TYPE Original Research

PUBLISHED 11 October 2023

DOI 10.3389/fimmu.2023.1223514

Kappa free light chains index

OPEN ACCESS in multiple sclerosis very
EDITED BY
Luisa Marı´a Villar,
Ramón y Cajal University Hospital, Spain
long-term prognosis
REVIEWED BY
Luca Massacesi, Pablo Arroyo-Pereiro 1, Lydia Garcı́a-Serrano 2,
University of Florence, Italy
Enric Monreal, Francisco Morandeira 2, Blanca Urban 2, Virginia Mas 2,
Ramón y Cajal University Hospital, Spain
Mario Framil 2, Isabel León 1, Albert Muñoz-Vendrell 1*,
*CORRESPONDENCE
Albert Muñoz-Vendrell
Elisabet Matas 1, Lucı́a Romero-Pinel 1,
[email protected] Antonio Martı́nez-Yélamos 1,3, Sergio Martı́nez-Yélamos 1,3
RECEIVED 16 May 2023 and Laura Bau 1
ACCEPTED 11 September 2023
PUBLISHED 11 October 2023
1
Multiple Sclerosis Unit, Department of Neurology, Hospital Universitari de Bellvitge- Institut
d’Investigació Biomèdica de Bellvitge (IDIBELL), L’Hospitalet de Llobregat, Spain, 2 Department of
CITATION Immunology, Hospital Universitari de Bellvitge – Institut d’Investigació Biomédica de Bellvitge
Arroyo-Pereiro P, Garcı´a-Serrano L, (IDIBELL), L’Hospitalet de Llobregat, Spain, 3 Departament of Clinical Sciences, Facultat de Medicina i
Morandeira F, Urban B, Mas V, Framil M, Ciències de la Salut, Universitat de Barcelona (UB), Barcelona, Spain
León I, Muñoz-Vendrell A, Matas E,
Romero-Pinel L, Martı´nez-Yélamos A,
Martı´nez-Yélamos S and Bau L (2023)
Kappa free light chains index in multiple
sclerosis very long-term prognosis. Introduction: The role of the kappa-free light chain (kFLC) in the diagnosis of
Front. Immunol. 14:1223514.
doi: 10.3389/fimmu.2023.1223514
multiple sclerosis (MS) and, to a lesser extent, its role as a medium-term
prognostic marker have been extensively studied. This study aimed to explore
COPYRIGHT
© 2023 Arroyo-Pereiro, Garcı´a-Serrano, its potential as a long-term prognostic marker for MS.
Morandeira, Urban, Mas, Framil, León,
Muñoz-Vendrell, Matas, Romero-Pinel, Methods: We performed an exploratory retrospective observational study by
Martı´nez-Yélamos, Martı´nez-Yélamos and
Bau. This is an open-access article selecting patients systemically followed up in our MS unit with available
distributed under the terms of the Creative cerebrospinal fluid and serum samples at the time of initial evaluation. Two
Commons Attribution License (CC BY). The
groups were defined: benign MS (bMS), defined as patients with Expanded
use, distribution or reproduction in other
forums is permitted, provided the original Disability Status Scale (EDSS) ≤ 3 at 10 years of follow-up, and aggressive MS
author(s) and the copyright owner(s) are (aMS), defined as patients with EDSS ≥ 6 at 15 years of follow-up. Clinical variables
credited and that the original publication in
this journal is cited, in accordance with were collected, and the immunoglobulin G (IgG) index, kFLC index, and
accepted academic practice. No use, oligoclonal bands (OCB) were determined for all patients and compared
distribution or reproduction is permitted
between the groups.
which does not comply with these terms.

Results: Twenty bMS and 15 aMS patients were included in this study. Sixty
percent (21/35) were female, and the mean age at the time of the first symptom
was 31.5 ± 9.45 years, with no statistical differences between groups. Median
follow-up time was 19.8 years (Interquartile range, IQR 15.9–24.6). The median
EDSS scores at the last follow-up were 1.5 and 7.5 in the bMS and the aMS group,
respectively. No statistically significant differences were found in the kFLC index
between the two groups (136.6 vs. 140.27, p=0.59). The IgG index was positive in
62.9% of patients (55% bMS vs. 73.3% aMS, p>0.05), and OCB was positive in
88.6% (90% bMS vs. 86.7% aMS, p>0.05). A significant positive correlation was
found between IgG and kFLC indices (rs = 0.85, p<0.001).

Frontiers in Immunology 01 frontiersin.org

Arroyo-Pereiro et al. 10.3389/fimmu.2023.1223514

Conclusion: Given the absence of differences between the two groups with
opposite disease courses, it is unlikely that the kFLC index is a reliable and
powerful marker of long-term prognosis in MS.

