IL-2, IL-6 and chitinase 3-like 2 might predict early relapse activity in multiple sclerosis.

BACKGROUND: The possibility to better predict the severity of the disease in a patient newly diagnosed with multiple sclerosis would allow the treatment strategy to be personalized and lead to better clinical outcomes. Prognostic biomarkers are highly needed.
OBJECTIVE: To assess the prognostic value of intrathecal IgM synthesis, cerebrospinal fluid and serum IL-2, IL-6, IL-10, chitinase 3-like 2 and neurofilament heavy chains obtained early after the onset of the disease.
METHODS: 58 patients after the first manifestation of multiple sclerosis were included. After the initial diagnostic assessment including serum and cerebrospinal fluid biomarkers, all patients initiated therapy with either glatiramer acetate, teriflunomide, or interferon beta. To assess the evolution of the disease, we followed the patients clinically and with MRI for two years.
RESULTS: The IL-2:IL-6 ratio (both in cerebrospinal fluid) <0.48 (p = 0.0028), IL-2 in cerebrospinal fluid ≥1.23pg/ml (p = 0.026), and chitinase 3-like 2 in cerebrospinal fluid ≥7900pg/ml (p = 0.033), as well as baseline EDSS ≥1.5 (p = 0.0481) and age <22 (p = 0.0312), proved to be independent markers associated with shorter relapse free intervals.
CONCLUSION: The IL-2:IL-6 ratio, IL-2, and chitinase 3-like 2 (all in cerebrospinal fluid) might be of value as prognostic biomarkers in early phases of multiple sclerosis.

Introduction

Multiple sclerosis (MS) is a chronic autoimmune and neurodegenerative disease of the central nervous system (CNS) with a highly variable course and there is still a lack of robust biomarkers that would clearly predict the future course of the disease in its early stages. Knowing whether a patient is going to develop a highly active type of MS, or rather will avoid long-term progression, could directly influence the treatment strategy. Several cerebrospinal fluid (CSF) and/or serum substances have already been investigated as possible prognostic biomarkers.

Intrathecal IgG and IgM synthesis

The oligoclonal IgG bands (OCGB), reflecting intrathecal synthesis in the IgG class, are included in 2017 McDonald diagnostic criteria [1]. In a meta-analysis, OCGB positivity was found in 87.7% of all MS patients [2]. Being a part of current diagnostic criteria makes mere OCGB positivity unusable as a prognostic marker. The prognostic value of OCGB count in CSF has already been studied, yielding conflicting results [3,4]. On the other hand, oligoclonal IgM band (OCMB) positivity was found in 20.8% of MS patients [5] and might therefore be considered as a potential biomarker. Some studies suggest a worse prognosis for patients with OCMB due to its association with a higher Expanded Disability Status Scale (EDSS) [6,7], a shorter time to reach EDSS 3, 4 [8] and 6 [9], a higher probability of conversion to clinically definite MS (CDMS) [10], a shorter time to secondary progressive MS (SP-MS) [9], and a higher relapse rate [10]. In contrast, other studies found no correlation with the time to reach EDSS 3 [5], nor with the time to the second relapse [11].

Interleukins IL-2, IL-6 and IL-10

IL-2 is the major autocrine and paracrine T cell growth factor [12], which is, above all, responsible for the clonal expansion of antigen-specific T cells [13]. It also participates in the growth, differentiation, and activation of B cells, NK cells, and cytotoxic T cells [12]. The expression of IL-2 by Th17 cells, lymphocytes with a key role in the pathogenesis of MS [14], was reported to be increased in serum of patients with MS compared to healthy controls [15]. Higher IL-2 concentrations in CSF compared to controls were also observed [16]. CSF concentrations of IL-2 were found to be higher during a relapse of MS [17].

IL-6 is a multifunctional cytokine produced by a wide range of immune cells and has a major role in the regulation of the immune system [13,18]. It is indispensable in the development of Th17 cells, antigen-specific cytotoxic T cells, and monocytes [12,18]. In a widely used animal model of MS, experimental autoimmune encephalomyelitis (EAE), mice with a homologous disruption of the gene encoding IL-6 were resistant to EAE induction [19]. In MS patients, the presence and predominant location of IL-6 in acute and chronic active plaques were demonstrated by immunohistochemistry methods [20]. Serum and CSF IL-6 levels are significantly higher in MS patients compared to non-inflammatory neurological controls [21-23].

IL-10, a pleiotropic cytokine produced mainly by Th2 cells, exerts a strong anti-inflammatory effect by counteracting many pro-inflammatory cytokines produced by Th1 cells, such as interferon gamma and tumor necrosis factor alpha [13]. In patients with non-inflammatory neurological diseases, the IL-10 gene expression is greater in CSF cells than in peripheral blood cells. In MS patients undergoing a relapse, this difference becomes less apparent [24]. The serum count of B cells producing IL-10 is lower in MS patients compared to controls [25]. Absence of IL-10 production in mice with EAE (i.e. IL-10 knockout mice) results in a severe course of the disease [26] and, for recovery, B cell-driven IL-10 production is necessary [27] and effective especially in the early stages [26].

Chitinase-like proteins

Chitinase-like proteins are expressed by astrocytes and microglial cells in reaction to pro-inflammatory conditions [28] and play a role in inherited and acquired immunity [29]. In brain regions where demyelination had taken place, chitinase-like proteins form part of a microenvironment that is required for neural stem cells to replace damaged oligodendrocytes [30]. Chitinase 3-like 1 (CHI3L1) and chitinase 3-like 2 (CHI3L2) show strong expression in the brain of MS patients compared to controls, as measured by analysis of the CSF proteome [31]. In MS, CHI3L1 is better explored than CHI3L2 [28]. CHI3L1 CSF levels are significantly higher in MS patients compared to healthy controls. CSF CHI3L1 levels are also higher during the remission phase of the disease, as compared to levels assessed during a relapse [32], unless the relapse is accompanied by extensive radiologic activity [33]. Finally, CSF CHI3L1 levels are higher in patients with primary progressive MS (PP-MS) than in those with relapse remitting MS (RR-MS) or SP-MS [32]. Concerning the prognostic value, elevated CSF CHI3L1 was found to predict the conversion from clinically isolated syndrome (CIS) to CDMS [34,35], development of disability [35], and long-term cognitive impairment [34]. On the other hand, the significance of CHI3L2, the closest homologue to CHI3L1, is still unknown. CHI3L2 CSF levels were elevated in patients shortly after optic neuritis as the first manifestation of the disease. In these patients, CHI3L2 CSF levels were higher than in healthy controls; correlated with markers of tissue damage such as neurofilament light chains, MBP, osteopontin and CHI3L1; and predicted long-term cognitive disability. In a multivariate analysis, CHI3L2 was found to predict the conversion of CIS to CDMS even better than CHI3L1 [36]. A recent study proposed CSF CHI3L2 as a prognostic biomarker associated with long-term disability progression in patients with PPMS [37].

