In vitro morphological profiling of T cells predicts clinical response to natalizumab therapy in patients with multiple sclerosis

Introduction

Very late antigen-4 (VLA-4) is a heterodimeric integrin composed of the α4 and β1 subunits¹. VLA-4 mediates adhesion, transmigration and intra-tissue trafficking of immune cells, by interacting with vascular cell adhesion molecule-1 (VCAM-1) and fibronectin². In addition, this integrin is involved in T-cell immune synapse assembly³ and activation^4,5. Because of its pivotal role in multiple aspects of T-cell biology, VLA-4 also contributes to mediate dysregulated T-cell responses, such as those underlying inflammatory diseases⁶. This integrin has been particularly studied in the context of multiple sclerosis (MS), in which lymphocytes overexpress VLA-4 on their surface⁷ facilitating transmigration to the central nervous system (CNS)^{8, 9–10}. The identification of this mechanism has been the rationale for targeting VLA-4 as a therapeutic approach to treat MS^11,12.

The humanized anti-α4 monoclonal antibody natalizumab, which blocks VLA-4-VCAM-1 interaction, is currently used to treat relapsing-remitting MS (RRMS) patients^13,14. This therapeutic antibody acts primarily by reducing the homing of pathogenic lymphocytes and activated monocytes to the CNS¹⁵. The NEDA endpoint (no evidence of disease activity) is commonly used to define controlled MS¹⁶. Currently, ~35% of MS patients present evidence of disease activity at 2 years¹⁷, such as new or enlarging lesions, relapses or disability progression, not satisfactorily responding to the treatment according to the NEDA criteria. VLA-4 expression on peripheral blood mononuclear cells (PBMCs)¹⁸, serum levels of IgM specific for phosphatidylcholine¹⁹, metalloproteinases²⁰, low percentage of CD5⁺ B cells²¹, age at prescription, and disease duration²² have been pointed as biological parameters related to natalizumab clinical response, although their robustness as predictive biomarkers remain to be established.

To date, a pretreatment test that could predict whether a given person with MS might benefit from the therapy is lacking. This would be particularly relevant to avoid exposing unresponsive patients to potential risks, such as the JC virus-associated progressive multifocal leukoencephalopathy (PML)²³, and to tailor treatment decisions toward other available therapeutic lines. In this context, we reasoned that profiling the in vitro response to natalizumab of leukocytes from MS patients might predict treatment response and therefore assist treatment decisions.

For this purpose,...