The debate about the relative importance of T cells and B cells in the pathogenesis of MS will be refueled with the long-awaited publication of the results of the failed ATAMS study of atacicept (Kappos et al. Lancet Neurol 2014;13:353-363). Preliminary results from the two atacicept studies, ATAMS in relapsing MS and ATON in optic neuritis, were presented at ECTRIMS 2011 (Kappos et al. Abstract 107; Sergott et al. Abstract P436) and signalled the death knell for the drug in MS.

Atacicept is a recombinant fusion protein combining the Fc domain of human immunoglobulin and the extracellular domain of the TACI (transmembrane activator and CAML interactor) receptor. TACI binds to two B cell stimulatory cytokines, BLyS (B cell lymphocyte stimulator) and APRIL (A proliferation-inducing ligand), which regulate B cell maturation, function and survival. Both BLyS and APRIL have been shown to be overexpressed in autoimmune disorders such as MS, lupus and rheumatoid arthritis (Krumbholz et al. J Exp Med 2005;201:195-200; Thangarajh et al. Scand J Immunol 2007;65:92-98; Roschke et al. J Immunol 2002;169:4314-4321).

In the ATAMS study, 255 patients with relapsing MS were randomized to one of three doses of subcutaneous atacicept (25 mg, 75 mg or 150 mg) or placebo for 36 weeks (Kappos 2014). Ninety patients completed the week 36 visit prior to early termination of the study. Atacicept reduced the number of mature B cells and serum Ig levels. However, the annualized relapse rate was higher in all three treatment groups (25 mg, 0.86; 75 mg, 0.79; 150 mg, 0.98) compared to placebo (0.38). The time to first relapse was also significantly shorter with atacicept 150 mg versus placebo. The mean number of gadolinium-enhancing T1 lesions per scan was similar with atacicept (2.26, 2.30, 2.49) and placebo (3.07).

Before the ATON study was terminated, atacicept did appear to reduce retinal nerve fibre layer (RNFL) loss in patients with optic neuritis (Sergott 2011). However, more atacicept-treated patients developed clinically-definite MS compared to those receiving placebo (6/17 vs. 3/17).

Thus, atacicept appeared to worsen clinical disease activity in relapsing MS although the reasons for this are unclear. It should be noted that unlike therapies that deplete the B cell pool (e.g. rituximab), atacicept selectively targets mature B cells, and has little or no effect on B cell progenitors and memory B cells. One suggestion is that atacicept may adversely affect regulatory B cell function, which could impair protective immune effects (Lulu & Waubant. Neurotherapeutics 2013;10:34-43). Another suggestion is that APRIL (unlike BLyS) does not play a significant role in B cell-mediated autoimmunity (Morel & Hahne. Arthritis Res Ther 2013;15:107-108).

Also noteworthy is the recent finding in a B cell-dependent model of EAE that atacicept promotes apoptosis of retinal ganglion cells and increases axonal loss of the optic nerve, suggesting that the drug may adversely affect neuronal survival through some unknown mechanism (Kretzschmar et al. J Neuroimmunol 2014;268:58-63).

Comment
Dr. François Grand’Maison: The B-cell versus T-cell debate will rage until the immunopathology of MS has been unravelled. Let us not forget that the immunology of neuromyelitis optica, an inflammatory disease clinically similar to MS, is now understood. Optimism with respect to disentangling MS immunology is therefore warranted. However, the description of at least four distinct MS pathological subtypes (Pittock et al. J Neurol Neurosurg Psychiatry 2005;76:1693-1697; free full text at www.ncbi.nlm.nih.gov/pmc/articles/PMC1739469/pdf/v076p01693.pdf) does suggest that MS immunology is also heterogeneous. It follows that patients with different immunological pathologies will likely not respond to the same drugs. Furthermore, B-cell and T-cell function are highly intertwined so that a B-cell inhibitor may greatly interfere with T-cell activation or proliferation, depending on the targeted B-cell subtype.