Neurofilaments

Neurofilaments (Nf) are major components of the axonal cytoskeleton [38]. They consist of light chain (NfL), medium chain, heavy chain (NfH), and α-internexin subunits. Nf are constantly released from axons of neuronal cells into the extracellular space. However, when axonal damage occurs, the quantities released rise markedly, rendering Nf a biomarker (yet nonspecific) of neurodegeneration [39]. Due to stability issues, only NfL and NfH are suitable for assessment in immunoassays [28]. Growing evidence supports NfL as a valuable biomarker of prognosis, disease activity, and treatment response in MS [39]. On the other hand, less attention has been paid to NfH, despite the potential complementary role of the two Nf forms in inflammatory and neurodegenerative processes [28]. NfH CSF levels were found to correlate with age, but are also elevated in CIS, RR-MS, SP-MS, and PP-MS after correction for age. Relatively higher CSF NfH levels were observed in progressive forms of MS as compared to RR-MS [40]. During a relapse, an elevation of CSF NfH levels occurs [41]. CSF NfH levels in patients with CIS and RR-MS correlate with EDSS in cross-sectional [41,42] and long term longitudinal studies [43] and also predict long term brain and spinal cord atrophy [44]. In contrast to NfL, CSF NfH levels do not correlate with serum levels in patients with MS [45].

According to the data summarized above, we hypothesized that intrathecal IgM synthesis; elevated values of CSF IL-2, IL-6, CHI3L2, NfH, and serum NfH; and lower values of CSF IL-10 detected early after the first manifestation of MS would predict an unfavorable disease course.

Methods

All treatment naïve patients who started first line treatment at the MS Centre of the Motol University Hospital between January 2017 and May 2018 were considered eligible for the study. In the Czech Republic, the first line treatment disease modifying drugs (DMD) comprise glatiramer acetate, teriflunomide, and the interferon beta group. Only patients who underwent a diagnostic lumbar puncture within four months from the onset of the symptoms were eligible. Patients who were treated with corticosteroids before the lumbar puncture (or in the preceding four months) were excluded. No patients started therapy with DMD before the diagnostic lumbar puncture. In total, 58 patients met the inclusion criteria, 54 with RR-MS, four with CIS. Diagnosis was made according to the 2017 McDonald criteria. During the follow-up period, eight patients were escalated to high efficacy treatment due to disease activity, while 50 patients continued with the first line treatment regardless of disease activity. Therapeutic decisions were made by a neurologist of the MS Centre, blinded to the results of the studied biomarkers. The study was approved by the Ethics Committee of the Motol University Hospital on the 30th of October 2018 and all participants (MS patients and controls) provided written informed consent. All data underlying our findings have been deposited in a public repository [46].

Cerebrospinal fluid

Patients underwent a lumbar puncture as part of a standard diagnostic protocol, in which intrathecal synthesis in the IgG class was determined 1) by calculation according to the Reiber’s formula for proper hyperbolic functions (IgG calc) [47], and 2) by assessing OCGB after isoelectric focusing (IEF) and immunoblotting. The CSF and serum samples were then stored at –80°C until used. For the purposes of the study, intrathecal synthesis in the IgM class was determined by both aforementioned methods (IgM calc, OCMB).

For the IEF of IgM, a gel pH range between four and eight was used and IgM pentamers were fragmented according to previously described techniques [48,49]. As primary and secondary antibodies, goat anti-human (AffiniPure Goat Anti-Human IgM, Fc5μ Fragment Specific, Jackson ImmunoResearch) and rabbit anti-goat (Biotin-SP (long spacer) AffiniPure Rabbit Anti-Goat IgG, Fc Fragment Specific, with minimal cross-reactivity to human serum proteins, Jackson ImmunoResearch) antibodies were used, respectively. Peroxidase conjugated Streptavidin (Jackson ImmunoResearch) and 3-amino-9-ethylcarbazole tablets (Sigma-Aldrich) diluted in methanol were used for visualization.

To determine IL-2, IL-6 and IL-10 levels in serum and CSF, a Luminex™ 200 instrument in magnetic bead mode (Magnetic Luminex® performance Assay–Human High Sensitivity Cytokine base Kit A, Magnetic Luminex® performance Assay beads, R&D Systems®), a method based on multiple simultaneous flow cytometry analyses, was used.

NfH were studied as phosphorylated forms (pNfH). The pNfH levels in CSF were assessed by ELISA (Euroimmun), whereas feasible results in serum analysis were only achieved after performing high sensitivity ELISA (Euroimmun).

CHI3L2 levels in serum and CSF were assessed by ELISA (CircuLex Human YKL-39 ELISA high sensitivity kit, MBL ltd.). The solution was diluted 13 and 25 times for CSF and serum, respectively.

As interleukins and CHI3L2 could originate both in serum and CNS, for further correlations, index values were used. The index of an analyte X was calculated according to the formula, (X/X)/(Alb/Alb) where Alb stands for albumin. Assuming pNfH in MS originates in CNS, we considered both values (from serum and CSF) and, in addition, we assessed the correlation between the two values. For IgG calc and IgM calc, positivity was considered for findings in area three or four according to Reiber’s diagram. For OCGB and OCMB, positivity was considered for findings corresponding with IEF patterns type two and three [50].

Follow-up

For all patients, the duration of the follow-up was two years, beginning the day of the administration of the first dose of DMD. Primarily, we recorded the time to the second relapse (referred to as relapse free interval, RFI) and the annual relapse rate (ARR). If a relapse occurred in the period between the first manifestation of MS and the first administration of DMD, it was included into the ARR for the first year. In addition, neurological status was assessed using EDSS every six months. The EDSS value was not taken into account if it was assessed during a relapse; in this case, a new assessment was performed after i.v. methylprednisolone infusion and clinical stabilization.

