Immune-inflammatory, metabolic and hormonal biomarkers are associated with the clinical forms and disability progression in patients with multiple sclerosis: A follow-up study

Journal Pre-proof Immune-inflammatory, metabolic and hormonal biomarkers are associated with the clinical forms and disability progression in patients with multiple sclerosis: A follow-up study

Wildéa Lice de Carvalho Jennings Pereira, Tamires Flauzino, Daniela Frizon Alfieri, Sayonara Rangel Oliveira, Ana Paula Kallaur, Andrea Name Colado Simão, Marcell Alysson Batisti Lozovoy, Damacio Ramón Kaimen-Maciel, Michael Maes, Edna Maria Vissoci Reiche PII:

S0022-510X(19)32395-0

DOI:

https://doi.org/10.1016/j.jns.2019.116630

Reference:

JNS 116630

To appear in:

Journal of the Neurological Sciences

Received date:

24 June 2019

Revised date:

5 December 2019

Accepted date:

10 December 2019

Please cite this article as: W.L. de Carvalho Jennings Pereira, T. Flauzino, D.F. Alfieri, et al., Immune-inflammatory, metabolic and hormonal biomarkers are associated with the clinical forms and disability progression in patients with multiple sclerosis: A followup study, Journal of the Neurological Sciences (2019), https://doi.org/10.1016/ j.jns.2019.116630

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier.

Journal Pre-proof

14

Immune- inflammatory, metabolic and hormonal biomarkers are associated with the clinical forms and disability progression in patients with multiple sclerosis: a follow-up study Wildéa Lice de Carvalho Jennings Pereira1,2 , Tamires Flauzino1 , Daniela Frizon Alfieri1 , Sayonara Rangel Oliveira1,3, Ana Paula Kallaur1 , Andrea Name Colado Simão1,3 , Marcell Alysson Batisti Lozovoy1,3 , Damacio Ramón Kaimen-Maciel2,4 , Michael Maes5,

6

, Edna

oo

1

f

Maria Vissoci Reiche1,3

Laboratory of Applied Immunology, Health Sciences Center, University of Londrina,

Outpatient Clinic for Neurology,

Londrina, Paraná, Brazil

Department of Pathology, Clinical Analysis and Toxicology, Health Sciences Center,

Pr

3

University Hospital, State University of Londrina,

e-

2

pr

Londrina, Paraná, Brazil;

al

University of Londrina, Londrina, Paraná, Brazil

Clinical Neurology, Santa Casa de Misericórdia de Londrina, Londrina, Paraná, Brazil

5

IMPACT Strategic Research Centre, School of Medicine, Deakin University, Geelong,

6

Department

of

Jo u

Victoria, Australia;

rn

4

Psychiatry,

King

Chulalongkorn

Memorial

Hospital,

Chulalongkorn,

Bangkok, Thailand

Corresponding author: Edna Maria Vissoci Reiche, Department of Pathology, Clinical Analysis and Toxicology, Health Sciences Center, Londrina State University, Av. Robert Koch, 60, CEP 86.038-440, Londrina, Paraná, Brazil. Phone/FAX number: + 55-43-3371-2619. e-mail: [email protected]; ORCID: 0000-0001-6507-2839 Abstract

Journal Pre-proof

15

The objective of this study was to evaluate the role of immune-inflammatory, metabolic, hormonal, and oxidative stress biomarkers in disability progression (DP) and clinical forms of multiple sclerosis (MS). The study evaluated 140 MS patients at admission (T0), and eight (T8) and 16 months (T16) later. The Expanded Disability Status Score (EDSS) and biomarkers were determined at T0, T8, and T16. A DP index (DPI) defined as an increase of ≥ 1 rank on the EDSS score indicated that 39.3% of the patients had significant DP. Quantification of the ordinal EDSS rank score was performed using optimal scaling methods.

oo

f

Categorical regression showed that the quantitative T16 EDSS score was predicted by T0 homocysteine (Hcy), T0 parathormone (PTH), T0 advanced oxidized protein products

pr

(AOPP) (all positively), low T0 vitamin D (<18.3 ng/mL) and T8 folic acid (<5 ng/mL)

e-

concentrations while higher T8 calcium concentrations (≥8.90 mg/dL) had protective effects.

Pr

Linear Mixed Models showed that the change in EDSS from T0 to T16 was significantly associated with changes in IL-17 (positively) and IL-4 (inversely) independently from the

al

significant effects of clinical MS forms, treatment modalities, smoking, age and systemic

rn

arterial hypertension. Hcy, PTH, IL-6, and IL-4 were positively associated with progressive

Jo u

versus relapsing-remitting MS while 25(OH)D was inversely associated. In conclusion, the ordinal EDSS scale is an adequate instrument to assess DP after category value estestimation. Aberrations in immune-inflammatory, metabolic and hormonal biomarkers are associated with DP and with the progressive form of MS.

Keywords: Multiple sclerosis; Inflammation; Disability; Homocysteine; Folic acid; Vitamin D

Journal Pre-proof

16

Introduction Multiple Sclerosis (MS) is a neuro-inflammatory, autoimmune-mediated disabling disease of the central nervous system (CNS) that affects, approximately, 2,5 million individuals worldwide [1]. The etiology of MS is not fully understood, but accumulating evidence suggests that the risk of MS, as well as the clinical course, are determined by interactions between a complex genetic background, breakdown in immune tolerance against myelin

and

neuronal

antigens,

dysfunction

of

the

blood-brain-barrier

(BBB)

and

oo

f

dysregulation of the immune response [2-4]. Moreover, several environmental risk factors for MS, such as vitamin D deficiency, tobacco smoking, Epstein–Barr virus (EBV) infection and

pr

dietary-intake are known to exert epigenetic effects [5-7].

e-

The presence of comorbidities, such as systemic arterial hypertension (SAH), type 2

Pr

diabetes mellitus (T2DM), and obstructive lung disease, but not hyperlipidemia showed an independent and cumulative impact on clinical disability measures [8,9]. Moreover, obesity,

al

metabolic syndrome (MetS) and tobacco smoking are associated with a delay in MS

rn

diagnosis and with the disability of MS patients [10].

inflammation,

Jo u

Different pathological mechanisms are involved in MS, including autoimmunity, demyelination,

neurodegeneration

with

axonal

and

neuronal

death,

astrogliosis, and metabolic alterations that are most likely responsible for the disease heterogeneity [11]. The innate immune system contributes to axonal loss in MS lesions through infiltrated macrophages, which produce pro-inflammatory cytokines, such as tumor necrosis factor (TNF)-α, interleukin (IL)-1 and IL-12, reactive oxygen species (ROS), glutamate and matrix metalloproteinases that exert effects on axonal integrity [12].

The

adaptive immune response in MS is characterized by increased expression of T helper (Th)1, Th2 and Th17 cytokines, as well as chemokines and their receptors [13]. While the demyelination and neurodegeneration occurring in MS are driven by Th1 and Th17 adaptive

Journal Pre-proof

17

immune response associated with disease initiation, Th2 and T regulatory (Treg) cells secreting anti-inflammatory cytokines (IL-4 and IL-10, respectively) are reported to be protective in MS [14]. The CNS is highly susceptible to ROS and reactive nitrogen species (RNS) and proteins, lipids and DNA are target of the oxidative and nitrosative stress initiating inflammatory processes, which contribute to myelin and oligodendrocytes destruction. These damage mechanisms also sustain the neurodegeneration in the chronic phase of disease [15-

oo

f

16].

The search for biomarkers in MS has been a very active field of research. Ferritin,

pr

albumin, lipid hydroperoxides, advanced oxidized protein products (AOPP), total antioxidant

e-

plasma capacity and nitric oxide metabolites (NOx) may predict MS diagnosis, whereas

Pr

albumin and AOPP may predict clinical forms of MS, such as relapsing-remitting MS (RRMS) and progressive clinical forms of MS (ProgMS) [17]. Dysregulation of the

al

concentration of molecules involved in different metabolic pathways, such as vitamin D [18],

rn

homocysteine (Hcy) [19-21] and folic acid [20] has been associated with MS. High levels of

Jo u

Hcy are associated with high disability scores [22]. However, other studies reported contradictory results [23,24]. The

evaluation

of

immune-inflammatory,

metabolic,

hormonal,

oxidative

and

nitrosative stress biomarkers during clinical, laboratory and treatment follow-up of patients with MS in different clinical forms are scarce. Likewise, the evaluation of a biomarker alone may not reflect the possible correlations between the different biomarkers, which could partly contribute to a better understanding of the pathophysiology of MS. Hence, the aims of the present study were to delineate the associations between immune-inflammatory, metabolic, hormonal, oxidative and nitrosative stress biomarkers and clinical forms of MS and disability progression (DP) during 16 months-follow-up.