KEYWORDS

multiple sclerosis, kappa free light chains, biomarkers, long-term, prognosis,

cerebrospinal fluid, disability

1 Introduction 2 Materials and methods

Multiple sclerosis (MS) is a chronic inflammatory immune- 2.1 Study design, patients and clinical
mediated neurological disease with a variable course derived from data collection
the accumulation of disability as a consequence of inflammatory
relapses and progression independent of relapses (1). In this proof-of-concept study, an exploratory retrospective
Thus, with the individual course of the disease being highly observational design was used, including patients diagnosed with
variable, there are currently only a few prognostic markers for Relapsing Remitting MS according to the McDonald criteria of 2017
practical use (2, 3). A recent approach to treatment using early high- (23) who were systematically followed up in the MS unit of a tertiary
efficacy treatments versus the more traditional approach in which hospital with available CSF and serum samples at the time of the
treatment is scaled up progressively to prevent the accumulation of initial patient evaluation.
disability in patients with aggressive forms has generated the need Two groups were defined with distinctive inclusion criteria:
for prognostic markers; therefore, stratification can be made to benign MS (bMS), defined as patients with Expanded Disability
guide treatment decisions (4–6). Status Scale (EDSS) ≤ 3 at 10 years of follow-up, and aggressive
MS is classically considered an organ-specific autoimmune MS (aMS), defined as patients with EDSS ≥ 6 at 15 years of
disease mediated mainly by T lymphocytes, although the role of B follow-up.
lymphocytes is considered important as evidenced in animal Patients were excluded if they had renal function impairment,
models (7, 8), by pathological studies in MS (9), by the presence the presence of mono-or polyclonal gammopathies, the presence of
of oligoclonal bands (OCB) in the cerebrospinal fluid (CSF) (10) other neoplasms, the presence of any systemic or inflammatory
and indirectly by the efficacy of targeted therapies against B cells disease (including sarcoidosis, lupus, rheumatoid arthritis, Sjögren
(11), among other data. syndrome, allergy, chronic rhinitis, asthma, idiopathic pulmonary
Free light chains (FLC) are a product of the physiological fibrosis, interstitial hypersensitivity pneumonitis, or HIV infection),
activity of plasma cells, which in their immunoglobulin presence of any other central nervous system diseases, or treatment
production process generate an excess of light chains (kappa or with corticosteroids in the month prior to lumbar puncture (LP)
lambda) that are not incorporated into their usual tetrameric performance and blood sampling. Patients in whom the samples did
structure. They can be interpreted as markers of the activity of not meet the technical criteria (presence of particles, lipemia, or
plasma cells and are detectable in various fluids (12–15). Several hemolyzed samples) were also excluded. The time from onset of the
metrics have been proposed to consider the permeability of the disease to LP was not consider as an exclusion criteria.
blood-brain barrier and therefore more accurately reflect the The European Database for Multiple Sclerosis software was
intrathecal fraction of FLC (16), with the FLC index being used for the prospective collection of clinical data (24). The
ultimately the most widely used. variables collected were sex, date of birth, date of MS onset, date
Lately, the usefulness of the kappa FLC (kFLC) index in the of LP, date of last relapse prior to LP, date of last dose of
diagnosis of MS has been extensively studied, and its equivalence to corticosteroid prior to LP, date of last follow-up, total number of
OCB has been suggested by various studies (17–19), although some relapses during follow-up, EDSS at the time of LP, EDSS at 10 years
studies disagree (19). A few recently published studies using the of onset for bMS, EDSS at 15 years of onset of aMS, EDSS at the end
kFLC index and modern nephelometry or turbidimetry techniques of follow-up for all patients, disease-modifying treatment (DMT)
have aimed to establish its potential role as a prognostic marker. received before LP and during follow-up, and progression start date.
However, its role in long-term prognosis has not been explored The following variables were calculated: age at onset, age at first
(20–22). LP, total follow-up time, time from MS onset to LP, and time from
We hypothesized that the kFLC index could predict long-term relapse to LP. The DMT variable was analyzed as: “no DMT”;
MS outcomes. Therefore, this study aimed to assess kFLC in two “moderate efficacy DMT” if any of the following was used at any
groups with opposite clinical courses to evaluate its potential role as moment of the follow-up and none of the high efficacy DMT:
a long-term prognostic marker in MS. interferon beta 1a, interferon beta 1b, peginterferon beta 1a,

Frontiers in Immunology 02 frontiersin.org

Arroyo-Pereiro et al. 10.3389/fimmu.2023.1223514

glatiramer acetate, teriflunomide, dimethyl fumarate, fingolimod, 2.4 Statistical analysis