Patients underwent an MRI examination of the brain after the first and second year of the follow-up. The MRI study was performed following a standardized protocol: 1) transversal FLAIR sequences with a 1.5 mm slice thickness, 100 slices, a 256x256 matrix size and a 1x1x1.5 mm voxel size, 2) transversal T1 sequences with a 1 mm slice thickness, 150 slices, a 256x256 matrix size and a 1x1x1 mm voxel size. Data were automatically processed by MATLAB and SPM12 Toolbox software [51]. Lesions were segmented by the lesion prediction algorithm [52] as implemented in the Lesion Segmentation Tool toolbox version 3.0.0. (www.statistical-modelling.de/lst.html) for Statistical Parametric Mapping. Brain volume (i.e. brain tissue volume without CSF volume, cm3/ml) and lesion load (volume and count) were recorded.

The abovementioned parameters allowed us to assess the No Evidence of Disease Activity 4 (NEDA 4) score [53,54]. Gadolinium was not administered at control MRI examinations, thus T1 gadolinium-enhancing lesions were not included in NEDA 4.

Controls

The control group consisted of 31 patients examined at our department in the last five years for other diagnoses than a demyelinating disorder of the CNS. These patients underwent lumbar puncture for diagnostic purposes because of either headache or back pain. The results of the basic examination of CSF (cytology, protein level, IgG calc) and brain imaging (if performed) must have been normal. Analysis of the studied parameters for CSF was performed in the control group in order to determine cut-off values for the patient group. Where appropriate, cut-off values were determined on the basis of results from the patient group directly.

Statistical analysis

Statistical analysis was performed using SAS software (SAS Institute Inc., Cary, NC, USA). For comparison of the distribution of variables between the tested groups (i.e. patients vs controls, ARR, lesion load, brain atrophy, NEDA 4), nonparametric tests such as the Wilcoxon two sample test and the median test were used. Differences in frequencies were tested by the chi-square test and Fisher’s exact test, while the clinical outcome was expressed as an odds ratio. In the analysis of controls vs patients, ROC curves were used to assess the selection capacity of the individual parameters, after which specificity, sensitivity, and odds ratios were sought in order to determine the cut-off values. Correlations were examined by Spearman’s rank correlation coefficient. The RFI and six months confirmed EDSS worsening were analyzed by means of Kaplan-Meier curves; differences were then tested using the log-rank test and the Gehan-Wilcoxon test. Here, the clinical impact was expressed by the hazard ratio. For the multivariate analysis, the Cox regression model was used. The normality of the data for the EDSS analysis was tested using the Kolmogorov-Smirnov test. Because normality was not proven, the Wilcoxon signed-rank test and the Wilcoxon two-sample test were applied. The level of statistical significance was set to alpha = 5%.

Results

At the baseline, the mean EDSS value of all the patients in the study was 1.5 ± 1. The mean time to diagnosis was 1.1 ± 1.3 months and the mean time to therapy initiation was 3.7 ± 2.0 months. Most of the patients initiated treatment with peginterferon beta-1a (27), others with glatiramer acetate (20), intramuscular interferon beta-1a (5), subcutaneous interferon beta-1a (5) and interferon beta-1b (1).

First, we evaluated differences between the baseline characteristics of the patient group and the control group (Table 1). No differences were found in the demographic data. In the patient group, 64% exhibited positive IgG calc (p<0.0001) and 95% exhibited positive OCGB (p<0.0001). Concerning the IgM class, 31% of patients exhibited positive IgM calc (p = 0.0004) and 36% exhibited positive OCMB (p = 0.0005). In contrast, none of the controls exhibited OCGB or IgM calc positivity, while one control subject (3%) showed OCMB positivity. Both IndexIL-2 and IndexIL-10 showed statistically significant differences between the studied groups (p = 0.0157 and p = 0.0143, respectively). An IndexIL-2 value lower than 0.34 increased the risk of MS 4.5-fold (p = 0.0063) with high sensitivity (87.7%), but low specificity (38.7%). Similarly, an IndexIL-10 value lower than 0.23 increased the risk of MS 4.6-fold (p = 0.0053), with 58.6% sensitivity and 71.4% specificity. No correlation was found between pNfH in CSF and pNfH in serum (rs = 0.05, p = 0.7341).

Table 1

Baseline characteristics.

	Patients	Controls	Wilcoxon TST / Fisher’s ET	Cut-off
				Group in risk	Fisher’s ET	Odds Ratio
			P-value		P-value	Value	Confidence Interval
n	58 (F 43; M 15)	31 (F 21; M 10)	n.s.	-	-	-	-
Age at clinical onset (y)	33.5 ± 10.5	33.8 ± 11.2	n.s.	-	-	-	-
OCGB positivity	55 (95%)	0	<0.0001	positive	-	568.3b	56.7–5700.4
IgM calc positivity	18 (31%)	0	0.0004	positive	-	13.5b	1.7–103.4
OCMB positivity	21 (36%)	1 (3%)	0.0005	positive	-	17.3	2.2–134.0
IndexIL-2	0.24 ± 0.10	0.31 ± 0.12	0.0157	<0.34	0.0063	4.5	1.6–13.2
IndexIL-6	0.41 ± 0.37	0.74 ± 1.41	n.s.	-	-	-	-
IndexIL-10	0.22 ± 0.12	0.29 ± 0.14	0.0143	<0.23	0.0053	4.3	1.6–11.6
IndexCHI3L2	2.45 ± 3.72	2.01 ± 3.46	n.s.	-	-	-	-
pNfH in CSF (pg/ml)	263.9 ± 377.3	432.4 ± 1447.5	n.s.	-	-	-	-
pNfH in serum (pg/ml)	24.2 ± 21.7	20.7 ± 17.2	n.s.	-	-	-	-

TST = two sample test; ET = exact test;— = does not apply; F = female; M = male; n.s. = non-significant; y = years; m = months; DMD = disease modifying drugs; EDSS = Expanded Disability Status Scale; OCGB = IgG oligoclonal bands; IgM calc = calculated IgM intrathecal synthesis; OCMB = IgM oligoclonal bands; IL-2 = interleukin 2; IL-6 = interleukin 6; IL-10 = interleukin 10; CHI3L2 = chitinase 3-like 2 protein; pNfH = phosphorylated neurofilament heavy chains; CSF = cerebrospinal fluid.

aValues in the table are absolute counts or means ± standard deviation unless otherwise stated.

bValues of odds ratio for OCGB and IgM calc are based on statistical simulation (0 patients in control group).

