845844 research-article2019 MSJ0010.1177/1352458519845844Multiple Sclerosis JournalCE Leurs, HAM Twaalfhoven MULTIPLE SCLEROSIS MSJ JOURNAL Original Research Paper Kappa free light chains is a valid tool in the Multiple Sclerosis Journal 2020, Vol. 26(8) 912­–923 diagnostics of MS: A large multicenter study DOI: 10.1177/ https://doi.org/10.1177/1352458519845844 1352458519845844 https://doi.org/10.1177/1352458519845844 © The Author(s), 2019. CE Leurs, HAM Twaalfhoven, BI Lissenberg-Witte, V van Pesch, I Dujmovic, J Drulovic, Article reuse guidelines: M Castellazzi , T Bellini, M Pugliatti, J Kuhle, LM Villar, JC Alvarez-Cermeño, sagepub.com/journals- R Alvarez-Lafuente , H Hegen, F Deisenhammer , LM Walchhofer, E Thouvenot, permissions M Comabella, X Montalban, L Vécsei, C Rajda, D Galimberti, E Scarpini, A Altintas, K Rejdak, JL Frederiksen, G Pihl-Jensen, PEH Jensen, M Khalil , MM Voortman , F Fazekas, A Saiz , D La Puma, M Vercammen, L Vanopdenbosch, BMJ Uitdehaag, J Killestein, C Bridel and C Teunissen Abstract Correspondence to: CE Leurs Objective: To validate kappa free light chain (KFLC) and lambda free light chain (LFLC) indices as a Department of Neurology, MS Center Amsterdam, diagnostic biomarker in multiple sclerosis (MS). VU University Medical Methods: We performed a multicenter study including 745 patients from 18 centers (219 controls and Center, De Boelelaan 1118, Amsterdam 1081 HZ, The 526 clinically isolated syndrome (CIS)/MS patients) with a known oligoclonal IgG band (OCB) status. Netherlands. KFLC and LFLC were measured in paired cerebrospinal fluid (CSF) and serum samples. Gaussian mix- c.leurs@vumc.nl CE Leurs ture modeling was used to define a cut-off for KFLC and LFLC indexes. BMJ Uitdehaag Results: The cut-off for the KFLC index was 6.6 (95% confidence interval (CI) = 5.2–138.1). The cut-off J Killestein Department of Neurology, for the LFLC index was 6.9 (95% CI = 4.5–22.2). For CIS/MS patients, sensitivity of the KFLC index MS Center Amsterdam, (0.88; 95% CI = 0.85–0.90) was higher than OCB (0.82; 95%CI = 0.79–0.85; p < 0.001), but specificity Amsterdam Neuroscience, Amsterdam UMC, location (0.83; 95% CI = 0.78–0.88) was lower (OCB = 0.92; 95% CI = 0.89–0.96; p < 0.001). Both sensitivity and VUmc, Amsterdam, The Netherlands specificity for the LFLC index were lower than OCB. HAM Twaalfhoven Conclusion: Compared with OCB, the KFLC index is more sensitive but less specific for diagnosing C Bridel C Teunissen CIS/MS. Lacking an elevated KFLC index is more powerful for excluding MS compared with OCB but Neurochemistry Laboratory the latter is more important for ruling in a diagnosis of CIS/MS. and Biobank, Department of Clinical Chemistry, MS Center Amsterdam, Amsterdam Neuroscience, Amsterdam UMC, location Keywords: Multiple sclerosis, KFLC, OCB, CSF, biomarkers VUmc, Amsterdam, The Netherlands BI Lissenberg-Witte Date received: 20 October 2018; revised: 11 February 2018; accepted: 28 February 2019 Department of Epidemiology and Biostatistics, Amsterdam UMC, location VUmc, Introduction the MS diagnosis in patients with a single clinical epi- Amsterdam, The Netherlands Cerebrospinal fluid (CSF) assessment is often part of sode suggestive of central nervous system (CNS) V van Pesch Department of Neurology, the diagnostic workup for multiple sclerosis (MS), inflammatory demyelinating disease. The OCB Cliniques Universitaires and its value is supported by the latest 2017 revisions assessment also has important prognostic value in Saint-Luc, Université Catholique de Louvain, of the McDonalds criteria.1 CSF examination is fre- CIS and MS.4,5 However, the assessment of OCB is Sint-Lambrechts-Woluwe, quently