cladribine; and “high efficacy DMT” if any of the following was used
at any moment of the follow-up: natalizumab, alemtuzumab, Data are described as mean ± standard deviation (SD) or as
ocrelizumab, rituximab, mitoxantrone. The diagnostic criteria for median and interquartile range (IQR), according to their
secondary progressive multiple sclerosis (SPMS) used to define the distribution. Categorical variables are described using frequencies.
progression date are those previously defined in 2016 and widely The distribution of the data was assessed using the Shapiro-Wilk W
used (25). test. For group comparisons, the chi-square test, Fisher’s exact test,
Mann-Whitney U test, and negative binomial regression were
applied as appropriate. Spearman’s rank correlation coefficient
2.2 Samples and laboratory analysis was used to assess the correlation between the kFLC and IgG
indices, the time from MS onset to LP, and the time from
Patient samples were provided by Biobank HUB-ICO-IDIBELL, previous relapse to LP. To assess the biomarkers accuracy in
funded by the Instituto de Salud Carlos III (PT20/00171), and by classifying patients into benign or aggressive forms, receiver
Xarxa de Bancs de Tumors de Catalunya, sponsored by Pla Director operating characteristic (ROC) curves and derived area under the
d’Oncologia de Catalunya (XBTC). Samples were collected as part curve (AUC) were calculated. ROC curve standard error
of the usual initial patient evaluation and stored at -80°C calculations and comparisons were performed with DeLong’s
until analysis. method All tests were conducted with 95% confidence intervals
The following parameters were determined in both the serum (CIs) and a significance level of 5%. Statistical analyses were
and CSF: albumin, total immunoglobulin G (IgG), kFLC, and OCB. performed using the Stata 13 software (StataCorp LLC,
IgG OCB was detected by isoelectric focusing on a Hydrasys Texas, USA).
Focusing System (Sebia Hispania, Barcelona, Spain) according to
the manufacturer’s instructions. OCB were considered positive if at
least two bands in the CSF were present without corresponding 2.5 Ethics approval
bands in the serum. IgG and albumin levels in the CSF and serum
were measured using a BN II nephelometer system (Siemens The study was approved by the Ethics Committee of the
Healthcare, Marburg, Germany). To account for the influence of Hospital Universitari de Bellvitge (reference PR134/22). Patient
blood barrier permeability, the IgG index was calculated using the information confidentiality was addressed in accordance with the
following formula: Spanish regulations.

CSF IgG  serum Albumin

IgG index =
serum IgG  CSF Albumin 3 Results
CSF and serum KFLC levels were measured by turbidimetry
using the Freelite Mx Free Kappa Optilite kit (Binding Site, 3.1 Groups characteristics
Birmingham, UK) on an Optilite Analyzer (Binding Site). The
kFLC quotient and index were calculated using the following Out of the 2,216 patients included in the EDMUS database, 206
formulae: patients met the criteria of aMS group, of which 16 cases had available
paired CSF and serum samples from the initial study. 1 had to be
CSF KFLC excluded due to medical history of chronic allergy. A total of 729
KFLC quotient =
serum KFLC patients met the criteria for bMS. Among these, 20 patients age-
matched patients with paired CSF and serum samples were selected.
CSF KFLC  serum Albumin Twenty patients were included as having bMS and 15 as having aMS.
KFLC index =
serum KFLC  CSF Albumin Sixty percent were female, and the mean age at MS onset was 31.5 ±
In those samples with a kFLC value below the detection 9.45 years old. They were followed up for a median of 19.8 years (IQR
threshold (<0.33 mg/L according to the manufacturer), a value of 15.9-24.6). No statistical differences were found in age at onset nor in
0.17 mg/L was assigned to them. follow-up time between groups. In the latter although deemed no
Values of IgG index > 0.7 were considered as positive. statistically significant, the median follow-up was approximately 4
years longer for the aMS group (17.33 in bMS vs. 21.6 years in aMS;
p=0.06). In this regard, despite the fact that the time from MS onset to
2.3 Study endpoints LP was not statistically different between groups (24.21 in bMS vs.
39.49 months in aMS; p=0.3), age at the time of LP was indeed higher
The primary endpoint was the difference in kFLC index for aMS (31.05 in bMS vs. 39.69 years in aMS; p=0.018). The median
between patients with bMS and those with aMS. As secondary EDSS score at the time of LP was 1.5 (1 in bMS vs. 3.5; aMS, p=0.002).
endpoints, differences in the IgG index and OCB status were The median EDSS at the last follow-up was 1.5 (IQR 1.25–1.5) for the
assessed between both groups, and the correlation between kFLC bMS group and 7.5 (IQR 7–8) for the aMS group. Twelve patients in
and IgG indices was analyzed. the aMS group developed Secondary Progressive MS (SPMS),