Relapses: Relapse free interval, ARR

In RFI analysis (Table 2), we found statistically significant differences among patients sorted according to IndexIL-2 (p = 0.0367) and also among patients sorted according to EDSS at the time of therapy initiation (p<0.0001). The statistically strongest cut-off for IndexIL-2 was 0.26 (p = 0.0277), with higher values indicating a 2.5-fold (p = 0.0347) higher risk of second relapse in the first two years (Fig 1). This corresponds to a 52.0% (month 12) and 40.0% (month 24) chance of survival (no relapse) in the high-risk group, compared to 78.1% and 71.9% in the low-risk group, respectively. Patients with an EDSS value higher than, or equal to 1.5 at the time of therapy initiation were at a 2.8-fold higher risk of relapse (p = 0.0426). A statistically significant difference was also observed when the patients were sorted according to serum pNfH values (p = 0.0445); however, in further analysis, no cut-off reached statistical significance. Being female also appeared to be a risk factor (p = 0.0466), but this was not confirmed in a Cox regression model.

Table 2

Relapse free interval.

	Univariate					Multivariate
	Kaplan-Meier	Cut-off				Cut-off
		Group in risk	K-M	Cox reg		Group in risk	Cox reg
	P-value		P-value	Hazard Ratio	P-value		Hazard Ratio	P-value
Sex	0.0466	F	-	3.2	n.s.	F	2.2	n.s.
Age at clinical onset (y)	n.s.	<22	n.s.	2.1	n.s.	<22	5.6	0.0465
Time to diagnosis (m)	n.s.	≥2	n.s.	1.4	n.s.	≥2	4.0	n.s.
Time to therapy initiation (m)	n.s.	<4	n.s.	1.5	n.s.	<4	3.2	n.s.
Type of the 1st DMD	n.s.	Rebif	n.s.	2.1	n.s.	Rebif	4.6	n.s.
EDSS at time of therapy initiation	<0.0001	≥1.5	0.0326	2.8	0.0426	≥1.5	3.4	0.0496
IgG calc	n.s.	positive	-	2.1	n.s.	positive	1.5	n.s.
OCGB	n.s.	positive	-	1.5	n.s.	positive	4.5	n.s.
IgM calc	n.s.	positive	-	1.7	n.s.	positive	6.6	0.0151
OCMB	n.s.	negative	-	1.3	n.s.	negative	3.9	n.s.
IndexIL-2	0.0367	≥0.26	0.0277	2.5	0.0347	≥0.26	1.2	n.s.
IndexIL-6	n.s.	≥0.25	n.s.	2.7	n.s.	≥0.25	3.3	n.s.
IndexIL-10	n.s.	≥0.20	n.s.	1.9	n.s.	≥0.20	2.0	n.s.
IndexCHI3L2	n.s.	≥1.79	n.s.	1.8	n.s.	≥1.79	1.4	n.s.
pNfH in CSF (pg/ml)	n.s.	≥95.0	n.s.	2.5	n.s.	≥95.0	2.1	n.s.
pNfH in serum (pg/ml)	0.0445	<23.3	n.s.	2.7	n.s.	<23.3	1.5	n.s.

K-M = Kaplan-Meier; Cox reg = Cox regression;— = does not apply; n.s. = non-significant, F = female; M = male; y = years; m = months; N/A = not available; DMD = disease modifying drugs; EDSS = Expanded Disability Status Scale; IgG calc = calculated IgG intrathecal synthesis; OCGB = IgG oligoclonal bands; IgM calc = calculated IgM intrathecal synthesis; OCMB = IgM oligoclonal bands; IL-2 = interleukin 2; IL-6 = interleukin 6; IL-10 = interleukin 10; CHI3L2 = chitinase 3-like 2 protein; pNfH = phosphorylated neurofilament heavy chains; CSF = cerebrospinal fluid.

aIndicators in brackets correspond to data in "Group in risk" columns.

Fig 1

Relapse free interval–IndexIL-2.

Fig 1 shows the survival analysis of the patients sorted according to IndexIL-2 cut-off 0.26. In the univariate analysis, the difference between the two groups was statistically significant in favor of those with lower values of IndexIL-2, (p = 0.0277).

In a multivariate analysis, age lower than 22 at clinical onset (p = 0.0465), an EDSS value higher than, or equal to 1.5 at the time of therapy initiation (p = 0.0496), and positive IgM calc (p = 0.0151) proved to be independent prognostic markers.

When analyzing the ARR for the first year of follow-up, EDSS at the time of therapy initiation, IndexIL-2 and IndexIL-6 values divided the patients into groups with higher and lower numbers of relapses (cut-offs of 1.5, 0.26 and 0.25, respectively, p = 0.043, p = 0.0086 and p = 0.0265, respectively, values higher than, or equal to cut-offs indicating risk) (S1 Table). This effect was not observed during the second year of follow-up, although it was still present when cumulative relapse rate for the first two years was considered (p = 0.0453, p = 0.0165 and p = 0.0327 respectively). There was also a significant increase in the cumulative relapse rate during the whole follow-up period (but not when considered ARR for the separate years) in the group of patients with serum pNfH values below 23.3pg/ml (p = 0.0406) and in the women group (p = 0.0375).

Relapse free interval: CSF values–ad hoc analysis

After the initial analysis, we hypothesized, that the prognostic value of the studied biomarkers could be independent of the origin of a specific biomarker. To test this, we included an analysis of the CSF values of IL-2, IL-6, IL-10 and CHI3L2. Furthermore, we analyzed the ratio of the CSF values of IL-2 and IL-6 (IL-2:IL-6) supposing a complementary role of the two interleukins [55]. We present only the results of the multivariate Cox regression model using a backward selection, where parameters with the p-value above 0.3 in the univariate analysis were eliminated stepwise (Table 3). The likelihood ratio of the model was 23.0677 (p = 0.0004). The relation between the risk of the second relapse and the number of positive biomarkers is shown in Fig 2.

