Treating Multiple Sclerosis With Monoclonal Antibodies

Mathias Buttman, MD; Peter Rieckmann, MD, FRCPC


Expert Rev Neurother. 2008;8(3):433-455. 

In This Article


Natalizumab is a humanized monoclonal IgG4κ Ab raised against human α4-integrin (CD49d) and is marketed as Tysabri (Biogen Idec and Elan Pharmaceuticals).

Natalizumab is the first therapeutic mAb specifically developed for MS therapy. It was first approved by the FDA in November 2004. After a voluntary market withdrawal in February 2005 when three PML cases had been identified,[1,2,3] it was finally approved by the FDA as a monotherapy for relapsing forms of MS in June 2006 under the restriction of a special safety program (see section ‘Tysabri safety program’ and FDA supplementary approval letter).[185]

In the EU, NATA was approved in June 2006 as a monotherapy for RRMS patients with high disease activity despite treatment with an IFN-β preparation, and in patients with rapidly evolving severe RRMS as a first-line therapy.

In August 2007, a FDA advisory committee recommended approval for moderate-to-severe Crohn’s disease, where NATA was superior to placebo in most of the trials conducted.[24,25,26] As of November 2007, an extended FDA review was ongoing.

α4-integrin is expressed on lymphocytes and monocytes but normally not on neutrophils.[27,28] However, under certain conditions neutrophils may also show functionally relevant α4-integrin expression.[29,30] Additionally, α4-integrin is found on hematopoietic cells including endothelial progenitor cells, where it is an important regulator of cell differentiation.[31,32]

α4-integrin forms heterodimers with β1-integrin (CD29) and β7-integrin. The α4β1-heterodimer, termed very late antigen (VLA)-4, is a counter receptor for vascular cell adhesion molecule-1 (VCAM-1, CD106) expressed on activated endothelial cells. VLA-4/VCAM-1 interaction mediates firm adhesion of immune cells to the vessel wall and is important for immune cell extravasation into CNS tissue. In addition, α4β1-integrin recognizes the extracellular matrix proteins fibronectin and osteopontin.[33,34,35]

The α4β7-heterodimer in particular, but also to a lower degree the α4β1-heterodimer, uses mucosal addressin cell adhesion molecule-1 as a counter receptor, which is expressed on high endothelial venules of mesenteric lymph nodes, Peyer’s patches and in the flat vascular endothelium of the lamina propria.[36,37] Accordingly, α4β7-integrin mediates homing of lymphocytes to gastrointestinal lymphoid tissue.[38]

In leukocytes, α4-integrin acts as a signal transducer into the cell upon ligand binding, which is mediated through PKA-dependent serine phosphorylation of α4-integrin.[39] In a migrating cell, phosphorylated α4-integrins are localized to the leading edge of the cell while dephosphorylated α4-integrins are found at the lateral and trailing edge.[40] Via the cyto-skeletal adapter protein paxillin and certain signaling molecules, α4-integrin phosphorylation determines the activation state of the small guanosine triphosphatase Rac1, which is a key regulator of actin polymerization and lamellipodia formation, and thereby of immune cell migration.[39,41,42,43,44,45] In conclusion, α4-integrin enables effective leukocyte extravasation, probably also into the CNS tissue, through spatial control of Rac1 activation in the migrating cell.[46]

Blockade of lymphocyte adhesion to CNS endothelium is currently regarded as the central mechanism of action mediating the therapeutic effect of NATA in MS (Figure 1).[47,48] Furthermore, NATA was reported to decrease the cerebro-spinal fluid (CSF) CD4+/CD8+ cell ratios in MS patients.[49] In vitro, NATA inhibited transmigration of CD3+ CD25+ Foxp3-expressing regulatory T cells (Tregs) but did not impair Treg function.[50] The therapeutic relevance of these effects is currently unclear. NATA has been reported to mobilize hemato-poietic cells from bone marrow to peripheral blood in MS patients.[51] As discussed later, this effect may be of special relevance for the pathogenesis of PML in patients treated with NATA.

Modes of therapeutic natalizumab action in multiple sclerosis.