Journal Pre-proof

18

Materials and Methods Study subjects and clinical characteristics An observational prospective study was carried out with 140 eligible MS adult patients, both sexes, enrolled from the Specialized Outpatient of University Hospital (AEHU) of Londrina, Southern Brazil, from December 2014 to March 2017. MS diagnosis was made according to the McDonald Criteria [25] and the MS patients were classified as RRMS,

oo

f

primary progressive MS (PPMS) and secondary progressive MS (SPMS) clinical forms [26]. The patients were evaluated at the time of the inclusion or baseline (T0), eight month-follow-

pr

up (T8), and 16 month-follow-up (T16). During the follow-up, the scheduling was planned by

e-

the study team who called directly to the patients.

Pr

All MS patients were in the remission clinical phase, defined as the period of recovery with no relapse episodes within the last three months prior to the time of study enrollment.

al

Phrased differently, patients who suffered from flare-ups were excluded to participate and

rn

consequently all patients included here were in a remission clinical phase during the follow-

Jo u

up period. All of them were treated according to the Brazilian MS treatment guideline [27] and the use of MS therapies was recorded in each clinical evaluation. Exclusion criteria were MS patients with diagnosis of other autoimmune or infectious diseases and treatment with any vitamin and/or antioxidant supplement that could alter the laboratory biomarkers that were evaluated. The disability was evaluated in each clinical evaluation (T0, T8 and T16) using the Expanded Disability Scale Status (EDSS) [28]. The (DP) was assessed as a the DP index (DPI) with an increase of at least 1 rank on the EDSS rank score indicating DP; and b) the confirmed disability progression (CDP), which includes a CDP diagnosis for patients with an ordinal rank EDSS change of ≥1.0 when
Journal Pre-proof

19

At the T0 visit (baseline), demographic, epidemiological, anthropometric and clinical data were taken by standard questionnaire. Ethnicity was classified according to the individual’s self-perception of skin color as Caucasian and non-Caucasian. Body mass index (BMI) and waist circumference (WC) were also recorded; systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured twice and the mean of these two measurements was used in the analysis. Moreover, use of antihypertensive medication was an indication of systemic arterial hypertension (SAH) [30]. T2DM was defined as a fasting

oo

f

serum glucose ≥126 mg/dL, a non-fasting serum glucose ≥ 200 mg/dL and/or use of hypoglycemic medication [31]. Dyslipidemia and MetS were defined as previously reported

pr

[32, 33].

e-

The protocol was approved by the Institutional Research Ethics Committees of

Pr

University of Londrina, Paraná, Brazil (CAAE: 22290913.9.0000.5231) and all of the individuals invited were informed in detail about the research and gave written Informed

rn

al

Consent.

Jo u

Blood samples and biomarkers

Venous blood samples were drawn after fasting for 12 hours, with and without anticoagulants at baseline (T0) and during the follow-up T8 and T16. The samples were consecutively and anonymously coded and centrifuged at 2,500 rpm for 15 min. Further, plasma and serum aliquots were stored at -800 C until used. The inflammatory biomarkers were evaluated through C-reactive protein, determined with high sensitivity assay (hsCRP) using turbidimetry (Architect C8000, Abbott Laboratory, Abbott Park, IL, USA),

ferritin,

determined with chemiluminescent microparticle assay (CMIA) (Architect i2000, Abbott Laboratory, Abbott Park, IL, USA) and plasma levels of cytokines IL-2, IL-6, IL-17, interferon (IFN)-γ, IL-4, and IL-10, determined using immunofluorimetric method with

Journal Pre-proof

20

microspheres multiplex immunoassay (Novex Life Technologies, Frederick, USA) for Luminex plataform in MAGPIX® instrument (Luminex Corp., TX, USA). Metabolic biomarkers were evaluated through glucose, total cholesterol, high-density lipoprotein (HDL)-cholesterol,

low-density lipoprotein (LDL)-cholesterol,

triglycerides,

and calcium

using a biochemical autoanalyzer (Dimension Dade AR Dade Behring, Deerfield, IL, USA); Hcy, parathormone (PTH), folic acid, and 25-hydroxyvitamin D [25(OH)D] serum levels were determined using CMIA (Architect i2000, Abbott Laboratory, Abbott Park, IL, USA).

oo

f

Oxidative stress was evaluated using advanced oxidative protein products (AOPP) as

pr

previously described [34].

e-

Statistical Analysis

Pr

We used analysis of variance (ANOVA) to assess intergroup differences in continuous variables and analysis of contingency tables (x2 -test) to check associations

al

between categorical variables. Because EDSS is measured as an ordinal scale we used

rn

category value estimation techniques to quantify the ordinal EDSS data employing optimal

Jo u

scaling, including category regression (CATREG), and an estimation based on the observed frequencies [35]. The latter method considers that the expected value of variable X in category j is the mean of the density functions corresponding to category j. Thus, the quantitative EDSS score of category j equals CFj-1 + CFj / 2 with CF being the cumulative frequency of the j category [35]. Linear Mixed Model (LMM) analysis, repeated measurements, was used to examine the associations between the three EDSS scores (from T0 to T8 to T16) and the biomarkers, while adjusting for the relevant extraneous variables that could interfere in the biomarkers included in the study. Toward this end we adjusted the results for time, age, sex, clinical MS forms, MS treatments, BMI, smoking and SAH. We employed a manual stepwise model selection approach and evaluated the goodness of fit of

Journal Pre-proof

21

the different models using the Bayesian information criterion (BIC). We used an autoregressive repeated covariance matrix and the maximum likelihood method. Two random effects were included namely intercept and patient identification. Residual plots (residual versus explanatory variables, residuals in order, residuals versus fitted values, histogram and probability plots) were inspected to evaluate the fitness of the model. CATREG analysis was used to examine the associations between T16 EDSS (dependent variable with numeric transformation as optimal scaling level) and selected biomarkers at T0 and T8 (entered in

oo

f

numerical or nominal transformations as the optimal scaling level), while adjusting for the relevant extraneous variables (toward this end we entered age, sex, clinical MS forms, MS

pr

treatments, BMI, smoking and SAH in the analysis). Transformation plots were used to

e-

interpret the quatifications of the explanatory variables and, therefore, their contributions to

Pr

the predicted dependent variable. We employed multivariate general linear model (GLM) analysis to assess the differences in biomarker data between subgroups (EDSS groups or

al

RRMS versus SPMS+PPMS) while controlling for extraneous variables (age, sex and BMI).

rn

Consequently, tests for between-subject effects were performed to examine the effects of

Jo u

significant explanatory variables on the dependent variables. We employed binary logistic regression analysis to define the most significant biomarkers predicting SPMS+PPMS versus RRMS while adjusting for the relevant confounders (including the extraneous variables listed above and self-declared ethnicity, T2DM, and duration of illness). Regression analyses were checked for multicollinearity. The data were transformed in z-scores (z) (computed on the population included in the analysis) to compare scores that are from different normal distributions, units or scales. All statistical analyses were performed employing IBM SPSS Windows version 25. Tests were 2-tailed, and an alpha level of 0.05 indicated a statistically significant effect.

Journal Pre-proof

22

Results Demographic and clinical characteristics of subjects At T0 (baseline), 140 patients were evaluated, 119 with RRMS, 18 with SPMS and 3 with PPMS; of them, 128 were evaluated at T8 (108 with RRMS and 20 with progressive

oo

f

clinical forms) and 12 did not respond to scheduling for clinical and laboratory evaluation and were discontinued in the study. At T16, 122 patients (103 with RRMS and 19 with

pr

progressive clinical forms) attended the appointment for the clinical and laboratory

e-

evaluation and six patients were discontinued in the study (one patient moved to another

Pr

country, the MS clinical diagnosis changed in one patient; one patient was hospitalized in the intensive care unit, while two others did not respond to the invitation). Therefore, we

al

evaluated 119 MS patients who had repeated measurements at T0, T8 and T16. At T0, five

rn

patients were without treatment for MS; 74 were using IFN-β1a, 31 were using glatiramer

Jo u

acetate, seven were using natalizumab, and two were using fingolimod. At T8, 5 patients were without treatment for MS; 58 were using IFN-β1a, 35 were using glatiramer acetate, 10 were using natalizumab, 10 were using fingolimod, and 1 was using dimethyl fumarate. At T16, 7 patients were without treatment for MS; 44 were using IFN-β1a, 28 were using glatiramer acetate, 12 were using natalizumab, 25 were using fingolimod, and 1 was using dimethyl fumarate and 2 were using teriflunomide.