Table 3

Relapse free interval: CSF values—multivariate Cox regression model.

	Group in risk	Hazard Ratio	P-value
Age at clinical onset (y)	<22	4.3	0.0312
EDSS at time of therapy initiation	≥1.5	3.0	0.0481
IL2:IL6	<0.48	7.0	0.0028
IL-2 CSF (pg/ml)	≥1.23	6.1	0.026
CHI3L2 CSF (pg/ml)	≥7900	3.1	0.033
Sex, type of the 1st DMD, IgG calc, IgM calc, IndexIL-2, IndexIL-6, IndexIL-10, IndexCHI3L2, pNfH in serum, IL-6 CSF, IL-10 CSF	-	-	n.s.

y = years; EDSS = Expanded Disability Status Scale; IL-2 = interleukin 2; IL-6 = interleukin 6; CSF = cerebrospinal fluid; CHI3L2 = chitinase 3-like 2 protein; DMD = disease modifying drugs; IgG calc = calculated IgG intrathecal synthesis; IgM calc = calculated IgM intrathecal synthesis; IL-10 = interleukin 10; pNfH = phosphorylated neurofilament heavy chains;— = does not apply; n.s. = non-significant.

aIndicators in brackets correspond to data in "Group in risk" column.

Fig 2

Relapse free interval—relation to the number of positive biomarkers.

Fig 2 shows five Kaplan-Meier curves that correspond to the number of positive biomarkers (0–4) defined by the multivariate analysis (i.e. age at clinical onset, EDSS at time of therapy initiation, IL2:IL6 CSF, IL-2 CSF, CHI3L2 CSF). RFI becomes shorter with every added positive biomarker. In our cohort, none of the patients tested positive for all five biomarkers.

EDSS: Change in EDSS, six months confirmed EDSS worsening

In this analysis, we focused on the change in EDSS values between the beginning (T = 0) and the end (T = 24) of the follow-up period (S2 Table). When comparing the initial EDSS values, no differences were found for any of the studied biomarkers. When we compared the change in EDSS values between T = 0 and T = 24, only IndexIL-2 could differentiate between the patients with better and worse outcome. The group with IndexIL-2 values higher than, or equal to 0.34 showed a decrease in EDSS values (p = 0.0239).

Next, we analyzed the capacity of the studied biomarkers to sort the patients into high and low risk groups with respect to six months confirmed progression of EDSS (as required in NEDA 4 [54]) (S3 Table). In this analysis, age lower than 22 at clinical onset (p = 0.0111) presented a 12.1-fold higher risk (p = 0.0117). No other statistically significant differences were observed, until searching for the strongest cut-off. The group of patients with a serum pNfH value higher than, or equal to 51.9pg/ml (p = 0.0068) was found to be at 9.2-fold higher risk (p = 0.0268). The group of patients with an EDSS value lower than 1.5 at the time of therapy initiation (p = 0.0496) were found to be at risk (p = 0.0042), but the hazard ratio could not be calculated.

MRI: Lesion load, brain atrophy

MRI analysis was performed on data obtained after one year (T = 12) and 2 years (T = 24) of follow-up. The initial MRI was not included in the analysis because most of the investigations had been performed in extramural institutions and did not correspond to the study protocol. For similar (technical) reasons, 14 patients (out of a total of 58) were excluded from any MRI analysis. In the lesion load analysis, we studied the capacity of the proposed biomarkers to distinguish between patients with or without new/enlarging T2 lesions in the control scans (S4 Table). There were no statistically significant differences observed, until searching for the strongest cut-off. The group of patients with a CSF pNfH value lower than 95.0pg/ml (p = 0.0423) was found to be most at risk; however, the odds ratio did not reach statistical significance.

Next, we analyzed the capacity of the studied biomarkers to distinguish between patients with an annualized rate of whole brain volume loss above or below 0.4% (S5 Table). In this analysis, an EDSS value lower than 2.0 at the time of therapy initiation (p = 0.0209) presented a 5.7-fold higher risk (p = 0.0246). No other statistically significant differences were observed, until searching for the strongest cut-off. Values of IndexIL-10 lower than 0.13 (p = 0.0368) and serum pNfH values greater than, or equal to 10.8pg/ml (p = 0.0428) were found to be risk factors. However, the odds ratios for these findings did not reach statistical significance.

NEDA 4

Because of the limited MRI data available, we were only able to assess NEDA 4 for the second year of follow-up in 44 patients. None of the studied biomarkers could distinguish patients that would reach NEDA 4 from those who would not (S6 Table).

Discussion

Hand in hand with the growing armamentarium of DMD approved for the treatment of MS, there is a rising demand for the personalization of treatment according to predicted severity of the disease. Our study aimed to assess the prognostic value of serum and CSF biomarkers obtained early after the onset of the disease: intrathecal IgM synthesis, IL-2, IL-6, IL-10, CHI3L2, and pNfH. For evaluation, we used well-established clinical (RFI, ARR, change in EDSS, six months confirmed EDSS worsening), neuroimaging (lesion load, brain atrophy) and complex (NEDA 4) outcomes. Even though some of the studied biomarkers proved to have significant predictive value in some of the measured outcomes (as discussed below), none of the studied biomarkers showed significant predictive value in all of the measured outcomes.

To the best of our knowledge, no studies have been conducted on CSF IL-2 as a potential predictor of disease course in early MS. In our setting, high IndexIL-2 (≥0.26) predicted shorter time to second relapse, higher ARR in the first year of follow-up, and also higher combined ARR for the whole follow-up period. However, higher values of IndexIL-2 (≥0.34) were found to be protective against the worsening of EDSS. To explain these seemingly discordant results, we propose the commonly accepted theory of a disbalance between regulatory and effector T cell adaptive immunity [55,56]. High IndexIL-2 may be an indicator of a physiologically high IL-2 concentration in the CNS, ensuring a properly functioning regulatory pathway (T regulatory cell survival). As for middle range values (≥0.26 and <0.34) marking MS patients exhibiting a worse clinical outcome, these might be a result of IL-2 acting as a Th1 pathway activator, while suppressing Th17 differentiation. As for the lowest values (<0.26) representing MS patients with a mild disease course, we suppose the activation of other cell subsets (i.e. Th17, Th2) or weaker Th1 activation. This theory is also consistent with the results obtained from IndexIL-6 analysis, where, similarly to IndexIL-2, higher values indicated higher ARR in the first year of follow-up and higher combined ARR for both years of follow-up. While the Th17 pathway might be partially inhibited in an abundance of IL-2, the next step in the cascade is co-activation by IL-6 and TGF beta [55]. In a recent study, ongoing IL-6 signaling was found to be required to maintain Th17 cells [57]. Thus, higher IndexIL-6 might imply a worse disease course by hyperactivation of the Th17 pathway. Our results are partially in line with a study in which higher CSF IL-6 values at the time of diagnosis (mean disease duration six months) were shown to correlate with a higher number of relapses, higher MRI activity and higher EDSS values after two and three years in a three year follow-up [58].

