Assessing the presence of oligoclonal IgM bands as a prognostic biomarker of cognitive decline in the early stages of multiple sclerosis

Abstract Background An association has been found between the presence of lipid‐specific oligoclonal IgM bands (LS‐OCMB) in cerebrospinal fluid and a more severe clinical multiple sclerosis course. Objective To investigate lipid‐specific oligoclonal IgM bands as a prognostic biomarker of cognitive impairment in the early stages of multiple sclerosis. Methods Forty‐four patients underwent neuropsychological assessment at baseline and 4 years. Cognitive performance at follow‐up was compared adjusting by age, education, anxiety–depression, and baseline performance. Results LS‐OCMB+ patients only performed worse for Long‐Term Storage in the Selective Reminding Test (p = .018). Conclusion There are no remarkable cognitive differences between LS‐OCMB– and LS‐OCMB+ patients in the early stages of MS.

Given the high prevalence and also the great impact on daily activities of CI it is important to identify risk prognosis factors in order to consider more active clinical and therapeutic actions from the onset of the disease. Certain demographic and clinical data have been identified as predictors of the disease clinical course (age, physical disability, disease subtype, number of relapses in the first 2 years, etc.) (Langer-Gould et al., 2006), but the evidence is controversial (DeLuca et al., 2015;Ruano et al., 2017) and for some of them such as number of relapses in the first 2 years, the time required is too long for prognostic purposes (Thangarajh et al., 2008). Some neuroimaging factors have been reported (e.g., lesion load and brain atrophy) as potential biomarkers (DeLuca et al., 2015;Eijlers et al., 2018); however, even though these findings are valuable in helping to explain the mechanisms underlying cognitive impairment, they have proved of little clinical applicability for predicting this impairment to date. Besides, some biomarkers present in the cerebrospinal fluid (CSF) have been investigated. Neurofilament light chain (NfL) is one of the most studied biomarkers in a variety of neurological disorders, its concentration in the CSF reflects the ongoing pathology driving to axonal damage (Gaetani, Blennow et al., 2019). In MS NfL has been related to overall cognitive impairment (Gaetani, Salvadori et al., 2019;Kalatha et al., 2019) and also with specific cognitive domain impairment, mainly information processing speed (Kalatha et al., 2019;Modvig et al., 2015;Quintana et al., 2018).
Chitinase 3-like 1 protein (CHI3L1) has as well been investigated as a potential biomarker for cognitive impairment and it has been negatively correlated with information processing speed (Modvig et al., 2015;Quintana et al., 2018). Current findings indicate correlations between both NfL and CHI3L1 levels and cognitive performance measured at the same moment but evidence about their value for predicting cognitive outcomes at middle or long term is scarce. Identification of new and useful biomarkers to predict cognitive impairment in MS patients is desirable.
Local production of antibodies has been described in MS patients, and some have been identified as biomarkers present in CSF (Thangarajh et al., 2008). Since oligoclonal IgG bands (OCGB) are present in around 95% of MS patients  its presence is widely accepted as a diagnostic biomarker. Some authors have considered OCGB as a prognostic tool for the disease (Gaetani et al., 2021).
Regarding cognitive impairment, few studies report worse cognitive performance of those patients showing OCGB+ compared to those OCGB-and healthy controls (Anagnostouli et al., 2015;Farina et al., 2017).
Most of those patients (70%) show lipid-specific OCMB (LS-OCMB), which are directed against myelin lipids (Beltrán et al., 2012) and are more strongly related to an aggressive disease course than those that are not lipid-specific (Magraner et al., 2012;Thangarajh et al., 2008;Villar et al., 2003Villar et al., , 2008. The presence of LS-OCMB has been related to early conversion from clinically isolated syndrome to relapsingremitting MS and to a greater likelihood of developing secondary progressive MS. It has also been linked to a higher number of relapses, leading to greater physical disability, lesion load, and greater brain atrophy (Boscá et al., 2010;Ferraro et al., 2013;Magraner et al., 2012;Pfuhl et al., 2019;Thangarajh et al., 2008;Villar et al., 2005Villar et al., , 2008.
Given that cognitive impairment is a symptom of MS and that the presence of LS-OCMB in CSF is consistently related to a more aggressive disease course it seems reasonable that LS-OCMB could also predict poorer cognitive outcomes. To the best of our knowledge, this is the first data to be published about LS-OCMB and cognitive outcomes.

OBJECTIVE
The aim of this study is to investigate the LS-OCMB in CSF as a prognostic biomarker for cognitive impairment in the early stages of MS. Data from all linear regression models performed is shown in Table 2, only data corresponding to the variable "group" is shown.

Materials and methods
No significant differences between groups were observed in cognitive performance at follow-up except for the Selective Reminding Test-Long-Term Storage score in which the performance of LS-OCMB+ patients was significantly lower (β = −9.14, 95%CI −16.61 to −1.68; p = .018).

DISCUSSION
Efforts to find useful biomarkers related to cognition are emerging due to its clinical relevance and the possibility they offer of moving On the other hand, when assessing clinical utility of OCGB as a potential cognitive prognostic biomarker, we should consider that its high prevalence (95%) among MS patients  makes its discriminative power to be limited. OCMB, and particularly LS-OCMB, has been identified as a prognostic biomarker for poor clinical and radiological evolution (Ferraro et al., 2013;Magraner et al., 2012;Perini et al., 2006;Pfuhl et al., 2019;Thangarajh et al., 2008;Villar et al., 2003).
For all these reasons, we aimed to assess LS-OCBM as prognostic biomarker for cognition with follow-up cognitive measures. To our knowledge, this is the first longitudinal study investigating cognitive outcomes related to LS-OCMB.
No differences were found regarding number of relapses or EDSS in our cohort, unlike findings previously reported. This could be explained by the small cohort size in the present study. A significant difference of age was observed in this population, being LS-OCMB+ patients younger than LS-OCMB-patients. This finding is similar to Pfuhl's et al. (2019) data and in contradiction with other previous reports (Magraner et al., 2012;Thangarajh et al., 2008;Villar et al., 2010). It is important to note that Pfuhl et al. (2019) (Rocca et al., 2015) or active strategies used by patients to improve their performance in tasks (Sumowski et al., 2009), which are especially active in early stages of the disease. Thus CI seems to appear when compensatory mechanisms can no longer compensate brain damage (Sumowski & Leavitt, 2013). The small number of participants and possible practice effects, despite efforts to avoid this phenomenon, are limitations of the study.
Further investigation with a larger cohorts and longer disease duration is needed to clarify whether the presence of LS-OCMB can predict differences in long-term cognitive outcomes.

CONCLUSION
Whereas reports comparing the clinical and radiological evolution of LS-OCMB+ and LS-OCMB-MS patients have found significant differences in these respects, the present cognitive study does not find remarkable differences in the cognitive evolution of these two types of patient 4 years after diagnosis. Future studies with a longer follow-up period and larger cohorts are needed to clarify whether the presence of LS-OCMB can predict differences in long-term cognitive outcomes in MS.

FUNDING
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

CONFLICT OF INTEREST
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1002/brb3.2405

DATA AVAILABILITY STATEMENT
Data could be available under request.