Quantification of the EDSS rank scores. Table 1 shows the quantification of the endpoint T16 EDSS score (as this is the final outcome variable of this research) in patients with MS using a quantification method based

Journal Pre-proof

23

on the observed frequencies. This table shows the T16 EDSS using the transformed category value estimates. There were highly significant Pearson’s product moment correlations between the ordinal EDSS rank and the quantitative EDSS scores (r=0.998).

DP from baseline to 16 months later Table 2 shows the demographic and clinical characteristics of the 119 MS patients divided into two groups according to the DPI. We found that 47 patients showed DP while 72

oo

f

patients did not. There were no significant differences in baseline T0 EDSS values between patients with and without a positive DPI while the T8 and T16 EDSS values were

pr

significantly higher in the DP group than in the non-DP group. There were no significant

e-

differences in duration of illness, BMI, WC, sex, ethnicity, MetS, SAH, and the

Pr

RRMS/SPMS+PPMS ratio between the two study groups. In the study sample with a positive

al

DPI, age was somewhat higher, and there were significantly more smokers.

rn

Effects of time on the biomarkers

Jo u

Table 3 shows the outcome of LMM analysis, repeated measurements, conducted in the 119 patients who had repeated measurement of clinical and biomarker data. The first LMM shows the quantitative EDSS score at T0, T8 and T16 as dependent variable and time, DPI and the time by DPI interaction as explanatory variables. We found a significant effect of time with significant differences in the quantitative EDSS scores among the three time points with increasing scores from T0  T8  T16. There was also a significant interaction time by DP indicating that in patients with DP, the quantitative EDSS score increased while in those without DP the quantitative EDSS score decreased over time. Table 3 also shows the results of LMM analyses with the laboratory biomarkers (from T0 to T8 to T16) as dependent variables and time as primary explanatory variables

Journal Pre-proof

24

(while adjusting for the significant explanatory variables including age, smoking, sex and BMI). We found that Hcy levels significantly decreased over time with significant differences between the three time points. Folic acid was significantly lower at T16 than at T0. PTH was significantly higher at T16 than at T0 and T8, while calcium was significantly different between the three time points and increased from T0  T8  T16.

IL-4 was

significantly lower at T16 than T0, while AOPP decreased from T0  T8  T16. There was

oo

f

no significant effect of time on vitamin D, hsCRP, IL-6 and IL-17.

Associations between biomarkers and EDSS values over time

pr

In order to examine the associations over time (from baseline T0 to T16) between the

e-

biomarkers (explanatory variables) and the quantitative EDSS values we have performed

Pr

LMM analysis with the EDSS scores over time as repeated measures, while adjusting for extraneous variables including clinical forms of MS, treatment groups, SAH, age, sex,

al

T2DM, BMI (only the significant predictors are included). Table 4 shows the results of

rn

LMM analysis, repeated measures, indicating that the quantitative EDSS score was

Jo u

significantly associated with SAH, age, IL-17 and AOPP (all positively) and IL-4 (negatively). The quantitative EDSS score was significantly higher in SAH (mean ±SE= 2.603 ±0.237) than in those without SAH (2.105 ±0.155). After deleting the clinical forms of MS from this analysis (these diagnosis are higher order constructs based on clinical symptoms, progression, and pathways) we found that the clinical forms were replaced by smoking (F=5.92, df=1/96.94, p=0.017), with higher levels in smokers (mean ±SE= 2.634 ±0.252) than in non-smokers (2.073 ±0.143).

Prediction of the quantitative T16 EDSS score using CATREG

Journal Pre-proof

25

Table 4 shows the outcome of different CATREG analyses with the quantitative T16 EDSS scores as dependent variables and T0/T8 biomarkers with and without treatment or T0 EDSS as additional explanatory variables. We built four CATREG models, the first two including only biomarkers, a third with biomarkers and the significant background variables (entered were age, sex, treatment, MS forms, HAS, BMI, smoking and T2DM), and a fourth CATREG with biomarkers and T0 EDSS. The first CATREG analysis showed that 24.3% of the variance in the optimized T0 EDSS score was explained by T0 PTH and T0 Hcy (both

oo

f

numerical and positively associated with the EDSS score) and T8 folic acid and T8 calcium (both scaled at a nominal level); very low folic acid levels (<5 ng/mL, n=14) have the highest

pr

quantifications and thus increase the predicted EDSS values while the highest calcium values

e-

(≥8.90 mg/dL, n=31) show the lowest quantifications and thus lower the predicted EDSS

Pr

score. The second CATREG shows that 25.2% of the variance in the optimized scaled EDSS score was explained by 4 biomarkers, namely T0 Hcy, T0 PTH and T0 AOPP (numerical,

al

and positively associated) and T8 calcium with the highest calcium values decreasing the

rn

quantifications and thus the predicted EDSS values. We also found that 6.8% of the variance

Jo u

in the optimal scaled EDSS score was explained by the regression on T0 25(OH)D levels (β=0.292, bootstrap error=0.097, F=8.98, p<0.001) with very low 25(OH)D concentrations (<18.3 ng/mL) strongly increasing the quantifications and, therefore, the predicted scaled EDSS score.

The third CATREG shows that 35.9% of the variance in the optimized numeric EDSS scores could be explained by the regression on treatment modalities and the same predictors as shown in regression #2. Age, sex, BMI, smoking, SAH, clinical forms and T2DM were not significant in this CATREG. The fourth CATREG regression shows that 62.2% of the variance in the quantitative T16 EDSS scores was explained by a) T0 EDSS, T0 PTH and T0

Journal Pre-proof

26

Hcy (all numerical and positively associated). In those 4 CATREGs all tolerance measures were very high (>0.90) indicating that no multicollinearity was present.

Differences between RRMS and SPMS+PPMS Table 6 shows the best prediction of SPMS+PPMS (dependent variable) using biomarkers and other possible clinical explanatory variables (ethnicity, sex, duration of illness, T2DM, and BMI) using stepwise logistic regression. We found that SPMS+PPMS

oo

f

was best predicted by ethnicity (being not Caucasian), male sex, duration of illness, T2DM, PTH and IL-6 (all positively associated) and BMI (negatively). With this model, 86.3% of the

pr

SPMS+PPMS patients were correctly classified with a sensitivity of 72.1% and specificity of

e-

89.4%. After removing both biomarkers and duration of illness as explanatory variables we

Pr

also detected that IL-4 and Hcy (both positively) and 25(OH)D (inversely) were associated with SPMS+PPMS. Figure 1 shows the z values of all biomarkers in those two MS clinical

al

forms. Hcy, PTH, IL-6 and IL-4 were significantly and positively associated with

Discussion

Jo u

rn

SPMS+PPMS, while 25(OH)D was negatively associated with SPMS+PPMS versus RRMS.

The first major finding of this study is that the EDSS score can reliably be employed to assess DP by computing quantitative EDSS scores and the DPI. In the current study we computed two quantitative EDSS scores which showed a uniform distribution all along the EDSS categories. Parametric statistical analyses may be applied on those quantitative EDSS scores, whereas applying these statistical methods on the ordinal EDSS rank score may be criticized. Importantly, here we propose that an increase of at least one rank on the ordinal EDSS rank score may be used as an adequate criterion for DP. Indeed, patients allocated to the DP subgroup show a significant increase in the quantitative EDSS score from baseline to

Journal Pre-proof

27

16 months later. Importantly, all patients included in our study were in a remission clinical phase during the follow-up period thereby precluding that the increases in the endpoint EDSS scores may be due to the effects of acute relapses. Previously, a similar disability index was employed by the IFNB Multiple Sclerosis Study Group and the University of British Columbia MS/MRI Analysis Group (1995) who considered that a patient with a persistent increase of one or more EDSS points shows progression in disability [36]. Previously, another index has been proposed, namely

the CDP [29]. However, the definitions of CDP vary across trials, both in the magnitude of

oo

f

the change in EDSS score that constitutes progression, and in the time over which this change must be sustained [29]. Furthermore, our results show that the quantitative EDSS scores are

pr

uniformly distributed in patients with lower as well as higher EDSS scores and, therefore,

e-

that the use of two different cut-offs (as the CPD criteria propose) cannot be validated.

