The Immune Signature of CSF in Multiple Sclerosis with and without Oligoclonal Bands: A Machine Learning Approach to Proximity Extension Assay Analysis

Lorenzo Gaetani; Giovanni Bellomo; Elena Di Sabatino; Silvia Sperandei; Andrea Mancini; Kaj Blennow; Henrik Zetterberg; Lucilla Parnetti; Massimiliano Di Filippo

doi:10.3390/ijms25010139

The Immune Signature of CSF in Multiple Sclerosis with and without Oligoclonal Bands: A Machine Learning Approach to Proximity Extension Assay Analysis

Int J Mol Sci. 2023 Dec 21;25(1):139. doi: 10.3390/ijms25010139.

Authors

Lorenzo Gaetani¹, Giovanni Bellomo¹, Elena Di Sabatino¹, Silvia Sperandei¹, Andrea Mancini¹, Kaj Blennow^{2

3}, Henrik Zetterberg^{2

3

4

5

6

7}, Lucilla Parnetti¹, Massimiliano Di Filippo¹

Affiliations

¹ Section of Neurology, Department of Medicine and Surgery, University of Perugia, 06132 Perugia, Italy.
² Institute of Neuroscience and Physiology, Department of Psychiatry and Neurochemistry, The Sahlgrenska Academy at the University of Gothenburg, 431 41 Mölndal, Sweden.
³ Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, 431 41 Mölndal, Sweden.
⁴ Department of Neurodegenerative Disease, UCL Institute of Neurology, Queen Square, London WC1N 3BG, UK.
⁵ UK Dementia Research Institute at UCL, London WC1E 6BT, UK.
⁶ Hong Kong Center for Neurodegenerative Diseases, Clear Water Bay, Hong Kong 518172, China.
⁷ Wisconsin Alzheimer's Disease Research Center, University of Wisconsin School of Medicine and Public Health, University of Wisconsin-Madison, Madison, WI 53792, USA.

Abstract

Early diagnosis of multiple sclerosis (MS) relies on clinical evaluation, magnetic resonance imaging (MRI), and cerebrospinal fluid (CSF) analysis. Reliable biomarkers are needed to differentiate MS from other neurological conditions and to define the underlying pathogenesis. This study aimed to comprehensively profile immune activation biomarkers in the CSF of individuals with MS and explore distinct signatures between MS with and without oligoclonal bands (OCB). A total of 118 subjects, including relapsing-remitting MS with OCB (MS OCB+) (n = 58), without OCB (MS OCB-) (n = 24), and controls with other neurological diseases (OND) (n = 36), were included. CSF samples were analyzed by means of proximity extension assay (PEA) for quantifying 92 immune-related proteins. Neurofilament light chain (NfL), a marker of axonal damage, was also measured. Machine learning techniques were employed to identify biomarker panels differentiating MS with and without OCB from controls. Analyses were performed by splitting the cohort into a training and a validation set. CSF CD5 and IL-12B exhibited the highest discriminatory power in differentiating MS from controls. CSF MIP-1-alpha, CD5, CXCL10, CCL23 and CXCL9 were positively correlated with NfL. Multivariate models were developed to distinguish MS OCB+ and MS OCB- from controls. The model for MS OCB+ included IL-12B, CD5, CX3CL1, FGF-19, CST5, MCP-1 (91% sensitivity and 94% specificity in the training set, 81% sensitivity, and 94% specificity in the validation set). The model for MS OCB- included CX3CL1, CD5, NfL, CCL4 and OPG (87% sensitivity and 80% specificity in the training set, 56% sensitivity and 48% specificity in the validation set). Comprehensive immune profiling of CSF biomarkers in MS revealed distinct pathophysiological signatures associated with OCB status. The identified biomarker panels, enriched in T cell activation markers and immune mediators, hold promise for improved diagnostic accuracy and insights into MS pathogenesis.

Keywords: biomarkers; cerebrospinal fluid; machine learning; multiple sclerosis; oligoclonal bands; proximity extension assay.

MeSH terms

Axons
Biological Assay
Humans
Multiple Sclerosis* / diagnosis
Multiple Sclerosis, Relapsing-Remitting* / diagnosis
Oligoclonal Bands

Substances

Oligoclonal Bands

Grants and funding

H.Z. is a Wallenberg Scholar supported by grants from the Swedish Research Council (#2022-01018 and #2019-02397), the European Union’s Horizon Europe research and innovation program under grant agreement No 101053962, Swedish State Support for Clinical Research (#ALFGBG-71320), the Alzheimer Drug Discovery Foundation (ADDF), USA (#201809-2016862), the AD Strategic Fund and the Alzheimer’s Association (#ADSF-21-831376-C, #ADSF-21-831381-C and #ADSF-21-831377-C), the Bluefield Project, the Olav Thon Foundation, the Erling-Persson Family Foundation, Stiftelsen för Gamla Tjänarinnor, Hjärnfonden, Sweden (#FO2022-0270), the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 860197 (MIRIADE), the European Union Joint Program—Neurodegenerative Disease Research (JPND2021-00694), the National Institute for Health and Care Research University College London Hospitals Biomedical Research Centre, and the UK Dementia Research Institute at UCL (UKDRI-1003). LP is supported by grants from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska–Curie grant agreement No 860197 (MIRIADE), the European Union Joint Program—Neurodegenerative Disease Research (JPND2021-00694).