Frontiers in Immunology 03 frontiersin.org

Arroyo-Pereiro et al. 10.3389/fimmu.2023.1223514

whereas none in the bMS group developed this clinical form. With treatment, or had received corticosteroids in the past 6 months.
regard to corticosteroid treatment before the LP, 6 out of the 35 Figure 1 shows the kFLC quotient and kFCL index distribution
received corticosteroids before the LP, with the last dose being within boxplot. The differences between groups were not statistically
the last 3 months in 3 cases (34, 76, and 79 days before LP, significant, neither in the case of the kFLC index (p=0.59) nor in
respectively). In the remaining cases, the time interval exceeded 6 the case of the kFLC quotient (p=0.64). The calculation of ROC
months. The patient who received corticosteroids 34 days before is curves showed an AUC of 0.55 (IQR 0.35-0.75) for the kFLC index
part of the bMS group. None of the patients were on DMTs at the and an AUC of 0.55 (IQR 0.35-0.75) for the kFLC ratio.Biomarkers
time of the LP; however, one patient in the aMS group was receiving and comparison tests between groups are summarized in Table 2.
azathioprine treatment. The IgG index was positive in 62.9% of the patients (55% for
The clinical characteristics and comparison tests between the bMS vs. 73.3% for aMS, p>0.05), and the OCB was positive in 88.6%
groups are summarized in Table 1. (90% for bMS vs. 86.7% for aMS, p>0.05). A strong correlation was
found between the IgG and kFLC indices (rs = 0.85, p<0.001), as
shown in Figure 2.
3.2 Biomarkers and prognostic role of No correlation was found between the kFLC index and the time
kFLC index from MS onset to LP or between the kFLC index and the time from
previous relapse to LP.
The median kFLC index was slightly higher in the aMS group (74 The statistical analysis was repeated, excluding outliers in the
[IQR 37.61-332.45]) than in the bMS group (67.27 [IQR 30.23- bMS group, and the results were identical (data not shown).
159.41]). Similarly, the kFLC quotient was also higher in the aMS
group (0.39 [IQR 0.18-0.67] vs. 0.26 [IQR 0.15-0.52]). It’s worth
noting that in the bMS group, there were 2 outlier values for the kFLC 4 Discussion
index and 3 for the kFLC quotient. Notably, one of these outliers,
pertaining to both metrics, corresponds to the patient who had Recently, the kFLC index has been shown to be an alternative to
received corticotherapy 34 days prior to the lumbar puncture. The OCBs for the diagnosis of MS (17, 18). It is a marker that reflects
remaining outliers were not receiving any DMT, immunosuppressive intrathecal immunoglobulin synthesis; however, unlike OCB, it is

TABLE 1 Clinical characteristics and group comparisons.

All patients Benign MS Aggressive MS p value

Number of patients 35 20 15

Age at MS onset, years (mean ± SD) 31.5 ± 9.45 28.33 ± 4.29 35.74 ± 12.58 0.15

Female, n (%) 21 (60%) 14 (70%) 7 (46.7%) 0.16

Total follow-up time, years [median (IQR)] 19.79 17.33 21.6 0.06
(15.89–24.57) (14.54–22.18) (18.55–24.77)

ARR [median (IQR)] 0.28 0.23 0.40 0.015*

(0.16-0.40) (0.15-0.30) (0.16-0.56)

EDSS at last follow-up [median (IQR)] 2 (1.5–7.5) 1.5 (1.25–1.5) 7.5 (7–8) <0.001*

MSSS at last follow-up (mean ± SD) 3.8 ± 3.68 0.73 ± 0.41 7.89 ± 1.14 <0.001*

ARMSS score at last follow-up (mean ± SD) 4.71 ± 3.52 1.84 ± 0.95 8.54 ± 1.2 <0.001*

DMT during follow-up, n (%): 5 (14.3%) 4 (20%) 1 (6.7%) <0.001*

-No DMT 19 (54.3%) 16 (80%) 3 (20%)
-Moderate efficacy 11 (31.4%) 0 (0%) 11 (73.3%)
DMT
-High efficacy DMT

SPMS, n (%) 12 (34.3%) 0 (0%) 12 (80%) <0.001*

Age at LP, years (mean ± SD) 34.75 ± 9.08 31.05 ± 4.85 39.69 ± 11.06 0.018*

EDSS at LP (median (IQR)) 1.5 (1–3.5) 1 (0.5–1.5) 3.5 (2–3.5) <0.001*

Time from MS onset to LP, months (median (IQR)) 29.96 24.21 39.49 0.3
(4.57–51.81) (3.53–42.45) (4.57-64.1)

Time from previous relapse to LP, months [median (IQR)] 2.99 2.14 3.48 0.14
(1.12–6.67) (0.92–5.8) (1.71–11.73)

MS, multiple sclerosis; SD, standard deviation; IQR, interquartile range; ARR, annualized relapse rate; EDSS, expanded disability status scale; MSSS, multiple sclerosis severity score; ARMSS
score, age related multiple sclerosis severity score; DMT, disease-modifying treatment; SPMS, secondary progressive multiple sclerosis; LP, lumbar puncture.
*statistically significant.

Frontiers in Immunology 04 frontiersin.org

Arroyo-Pereiro et al. 10.3389/fimmu.2023.1223514

A B

FIGURE 1
kFLC index (A) and kFLC quotient (B) distribution boxplots for benign and aggressive MS courses. A cases scatter plot is represented by dots and
outliers by triangles.

quantitative. This fact raises the question of whether, in addition to aimed at determining whether the FLC index is able to predict the
its role as a diagnostic marker, FLC could also have a role as a second relapse of the disease and when it occurs; in other words, the
prognostic marker, namely, whether high FLC index values are conversion from clinically isolated syndrome to clinically definite
related to a worse disease course. However, this hypothesis has not multiple sclerosis (26–29). The second group focused on
yet been fully explored. inflammatory activity in the early years (beyond the first relapse)
This study aimed to determine the utility of the kFLC index in and cumulative disability during this period.
long-term prognosis. The rationale for its design was based on the Focusing on the second group, the oldest studies used the
fact that if the kFLC index is a useful and reliable prognostic absolute value of KFLC in the CSF or CSF/serum FLC quotient as
biomarker, it should be able to differentiate patients with MS with a metric of FLC and ELISA and radioimmunoassay techniques for
opposite courses. No differences were observed between the two its measurement (30–32).
groups. Therefore, our results suggest that this index is not a useful Although they used a different metric, it is worth mentioning a
clinical marker for long-term prognosis. study that demonstrated the relationship between the kFLC:LFLC
To date, some studies have assessed the relationship between ratio and EDSS at 5 years of follow-up (33).
the FLC index and prognosis. These can be divided into two main Among the most recent studies that used the kFLC index and
groups based on their objectives. The first group includes studies modern turbidimetry or nephelometry methods, three found no

TABLE 2 Biomarkers and groups comparison.