Pr

The second major finding of the present study is that a set of immune-inflammatory, metabolic, hormonal, and oxidative stress biomarkers was strongly associated with the

al

quantitative T16 EDSS score and changes in disability over time from baseline to endpoint

rn

(as assessed by the regression of T16 EDSS on T0 EDSS). The most important biomarkers of the endpoint EDSS score were in descending order of importance (based on partial

Jo u

correlations with the T16 EDSS score) T0 Hcy, T0 PTH, T8 calcium, T8 folic acid, and T0 AOPP. Important biomarkers of the changes in the quantitative EDSS score (as assessed using LMM or regression analysis with T0 EDSS as covariate) were Hcy, PTH, IL-17, AOPP (all positively) and IL-4 (negatively). The association between increased IL-17 and Hcy and lowered IL-4 and folic with the changes in disability over time are consistent with the immune imbalance and inflammatory and metabolic mechanisms that modulate pathophysiological cascades in the CNS followed by the destruction of myelin and axonal death. Other important findings of the present study are that some baseline characteristics of MS patients, such as ethnicity (no Caucasian), sex (male), duration of illness and the presence

Journal Pre-proof

28

of T2DM together with high levels of PTH and IL-6, as well as higher IL-4 and Hcy, but lower 25(OH)D are associated with ProgMS. These results deserve to be discussed in detail. The immune-inflammatory response mediated by Th17 cells plays a key role in the pathogenic mechanisms of MS [37]. Increased IL-17 expression has been detected in peripheral blood mononuclear cells of MS patients during disease relapses. Human endothelial cells from patients with MS express high levels of IL-17 receptors, which could facilitate Th17 infiltration into the CNS [38]. Higher IL-17-secreting T cells were detected in

oo

f

the peripheral blood of patients with MS during the clinical remission phase as compared with healthy individuals, and in vitro IL-17 levels were directly associated with disability

pr

[39].

e-

IL-6 has been implicated in the induction of pathogenic IL-17-producing T cells in

Pr

autoimmune diseases. In patients with active RRMS, IL-6 signaling was shown to support T effector cell resistance to regulation by regulatory T cells (Treg), which may contribute to

al

disease aggravation [40]. In both EAE and MS, IL-6 seems to affect the disease pathogenesis

rn

through its activity in the peripheral lymphoid organs [41].

Jo u

The present study showed that there were significantly more smokers in the group with increasing DP over time. These findings are in agreement with those of previous reports [10,42, 43]. Smoking was associated with the development of MS and disability, as well as DP [42]. Moreover, smoking has been demonstrated to impress epigenetic effects promoting or hindering the kinetics of gene expresion [44]. Another important finding of the present study was that higher T0 Hcy and lower T8 folic acid levels significantly predicted the endpoint T16 EDSS score, while increasing Hcy concentrations are associated with the changes in the EDSS score from baseline (T0) to 16 months later (T16). These findings are in agreement with previous studies, in which patients with ProgMS showed higher plasma levels of Hcy than those with RRMS [18]. Patients with

Journal Pre-proof

29

higher Hcy showed a faster progression of MS and more disability compared to patients with lower Hcy [45]. Hcy can induce BBB disruption [46] and may cause neuronal damage through increased production of ROS [47], promote excitotoxicity via stimulation of Nmethyl-D-aspartate receptors (NMDA), damage neuronal DNA, and trigger apoptosis [48]. Hyperhomocysteinemia resulting from poor reconversion to methionine is related to low methionine availability. As methionine is an important methyl group donor in several biochemical processes, the hypomethylation of arginine of myelin basic protein (MBP)

enhancing degeneration of the myelin sheath [21].

oo

f

decreases the hydrophobicity of MBP and gives rise to less stable myelin structures thereby

pr

The association of low T8 folic acid levels and endpoint EDSS score in our MS

by

this compound

controlling the inflammatory response in a variety of

Pr

exerted

e-

patients with increased disability over time may underscore the anti-inflammatory effects

inflammatory-related diseases [49]. Like methionine, folic acid acts as a methyl donor

al

precursor for DNA methylation, a key regulatory mechanism behind some inflammatory

rn

processes. Folic acid may modulate the inflammatory response in microglia shifting them

Jo u

toward an anti-inflammatory phenotype through regulating NF-kB related pathways [50]. In the present study, the higher serum levels of PTH among patients with ProgMS (compared with those with RRMS) and the association between increased PTH and increasing disability over time are consistent with the findings of a previous study [51]. These authors showed that newly diagnosed MS patients had the lowest serum levels of PTH compared to those observed in patients at a more advanced stage of the disease as well as in the control group while an increase in serum PTH was observed with increased relapses. PTH stimulates IL-6 production by osteoblasts and liver cells, and IL-6 may modulate acute phase protein synthesis in the liver. Patients with hyperparathyroidism theoretically have higher levels of IL-6, CRP, or TNF-α than those with normal PTH levels. However, the casual

Journal Pre-proof

30

relationship between PTH levels and inflammation remains to be elucidated [52]. In theory, many factors may interfere with PTH values by impairing renal functions and thus PTHcalcium-phosphate

metabolism

including

age,

smoking,

and

the

tendency

towards

comorbidities including SAH and T2DM [53,54]. Nevertheless, our PTH results were adjusted for the effects of these background variables indicating that PTH may have a genuine effect on DP. Comorbidities, such as hypertension and T2DM, are known to affect MS patients in a

oo

f

number of ways, including delaying time to diagnosis and reducing health-related quality of life [9]. Underscoring this statement, another significant finding of the present study was that

pr

changes in disability over time were associated with SAH, which is in agreement with

e-

previous reports [9, 55, 56]. One of the mechanisms that may explain the unfavorable

Pr

outcome of hypertension in MS patients is the increased risk of neurodegenerative processes as evidenced by accelerated brain atrophy [57]. Moreover, patients with hypertension showed

al

a higher percentage of lateral ventricle volume change compared to non-hypertensive patients

rn

[8] while anti-hypertensive medications may influence the disease course and hypertension

Jo u

may decrease the effectiveness of MS therapies [9]. Regarding T2DM, the results obtained here are in agreement with our previous study in which 40% of MS patients had insulin resistance and showed higher disability than those without insulin resistance [58]. In the present study we found that changes in EDSS scores were significantly associated with IL-17 (positively) and IL-4 (negatively) independently of hypertension, age, MS treatment and clinical forms. These results extend those of previous studies, which reported higher IL-4 levels among RRMS patients with mild disability compared with those with moderate/severe disability [14]. Furthermore, we also reported that patients with ProgMS presented higher levels of IL-1β, IL-6, TNF-α, IFN-γ, IL-17, IL-4, and IL-10 than controls [59].

Journal Pre-proof

31

In the present study, the T16 EDSS score was significantly predicted by very low 25(OH)D concentrations (<18.3 ng/mL), which therefore is a risk factor for DP. Our findings underscore the possible use of this biomarker as a predictor of DP, as previously reported [17,60]. Moreover, vitamin D levels were inversely associated with MS activity as measured by brain magnetic resonance imaging [61]. It was also shown that vitamin D deficiency may contribute to increased EDSS scores and may be one of the predictors of disability in MS patients independently of the redox status [17]. In addition, low MS relapse rate was observed

oo

f

in patients with higher levels of vitamin D than those with lower levels of vitamin D [62,63]. Some mechanistic explanations are that 1,25(OH)D inhibits Th17 cells differentiation via

pr

regulating NF-kB activity and expression of IL-17 [64]; both 25(OH)D and 1,25(OH)D

e-

modulate Th17 effector responses and enhance regulatory function of CD4 +T cells, as well as

Pr

that vitamin D, drives CD4+ T cells to a CD25hi FoxP3+CTLA+ phenotype, which has immunosuppressive and regulatory functions [65]. Vitamin D also promotes the secretion of

al

the anti-inflammatory cytokine transforming growth factor (TGF)-β1 and suppresses the pro-

Jo u

producing cells [65].

rn

inflammatory cytokines IL-17 and IFN-γ, reducing the frequency of IL-17A and IFN-γ

In the present study, we found that the ProgMS clinical forms (SPMS+PPMS) were positively associated with male sex and the presence of T2DM. MS is more common in women than men, with the female to male ratio as high as 3:1 and females have more relapses than men [66, 67]. However, male sex has been associated with a poorer clinical outcome in relapse-onset cohorts and men are reported to have a more rapid accumulation of disability [68, 69]. Regarding BMI, patients of the present cohort with low BMI values would be less likely to develop ProgMS. While obesity in childhood and during adolescence been reported as a risk factor for MS [70,71], data are conflicting concerning the relationship between BMI