Blockade of α4β7/mucosal addressin cell adhesion molecule-1 interaction is currently thought to mainly mediate the thera-peutic effect of NATA in Crohn’s disease.[52]

Two excellent articles in Expert Review of Neurotherapeutics, one by Rudick and Sandrock,[53] and one by Kachuck,[54] summarized the early development of NATA for MS therapy, including detailed information on the pharmacology of the drug.[53] Therefore, we do not discuss these aspects here. We update these reviews by starting with the results of two randomized, double-blind, placebo-controlled multicenter Phase III clinical trials, performed to determine the efficacy and safety of NATA treatment in patients with RRMS. The AFFIRM (Natalizumab Safety and Efficacy in Relapsing-Remitting MS) study investigated NATA as a monotherapy while the SENTINEL (Safety and Efficacy of Natalizumab in Combination with Interferon β-1a in Patients with Relapsing-Remitting Multiple Sclerosis) trial studied NATA as a combination therapy with IFN-β1a (Avonex®) in RRMS.[55,56,57]

The AFFIRM trial was a randomized, double-blind, placebo-controlled, multicenter trial investigating safety and efficacy of NATA mono-therapy in patients with RRMS.[55,56] Patients were between 18 and 50 years of age, had a diagnosis of RRMS according to the McDonald criteria,[15] an expanded disability status scale (EDSS) score between 0 and 5.0 at baseline and at least one medically confirmed relapse in the year prior to study entry. A total of 942 patients were randomized in a 2:1 ratio (stratified according to study site) to receive either intravenous infusions of 300 mg NATA (627 of 942) or placebo (315 of 942) every 4 weeks over a treatment period of 116 weeks. Clinical examinations were performed every 12 weeks, and MRI scans were obtained at baseline, at week 52 and at week 104, and centrally analyzed at the University College London (London, UK) by blinded investigators. Annualized relapse rate after 1 year and sustained disability progression after 2 years were defined as co-primary endpoints.

Natalizumab reduced the annualized relapse rate after 1 year from 0.81 to 0.26 (p < 0.001), which corresponds to a relative reduction of 68%. At 2 years, the cumulative probability of sustained disability progression was 29% in the placebo and 17% in the NATA group (hazard ratio 0.58; p < 0.001). More data on clinical and MRI measures after 2 years are provided in Figure 2. An open-label extension phase revealed that after 3 years, NATA showed a sustained effect on annualized relapse rate and confirmed disease progression.[58] NATA treatment significantly improved health-related quality of life, as measured by the Short Form-36 survey and subject global assessment visual analog scale, administered at baseline and weeks 24, 52 and 104.[59]

AFFIRM 2-year efficarcy data.
EDSS: Expanded disability status scale; Gd: Gadolinium

Natalizumab was generally safe and well tolerated. After 2 years, the only adverse events more frequently observed in the NATA-treated than in the placebo-treated patients were fatigue (27 vs 21%; p = 0.048) and allergic reactions (9 vs 4%; p = 0.012). Although most allergic reactions occur within 2 h of infusion, corresponding to a type I allergy, a delayed, serum sickness-like, type III systemic allergic reaction to NATA in a 23-year-old man who consecutively developed nAbs was recently reported.[60] He did not participate in the AFFIRM trial. The issue of nAbs will be discussed later. No PML case was observed in the AFFIRM trial.

The SENTINEL study was a randomized, double-blind, placebo-controlled, multicenter trial investigating safety and efficacy of NATA as an add-on therapy to weekly intramuscular injections of IFN-β1a 30 μg in RRMS patients clinically not stable on IFN-β1a.[59] A total of 1171 IFN-β1a-treated patients having experienced at least one relapse within the year prior to study entry were randomized in a 1:1 ratio to receive either NATA 300 mg (589 of 1171) or placebo (582 of 1171) intravenously every 4 weeks over a planned treatment period of 116 weeks. At entry, the patients were 18-55 years of age, had RRMS according to the McDonald criteria,[15] an EDSS score between 0 and 5.0, and had been pretreated with IFN-β1a for at least 12 months prior to randomization. As in the AFFIRM trial, clinical evaluations were performed every 12 weeks, MRI scans were obtained at baseline, weeks 52 and 104, and annualized relapse rate after 1 year and cumulative rate of sustained disability progression after 2 years were defined as co-primary endpoints.

