Opinion

Monoclonal Antibodies, Blood-Brain Barrier and Disability in Multiple Sclerosis: Time for Combination Therapies

Jagannadha Avasarala, MD, PhD

From the Department of Medicine, Division of Neurology at Greenville Health System, Greenville, SC (J.A.), and University of South Carolina School of Medicine Greenville, Greenville, SC (J.A.)

Abstract The treatment of multiple sclerosis (MS) continues to evolve. However, even with the introduction of B-cell–depleting monoclonal antibodies (MAbs), disability progression continues unabated since B-cell therapies with MAbs do not address disease that lurks in the brain. The principal reason MAbs cannot enter the brain is because of their considerable size—limitations in size prevent MAbs from crossing the blood-brain barrier (BBB). To counter this problem, combining drugs that cross the BBB, such as siponimod, cyclophosphamide, or laquinimod with peripherally acting MAbs, could be one option. The concept of combining a small molecule with MAbs is novel from a pharmacolog- ic perspective and follows disease pathophysiology. In this brief report, a fundamentally new idea based on combination therapies designed to address pathology within and outside of the brain/ cord is presented. Limitations to such strategies include scant knowledge of the effects of small molecules in suppressing disease in the brain/cord, but it has been 23 years since the first MS drug was approved by the Food and Drug Administration; not a single drug, thus far, halts disability or the inevitable march of disease progression. It is time for scientists and clinicians to consider a different approach in MS therapeutics.

apid strides have been made in the treat- bodies (MAbs) in particular. Currently, 3 MAbs ment of multiple sclerosis (MS) since the have been approved for MS therapy: natalizumab, RFood and Drug Administration (FDA) alemtuzumab, and daclizumab. approved the first drug for use in 1993. Despite these advances, treatment options for worsening Radiologic stability in patients with MS requires disease characterized by accumulating disability that no new T1-gadolinium–enhanced lesions or on expanded disability status scores (EDSS) or expanding/new T2 lesions appear in the brain or cognitive decline are disappointing and unsuc- spinal cord, a cornerstone of NEDA (no evidence cessful. These outcomes are particularly worri- of disease activity) concept in MS. If MAbs are some for African American (AA) patients who being considered for therapy in the face of wors- have worse EDSS scores and gait abnormalities ening disability, it can be a challenge since they at diagnosis, and they likely have more primary do not penetrate the BBB. progressive MS (PPMS) variants.1 Unlike nearly all other blood vessels in the body, The reason for the relentless progression of dis- the endothelial cells of the BBB are held together by ability in relapsing-remitting MS (RRMS) across tight junctions, and for a drug to enter the central all racial and ethnic backgrounds that eventu- nervous system (CNS), it must take the transcellu- ally transitions into secondary progressive MS lar route. The tight junctions, coupled with numer- (SPMS) is probably linked to the poor penetration ous efflux transporters and metabolizing enzymes, of the blood-brain barrier (BBB) by drugs used in constitute a barrier to the movement of both mole- MS therapies in general and monoclonal anti- cules and cells from the bloodstream into the CNS.

GHS Proc. November 2016; 1 (2): 89-91 89 The BBB plays a role in MS treatment at multi- If drugs are to act in the brain, CSF concentra- ple levels, and MAbs that have been listed herein tions must be at least as high as in serum. As an cannot cross the BBB owing to their large size/ example, the poor BBB penetration of IV rit- molar mass or molecular size. The molar mass uximab became apparent in treatment of CNS or the mass of one mole (6.02 x 10^23 molecules, lymphoma that produced concentrations that Avogadro’s number) of the most commonly used were 1000-fold less in CSF than seen in serum.6 MAbs in neuroimmunology, including those that Additionally, poor BBB penetration of IV ritux- are non-FDA approved are natalizumab (146 kDa), imab was revealed in a study of 15 patients with rituximab (143 kDa), alemtuzumab (145 kDa), MS that followed IgG levels, IgG index, and oli- ocrelizumab (148 kDa), and daclizumab (142 kDa), goclonal band numbers in CSF and noted that respectively. These data are critical since BBB pen- they were not consistently altered at 24 weeks; etration is inversely related to the square root of the however, B and T cells in the CSF compartment molecular weight (or molar mass) of a substance2; were reduced although the mechanism remains typically, few molecules >0.5 kDa cross the BBB. unclear.7