All patients Benign MS Aggressive MS p value

Number of patients 35 20 15

Serum kFLC, mg/L (median (IQR)) 14.8 14.77 15 0.69

(14.19–16.24) (13.46–16.75) (14.36–16.24)

CSF kFLC, mg/L [median (IQR)] 5.18 4.51 5.47 0.71

(2.49–9.21) (2.14–9.26) (2.86–8.12)

kFLC quotient [median (IQR)] 0.35 0.26 0.39 0.64

(0.18–0.61) (0.15–0.52) (0.18–0.67)

kFLC index [median (IQR)] 73.34 67.27 74 0.59

(32.15–196.17) (30.23–159.41) (37.61–332.45)

kFLC index >50, n (%) 21 (60%) 11(55%) 10(66.7%) 0.49

kFLC index>100, n (%) 15(42.9%) 8 (40%) 7(46.7%) 0.69

IgG index [median (IQR)] 0.83 0.72 0.96 0.2

(0.62–1.4) (0.62–1.14) (0.67–1.49)

Positive IgG index (≥0.7), n (%) 22 (62.9%) 11(55%) 11(73.3%) 0.312

OCB, n (%) 31(88.6%) 18(90%) 13(86.7%) 1

kFLC, kappa free light chains; IQR, interquartile range; CSF, cerebrospinal fluid; IgG, immunoglobulin G; OCB, oligoclonal bands.

Frontiers in Immunology 05 frontiersin.org

Arroyo-Pereiro et al. 10.3389/fimmu.2023.1223514

smoldering plaques, macrophages, and microglia (9, 37).

Ultimately, this could explain the difference in results between
previous studies and our long-term study.
Another distinguishing factor worth highlighting in
comparison to numerous previous studies, which could explain
the difference in results, is the fact that in our study, only patients
diagnosed with MS according to the McDonald 2017 criteria have
been included. This was done to avoid mixing the diagnostic
capacity of kFLC (its ability to predict whether CIS will convert
to MS) with its prognostic capacity for disability. In fact, out of the 4
studies that analyze the prognostic capacity of disability without
including CIS or excluding the CIS subgroup from the prognostic
analysis, only one finds a relationship with the kFLC index
employing modern detection tecnhiques, but with a median
FIGURE 2
Correlation between kFLC and IgG indices. The linear regression fit
follow-up time of less than 5 years (22).
is represented by a 95% confidence interval. Whether the proximity of a recurrence or the use of
immunosuppressive drugs affects kFLC index values has been
discussed, and there are a few previous studies that provide some
relationship between the kFLC index and disability progression data in this regard (27). In our cohort, owing to the selection
(33–35), although one showed a shorter time to DMT initiation in criteria, none of the patients received corticosteroids in the month
patients with a higher kFLC index (35). The remaining studies before sample collection. We were unable to demonstrate a
found an association between a higher kFLC index and a higher relationship between the time from the previous relapse to sample
confirmed EDSS (20–22), but the follow-up time in these studies collection and the kFLC index.
was focused on the mid-term. With regard to the limitations of this study, this was conceived
It is important to highlight that out of all these studies aimed at as an exploratory observational study. This could lead to a bias
assessing the FLC prognostic capacity of disability accumulation, associated with observational studies. The sample size represents
only two studies did not include CIS in their cohorts (22, 31), and another limitation of the study as it can result in both a lack of
another two analyzed the prognostic capacity of FLC in the MS statistical power and potential biases.
subgroup, excluding CIS from the analysis (32, 34). Nevertheless, as we have mentioned previously, we believe that
Due to the design of this study, a negative result does not due to the study’s design, its outcome is still valuable as it
completely rule out the possibility that differences in the kFLC index reasonably allows us to rule out the actual clinical utility of FLC
due to the disease course can be found with a larger sample size. But as a long-term prognostic marker. Although no significant
this design allows us to state that, even in this case, it is unlikely that differences were observed in the age at onset between the groups,
this result would be relevant, and that the kFLC index would be a a significant difference was noted between the ages of the patients at
strong prognostic and clinically useful marker with good sensitivity, the time of LP, which was higher in the aMS group than in the bMS
specificity, and positive and negative predictive values. As an group. According to FLC kinetics, older age should contribute to an
additional post-hoc analysis, we tested whether the bMS and aMS increase in FLC serum levels, as multiple comorbidities as well as
groups had unequal numbers of high kFLC index values; we did so renal function impairment can raise them (38–40), although it is
considering kFLC index cutoff values of 50 and 100, based on two unclear if this would influence the kFLC index. In any case, this
recent positive studies (22, 35). There was no statistically significant would favor finding higher levels of the kFLC index in the aMS
difference in the number of high kFLC index values between group and lower levels in the bMS group, that is, it would facilitate
the groups. finding differences between groups. Given that no such differences
Moreover, we must consider that studies with a shorter follow- were observed in our study, it is unlikely that this played a role. MRI
up time mainly assessed the accumulated disability derived from data were not included because they were not uniformly available
inflammatory activity, while in our study, with a median follow-up across the sample, which constitutes another limitation of
of approximately 20 years, it is more likely that the progressive this study.
phase of the disease had a greater influence on the disability of our In conclusion, the levels of kFLC did not differ between patients
patients. It is worth mentioning that 12 out of the 15 patients from with aggressive and benign MS in the long term. Our approach can
the aMS group in our study progressed to the secondary be used as an efficient way to select candidate biomarkers for the
progressive phase during follow-up. In fact, some studies have long-term prognosis of MS.
suggested that the kFLC index could be related to the CNS-
resident B lymphocyte load, which gives it biological plausibility
(22, 33, 36). This supports the hypothesis that the kFLC index Data availability statement
can predict the more inflammation-dependent phase of the disease
but not the progressive phase, which is more dependent on The raw data supporting the conclusions of this article will be
neurodegeneration pathogenically associated with slow or made available by the authors, without undue reservation.