Add-on therapy of NATA reduced the annualized relapse rate after 1 year from 0.82 in the patients receiving IFN-β1a monotherapy to 0.38 in the patients additionally receiving NATA, which represented a relative reduction of 54% (p < 0.001). After 2 years, the annualized relapse rate was 0.75 with IFN-β1a alone and 0.34 with combination therapy (relative reduction 55%; p < 0.001). While 29% of the monotherapy patients demonstrated sustained EDSS progression after 2 years of treatment, only 23% of the patients receiving combination therapy had experienced sustained disability progression, which corresponded to a relative decrease of 24% (p = 0.02). The mean number of new or enlarged T2 lesions over 2 years was reduced from 5.4 in the monotherapy group to 0.9 in the combination therapy group, representing a relative reduction of 83% (p < 0.001). The mean number of gadolinium (Gd)+ lesions after 2 years was 0.9 in the group receiving only IFN-β1a and 0.1 in patients receiving the combination with NATA (relative reduction 89%; p < 0.001). Therefore, NATA appeared to be an effective add-on treatment in RRMS patients not stable on IFN-β1a.

Treatment with NATA was well tolerated in most patients. Adverse events significantly associated with combination therapy of NATA with IFN-β1a included anxiety (12 vs 8%; p < 0.01), pharyngitis (7 vs 4%; p < 0.05) and peripheral edema (5 vs 1%; p < 0.001). However, when two PML cases, one of which was fatal, were identified in the combination group,[1,2] treatment was stopped on 28th February 2005, approximately 1 month early. At this time point 86% (1003 of 1171) of patients had completed the 120-week study. A careful investigation was initiated where no further PML cases were identified in the AFFIRM and the SENTINEL trial.[4] A third NATA-treated patient in whom PML occurred was retro-spectively identified in a Crohn’s disease trial when treatment was stopped.[3]

Progressive multifocal leukoencephalopathy is a rare and rapidly progressive, in most cases fatal, opportunistic virus infection of the CNS caused by John Cunningham virus (JCV), a human polyomavirus.[61,62] At least 50% of the healthy population will become JCV seropositive during their lifetime.[63] However, the rate of people carrying replication-competent virus is unknown. The virus may persist in renal tubular epithelium, lymph nodes and in bone marrow cells.[64,65] PML only develops in immunocompromised individuals, especially in HIV-infected patients,[66] where the virus may replicate without hindrance.

The reason for the development of PML in NATA-treated patients is not known. It has been hypothesized that NATA may promote the premature exit of infected bone marrow cells into the circulation. These cells, now no longer under the control of the bone marrow stroma, may enter the CNS where JCV may replicate in oligodendrocytes resulting in development of PML, additionally owing to reduced immune surveillance of the CNS as a result of NATA treatment.[67] However, it is unclear whether JCV needs a cellular carrier to enter the CNS and whether it is not already there when NATA treatment starts, although there is no positive indication.[68]

Clinical distinction between PML and a MS relapse may be very difficult. It has been emphasized that in contrast to PML a MS relapse usually shows a rapid onset and that it tends to stabilize and resolve. While optic neuritis and myelopathic signs are frequently observed in MS but not in PML, the classical triad of PML comprises progressive dementia, motor dysfunction and bilateral loss of vision.[69] When PML cannot be ruled out in a patient with new onset of neurological symptoms, Gd-enhanced MRI should be performed. While MS lesions usually show a focal distribution with sharp edges in a typical localization (periventricular, corpus callosum or cerebellar peduncles) and orientation (so-called Dawson’s fingers), PML lesions tend to show an ill-defined and diffuse involvement of the sub-cortical white matter. In contrast to MS, Gd enhancement is rarely observed in PML.[69] However, it is impossible to distinguish between PML and a MS relapse solely based on MRI findings. Therefore, a lumbar puncture including determination of JCV DNA in the CSF by PCR may be required, which may confirm or help to rule out the diagnosis.

