficking mechanisms induce BBB damage leading to seizure generation [
10]. These results were supported by studies performed in an acute viral meningitis model in which cytotoxic T lymphocytes and massive recruitment of monocytes and neutrophils were required for vascular leakage and seizure-induced death [
11]. Importantly, white matter angiopathy and increased number of CD68-positive cells and CD3-positive T cells in perivascular cavities were documented in a subpopulation of young patients with refractory epilepsy [
12]. In addition, increased number of leukocytes was observed in brain parenchyma of epileptic patients, independently on the disease etiology [
10]. However, despite growing evidence showing a role for leukocyte trafficking and BBB damage in seizure generation, clinical trials with anti-adhesion therapies have not been performed yet in patients with epilepsy.
Current anti-inflammatory and immunosuppressive MS-treatments include β-interferons, glatiramer acetate (GA) and different chemotherapies. Recently, natalizumab, a monoclonal antibody directed against the α4 chain of integrin VLA-4, an adhesion molecule controlling leukocyte adhesion to brain endothelium, was approved by the U.S. Food and Drug Administration and the European Medicines Agency as monotherapy for highly active relapsing-remitting MS. Despite the occurrence of progressive multifocal leukoencephalopathy (PML) as adverse reaction, natalizumab represents the most potent drug approved thus far for the treatment of relapsing-remitting MS [
13,
14].
Here we describe a dramatic reduction of seizures after treatment with natalizumab in a patient with severe refractory epilepsy and MS. For better clarity we split the description of seizures and epilepsy from the non-epileptic MS course.
Discussion
Epilepsy affects between 0.5-1% of world population and, despite medical therapy, about one third of patients develop refractory epilepsy [
16‐
18]. Among other factors, MS represents a risk factor for refractory epilepsy [
19]. The pathophysiology of seizures in MS remains to be elucidated although cortical and subcortical lesions (as in our case, Figure
1C) may reasonably explain their increased frequency in MS [
5]. Seizures can occur in MS patients as the presenting symptom or as a relapse, being either related or unrelated to other, non epileptic clinical relapses [
5]. In our case, seizures occurred in both conditions as indicated on Table
1. Epileptic and non-epileptic symptoms were concomitant at disease onset (seizures followed ophthalmoparesis by 24 hours), and during 2004, and 2007. In contrast, seizures appeared to be clinically and temporally isolated in 2001, 2002, 2003, 2005, 2006 and 2008 and during the course of natalizumab therapy.
Experimental and clinical studies have shown that inflammation mechanisms are activated in epilepsy [
7]. Proinflammatory cytokines such as IL-1β, TNF-α and IL-6 have been shown to be overexpressed in experimental models of seizures, prominently by glia [
20], suggesting that glia activation may contribute to vascular inflammation in epilepsy. In addition, increased proinflammatory cytokines were found in the serum and CSF in patients with epilepsy, whereas the analysis of human brain specimens from drug-refractory epileptic patients showed strong activation of the IL-1b/IL-1R1 system in brain resident cells, such as in glia and neurons [
21,
22]. Functional interactions between cytokines and classical neurotransmitters such as glutamate and GABA have been also described, pointing out at novel glio-neuronal communications in epilepsy and suggesting that cytokine-mediated changes in neuronal excitability may promote seizures [
20]. Innate immune mechanisms such as complement have been also shown to be potentially involved in epilepsy [
22]. The role of innate immunity was further supported by a recent study performed in spontaneously epileptic mice and in human TLE showing a new proconvulsant pathway involving high-mobility group box-1 (HMGB1) release from neurons and glia and its interaction with Toll-like receptor 4 (TLR4) [
23]. TLR4 and HMGB1 blockade by pharmacologic inhibitors or by genetic deficiency, potentially interrupted glia-neuron communication and reduced seizure generation in animal models of seizures in this study [
23].
In addition to the inflammation mechanisms described above, recent studies performed in experimental animal models with relevance to human disease show a role for vascular inflammatory mechanisms and leukocyte-endothelial adhesion in the induction of BBB leakage and seizure generation [
10,
11]. BBB breakdown has been implicated both in the induction of seizures and in the progression to epilepsy with chronic seizure generation by exposure of neuronal cells to blood albumin and potassium ions [
9,
24‐
26]. Moreover, inhibition of BBB breakdown by blockade of leukocyte-endothelial interaction with an anti-VLA-4 antibody has preventive as well as therapeutic effects in a mouse model of epilepsy [
10]. VLA-4 mediates adhesion of lymphocytes [
27], monocytes [
28] and under inflammatory conditions also of neutrophils [
29], suggesting that leukocytes from innate and adaptive immunity may both contribute to seizure generation.
