Altered Immune Phenotypes and HLA-DQB1 Gene Variation in Multiple Sclerosis Patients Failing Interferon β Treatment

• Published in: Frontiers in immunology Vol. 12; p. 628375
• Main Authors: Devi-Marulkar, Priyanka, Moraes-Cabe, Carolina, Campagne, Pascal, Corre, Béatrice, Meghraoui-Kheddar, Aida, Bondet, Vincent, Llibre, Alba, Duffy, Darragh, Maillart, Elisabeth, Papeix, Caroline, Pellegrini, Sandra, Michel, Frédérique
• Format: Journal Article
• Language: English
• Published: Frontiers 25.05.2021; Frontiers Media S.A
• Subjects: HLA class II genes; immune phenotypes; Immunology; interferon-stimulated genes; Life Sciences; multiple sclerosis; T cells; type I interferon
• ISSN: 1664-3224; 1664-3224
• DOI: 10.3389/fimmu.2021.628375

Interferon beta (IFNβ) has been prescribed as a first-line disease-modifying therapy for relapsing-remitting multiple sclerosis (RRMS) for nearly three decades. However, there is still a lack of treatment response markers that correlate with the clinical outcome of patients.BackgroundInterferon beta (IFNβ) has been prescribed as a first-line disease-modifying therapy for relapsing-remitting multiple sclerosis (RRMS) for nearly three decades. However, there is still a lack of treatment response markers that correlate with the clinical outcome of patients.To determine a combination of cellular and molecular blood signatures associated with the efficacy of IFNβ treatment using an integrated approach.AimTo determine a combination of cellular and molecular blood signatures associated with the efficacy of IFNβ treatment using an integrated approach.The immune status of 40 RRMS patients, 15 of whom were untreated and 25 that received IFNβ1a treatment (15 responders, 10 non-responders), was investigated by phenotyping regulatory CD4+ T cells and naïve/memory T cell subsets, by measurement of circulating IFNα/β proteins with digital ELISA (Simoa) and analysis of ~600 immune related genes including 159 interferon-stimulated genes (ISGs) with the Nanostring technology. The potential impact of HLA class II gene variation in treatment responsiveness was investigated by genotyping HLA-DRB1, -DRB3,4,5, -DQA1, and -DQB1, using as a control population the Milieu Interieur cohort of 1,000 French healthy donors.MethodsThe immune status of 40 RRMS patients, 15 of whom were untreated and 25 that received IFNβ1a treatment (15 responders, 10 non-responders), was investigated by phenotyping regulatory CD4+ T cells and naïve/memory T cell subsets, by measurement of circulating IFNα/β proteins with digital ELISA (Simoa) and analysis of ~600 immune related genes including 159 interferon-stimulated genes (ISGs) with the Nanostring technology. The potential impact of HLA class II gene variation in treatment responsiveness was investigated by genotyping HLA-DRB1, -DRB3,4,5, -DQA1, and -DQB1, using as a control population the Milieu Interieur cohort of 1,000 French healthy donors.Clinical responders and non-responders displayed similar plasma levels of IFNβ and similar ISG profiles. However, non-responders mainly differed from other subject groups with reduced circulating naïve regulatory T cells, enhanced terminally differentiated effector memory CD4+ TEMRA cells, and altered expression of at least six genes with immunoregulatory function. Moreover, non-responders were enriched for HLA-DQB1 genotypes encoding DQ8 and DQ2 serotypes. Interestingly, these two serotypes are associated with type 1 diabetes and celiac disease. Overall, the immune signatures of non-responders suggest an active disease that is resistant to therapeutic IFNβ, and in which CD4+ T cells, likely restricted by DQ8 and/or DQ2, exert enhanced autoreactive and bystander inflammatory activities.ResultsClinical responders and non-responders displayed similar plasma levels of IFNβ and similar ISG profiles. However, non-responders mainly differed from other subject groups with reduced circulating naïve regulatory T cells, enhanced terminally differentiated effector memory CD4+ TEMRA cells, and altered expression of at least six genes with immunoregulatory function. Moreover, non-responders were enriched for HLA-DQB1 genotypes encoding DQ8 and DQ2 serotypes. Interestingly, these two serotypes are associated with type 1 diabetes and celiac disease. Overall, the immune signatures of non-responders suggest an active disease that is resistant to therapeutic IFNβ, and in which CD4+ T cells, likely restricted by DQ8 and/or DQ2, exert enhanced autoreactive and bystander inflammatory activities.