Frontiers in Immunology 06 frontiersin.org

Arroyo-Pereiro et al. 10.3389/fimmu.2023.1223514

Ethics statement and The Binding Site for providing the reagents needed to perform
the kFLC analysis. Study supported by Instituto de Salud Carlos III
The studies involving humans were approved by Ethics (PT20/00171) and Xarxa de Bancs de Tumors de Catalunya,
Committee of the Hospital Universitari de Bellvitge (reference sponsored by Pla Director d’Oncologia de Catalunya (XBTC). We
PR134/22). The studies were conducted in accordance with the thank the CERCA Programme/Generalitat de Catalunya for their
local legislation and institutional requirements. The participants institutional support.
provided their written informed consent to participate in this study.

Conflict of interest
Author contributions
PA-P, LB, EM, IL, AM-V, LR-P, AM-Y and SM-Y received
PA-P contributed to the collection, analysis, and interpretation of
honoraria for participating on advisory boards and for
the data and writing of the report. LG-S, FM, BU, VM, and MF
collaborations as consultants and scientific communications; they
contributed to the analysis and interpretation of the data, and
also received research support as well as funding for travel and
correction of the report. IL contributed to data collection. AM-V,
congress expenses from Roche, Biogen Idec, Novartis, TEVA,
EM, and LR-P contributed to the data collection and report correction.
Merck, Genzyme, Sanofi, Bayer, Almirall, and Celgene.
AM-Y and SM-Y contributed to the study design, data interpretation,
LG-S and VM received funding for travel and congress expenses
and report correction. LB contributed to the study design; collection,
from The Binding Site.
analysis, and interpretation of data; and correction of the report. All
The remaining authors declare that the research was conducted
authors contributed to the article and approved the submitted version.
in the absence of any commercial or financial relationships that
could be construed as a potential conflict of interest.

Funding
The authors declare that no financial support was received for Publisher’s note
the research, authorship, and/or publication of this article.
All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their affiliated
Acknowledgments organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
The authors thank Susana Pobla for the technical support. We claim that may be made by its manufacturer, is not guaranteed or
would also like to thank all patients who participated in this study endorsed by the publisher.