performed in diagnosing MS for excluding labor-intensive, requires trained personnel, and is in Belgium alternative diagnoses, although, in the current MS cri- some cases examiner- and method-dependent, which I Dujmovic J Drulovic teria,2 the assessment of oligoclonal IgG bands (OCB) may affect its reliability. Clinic of Neurology, Clinical plays a limited role. However a recent study showed Centre of Serbia, School of Medicine, University of the added value of OCB in the MS diagnostic crite- Alongside intact immunoglobulins, which are com- Belgrade, Belgrade, Serbia ria.3 In the 2017 revisions, the OCB have a prominent posed of two heavy and two light chains, plasma cells M Castellazzi role in patients with a clinically isolated syndrome produce and secrete immunoglobulin free light T Bellini M Pugliatti (CIS). The presence of both magnetic resonance chains (FLC) of either kappa (KFLC) or lambda Department of Biomedical imaging (MRI) criteria for dissemination in space (LFLC) chains. KFLC and LFLC can be detected in and Specialty Surgical Sciences, University of (DIS) and CSF-specific OCB will enable to establish both CSF and serum.6–10 Since the late 1970s, Ferrara, Ferrara, Italy 912 journals.sagepub.com/home/msj CE Leurs, HAM Twaalfhoven et al. J Kuhle multiple studies have reported increased CSF levels CSF and serum samples Neurologic Clinic and of KFLC in MS.6–14 The analytical specificity of the Only CSF samples that were immediately centrifuged Policlinic, Departments of Medicine, Clinical Research earlier methods (e.g. radioimmunoassay,15,16 quanti- and stored in polypropylene tubes within 2 hours at and Biomedicine, University tative enzyme-linked immunosorbent assay)8,17 was −80°C, at the local center, were included. The assess- Hospital Basel, University of Basel, Basel, Switzerland insufficient, but with the recent emergence of the ment of OCB had been performed by isoelectric LM Villar more sensitive nephelometric and turbidimetric FLC focusing (on agarose or polyacrylamide gel), fol- Department of Immunology, assays, research in this field has been revived. lowed by immunofixation by the centers as part of the Hospital Ramón y Cajal, IRYCIS, Madrid, Spain/ Nephelometric (and turbidimetric) FLC level deter- diagnostic workup. Red Española de Esclerosis mination has the additional advantage compared to Múltiple (REEM), Madrid, Spain OCB of being assessed by an automated procedure Samples were taken between 2005 and 2016, with a JC Alvarez-Cermeño and being quantifiable.18 median age of 2.9 years (IQR = 1.7–5.7). Red Española de Esclerosis Múltiple (REEM), Madrid, Spain/Department of Using the FLC assay,19 recent studies showed that We used fresh aliquots, and in our lab, we did not Neurology, Hospital Ramón both CSF KFLC levels and the KFLC index are freeze and thaw the samples during the analyses. As y Cajal, IRYICIS, Madrid, Spain increased in patients with CIS or relapsing-remitting far as we know, no effects of freezing and thawing R Alvarez-Lafuente MS (RRMS) compared with controls.8,14,18,20–22 The have been reported. Red Española de Esclerosis use of an index measure is necessary, for example Múltiple (REEM), Madrid, Spain/Grupo de Investigación [CSF KFLC/serum KFLC]/[CSF albumin/serum de Esclerosis Múltiple, albumin], to include blood-CSF barrier permeabil- KFLC, LFLC, and albumin analysis Hospital Clínico San Carlos, Instituto de Investigación ity.10,14 The KFLC index has comparable sensitivity KFLC, LFLC, and albumin concentrations in CSF and Sanitaria San Carlos and specificity to OCB for diagnosis of MS and serum samples were analyzed using the turbidimetric (IdISSC), Madrid, Spain CIS.14,21,23 