Journal Pre-proof

32

and disease severity in MS. A cross-sectional study found that BMI had a modest correlation with MS severity and symptom in women [72] while a larger study showed that obesity was correlated with worse disability in women, but not in men [73]. Adipose tissue is a major source of pro-inflammatory and anti-inflammatory adipokines; therefore, can have both harmful and protective effects, respectively. The relationship between BMI and mortality rate results in a J-shaped curve, which shows that both high BMI and very low BMI are associated with increased mortality [74]. While BMI is an indicator of the overall body

oo

f

adiposity, it does not reflect the adipose capacity to shift toward to an anti-inflammatory status through adiponectin, one of the adipokines that may modulate the inflammatory

pr

response [74, 75].

e-

The present study has some limitations that need to be addressed. First, the possible

Pr

effects of uncontrolled confounder variables including variant alleles in genes coding for immune-inflammatory, oxidative and metabolic molecules evaluated in this study, as well as

al

lifestyle conditions (including the individual variability of alcohol consumption, physical

rn

activity, dietary habits, and sun exposure). Second, although we excluded patients with acute

Jo u

flare-ups during the study period, it is possible that a temporary worsening of the symptoms in some patients could have affected the endpoint EDSS scores. Nevertheless, such effects, if present, would have induced more variability in the T16 EDSS data thereby decreasing the effect sizes, which estimate the prediction of the T16 EDSS score by biomarkers measured 16 and 8 months earlier. Third, it would have been even more interesting if we had measured neurotoxic

chemokines (including CCL2 and CCL11) and vitamin D receptor levels. On the other hand, the strengths are the longitudinal design with a 16-month follow- up.

Conclusions

Journal Pre-proof

33

Taken together, the results showed that immune-inflammatory, metabolic, hormonal, and oxidative biomarkers are associated with the changes in disability over time of MS patients when evaluated during a 16-month follow-up. DP was predicted by higher levels of PTH, AOPP, IL-17, Hcy and by lower levels of folic acid, 25(OH)D and IL-4, while higher calcium levels have a protective effect. Moreover, high levels of PTH, IL-6 and IL-4 and low levels of 25(OH)D are significantly associated with ProgMS. Our results suggest that those immune-inflammatory, metabolic and hormonal pathways are possible new targets for

oo

f

individualized therapies, which when associated with the classical treatments for MS may

e-

pr

modulate the pathophysiological mechanisms underpinning disease progression in MS.

Pr

Conflict of interest

al

The authors declare that they have no conflict of interest.

rn

Financial support

development

Jo u

This study is partially financially supported by Novartis Biosciences S.A for the of

the

research

according

to

the

Researcher's

Initiative

Study

CFTY720DBR07T. The authors do not receive any reimbursement or financial benefits and declare that they have no competing interests. Novartis Biosciences S.A. played no role in the design, methods, data management or analysis or in the decision to publish. This study was also financed, in part, by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES), Finance Code 001.

Ethical approval

Journal Pre-proof

34

The protocol was approved by the Institutional Research Ethics Committees of University of Londrina, Paraná, Brazil (CAAE: 22290913.9.0000.5231) and all of the individuals invited were informed in detail about the research and gave written Informed Consent.

References

T, Katrajian

R, Danoff R

Prim

pr

Care 42(2):159-75. doi: 10.1016/j.pop.2015.01.007.

(2015) Multiple sclerosis.

f

Files DK, Jausurawong

oo

1.

up-to-date review.

doi: 10.1111/j.1600-

NA, Xifara

DK, Davis

MF, Kemppinen

A,

International

rn

Beecham AH, Patsopoulos

al

065X.2012.01134.x.

3.

Immunol Rev 248(1):87-103.

Pr

sclerosis: an

e-

2. Gourraud PA, Harbo HF, Hauser SL, Baranzini SE (2012) The genetics of multiple

Jo u

Multiple Sclerosis Genetics Consortium (IMSGC), et al (2013) Analysis of immune-related loci identifies 48 new susceptibility variants for multiple sclerosis. Nat Genet 45(11):135360. doi: 10.1038/ng.2770.

4. Olsson T, Barcellos LF, Alfredsson L (2017) Interactions between genetic, lifestyle and environmental risk factors for multiple sclerosis. Nat Rev Neurol 13(1):25-36. doi: 10.1038/nrneurol.2016.187.

5. Swank RL, Dugan BB (1990) Effect of low saturated fat diet in early and late cases of multiple sclerosis. Lancet 336(8706):37–9.doi: 10.1016/0140-6736(90)91533-g

Journal Pre-proof

35

6. Ramagopalan SV, Dobson R, Meier UC, Giovannoni G (2010) Multiple sclerosis: Risk factors, prodromes, and potential causal pathways. Lancet Neurol 9(7):727–39. doi: 10.1016/S1474-4422(10)70094-6.

7.

Küçükali

Cİ, Kürtüncü

M, Çoban

A, Çebi

M, Tüzün

E

(2015)

Epigenetics of multiple sclerosis: an updated review. Neuromolecular Med 17(2):83-96. doi:

oo

f

10.1007/s12017-014-8298-6.

pr

8. Jakimovski D, Gandhi S, Paunkoski I, et al (2019) Hypertension and heart disease are

e-

associated with development of brain atrophy in multiple sclerosis: a 5-year longitudinal

Pr

study. Eur J Neurol 26(1): 87-e8. doi: 10.1111/ene.13769.

al

9. Conway DS, Thompson NR, Cohen JA (2017) Influence of hypertension, diabetes,

rn

hyperlipidemia, and obstructive lung disease on multiple sclerosis disease course. Mult Scler

Jo u

23(2):277-85. doi: 10.1177/1352458516650512.

10. Overs S, Hughes CM, Haselkorn JK, Turner AP (2012) Modifiable comorbidities and disability in multiple sclerosis. Curr Neurol Neurosci Rep 12(5):610-7. doi: 10.1007/s11910012-0293-4.

11. Sospedra M, Martin R (2016) Immunology of Multiple Sclerosis. Semin Neurol 36(2):115–27. doi: 10.1055/s-0036-1579739.

Journal Pre-proof 12.

36

Hendriks JJ, Teunissen CE, de Vries HE, Dijkstra CD (2005) Macrophages and

neurodegeneration.

Brain

Res

Brain

Res

Rev

48(2):185-95.

doi:

10.1016/j.brainresrev.2004.12.008

13. Mikulkova Z, Praksova P, Stourac P, Bednarik J, Michalek J (2011) Imbalance in T-cell and cytokine profiles in patients with relapsing-remitting multiple sclerosis. J Neurol Sci

oo

f

300(1-2):135-41. doi: 10.1016/j.jns.2010.08.053.

14. Kallaur AP, Oliveira SR, Colado Simão AN, et al (2013) Cytokine profile in relapsing-

pr

remitting multiple sclerosis patients and the association between progression and activity of

Pr

e-

disease. Mol Med Rep 7(3):1010-20. doi: 10.3892/mmr.2013.1256.

15. Adamczyk B, Adamczyk-Sowa M (2016) New insights into the role of oxidative stress

al

mechanisms in the pathophysiology and treatment of multiple sclerosis. Oxid Med Cell

Jo u

rn

Longev 2016:1973834. doi.org/10.1155/2016/1973834.

16. Ohl K, Tenbrock K, Kipp M (2016) Oxidative stress in multiple sclerosis: Central and peripheral mode of action. Exp Neurol 277:58-67. doi: 10.1016/j.expneurol.2015.11.010.

17. Oliveira SR, Kallaur AP, Reiche EMV, et al (2017) Albumin and Protein Oxidation are Predictors

that

Differentiate

Relapsing-Remitting

from Progressive Clinical Forms of

Multiple Sclerosis. Mol Neurobiol 54(4):2961-8. doi: 10.1007/s12035-016-9860-z.

Journal Pre-proof

37

18. Oliveira SR, Simão ANC, Alfieri DF, et al (2017) Vitamin D deficiency is associated with disability and disease progression in multiple sclerosis patients independently of oxidative and nitrosative stress. J Neurol Sci 381:213-9. doi: 10.1016/j.jns.2017.07.046.

19. Oliveira SR, Flauzino T, Sabino BS, et al (2018) Elevated plasma homocysteine levels

oo

f

are associated with disability progression in patients with multiple sclerosis. Metab Brain Dis

Moghaddasi M,

Mamarabadi M,

Mohebi N,

e-

20.

pr

33(5):1393-9.