References
1. Dobson R, Giovannoni G. Multiple sclerosis-a review. Eur J Neurol (2019) 26:27– 10. Obermeier B, Lovato L, Mentele R, Brück W, Forne I, Imhof A, et al. Related B cell
40. doi: 10.1111/ene.13819 clones that populate the CSF and CNS of patients with multiple sclerosis produce CSF
2. Paul A, Comabella M, Gandhi R. Biomarkers in multiple sclerosis. Cold Spring immunoglobulin. J Neuroimmunol (2011) 233(2):245–8. doi: 10.1016/j.jneuroim.2011.01.010
Harb Perspect Med (2019) 9(3). doi: 10.1101/cshperspect.a029058 11. Greenfield AL, Hauser SL. B-cell therapy for multiple sclerosis: entering an era.
3. Tintore M, Rovira À, Rı́o J, Otero-Romero S, Arrambide G, Tur C, et al. Defining Ann Neurol (2018) 83(1):13–26. doi: 10.1002/ana.25119
high, medium and low impact prognostic factors for developing multiple sclerosis. 12. Hannam-Harris AC, Smith JL. Free immunoglobulin light chain synthesis by
Brain (2015) 138(Pt 7):1863–74. doi: 10.1093/brain/awv105 human foetal liver and cord blood lymphocytes. Immunology (1981) 43(3):417–23.
4. Harding K, Williams O, Willis M, Hrastelj J, Rimmer A, Joseph F, et al. Clinical 13. Hopper JE, Papagiannes E. Evidence by radioimmunoassay that mitogen-
outcomes of escalation vs early intensive disease-modifying therapy in patients with activated human blood mononuclear cells secrete significant amounts of light chain
multiple sclerosis. JAMA Neurol (2019) 76(5):536–41. doi: 10.1001/jamaneurol.2018.4905 Ig unassociated with heavy chain. Cell Immunol (1986) 101(1):122–31. doi: 10.1016/
5. Ontaneda D, Tallantyre E, Kalincik T, Planchon SM, Evangelou N. Early highly 0008-8749(86)90191-7
effective versus escalation treatment approaches in relapsing multiple sclerosis. Pers 14. Abe M, Goto T, Kosaka M, Wolfenbarger D, Weiss DT, Solomon A. Differences
View Lancet Neurol (2019) 18:973–80. doi: 10.1016/S1474-4422(19)30151-6 in kappa to lambda (k:l) ratios of serum and urinary free light chains. Clin Exp
6. Weideman AM, Tapia-Maltos MA, Johnson K, Greenwood M, Bielekova B. Meta- Immunol (1998) 111(2):457–62. doi: 10.1046/j.1365-2249.1998.00487.x
analysis of the age-dependent efficacy of multiple sclerosis treatments. Front Neurol 15. Bradwell AR, Carr-Smith HD, Mead GP, Tang LX, Showell PJ, Drayson MT,
(2017) 8(NOV). doi: 10.3389/fneur.2017.00577 et al. Highly sensitive, automated immunoassay for immunoglobulin free light chains in
7. Genain CP, Hauser SL. Allergic encephalomyelitis in common marmosets: serum and urine. Clin Chem (2001) 47(4):673–80. doi: 10.1093/clinchem/47.4.673
pathogenesis of a multiple sclerosis-like lesion. METHODS: A Companion to 16. Ramsden DB. Multiple sclerosis: assay of free immunoglobulin light chains. Ann
Methods Enzymol (1996) 10:420–34. doi: 10.1006/meth.1996.0120 Clin Biochem (2017) 54(1):5–13. doi: 10.1177/0004563216652175
8. Ray A, Mann MK, Basu S, Dittel BN. A case for regulatory B cells in controlling 17. Arrambide G, Espejo C, Carbonell-Mirabent P, Dieli-Crimi R, Rodrı́guez-
the severity of autoimmune-mediated inflammation in experimental autoimmune Barranco M, Castillo M, et al. The kappa free light chain index and oligoclonal
encephalomyelitis and multiple sclerosis. J Neuroimmunol (2011) 230(2):1–9. doi: bands have a similar role in the McDonald criteria. Brain (2022) 145(11):3931–42.
10.1016/j.jneuroim.2010.10.037 doi: 10.1093/brain/awac220
9. Lassmann H. Pathology and disease mechanisms in different stages of multiple 18. Hegen H, Walde J, Berek K, Arrambide G, Gnanapavan S, Kaplan B, et al.
sclerosis ☆. J Neurol Sci (2013) 333:1–4. doi: 10.1016/j.jns.2013.05.010 Cerebrospinal fluid kappa free light chains for the diagnosis of multiple sclerosis: A