In our study, however, neither IndexIL-2 nor IndexIL-6 retained statistical significance in the multivariate analysis of RFI and different independent prognostic factors emerged: age at clinical onset <22 years, EDSS at therapy initiation ≥1.5, and positive IgM calc. In the ad hoc multivariate analysis of RFI, where separate CSF values were also included, we confirmed age at clinical onset and EDSS at therapy initiation as independent prognostic markers. The prognostic role of age at clinical onset is further supported by the six months confirmed EDSS progression analysis, where an age <22 years presented an increased risk. As for EDSS, higher initial values also predicted a higher ARR, although they seemed to be protective against cumulating brain atrophy and six months confirmed EDSS worsening (not enough data to support the latter). We explain this by a short-term follow-up in our study in the face of long-term outcomes. Moreover, the IL-2:IL-6 ratio (both in CSF) and CSF IL-2 also proved to be independent prognostic markers. This result supports the abovementioned theory, whereas it might be assumed, that higher values of IL-6 in CSF might predict poor prognosis despite low CSF values of IL-2. IndexIL-10 did not prove valuable as a prognostic marker.

We found significant differences for CSF pNfH concerning lesion load and for serum pNfH concerning RFI, brain atrophy, and EDSS worsening, but could not confirm these results in all statistical tests. In addition, lower serum pNfH was found to be associated with a higher number of relapses during the whole follow-up period. This result is surprising with regard to our hypotheses and we cannot provide a feasible explanation based on previous research; therefore, false statistical significance should be considered. Although we did not confirm the poor prognosis for patients with elevated CSF pNfH reported in long-term studies concerning EDSS worsening and brain atrophy [43,44], this is probably the first study to evaluate the RFI and ARR in such patients.

IndexCHI3L2 did not reach statistical significance in any of the measured outcomes. In the ad hoc multivariate analysis of RFI, CHI3L2 in CSF proved to be an independent prognostic marker. This result is in line with a previous study investigating it’s capacity to predict the development of CIS into CDMS [36].

In some previous studies, OCMB were found to be a prognostic biomarker in MS [6-10], but in others the prognostic value could not be proven [5,11]. In our study, OCMB positivity was predictive of neither relapses, EDSS, MRI nor NEDA 4. This might be partially because of the short-term follow-up in our study in the face of long-term outcomes evaluated in other studies such as time to reach certain EDSS [8,9] and time to SP-MS [9], or different intended outcomes such as probability of conversion from CIS to CDMS [10]. The significance of long-term effects is supported by the clonal stability of the humoral immune response in MS over long periods, meaning that, once present, the OCMB pattern (as well as OCGB and IgA oligoclonal bands) persists [9,59]. One study reported a higher ARR in an OCMB positive group; however, it was conducted only on a small population of 22 MS patients [10].

In our cohort, significant differences between the group of MS patients and the control group were observed in IgM calc, OCGB positivity, OCMB positivity, IndexIL-2 and IndexIL-10. A level of OCGB positivity of 95% in the patient group vs 0% in the control group and a level of OCMB positivity of 36% in the patient group vs 3% in the control group are both consistent with former studies [2,5,60,61]. We observed lower IndexIL-10 to raise the risk of developing MS, which complies with indirect data suggesting its lower values in MS patients [24-27]. Surprisingly, low values of IndexIL-2 were also found to be associated with MS, yet we assume that these results might also be explained by the abovementioned “disbalance of T cell immunity” theory. In the other studied biomarkers, no differences were found between MS patients and the control group, which might be partially due to the use of indexed values in our study, and/or, in the case of pNfH, the presence of an age-matched control group.

We see the limitations of our study in the relatively small sample of patients, the short follow-up time, and incomplete MRI data. Also, multivariate analyses were performed only for RFI, since mostly negative results were obtained from univariate analyses of the other outcomes. There were a small number of patients who were escalated to high efficacy DMD treatment during the follow-up period, which might have influenced the results in the sense of underestimating the potential prognostic value. In case of chitinase proteins, we did not investigate CHI3L1, which might have allowed direct comparison with CHI3L2. Similarly, assessment of NfL, which is a more established biomarker than pNfH, was not a part of the study. Comparative studies with CHI3L1 and CHI3L2, as well as NfL and pNfH, might be good subjects for further investigation.

On the other hand, our study analyzed a relatively homogenous population of MS patients with only RR-MS or CIS subtypes, both on first line DMD treatment (as opposed to some previous studies that included treated and untreated patients), diagnosed according to the latest 2017 McDonald criteria and followed longitudinally by means of both clinical and MRI measures. In addition, we used indexed values of the studied biomarkers that we believe would better reflect their CNS origin. Ad hoc, we confirmed the results in an analysis of CSF values.

Conclusion

Overall, assessment of some of the studied biomarkers at the time of diagnostic lumbar puncture might be useful to estimate early relapse activity in MS. The most promising predictors of a worse outcome seem to be CSF IL-2, IL2:IL6 ratio (both in CSF) and CSF CHI3L2. Concerning clinical characteristics, age at clinical onset and EDSS at therapy initiation appear to be of value. Studies with larger cohorts of patients and longer follow-up periods, as well as more effective and standardized detection methods such as SIMOA, might help to achieve statistically significant results in long-term outcomes and also for the other studied biomarkers.

Relapse rate.

(PDF)

Click here for additional data file.

Change in EDSS.

(PDF)

Click here for additional data file.

Six months confirmed EDSS worsening.

(PDF)

Click here for additional data file.