There is no established therapy for PML.[70] Immediate discontinuation of NATA in a PML patient seems obvious. Furthermore, plasma exchange has been shown to accelerate NATA clearance in MS patients, thereby possibly helping to restore immune function in a PML patient.[71]

John Cunningham virus serological studies of serum and CSF are not considered useful to establish the risk for PML before NATA treatment initiation. Regular blood and CSF PCR examinations for the presence of JCV DNA during NATA treatment are not considered helpful either, since detectability of JCV DNA does not seem to precede the clinical manifestation of PML in a clinically useful time frame.[4] Therefore, the most important safety measure for PML prevention is not to initiate NATA treatment in immunocompromised individuals. There is a need for the development of widely applicable prognostic markers allowing risk assessment in individual patients, which may best be achieved based on a cooperative effort between industrial and academic researchers.

In the USA, NATA is only available under the Tysabri Outreach: Unified Commitment to Health (TOUCH®) prescribing program which all US prescribers and patients have to enroll into.[5] In August 2007, approximately 14,000 patients were enrolled. This program was developed in cooperation with the FDA to assure a high level of information and vigilance on both sides for prescribers and patients. Special attention is given to the early recognition of severe opportunistic infections including PML.[72] Details about the TOUCH program are provided by the manufacturer at[186] or from the toll free number +1 800 456 2255 (Monday to Friday, 08:30 to 20:00, ET). As of August 2007, no new PML cases were identified. However, postmarketing experience indicates that there is an increased frequency of herpes infections in NATA-treated patients.[5] Cumulative safety data were comparable to those observed in AFFIRM and SENTINEL.

While TOUCH is a US program, TYsabri Global Observation Program In Safety (TYGRIS) is a voluntary observational cohort study including patients from all over the world. It is planned to include 5000 patients, including 3000 patients and will follow them over 5 years. As of 23 August 2007, 654 patients were enrolled, approximately two thirds of whom were from Germany. Patients will be evaluated every 6 months. All immunomodulatory pretreatments and all serious adverse events including severe infections as well as malignancies are registered.[72]

The international TYSABRI Pregnancy Exposure Registry is another part of the Tysabri safety program.[5] The coordinating center monitors patients exposed to NATA within 3 months prior to conception throughout their pregnancies and monitors the infants until 8-12 weeks of age in the USA and Canada, and within 4 weeks after the estimated date of delivery in the rest of the world. It is planned to enroll 300 patients into this observational cohort study. As of August 2007, 24 pregnancies were enrolled in the registry. At that time, 21 pregnancies were ongoing, and one live birth, one spontaneous abortion and one elective termination were reported. The coordinating center can be contacted under the following address: Pregnancy Exposure Coordinating Center, 3168 Collins Ferry Road, Morgantown, WV 26505-3352; Tel.: +1 866 831 2358; Fax: +1 866 718 6927; Email:

The Safety of TYSABRI Redosing and Treatment (STRATA) study is an ongoing open-label, multi-national, safety extension study, mainly of AFFIRM and SENTINEL, evaluating the safety of NATA re-dosing after transient suspension.[73] No unexpected adverse events in this patient cohort were identified during the 48-week re-treatment period, which will be followed by a 4-year follow-up period. However, patients with nAb positivity in prior evaluations were not included into the STRATA study. Therefore, the incidence of hypersensitivity and infusion reactions upon re-dosing in this patient subgroup of special interest is not elucidated by the STRATA study. The safety profile in patients who switched to NATA from other disease-modifying drugs received during the treatment gap was comparable with that in patients who did not receive MS treatments during that time.

Clinical neurologists are well aware of increasing evidence that the development of nAbs against IFN-β might reduce the clinical efficacy of this recombinant cytokine.[74] However, several other protein therapeutics, including mAbs, may also induce the development of nAbs.[75] Therefore, both the AFFIRM and the SENTINEL trial were designed to assess the presence of nAbs directed against NATA.[76] Blood samples for nAb measurement by ELISA were obtained at baseline and every 12 weeks during the treatment period of 2 years. Patients were defined as ‘transiently positive’ if nAbs were detectable at a single time point, while patients with two or more positive blood samples at least 6 weeks apart were defined as ‘persistently positive’.