However, large-scale studies comparing analyzer SPAplus® (The Binding Site, Birmingham, H Hegen diagnostic performance of the two methods and to UK) with the serum free light chain immunoassay F Deisenhammer LM Walchhofer define the cut-off of FLC are lacking. The main aim (Freelite®, The Binding Site, Birmingham, UK) Department of Neurology, of this study was to validate KFLC and LFLC indices according to the manufacturer’s instructions. All sam- Medical University of Innsbruck, Innsbruck, Austria as a diagnostic biomarker in MS compared with OCB ples were measured centrally in the Neurochemistry E Thouvenot in a large multicenter study including samples from Laboratory of the Department of Clinical Chemistry Department of Neurology, 18 MS centers across Europe. of the VU University Medical Center (VUmc), CHU Nîmes, Hôpital Caremeau, Nîmes, France/ Amsterdam, the Netherlands. All samples were run Institut de Génomique blinded for the clinical data. Fonctionnelle, UMR5203, Université Montpellier, Methods Montpellier, France To verify the QC data supplied by the manufacturer, M Comabella Patients and controls we calculated intra-assay coefficient of variation X Montalban Servei de Neurologia- Eighteen MS centers participated, located in the (CV) by taking the mean CVs of four replicates of Neuroimmunologia, Centre Netherlands, Spain, France, Belgium, Hungary, Italy, five samples (CSF/serum) within one run. We calcu- d’Esclerosi Múltiple de Catalunya (Cemcat), Institut Poland, Turkey, Denmark, Serbia, Austria, and lated inter-assay CV based on n = 5 samples (CSF/ de Recerca Vall d’Hebron Switzerland. We selected 745 paired CSF/serum sam- serum) measured in 5 different days. CV values for (VHIR), Hospital Universitari Vall d’Hebron, Universitat ples from patients with known OCB status, diagnosed KFLC and LFLC were all found to be lower than Autònoma de Barcelona, as CIS (n = 242), RRMS (n = 235), primary-progres- those supplied by the manufacturer (Supplemental Barcelona, Spain sive MS (PPMS) (n = 41), and secondary-progressive Table 1). CV values for albumin were comparable to L Vécsei Department of Neurology, MS (SPMS) (n = 8). We also included inflammatory those supplied by the manufacturer. Assay linearity University of Szeged, neurological disease controls (INDC) (n = 67), non- was experimentally confirmed for albumin (serum Szeged, Hungary/MTA- SZTE Neuroscience Research inflammatory neurological disease controls (NINDC) and CSF) and the FLC assays in serum and showed Group, Szeged, Hungary (n = 76), symptomatic controls (SC) (n = 49), and that recalculated values varied by 25.7% (KFLC) and C Rajda Department of Neurology, healthy controls (HC) (n = 27) as defined previously.24 14.2% (LFLC) from the original value. University of Szeged, The different control groups were pooled into one Szeged, Hungary control group (n = 219). The CIS and MS patients For 29 samples, CSF albumin results were below D Galimberti E Scarpini were also pooled (CIS/MS) (n = 526). detection. Here, we assigned a random uniform value Multiple Sclerosis Centre, between 35 mg/mL (lowest detected value in undi- University of Milan, Dino Ferrari Centre, Fondazione The large majority (84%) of the CIS/MS patients luted rerun) and 175 mg/mL (formal detection limit). Ca’ Granda, IRCCS fulfilled the 2010 McDonald criteria2 but, in some Ospedale Policlinico, Milan, Italy cases, the patients were diagnosed according to the A Altintas 2005 McDonald criteria25 (16%). Table 1 presents FLC indices Koc University, School the demographic and clinical characteristics of the We determined the CSF/serum quotients (Q FLC) of of Medicine, Neurology Department, Istanbul, patients