Razjoouyan,

Aghaei M (2013)

Pr

Homocysteine, vitamin B12 and folate levels in Iranian patients with Multiple Sclerosis: A

al

case control study. Clin Neurol Neurosurg 115(9):1802–5.

rn

21. Ramsaransing GS, Fokkema MR, Teelken A, Arutjunyan AV, Koch M, De Keyser J

77:189-92.

Jo u

(2006) Plasma homocysteine levels in multiple sclerosis. J Neurol Neurosurg Psychiatry

22. Guzel I, Mungan S, Oztekin ZN, Ak F, et al (2016) Is there an association between the Expanded Disability Status Scale and inflammatory markers in multiple sclerosis? J Chin Med Assoc 79(2):54–7.

23.

Kararizou

E,

Paraskevas

G, Triantafyllou

Plasma homocysteine levels in patients with multiple sclerosis in Chin Med Assoc 76(11):611-4.

N,

et

al.

the Greek population.

(2013). J

Journal Pre-proof

38

24. Teunissen CE, Killestein J, Kragt JJ, et al (2008) Serum homocysteine levels in relation to clinical progression in multiple sclerosis. J Neurol Neurosurg Psychiatry 79(12):1349-53.

25. Polman CH, Reingold SC, Banwell B, et al (2011) Diagnostic Criteria for Multiple Sclerosis: 2010 Revision to the McDonald Criteria. Ann Neurol 69:292-302.

26. Lublin FD, et al. (2014) Defining the clinical course of multiple sclerosis: The 2013

oo

f

revisions. Neurology 83(3):278-86.doi: 10.1212/WNL.0000000000000560.

pr

27. Marques VD, Passos GRD, Mendes MF, et al (2018) Brazilian Consensus for

e-

the Treatment of Multiple Sclerosis: Brazilian Academy of Neurology and Brazilian Committ Arq

Neuropsiquiat

76(8):

539-54.

doi:

Pr

ee Treatment and Research in Multiple Sclerosis.

al

10.1590/0004-282X20180078.

rn

28. Kurtzke JF (1983) Rating neurologic impairment in multiple sclerosis: an expanded

Jo u

disability status scale (EDSS). Neurol 33(11):1444-52. doi: 10.1212/wnl.33.11.1444.

29. Wiendl H, Meuth SG (2015) Pharmacological Approaches to delaying disability progression in patients with multiple sclerosis. Drugs 75(9):947-77. doi: 10.1007/s40265015-0411-0.

30. James PA, Oparil S, Carter BL, et al (2014) Evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth

Joint

National

10.1001/jama.2013.284427.

Committee

(JNC

8).

JAMA

311(5):507–20.

doi:

Journal Pre-proof

39

31. American Diabetes Association (2014) Standards of medical care in diabetes--2014. Diabetes Care 37(Suppl 1):S14-80. doi: 10.2337/dc14-S014.

32. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High blood cholesterol in adults (Adult Treatment Panel III) (2002) Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection,

oo

f

Evaluation, and Treatment of high blood cholesterol in adults (Adult Treatment Panel III)

pr

final report. Circulation 106 (25): 3143-421.

e-

33. Grundy SM, Cleeman JI, Daniels SR, et al (2005) Diagnosis and management of the

Institute

scientific

Pr

metabolic syndrome: An American Heart Association/National Heart, Lung, and Blood statement.

Circulation

112(17):2735–52.

doi:

rn

al

10.1161/CIRCULATIONAHA.105.169404.

Jo u

34. Witko-Sarsat V, Friedlander M, Nguyen KT, et al (1998) Advanced oxidation protein products as novel mediators of inflammation and monocyte activation in chronic renal failure. J Immunol 161(5):2524–32.

35. Casacci S, Pareto A (2015) Methods for quantifying ordinal variables: a comparative study. Qual Quant 49(5): 1859-1872. doi.org/10.1007/s11135-014-0063-2.

36. The IFNB Multiple Sclerosis Study Group (1995) Interferon beta-1b in the treatment of multiple sclerosis: final outcome of the randomized controlled trial. The IFNB Multiple

Journal Pre-proof

40

Sclerosis Study Group and The University of British Columbia MS/MRI Analysis Group. Neurology 45(7):1277-85. 37. Zepp J, Wu L, Li X (2011) IL-17 receptor signaling and T helper 17-mediated autoimmune

demyelinating

disease.

Trends

Immunol

32(5):232–9.

doi:

10.1016/j.it.2011.02.007.

oo

f

38. Ferreira TB, Hygino J, Barros PO, et al (2014) Endogenous interleukin-6 amplifies interleukin-17 production and corticoid-resistance in peripheral T cells from patients with

e-

pr

multiple sclerosis. Immunology 143(4):560-8. doi: 10.1111/imm.12334.

39. Teixeira B, Bittencourt VC, Ferreira TB, et al (2013) Low sensitivity to glucocorticoid

elevated

plasma

lipopolysaccharide

Clin

Immunol

148(2):209-18.

doi:

rn

10.1016/j.clim.2013.05.012.

levels.

al

to

Pr

inhibition of in vitro Th17-related cytokine production in multiple sclerosis patients is related

Jo u

40. Schneider A, Long SA, Cerosaletti K, et al (2013) In active relapsing-remitting multiple sclerosis, effector T cell resistance to adaptive T(regs) involves IL-6-mediated signaling. Sci Transl Med 5(170):170ra15. doi: 10.1126/scitranslmed.3004970.

41. Petković F, Castellano B (2016) The role of interleukin-6 in central nervous system demyelination Neural Regen Res 11(12):1922–3. doi: 10.4103/1673-5374.195273.

Journal Pre-proof

41

42. Zivadinov R, Weinstock-Guttman B, Hashmi K, et al (2009) Smoking is associated with increased lesion volumes and brain atrophy in multiple sclerosis. Neurology 73(7):504–10. doi: 10.1212/WNL.0b013e3181b2a706.

43. Dagan A, Gringouz I, Segal G (2016) Disability progression in multiple sclerosis is affected by the emergence of comorbid arterial hypertension. J Clin Neurol 12(3):345-50.

oo

f

doi: 10.3988/jcn.2016.12.3.345.

44. Aslani S, Jafari N, Javan MR, Karami J, Ahmadi M, Jafarnejad M (2017) Epigenetic and

therapy

in

multiple

sclerosis.

Neuromol Med

pr

modifications

doi:

Pr

e-

10.1007/s12017-016-8422-x.

19(1):11-23.

45. Skovierová H, Mahmood S, Blahovcová E, et al (2015) Effect of homocysteine on

rn

al

survival of human glial cells. Physiol Res 64(5):747–54.

Jo u

46. Kamath AF, Chauhan AK, Kisucka J, et al (2006) Elevated levels of homocysteine compromise blood-brain barrier integrity in mice. Blood 107(2):591-3. doi: 10.1182/blood2005-06-2506.

47. Faraci FM, Lentz SR (2004) Hyperhomocysteinemia, oxidative stress, and cerebral vascular dysfunction. Stroke 35(2):345-7. doi: 10.1161/01.STR.0000115161.10646.67

48. Ho PI, Ortiz D, Rogers E, Shea TB (2002) Multiple aspects of homocysteine neurotoxicity: Glutamate excitotoxicity, kinase hyperactivation and DNA damage. J Neurosci

Journal Pre-proof

42

Res 70(5):694–702 . doi: 10.1002/jnr.10416

49. Samblas M, Martínez JA, Milagro F (2018) Folic acid improves the inflammatory response in LPD-activated THP-1 macrophages. Mediators Inflamm. 2018:1312626. doi: 10.1155/2018/1312626.

50.

Cianciulli

A, Salvatore

R, Porro

C, et

al

Regulating

Multiple

Signaling

Pathways.

Mediators

Inflamm

2016:

Microglia 5240127.

by doi:

e-

pr

10.1155/2016/5240127.

oo

f

Folic Acid Is Able to Polarize the Inflammatory Response in LPS Activated

(2016)

Pr

51. Kubicka-Baczyk K, Labuz-Roszak B, Pierzchala K, Adamczyk-Sowa M, MachowskaMajchrzak M (2015) Calcium–phosphate metabolism in patients with multiple sclerosis. J

rn

al

Endocrinol Invest 38(6):635–42. doi: 10.1007/s40618-014-0235-x.