Lesion load.

(PDF)

Click here for additional data file.

Brain atrophy.

(PDF)

Click here for additional data file.

NEDA 4 (year 2).

(PDF)

Click here for additional data file.

13 May 2022

7 Jun 2022

PONE-D-22-08909

IL-2, IL-6 and chitinase 3-like 2 might predict early relapse activity in multiple sclerosis

Dear Editor, dear Reviewers,

First, we would like to thank you for the time you have spent with our manuscript and for all the comments you have made. We believe that your kind input will surely be reflected in the improved quality of our article. Below, you will find all your comments addressed one by one.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

The manuscript and all the submitted files have been rechecked to comply with the journal’s requirements and edited accordingly.

2. Thank you for stating the following in the Competing Interests section:

"I have read the journal's policy and the authors of this manuscript have the following competing interests: MP received publication honorarium and compensations for travel and conference registration fees from Novartis, Merck Serono and Sanofi Genzyme; all outside the submitted work. EM received speaker honoraria and consultant fees from Novartis, Merck Serono, Sanofi Genzyme, Roche, Biogen Idec and Teva; all outside the submitted work. JL received compensations for travel, speaker honoraria and consultant fees from Novartis, Merck Serono, Sanofi-Genzyme, Roche, Biogen Idec, Teva and Bayer Healthcare; all outside the submitted work. HM received compensations for travel and conference registration fees from Novartis, Merck Serono, Sanofi Genzyme and Roche; all outside the submitted work. ME received publication honorarium, speaker honoraria and compensations for travel and conference registration fees from Novartis, Merck Serono, Roche, Teva and UCB; all outside the submitted work. JH, PR, SK, JŠ, OS, and PM declare that they have no competing interests."

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

The updated Competing Interests statement has been included in the cover letter.

3. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

The reference list has been rechecked for any inconsistencies. We used scite_ (available at https://scite.ai/) to check for any retracted articles. This tool was able to identify 44 out of 55 references, no retracted articles were found. We checked the rest of the references manually: 1) at the web page of the journal where the article had been published, 2) at http://retractiondatabase.org/RetractionSearch.aspx?. No retracted articles were found either. Furthermore, we found some typos in the reference list and few references had to be corrected to comply with the journal’s requirements. One reference has been replaced by a more proper one. Also, reference to our raw data has been included (see responses to Reviewer #2). Below, you will find all the changes made to the reference list; numbers correspond to the revised manuscript (with track changes).

12. Peakman M, Vergani D. Appendix 2: Major cytokines, cells releasing them, targets and functions. In: Peakman M, Vergani D, editors. Basic and Clinical Immunology. London: Elsevier Health Sciences; 2009. p. 342-4. edited according to PLOS recommended format

13. Schroeter M, Jander S. T-cell cytokines in injury-induced neural damage and repair. Neuromolecular Med. 2005;7(3):183-95. abbreviation of the journal name corrected

21. Stampanoni Bassi M, Iezzi E, Drulovic J, Pekmezovic T, Gilio L, Furlan R, et al. IL-6 in the Cerebrospinal Fluid Signals Disease Activity in Multiple Sclerosis. Front Cell Neurosci. 2020;14:120. page numbers corrected

22. Matsushita T, Tateishi T, Isobe N, Yonekawa T, Yamasaki R, Matsuse D, et al. Characteristic cerebrospinal fluid cytokine/chemokine profiles in neuromyelitis optica, relapsing remitting or primary progressive multiple sclerosis. PLoS One. 2013;8(4):e61835. page numbers corrected

23. Stelmasiak Z, Kozioł-Montewka M, Dobosz B, Rejdak K, Bartosik-Psujek H, Mitosek-Szewczyk K, et al. Interleukin-6 concentration in serum and cerebrospinal fluid in multiple sclerosis patients. Med Sci Monit. 2000;6(6):1104-8. corrections to year of publication, authors’ names

24. Romme Christensen J, Börnsen L, Hesse D, Krakauer M, Sørensen PS, Søndergaard HB, et al. Cellular sources of dysregulated cytokines in relapsing-remitting multiple sclerosis. J Neuroinflammation. 2012;9:215. page numbers corrected

33. Ferreira-Atuesta C, Reyes S, Giovanonni G, Gnanapavan S. The Evolution of Neurofilament Light Chain in Multiple Sclerosis. Front Neurosci. 2021;15:642384. page numbers corrected

36. Teunissen CE, Iacobaeus E, Khademi M, Brundin L, Norgren N, Koel-Simmelink MJA, et al. Combination of CSF N-acetylaspartate and neurofilaments in multiple sclerosis. Neurology. 2009;72(15):1322. title corrected

39. Eikelenboom MJ, Uitdehaag BMJ, Petzold A. Blood and CSF Biomarker Dynamics in Multiple Sclerosis: Implications for Data Interpretation. Mult Scler Int. 2011;2011:823176. page numbers corrected

40. Petržalka M. CSF Biomarkers in early MS; 2022 [cited 2022 March 25]. Database: G-Node [Internet]. Available from: https://doi.org/10.12751/g-node.74jj3f. new

41. Reiber H, Otto M, Trendelenburg C, Wormek A. Reporting Cerebrospinal Fluid Data: Knowledge Base and Interpretation Software. Clin Chem Lab Med. 2001;39(4):324-32. year and journal title corrected

45. The MathWorks I, Natick, Massachusetts, United States. MATLAB and SPM12 Toolbox. Release 2020a [Software]. 2020 [cited 2022 March 25]. edited according to PLOS recommended format

46. Schmidt P. Bayesian inference for structured additive regression models for large-scale problems with applications to medical imaging. Dissertation, LMU München. 2017. Available from: https://edoc.ub.uni-muenchen.de/20373/1/Schmidt_Paul.pdf. edited according to PLOS recommended format

48. Beadnall HN, Wang C, Van Hecke W, Ribbens A, Billiet T, Barnett MH. Comparing longitudinal brain atrophy measurement techniques in a real-world multiple sclerosis clinical practice cohort: towards clinical integration? Ther Adv Neurol Disord. 2019;12:1756286418823462-. cancelled

48. Kappos L, De Stefano N, Freedman MS, Cree BA, Radue E-W, Sprenger T, et al. Inclusion of brain volume loss in a revised measure of 'no evidence of disease activity' (NEDA-4) in relapsing-remitting multiple sclerosis. Mult Scler. 2016;22(10):1297-305. new

50. Höfer T, Krichevsky O, Altan-Bonnet G. Competition for IL-2 between Regulatory and Effector T Cells to Chisel Immune Responses. Front Immunol. 2012;3:268. page numbers corrected

Reviewers' comments:

Reviewer #1: Petržalka M et al., have assessed the prognostic value of intrathecal IgM synthesis, cerebrospinal fluid and serum IL-2, IL-6, IL-10, chitinase 3-like 2 and neurofilament heavy chains obtained early after the onset of the disease.