There is minimal or no evidence that any of the The treatment landscape for MS continues to MAbs impact neurodegeneration in a clinically evolve. Impressive phase III data for ocrelizumab relevant manner over the long course of MS, in clinical trials, OPERA I and OPERA II for which is worrisome, particularly if MAbs are used RRMS, and ORATORIO for PPMS, have been as the sole therapeutic regimen. Additionally, the reported, particularly for disability scores. In the lipophilic or hydrophilic nature of molecules or latter study, 24% reduced risk of clinical disability size may not be the only limiting factors for drugs compared to placebo. The drug also reduced the 3 to cross the BBB. time required to walk 25 feet by 29%, the volume of chronic inflammatory brain lesions by 3.4%, As patients transition into SPMS from RRMS, fac- and the rate of brain volume loss by 17.5% com- tors that drive the process include autoreactive T pared to placebo. It is unclear how the beneficial cells that cross the BBB and participate in demye- effects on disability and MRI data on the brain lination, axonal transection, gliosis, and subsequent axonal degeneration4; the cascading neuroinflam- occurred, given the size of the molecule and its mation and neurodegeneration that follows con- poor penetration of the BBB, but certainly does tinues unabated as the BBB forms an impregnable not appear to be related to the phenomenon of barrier to drugs. Additionally, memory B cells and “regression to the mean.” plasma cells, central to humoral immunity, are Some illustrative examples of drugs—and this found in lesions and cerebrospinal fluid (CSF) of report makes no attempt to list all the possible patients with MS behind a protected BBB. combinations that can cross the BBB—are cyclo- Ectopic lymphoid follicles located in the meninges phosphamide (CYC), an alkylating agent that probably drive the pathology of MS and have been penetrates the BBB, and CNS parenchyma, which shown to house B cells and plasma cells,5 indicat- has been shown to decrease pro-inflammatory T ing that B cells migrate to the brain and can be helper Th1 cytokine interferon-gamma and inter- sustained locally within the CNS. Plasma cells do leukin-12 in MS. It also increases the secretion of not carry CD20 cell surface molecules and are not the anti-inflammatory Th2 cytokines IL-4 and 8 affected by anti-CD20 MAbs. Although appar- IL-10 in CSF and peripheral blood. It suppresses ently restricted to late disease phases, the develop- cell-mediated and humoral immunity through its ment of lymphoid tissue-like structures in brains actions of T and B cells.9 Cyclophosphamide may of patients with MS suggests a pathophysiological be useful in advanced disease but has not been role of B cells in MS that is possibly perpetual, studied in a randomized clinical trial setting. It is eventually being one of the many contributors perhaps time to revive an old drug. leading to SPMS. Second, laquinimod, an investigational CNS-ac- The progressive phase of MS is an autoimmune tive immunomodulator and a small molecule, can disorder characterized by an intrathecal com- diffuse freely across the BBB without any active partmentalization of inflammation resisting transport, but cardiotoxicity at higher doses (1.2 immunosuppressive treatments. The presence of mg/day) halted its progress in clinical trials and lymphoid follicle-like structures in the menin- is clearly a limitation.10 Can it be used at lower ges of some MS patients indicates that B cells doses that do not produce cardiotoxicity and yet can mature and perpetuate a compartmentalized be beneficial when combined with MAbs? That is humoral immune response. a question only clinical trials can address.