Jo u

52. Cheng SP, Liu CL, Liu TP, et al (2014) Association between Parathyroid Hormone Levels and Inflammatory Markers among US Adults. Mediators Inflamm 2014:709024. doi: 10.1155/2014/709024.

53. Hussain A, Latiwesh O B, Ali A, et al (2018) Parathyroid gland response to vitamin D deficiency in type 2 diabetes mellitus: an observational study. Cureus 10(11): e3656. doi 10.7759/cureus.3656.

54. Zhang Y, Zhang D (2018) Circulating parathyroid hormone and risk of hypertension: A meta-analysis. Clin Chim Acta 482: 40–45. doi:10.1016/j.cca.2018.03.028.

Journal Pre-proof

43

55. Marrie RA, Cutter G, Tyry T (2011) Substantial adverse association of visual and vascular comorbidities on visual disability in multiple sclerosis. Mult Scler 17(12):1464–71. doi: 10.1177/1352458511414041.

56. Tettey P, Simpson SJ, Taylor BV, et al (2014) Vascular comorbidities in the onset and progression

of

multiple

sclerosis.

J

Neurol

Sci

347(1-2):23-33.

doi:

oo

f

10.1016/j.jns.2014.10.020.

57. Nagai M, Hoshide S, Ishikawa J, Shimada K, Kario K (2008) Ambulatory blood pressure

pr

as an independent determinant of brain atrophy and cognitive function in elderly

Pr

e-

hypertension. J Hypertens 26(8):1636- 41. doi: 10.1097/HJH.0b013e3283018333.

58. Oliveira SR, Simão AN, Kallaur AP, et al (2014) Disability in patients with multiple

al

sclerosis: influence of insulin resistance, adiposity, and oxidative stress. Nutrition 30(3):268-

Jo u

rn

73. doi: 10.1016/j.nut.2013.08.001.

59. Kallaur AP, Oliveira SR, Simão ANC, et al (2017) Cytokine Profile in Patients with Progressive Multiple Sclerosis and Its Association with Disease Progression and Disability. Mol Neurobiol 54(4):2950–60. doi: 10.1007/s12035-016-9846-x. 60. Harandi AA, Shahbeigi S, Pakdaman H, Fereshtehnejad SM, Nikravesh E, Jalilzadeh R (2012) Association of serum 25(OH) vitamin D3 concentration with severity of multiple sclerosis. Iran J Neurol 11(2):54-8.

Journal Pre-proof

44

61. Mowry EM, Waubant E, McCulloch CE, et al (2012) Vitamin D status predicts new brain magnetic resonance imaging activity in multiple sclerosis. Ann Neurol 72(2):234-40. doi: 10.1002/ana.23591.

62. Smolders J, Menheere P, Kessels A, et al (2008) Association of vitamin D metabolite levels with relapse rate and disability in multiple sclerosis. Mult Scler 14(9):1220-4. doi:

oo

f

10.1177/1352458508094399.

63. Simpson SJ, Taylor B, Blizzard L, et al (2010) Higher 25-hydroxyvitamin D is associated lower

relapse

risk

in

multiple

Sun

D, Luo

F, Xing

Pr

64.

Ann

Neurol

68(2):193-203.

doi:

e-

10.1002/ana.22043.

sclerosis.

pr

with

JC,

et

al

(2018)

1,25(OH)2 D3 inhibited Th17

al

cells differentiation via regulating the NF-κB activity and expression of IL-17. Cell Prolif

Jo u

rn

51(5):e12461.doi: 10.1111/cpr.12461.

65. Fawaz L, Mrad MF, Kazan JM, et al (2016) Comparative effect of 25(OH)D3 and 1,25(OH)2D3 on Th17 cell

differentiation.

Clin

Immunol

166-167:59-71.

doi:

10.1016/j.clim.2016.02.011.

66. Bove R, Chitnis T (2013) Sexual disparities in the incidence and course of MS. Clin Immunol 149(2):201–10. doi: 10.1016/j.clim.2013.03.005.

Journal Pre-proof

45

67. Ribbons KA, McElduff P, Boz C, et al (2015) Male sex is independently associated with faster disability accumulation in relapse-onset MS but not in primary progressive MS. PLoS One10(6):e0122686. doi: 10.1371/journal.pone.0122686

68. Bove RM, Healy B, Augustine A, et al (2012) Effect of gender on late onset multiple sclerosis. Mult Scler 18(10):1472–9.doi: 10.1177/1352458512438236

oo

f

69. Shirani A, Zhao Y, Kingwell E, et al (2012) Temporal trends of disability progression in

e-

18(4):442–50. doi: 10.1177/1352458511422097.

pr

multiple sclerosis: findings from British Columbia, Canada (1975–2009). Mult Scler

Pr

70. Munger KL, Chitnis T, Ascherio A (2009) Body size and risk of MS in two cohorts of US

al

women. Neurology 73(19): 1543-50. doi: 10.1212/WNL.0b013e3181c0d6e0.

rn

71. Hedström AK, Olsson T, and Alfredsson L (2016) Body mass index during adolescence,

Jo u

rather than childhood, is critical in determining MS risk. Mult Scler 22(7): 878-83. doi: 10.1177/1352458515603798.

72. Richter B, Cutter G, Pandey K, Mohn J, Bacon TE, Kister I (2017) Body mass index correlates with multiple sclerosis disease and symptom severity in women, but not in men. Neurol Disord Therap 1(1): 1-5. doi: 10.15761/NDT.1000104.

73. Bove R, Musallam A, Xia Z, Baruch N, Messina S, et al. (2016) Longitudinal BMI trajectories in multiple sclerosis: Sex differences in association with disease severity. Mult Scler Relat Disord 8: 136-40. doi: 10.1016/j.msard.2016.05.019.

Journal Pre-proof

46

74. Goyal A, Nimmakayala KR, Zonszein J (2014) Is there a paradox in obesity? Cardiol Rev 22(4):163-70. doi: 10.1097/CRD.0000000000000004.

75. Corrales-Medina VF, Valayam J, Serpa JA, et al (2011) The obesity paradox in community-acquired

bacterial

pneumonia.

Int

J

Infect

Dis

Jo u

rn

al

Pr

e-

pr

oo

f

10.1016/j.ijid.2010.09.011.

15(1):e54-7.

doi:

Journal Pre-proof

47

Table 1 Quantification of the endpoint Expanded Disability Status Scale (EDSS) score in 119

patients with multiple sclerosis using transformed category value estimation Estimation from frequencies

0.0

1

1.0

1.234

1.5

1.502

2.0

1.634

2.5

1.836

oo

f

Ordinal EDSS rank

2.003

pr

3.0 3.5

e-

4.0

Pr

4.5

6.0

Jo u

6.5

rn

5.5

al

5.0

2.220 2.420 2.485 2.554 2.688 3.004 3.305

7.0

3.422

7.5

3.590

8.0-9.0

3.807

EDSS: Expanded Disability Status Scale

Journal Pre-proof Table 2

48

Demographic and clinical data of multiple sclerosis patients divided in to two

groups based on the disability progression index (DBI) no DP

DP df

P value

(n=47)

40.8 (12.7)

46.1 (12.2)

5.12

1/117

0.025

EDSS T0

1.876 (1.041)

2.078 (1.032)

1.05

1/115

0.307

EDSS T8

1.700 (1.023)

2.507 (0.898)

18.06

1/110

<0.001

EDSS T16

1.642 (0.825)

2.758 (0.748)

54.87

1/115

<0.001

Duration of illness (year)

6.7 (5.6)

8.2 (5.0)

2.40

1/117

0.124

BMI (kg/m2 )

25.5 (4.8)

26.2 (5.5)

0.57

1/117

0.453

WC (cm)

88.8 (13.1)

93.7 (13.3)

3.70

1/112

0.057

Sex (F/M)

50/22

31/16

0.16

1

0.690

Caucasian/Non-Caucasian

60/12

36/11

0.83

1

0.363

Tobacco use (No/Yes)

68/4

36/11

8.22

1

0.004

MetS (No/Yes)

56/15

29/17

3.52

1

0.061

SAH (No/Yes)

32/15

3.77

1

0.052

68/4

41/6

Ψ=0.127

-

0.166

63/9

37/10

1.69

1

0.194

rn

60/12

Jo u

T2DM (No/Yes) Clinical forms

RRMS/SPMS + PPMS

oo

e-

Pr

Age (years)

f

(n= 72)

pr

F/Ψ/X2

al

Variable

All results of analyses of variance or linear mixed models (F values). χ2: results of analyses of contingency tables. Continuous variables are expressed as mean and standard deviation. T0, T8, T16: measurements at baseline (T0), eight-month follow-up (T8) and 16-month followup (T16); EDSS: Expanded Disability Status Scale; DP: disability progression is assessed as an increase of at least 1 rank on the EDSS rank score. BMI: body mass index; WC: waist circumference; F: female; M: male; MetS: metabolic syndrome; SAH: Systemic arterial hypertension; 2TDM: type 2 diabetes mellitus; RRMS: relapsing-remitting multiple sclerosis; SPMS: secondary progressive multiple sclerosis; PPMS: primary progressive multiple sclerosis.