Frontiers in Immunology 07 frontiersin.org

Arroyo-Pereiro et al. 10.3389/fimmu.2023.1223514

systematic review and meta-analysis. Mult Scler J (2022) 29(2):169–81. doi: 10.1177/ 30. Rudick RA, Namey M, Fischer J, Medendorp SV, Boyle S. Multiple sclerosis
13524585221134213 progression in a natural history study: Predictive value of cerebrospinal fluid free kappa
19. Mené ndez-Valladares P, Garcı́a-Sá nchez MI, Cuadri Benı́tez P, Lucas M, Adorna light chains. Multiple Sclerosis (1995) 1(3):150–5. doi: 10.1177/135245859500100303
Martı́nez M, Carranco Galá n V, et al. Free kappa light chains in cerebrospinal fluid as a 31. Rinker JR, Trinkaus K, Cross AH. Elevated CSF free kappa light chains correlate
biomarker to assess risk conversion to multiple sclerosis. Mult Scler J Exp Transl Clin with disability prognosis in multiple sclerosis. Neurology (2006) 67(7):1288–90. doi:
(2015) 1:1–9. doi: 10.1177/2055217315620935 10.1212/01.wnl.0000238107.31364.21
20. Vecchio D, Crespi I, Virgilio E, Naldi P, Campisi MP, Serino R, et al. Kappa free 32. Makshakov G, Nazarov V, Kochetova O, Surkova E, Lapin S, Evdoshenko E.
light chains could predict early disease course in multiple sclerosis. Mult Scler Relat Diagnostic and prognostic value of the cerebrospinal fluid concentration of
Disord (2019) 30(January):81–4. doi: 10.1016/j.msard.2019.02.001 immunoglobulin free light chains in clinically isolated syndrome with conversion to
21. Salavisa M, Paixão P, Ladeira AF, Mendes A, Correia AS, Viana JF, et al. Prognostic mul tip le s clerosis . Pl oS One (2 015 ) 10 (1 1) :e 014 337 5. doi : 10 .137 1/
value of kappa free light chains determination in first-ever multiple sclerosis relapse. J journal.pone.0143375
Neuroimmunol (2020) 347(May):577355. doi: 10.1016/j.jneuroim.2020.577355 33. Rathbone E, Durant L, Kinsella J, Parker AR, Hassan-Smith G, Douglas MR,
22. Castillo-Villalba J, Gil-Perotı́n S, Gasque-Rubio R, Cubas-Nuñez L, Carratalà- et al. Cerebrospinal fluid immunoglobulin light chain ratios predict disease progression
Boscà S, Alcalá C, et al. High levels of cerebrospinal fluid kappa free light chains relate in multiple sclerosis. J Neurol Neurosurg Psychiatry (2018) 89(10):1044–9. doi: 10.1136/
to igM intrathecal synthesis and might have prognostic implications in relapsing jnnp-2018-317947
multiple sclerosis. Front Immunol (2022) 13:1. doi: 10.3389/fimmu.2022.827738 34. Presslauer S, Milosavljevic D, Huebl W, Parigger S, Schneider-Koch G, Bruecke
23. Thompson AJ, Banwell BL, Barkhof F, Carroll WM, Coetzee T, Comi G, et al. T. Kappa free light chains: Diagnostic and prognostic relevance in MS and CIS. PloS
Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol One (2014) 9(2):e89945. doi: 10.1371/journal.pone.0089945
(2018) 17:162–73. doi: 10.1016/S1474-4422(17)30470-2 35. Cutellè C, Balducci C, Cereda D, Fusco ML, Iacobucci D, Perugini J, et al. K
24. Confavreux C, Compston DAS, Hommes R, Mcdonald WI, Thompson AJ. index utility as diagnostic and prognostic biomarker in the assessment of patients with
EDMUS, a European database for multiple sclerosis. Neurosurgery Psychiatry (1992) suspected Multiple Sclerosis. J Neuroimmunol (2022) 373:577992. doi: 10.1016/
55:671–6. doi: 10.1136/jnnp.55.8.671 j.jneuroim.2022.577992
25. Lorscheider J, Buzzard K, Jokubaitis V, Spelman T, Havrdova E, Horakova D, 36. Cepok S, Rosche B, Grummel V, Vogel F, Zhou D, Sayn J, et al. Short-lived
et al. Defining secondary progressive multiple sclerosis. Brain (2016) 139(9):2395–405. plasma blasts are the main B cell effector subset during the course of multiple sclerosis.
doi: 10.1093/brain/aww173 Brain (2005) 128(Pt 7):1667–76. doi: 10.1093/brain/awh486
26. Senel M, Tumani H, Lauda F, Presslauer S, Mojib-Yezdani R, Otto M, et al. 37. Correale J, Gaitá n MI, Ysrraelit MC, Fiol MP. Progressive multiple sclerosis:
Cerebrospinal fluid immunoglobulin kappa light chain in clinically isolated syndrome from pathogenic mechanisms to treatment. Brain (2017) 140(3):527–46. doi: 10.1093/
and multiple sclerosis. PloS One (2014) 9(4):e88680. doi: 10.1371/journal.pone.0088680 brain/aww258
27. Voortman MM, Stojakovic T, Pirpamer L, Jehna M, Langkammer C, Scharnagl 38. Maack T, Johnson V, Kau ST, Figueired J, Sigulem D. Renal filtration, transport,
H, et al. Prognostic value of free light chains lambda and kappa in early multiple and metabolism of low-molecular-weight proteins: A review. Kidney Int (1979) 16:70.
sclerosis. Multiple Sclerosis (2017) 23(11):1496–505. doi: 10.1177/1352458516681503 doi: 10.1038/ki.1979.128
28. Gaetani L, Di Carlo M, Brachelente G, Valletta F, Eusebi P, Mancini A, et al. 39. Hutchison CA, Harding S, Hewins P, Mead GP, Townsend J, Bradwell AR, et al.
Cerebrospinal fluid free light chains compared to oligoclonal bands as biomarkers in Quantitative assessment of serum and urinary polyclonal free light chains in patients
multiple sclerosis. J Neuroimmunol [Internet] (2020) 339:577108. doi: 10.1016/ with chronic kidney disease. Clin J Am Soc Nephrol (2008) 3(6):1684. doi: 10.2215/
j.jneuroim.2019.577108 CJN.02290508
29. Berek K, Bsteh G, Auer M, Di Pauli F, Grams A, Milosavljevic D, et al. Kappa- 40. Basile U, Gulli F, Gragnani L, Napodano C, Pocino K, Rapaccini GL, et al. Free
free light chains in CSF predict early multiple sclerosis disease activity. Neurology(R) light chains: Eclectic multipurpose biomarker. J Immunol Methods (2017) 451
neuroimmunol Neuroinflamm (2021) 8(4):e1005. doi: 10.1212/NXI.0000000000001005 (April):11–9. doi: 10.1016/j.jim.2017.09.005

Frontiers in Immunology 08 frontiersin.org