In the AFFIRM trial, 6% (37 of 625) of the NATA-treated patients were persistently nAb-positive and an additional 3% (20 of 625) were only transiently positive. Of note, 88% (50 of 57) of the nAb-positive patients developed nAbs within the first 12 weeks of the 2-year treatment period, and another 9% (5 of 57) within 24 weeks of therapy initiation. Interestingly, 53% (10 of 19) of the so-called persistently positive patients who were treated with NATA for the full period of 2 years returned to nAb negativity within these 2 years. Importantly, persistently positive patients showed reduced clinical efficacy of the drug in comparison to nAb-negative patients ( Table 3 ). By contrast, transiently positive patients only experienced a slightly higher annualized relapse rate in comparison to nAb-negative patients during the first 6 months of treatment (0.42 vs 0.26; p = 0.38). However, at the end of the 2-year treatment period annualized relapse rate, disability progression and MRI measures were comparable in the transiently positive and the negative patients.

With one exception, incidence and types of adverse reactions were comparable in nAb-negative and persistently positive patients: infusion-related adverse events, defined as any adverse effect occurring within 2 h after starting the 1-h NATA infusion, were more frequently observed in persistently nAb-positive patients (76%) than in nAb-negative patients (20%). Symptoms only rarely observed in the nAb-negative patients included headache (19%), nausea (16%), urticaria (14%) and rigors (11%) in the persistently positive patients.

Like in the AFFIRM trial, 6% of NATA patients were persistently nAb-positive in the SENTINEL trial. Another 5% were found to be transiently positive. By contrast to the AFFIRM trial, persistently positive patients in SENTINEL did not show significantly increased disability progression in comparison to the nAb-negative patients after 2 years (p = 0.503). However, they had a higher annualized relapse rate (0.65 vs 0.31; p < 0.001) and higher MRI disease activity concerning the development of Gd+ lesions (p < 0.001) and new or enlarging T2 lesions (p < 0.001) over 2 years. Comparable with the AFFIRM trial, infusion-related adverse reactions were the only adverse events more frequently observed in persistently positive (79%) than in nAb-negative patients (19%). In general, symptoms of these infusion reactions were similar to those observed in the AFFIRM trial, only rigors were reported more frequently (29%), and flushing (13%), pruritus (11%) and tachycardia (11%) were additionally reported.

In both the AFFIRM and the SENTINEL study, nAb presence was associated with largely reduced NATA serum levels, indicating that nAbs induce NATA removal from the circulation. Parallel nAb detection by ELISA and flow cytometry revealed that most, if not all, nAbs detected by ELISA were raised against the α4-integrin-recognizing domain of NATA, thereby affecting its blocking capacity.

In summary, in both trials 6% of patients were persistently nAb-positive and only these patients but not the transiently positive patients showed reduced clinical and MRI efficacy of the drug in comparison to the nAb-negative patients after a 2-year treatment period. From a clinical point of view, it could be concluded that routine nAb testing in NATA-treated patients is not necessary owing to the low rate of persistently positive patients. Even more, there may be cases where routine testing could pose difficult questions to the treating neurologist since ‘persistent’ nAb positivity of a clinically stable patient would raise questions concerning further treatment: according to current knowledge approximately half of the ‘persistently’ positive patients become negative again within 2 years.

Since transient nAb positivity occurs in a significant proportion of patients during the first 6 months of treatment and since transiently positive nAbs do not seem to hamper the long-term effect of the drug, the clinical usefulness of nAb testing during the first 6 months of treatment is especially limited, or at least requires repeated testing to identify persistently positive patients. In our opinion, nAb testing could be recommended in clinically unstable NATA-treated patients or in patients experiencing infusion-related adverse events, especially after at least 6 months of NATA treatment, when most of the transiently positive patients returned to nAb negativity.

A recent report indicated an increase in the median annualized number of new or enlarging MRI T2 lesions from 3.43 in the pretreatment interval to 10.32 in the postwithdrawal interval (p = 0.014) in a subcohort of 21 AFFIRM and SENTINEL patients.[77] The increase was much more pronounced in placebo/NATA patients (from 2.24 to 10.37) than in NATA/NATA patients (from 3.47 to 4.81), indicating that short exposure to NATA versus longer exposure might increase the risk of a postwithdrawal rebound of disease activity. These important observations await further confirmation before conclusions affecting clinical routine can be drawn.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.