Acute seizure activity induces expression of adhesion molecules on brain endothelium [
10,
30]. Importantly, it has been recently shown that also recurrent seizures lead to chronic expression of VCAM-1, the ligand for VLA-4 integrin, potentially contributing to BBB permeability, neuroinflammation and brain damage potentially contributing to the evolution of chronic disease [
10]. Vascular inflammation induced by each seizure (eventually also in the absence of concomitant infection and autoimmunity) may allow adhesion and transmigration of myeloid cells and activated lymphocytes, increasing local inflammation and potentially favoring the generation of new seizures. However, whether vascular inflammation, leukocyte trafficking mechanisms and BBB leakage are involved in all types of epilepsy need to be clarified in further studies.
In support to the results obtained from animal models of epilepsy, it has been shown that BBB disruption in patients with cerebral lymphoma induces focal motor seizures [
9]. Vascular alterations and lymphocyte accumulation into the brain parenchyma were documented in a study performed on 87 young patients with refractory epilepsy [
12]. In addition, perivascular and parenchymal T lymphocytes with a predominance of CD8 cytotoxic cells were found in grey and white matter in samples obtained from patients with tuberous sclerosis complex [
31]. Cells of the microglia/macrophage cell system and scarce CD3 lymphocytes were also found to accumulate in brain samples obtained from patients with of TLE and hippocampal sclerosis [
21]. In addition, recent results showed increased number of leukocytes in brain parenchyma of patients with epilepsy independently on the disease etiology [
10]. In line with these previous data showing that leukocyte subpopulations may accumulate in the brain of patients with epilepsy, the results described in the present manuscript show successful treatment of epilepsy with natalizumab in a patient with MS. Notably, MS and epilepsy started concomitantly, but disease courses were relatively divergent during a 6-year period in which MS was under relative control with GA treatment whereas seizure frequency, duration and severity highly increased. Interestingly, generalized seizures preceded most of the new MS relapses leading us to speculate that, as seizure activity
per se induces vascular inflammation [
10], seizure activity may contribute to MS worsening. Thus, our results suggest that MS contribution to epilepsy induction, together with seizure activity potentially favoring MS relapses, may create a positive feedback loop between the two disease conditions.
The generalization of the remarkable effect of natalizumab on refractory epilepsy observed in the present clinical case to other more common types of epilepsy requires further studies. In contrast to the anti-adhesion therapy with natalizumab, a specific and sustained anti-adhesive activity exerted by GA and steroids in the present clinical case seems rather unlikely. In fact, GA treatment causes in vivo changes of the cytokine secretion pattern and effector function of GA-specific T cells but increases the migration rate of Th2 cells and do not affect the migration of Th1 cells [
32,
33], whereas steroids may reduce endothelial activation in a transient and non-selective manner during its short clinical use after MS relapses.
Conclusions
Current pharmacological treatments for epilepsy do not address the inflammatory component of pathogenesis highlighted by recent studies and most AED drugs aim to depress aberrant neuronal excitation. However, noncompliant and refractory epilepsy cases demand investigation into alternative mechanisms and corresponding treatments, and interfering with the adhesion of immune cells to the cerebral vasculature may potentially open new avenues for epilepsy treatment [
34].
Our results indicate treatment with natalizumab as highly effective in patients with MS and epilepsy. As no interactions with AED drugs were described for natalizumab and recent clinical data show that safety may be increased for PML associated with natalizumab therapy [
35], our results suggest that anti-adhesion therapies such as natalizumab may complement traditional therapies, and might be useful in treating refractory epilepsy and epilepsy following MS or inflammatory inciting events such as trauma, stroke and infections.
Taking into account our current understanding of the pathogenesis of seizures and epilepsy and the emerging key role of inflammation mechanisms in epilepsy, our data suggest a more general application of natalizumab and anti-adhesion therapy in other types of epilepsy.
Competing interests
Only public funds from the University of Sassari and the University of Verona were used for the present paper. No conflicts of interest exist with the companies whose products are mentioned and/or discussed in this article. GC is co-author in a patent owned by Stanford University USA and entitled "Anti-leukocyte recruitment therapy for the treatment of seizures and epilepsy" (U.S. Patent Application Serial No. 11/811,245).
Authors' contributions
All authors fulfill the authorship criteria because of their substantial contributions to the conception, design, analysis and interpretation of the data. SS and GC wrote the manuscript, MRM made the clinical follow-up of the patient. All authors gave their final approval of the version to be published.