The article could be of interest before that there are following changes that need to be made.

1. Efforts should be made to improve the written style of the article. It is very difficult to follow.

Before the initial submission, the manuscript had been language-edited by a professional language editor, an English native speaker. Despite this fact, to avoid any misunderstandings, we tried to improve the style and rephrase some of the sentences/paragraphs that appeared less clear to us.

2. Graphical representation of the main findings should be done.

We added Fig 2. Relapse free interval - relation to the number of positive biomarkers.

3. It is not clear whether the treated patients were used for the experiments or untreated patients.

The lumbar puncture (and the blood draws) was performed as a diagnostic measure, thus none of the patients were treated at time of the lumbar puncture. Later on, all patients started treatment with DMD. This is stated in the Methods section of the Abstract and in more detail in the Methods section of the main article. The confusion might have been caused by the first sentence of this section, where we state “All treatment naïve patients who started first line treatment at the MS Centre of the Motol University Hospital between January 2017 and May 2018 were considered eligible for the study.”. This is meant to say that we included only patients who started with the first line treatment; those who started directly with the high efficacy therapy were excluded. To further clarify this, we added an explanatory sentence, see line 163.

4. Standard nomenclature should be used. IgM oligoclonal bands (OBM), IgG oligoclonal bands (OBG) are not standard nomenclature. Similarly, Relapse free interval (RFI) is not a standardly used terminology etc.

Instead of OBG, we now use OCGB, instead of OBM, we now use OCMB. As for RFI, we performed a short literature search and found several possible alternatives to it, none of which seemed to be more frequently used than the others: relapse free interval, recurrence free interval, relapse free period, relapse-free survival, disease-free survival, time to first relapse, time to second relapse, time to conversion to clinically definite MS. In an article by Havrdova et al, 2009 (Havrdova E, Galetta S, Hutchinson M, et al. Effect of natalizumab on clinical and radiological disease activity in multiple sclerosis: a retrospective analysis of the Natalizumab Safety and Efficacy in Relapsing-Remitting Multiple Sclerosis (AFFIRM) study. Lancet Neurol. 2009;8(3):254-260. doi:10.1016/S1474-4422(09)70021-3), which is one of the first papers that lead to the establishment of the NEDA concept, no specific terminology or abbreviation is used to address the clinical aspect of NEDA. Therefore, we prefer to continue using “relapse free interval” or “RFI”.

5. There are discordances too. For example oligoclonal bands (OBG) and IgG oligoclonal bands (OBG) have the same short form etc.

This was a typo; in both cases we referred to IgG oligoclonal bands. Corrected accordingly.

Reviewer #2: Petržalka et al test the predictive value of various biomarkers for disease progression including IL-2, IL-6, chitinase 3-like 2 and IgM in Serum and CSF in patients with early MS. The content of this study is of value to the community.

There are minor points that should be included into the manuscript:

- the Authors discuss that the function of Chitinase-like proteins are still widely unknown an cite a review from 2011. Due to the growing interest in these proteins as biomarkers in MS, also the literature on their function has grown. Please include some of the recent findings in the introduction (eg. Starossom et al 2019, Cubas-Nunes L et al 2021 and more)

In the section about chitinase proteins, we had aimed to introduce mainly CHI3L2. To provide a better understanding of the involvement of these proteins in MS, we have now included some of the more recent findings, as advised.

- a table describing the changes in clinical parameters (relapse free interval, ARR, change in EDSS, six months confirmed

EDSS worsening) & neuroimaging (lesion load, brain atrophy) during the duration of the study for the entire patient group will be helpful to better capture the characteristics of the patient group, that was studied.

We believe that all the information mentioned above is included in the table with our raw data in a very orderly manner. Although freely available at https://doi.org/10.12751/g-node.74jj3f, this table was not mentioned in the manuscript. We have added a reference to it, see line 170. Also, we have added tables describing each measured outcome as Supporting information (see the end of the manuscript).

- The authors discuss that neither of the studied biomarkers proved to have significant predictive value in all of the measured

outcomes. Yet the title of the manuscript suggest exactly thatl. Please change the title to something more neutral/descriptive.

We agree that none of the studied biomarkers proved to have significant predictive value in ALL of the measured outcomes. The title, however, states, that only some of the studied biomarkers (IL-2, IL-6 and chitinase 3-like 2) might predict worse outcome only in some of the measured outcomes (early relapse activity). To further clarify this, we have edited the mentioned sentence in the Discussion, see lines 390-393. To this end, we have also edited some parts of the Conclusion. Thus, we would like to maintain the current title.

- Legend to Figure 1 is missing

Legend has been added.

- several typos in the manuscript should be corrected

Addressed.

- are more stablished Biomarkers such as serum of CSF NfL available for the study group'? If so, these should be included here. If not, please include a brief discussion that this could be helpful also in comparison to the herein studied biomarker.

Neither CSF NfL, nor other biomarkers are available for the study group. A brief discussion was included as advised, see lines 474-477.

Yours Faithfully,

Marko Petržalka and co-authors

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

14 Jun 2022

IL-2, IL-6 and chitinase 3-like 2 might predict early relapse activity in multiple sclerosis

PONE-D-22-08909R1

Dear Dr. Petržalka,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Carmen Infante-Duarte, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: N/A

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: Alll concerns have been adequately addressed. All required questions have been answered and all responses meet formatting specifications. I now recommend the paper for publication.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

16 Jun 2022

PONE-D-22-08909R1

IL-2, IL-6 and chitinase 3-like 2 might predict early relapse activity in multiple sclerosis

Dear Dr. Petržalka:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Carmen Infante-Duarte

Academic Editor

PLOS ONE