Journal Pre-proof Table 3

1

Results of Linear Mixed Model analysis, repeated measures, with the Expanded Disability Status Scale (EDSS) score and the

biomarkers as dependent variables. Variables

Time F T0

A

T8

B

T16

df

p value

C

qEDSS (no DP) n=72$

-0.137 (0.116)

-0.306 (0.110)

-0.368 (0.093)

qEDSS (DP) n=47$

0.061 (0.144)

0.487 (0.137)

0.731 (0.115)

Homocysteine (µmol/L)

13.18 (0.67)B,C

12.42 (0.64)A,C

Folic Acid (ng/mL)

11.23 (0.94)C

9.89 (0.56)

Vitamin D (ng/mL)

31.55 (2.47)

36.77 (2.75)

Parathormone (pg/mL)

59.99 (3.52)C

Calcium (mg/dL)

8.48 (0.06)B,C

hsCRP* (mg/L)

4.73 (1.59)

IL-6* (pg/mL)

f o

8.23#

2/111.8

<0.001

34.03##

2/111.8

<0.001

10.97 (0.62)A,B

21.09

2/112.1

<0.001

9.24 (0.51)A

3.13

2/105.5

0.048

35.20 (2.75)

2.98

2/121.9

0.054

57.29 (3.67)C

65.46 (3.66)A,B

6.32

2/112.8

0.003

8.64 (0.05)A,C

8.74 (0.06)A,B

9.98

2/118.2

<0.001

4.18 (0.70)

4.94 (0.85)

0.62

2/123.5

0.539

27.04 (8.83)

17.49 (4.16)

13.80 (3.34)

1.93

2/110.2

0.150

IL-17* (pg/mL)

35.47 (15.25)

25.22 (9.65)

20.06 (6.44)

1.17

2/111.3

0.314

IL-4* (pg/mL)

80.60 (48.80)C

49.36 (24.61)

27.43 (17.65)A,B

6.30

2/109.9

0.003

AOPP (µmol/L of

135.35 (5.21)B,C

116.13 (4.26)A,C

103.11(3.87)A,B

26.03

2/119.8

<0.001

chloramine-T equivalents)

r u o

a n

J

-p

re

P l

ro

Journal Pre-proof qEDSS: quantitative EDSS score; DP: disease progression.

$

2

The EDSS data are shown in z-scores with SE values. hsCRP: C-reactive

protein with high sensitivity assay; IL-6: interleukin-6; IL-17: interleukin-17; IL-4: interleukin-4; AOPP: advanced oxidation protein products; *Processed in natural logarithm (Ln) transformation; the data were adjusted for age, sex and smoking. #: effects of time, ## effects

of

the

interaction

l a n

J

r u o

time

by

r P

f o

o r p

e

disease

progression.

Journal Pre-proof

14

Results of linear mixed model analysis (LMM) analysis, repeated

Table 4

P value

Systemic arterial hypertension

5.27

1/97.9

0.024

Treatment

2.72

6/198.2

0.015

Clinical forms

22.52

2/104.3

<0.001

Age

10.63

1/104.6

0.002

AOPP

4.00

1/215.8

0.047

IL-17

5.56

1/237.8

0.019

IL-4

1/264.1

0.033

f

df

oo

Explanatory variables

4.60

e-

LMM

F

pr

measures, with the quantitative EDSS scores over time as dependent variables

four

Pr

All the results of analyses of variance (F values); df: degree of freedom; treatment: different treatment of multiple sclerosis; clinical forms: relapsing-remitting

al

multiple sclerosis and progressive forms of multiple sclerosis; LMN: linear mixed

Jo u

IL: interleukin

rn

model; z: results expressed as z scores; AOPP: advanced oxidation protein product;

Journal Pre-proof

15

Table 5. Results of categorical regression (CATREG) with the T16 EDSS scores as dependent variables and biomarker at T0 and T8 as explanatory variables Explanatory β

error

F

variables

0.087

14.41

<0.001

0.339

T8 Folic acid

0.161

0.084

3.65

0.015

0.173

T0 PTH

0.187

0.088

4.55

0.035

0.208

T8 Calcium

0.166

0.083

4.01

T0 Hcy

0.370

0.084

T0 PTH

0.189

0.080

T8 Calcium

0.176

0.079

T0 AOPP

0.160

0.080

oo 0.021

pr

Model

f

0.329

19.54

0.184 0.252

<0.001

0.392

5.50

0.021

0.208

4.99

0.003

0.191

4.05

0.047

0.177

0.076

27.25

<0.001

0.439

T0 PTH

0.169

0.079

4.57

0.035

0.201

0.180

0.077

5.43

0.002

0.209

0.147

0.074

4.00

0.048

0.176

0.329

0.069

24.52

<0.001

0.379

Jo u

T0 AOPP

rn

0.395

al

0.359

T0 Hcy

T8 Calcium

MS treatment #4

Square

T0 Hcy

Model # #3

correlation

0.243

e-

#2

Partial Eta

Model

Pr

#1

Partial P value

Model

0.622

T0 EDSS

0.687

0.057

147.08

<0.001

0.732

T0 Hcy

0.159

0.058

7.57

<0.001

0.250

T0 PTH

0.179

0.062

8.39

<0.001

0.269

Journal Pre-proof EDSS:

Expanded

Disability

Status

Scale;

16

25(OH)D:

25-hydroxyvitamin

D;

Hcy:

homocysteine; PTH: parathormone. T0, T8, T16: measurements at baseline (T0), eight-

Jo u

rn

al

Pr

e-

pr

oo

f

month follow-up (T8) and 16-month follow-up (T16); MS: multiple sclerosis

Journal Pre-proof Table 6.

17

Results of binary logistic regression analyses with primary progressive

multiple sclerosis and secondary progressive multiple sclerosis clinical forms as dependent variables Explanatory variable

Wald

df

P value

OR

95% CI

4.57

1

0.033

3.54

1.11-11.30

Sex (male)

12.05

1

0.001

5.34

2.07-13.76

Duration of illness

26.22

1

<0.001

1.27

1.16-1.39

Type 2 diabetes mellitus

4.97

1

0.026

Body mass index

14.55

1

<0.001

Parathormone

3.92

1

Interleukin 6

20.57

1

Interleukin-4

9.67

1

Homocysteine

7.97

25(OH)D

5.96

1.23-25.39

0.82

0.74-0.91

0.048

1.54

1.00-2.36

<0.001

e-

3.36

1.99-5.66

0.002

1.61

1.19-2.18

0.005

1.12

1.03-1.20

0.015

0.68

0.49-0.93

Pr

pr

oo

5.59

rn

al

1 1

f

Non Caucasian

Jo u

OR: odds ratio; CI: confidence interval; 25(OH)D: 25-hydroxyvitamin D

Journal Pre-proof

Pr

e-

pr

oo

f

18

al

Figure 1 The z transformed values and standard erros (SE) of the biomarkers in patients with

sclerosis

(SPMS)

Jo u

multiple

rn

relapsing remitting multiple sclerosis (RRMS) versus those with secondary progressive plus

primary

progressive

multiple

sclerosis

(PPMS).

Homocysteine, parathormone (PTH), interleukin (IL)-6 and IL-4 were positively associated with SPMS+PPMS as compared with RRMS, while vitamin D was negatively associated with SPMS+PPMS.

Journal Pre-proof

19

Highlights 

Immune-inflammatory and metabolic biomarkers were associated with the changes in disability over time in patients with multiple sclerosis



Disability progression index (DPI) was associated with changes in interleukin (IL)-17 (positively) and IL-4 (negatively)



Homocysteine, parathormone, IL-6, and IL-4 were positively associated with

pr

Vitamin D was inversely associated with progressive versus relapsing-remitting

rn

al

Pr

e-

multiple sclerosis

Jo u



oo

f

progressive versus relapsing-remitting multiple sclerosis