90 GHS Proc. November 2016; 1 (2): 89-91 MONOCLONAL THERAPIES IN MULTIPLE SCLEROSIS

Last, sphingosine-1-phosphate (S1P) receptors proof-of-concept studies, combination trials need Correspondence blockers cross the BBB but fingolimod failed in to identify patients with worsening disability Address to: the PPMS study.10 However, the new S1P blocker scores on EDSS, and selection of appropriate tar- Jagannadha Avasarala, such as siponimod has a relatively short half-life get population is key. Enrollment of failed RRMS MD, PhD, Greenville compared to fingolimod. It has fewer cardiac side patients on a standard drug based on NEDA cri- Health System, effects and comes with BBB penetration capabil- teria and whose EDSS is worsening perhaps forms Neuroscience ities. A phase III trial of siponimod in SPMS is the most eligible patient cohort to initiate combi- Associates, 200 ongoing. Could drugs such as CYC, fingolimod, or nation therapy trials. Patewood Dr, laquinimod be the future of MS therapies in com- Greenville, SC 29615 bination with MAbs? Some other strategies that A seminal study analyzed immunological, histo- ([email protected]) have been tried in oncology, for example, include chemical, and morphometric data on post-mor- the use of nanoparticles, immunoliposomes, pep- tem brain tissue samples from 29 SPMS and 7 11 tide vectors, and design of influx transporters and primary progressive PPMS cases. B-cell fol- could perhaps serve as model strategies for MS licles were detected in the meninges in 41.1% of therapies moving forward. the SPMS cases but not in PPMS cases, suggest- ing that the diseases are fundamentally different The premise of this report lies in the fact that strat- and treatment ought to begin at the RRMS or CIS egies to treat MS ought to begin early, before B-cell stage of the disease. This strategy could be partic- follicles have had time to develop. To address dis- ularly useful for the AA patients with MS. ease within the CNS, molecules that cross the BBB are critical. Alternatively, the best responsiveness We will never know if combining a small mole- to MAbs could perhaps be found in strategies that cule with MAbs at disease onset could work if destroy encephalitogenic B-cell populations in we do not use scientific reasoning to address the the periphery at the clinically isolated syndrome disease mechanisms both inside and outside of (CIS) stage and not as second-line drugs. the CNS. It must also be recognized that SPMS cases with follicles demonstrate younger age at The idea of combination therapy in MS is not onset, irreversible disability, death, and more pro- new. Many studies have explored concepts such nounced demyelination, microglia activation, and as safety, tolerability, and efficacy of several com- loss of neurites in the cerebral cortex.11 Without bination regimens but are underpowered and/or addressing such fundamental pathological phe- poorly designed. The key to success is to combine nomena residing in a compartmentalized “zone,” agents that have been shown to penetrate BBB MS therapies will continue to scratch the surface with drugs that work effectively and show robust in treating worsening disability status in patients efficacy outside of the CNS, such as MAbs. For and fail summarily.

References 1. Naismith RT, Trinkaus K, Cross AH. Phenotype imab therapy for CNS lymphomas: targeting the lep- and prognosis in African-Americans with multiple tomeningeal compartment. Blood. 2003;101:466-8. sclerosis: a retrospective chart review. Mult Scler. 7. Cross AH, Stark JL, Lauber J, Ramsbottom MJ, Lyons 2006;12:775-81. JA. Rituximab reduces B and T cells in cerebrospinal 2. Banks WA. Characteristics of compounds that cross fluid of multiple sclerosis patients. J Neuroimmunol. the blood-brain barrier. BMC Neurol. 2009;9:S3. 2006;180:63-70. 3. Cheng Z, Zhang J, Liu H, Li Y, Zhao Y, Yang E. Cen- 8. Smith DR, Balashov KE, Hafler DA, Khoury SJ, tral nervous system penetration for small molecule Weiner HL. Immune deviation following pulse cyclo- therapeutic agents does not increase in multiple scle- phosphamide/methylprednisolone treatment of mul- rosis and Alzheimer’s disease-related animal models tiple sclerosis: increased interleukin-4 production and despite reported blood-brain barrier disruption. Drug associated eosinophilia. Ann Neurol. 1997;42:313-8. Metab Dispos. 2010;38:1355-61. 9. Brück W, Wegner C. Insight into the mechanism of 4. Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mörk laquinimod action. J Neurol Sci. 2011;306:173-9. S, Bö L. Axonal transection in the lesions of multiple 10. Lublin F, Miller DH, Freedman MS, et al. Oral fin- sclerosis. N Engl J Med. 1998;338:278-85. golimod in primary progressive multiple sclerosis 5. Serafini B, Rosicarelli B, Magliozzi R, Stigliano E, (INFORMS): a phase 3, randomized, double-blind, Aloisi F. Detection of ectopic B-cell follicles with ger- placebo-controlled trial. Lancet. 2016;387:1075-84. minal centers in the meninges of patients with sec- 11. Magliozzi R, Howell O, Vora A, et al. Meningeal B-cell ondary progressive multiple sclerosis. Brain Pathol. follicles in secondary progressive multiple sclerosis 2004;14:164-74. associate with early onset of disease and severe corti- 6. Rubenstein JL, Combs D, Rosenberg J, et al. Ritux- cal pathology. Brain. 2007:130:1089-104.

GHS Proc. November 2016; 1 (2): 89-91 91