and controls. KFLC and LFLC and calculated indices in order to Turkey journals.sagepub.com/home/msj 913 914 A Saiz Belgium M Khalil K Rejdak F Fazekas D La Puma PEH Jensen Graz, Austria Lublin, Poland G Pihl-Jensen M Vercammen MM Voortman JL Frederiksen d’Investigacions Barcelona, Spain L Vanopdenbosch AZ Sint-Jan Brugge- Service of Neurology, Centre, Department of Medicine, AZ Sint-Jan Copenhagen, Denmark Copenhagen, Denmark Hospital Clinic, Institut Biomèdiques August Pi i Sunyer (IDIBAPS) and Universitat de Barcelona, Danish Multiple Sclerosis Department of Laboratory Department of Neurology, Department of Neurology, Department of Neurology, Department of Neurology, Brugge-Oostende, Brugge, Neurology, Rigshospitalet, University of Copenhagen, University of Copenhagen, Oostende, Brugge, Belgium Rigshospitalet Glostrup and Medical University of Graz, Center of Neuroimmunology, Medical University of Lublin, Multiple Sclerosis Journal 26(8) Table 1. Demographic and clinical characteristics of the included patients and controls at the time of lumbar puncture. Disease N = 745 Sex (female) Age (years) Stage/type Disease duration No. of No. of EDSS OCB group N (%) Mean ± SD of disease (months) patients patients on Median (IQR) positive RR/SP/PP Median (IQR) using corticosteriods N (%) treatment Median (IQR) CIS 242 177 (73.1) 35 ± 10 n.a. 0.8 (0.3–3.0) 1 4 2.0 (1.0–2.5)a 186 (76.9) MS 284 170 (59.9) 38 ± 11 235/8/41 13.1 (2.4–48.1) 3b 10 2.0 (1.5–3.5)c 245 (86.3) Controls 219 130 (59.4)d 42 ± 12 n.a. n.d. 1 1 n.a. 17 (7.8) - NINDC 76 49 (64.5) 45 ± 13 n.a. n.d. 0 0 n.a. 4 (5.3) - INDC 67 33 (49.3) 42 ± 13 n.a. n.d. 1e 1e n.a. 13 (19.4) - SC 49 32 (65.3) 39 ± 10 n.a. n.d. 0 0 n.a. 0 (0) - HC 27 16 (59.3) 41 ± 10 n.a. n.d. 0 0 n.a. 0 (0) RR: relapsing-remitting; SP: secondary progressive; PP: primary progressive; disease duration: time between CSF lumbar puncture and date of onset first neurological complaints; EDSS: Expanded Disability Status Scale, OCB: oligoclonal IgG bands; CIS: clinically isolated syndrome; MS: multiple sclerosis; n.a.: not applicable; n.d.: not determined; NINDC: non-inflammatory neurological disease control; INDC: inflammatory neurological disease controls; SC: symptomatic control; HC: healthy control; CSF: cerebrospinal fluid; CLIPPERS: chronic lymphocytic inflammation with pontine perivascular enhancement responsive to steroids. All participants gave written informed consent at the center where the CSF/serum was conducted. aIn 178 CIS patients, an EDSS was available. bDisease-modifying treatment in combination with corticosteroids was used in two MS patients. In one patient, only disease-modifying treatment. cIn 107 MS patients, an EDSS was available. dIn six control patients, the sex was unknown. eOne INDC (CLIPPERS syndrome) patient had steroids and methotrexate at the time of lumbar puncture. journals.sagepub.com/home/msj CE Leurs, HAM Twaalfhoven et al. Table 2. FLC in CIS/MS and controls. A CIS/MS (n = 526) Controls (n = 219) p-value CSF KFLC (mg/L) 3.7 (1.0–10.1) 0.2 (0.1–0.4) <0.001 Serum KFLC (mg/L) 12.4 (9.8–15.7) 13.7 (10.6–16.9) 0.001 KFLC index 75.6 (21.8–197.0) 2.8 (2.2–4.9) <0.001 KFLC index ⩾ 6.6 460 (87.5) 38 (17.4) – B CIS/MS (n = 543) Controls (n = 202) p-value CSF LFLC (mg/L) 0.5 (0.2–1.3) 0.2 (0.2–0.3) < 0.001 Serum LFLC (mg/L) 11.3 (9.2–13.7) 12.1 (9.8–15.4) 0.001 LFLC index 11.6 (5.3–33.3) 3.3 (2.5–4.7) < 0.001 LFLC index ⩾ 6.9 359 (68.3) 30 (13.7) – FLC: free light chains; CIS: clinically isolated syndrome; MS: multiple sclerosis; A: KFLC in CIS/MS and controls; CSF: cerebrospi- nal fluid; KFLC: kappa free light chain; index: FLC quotient/albumin quotient; B: LFLC in CIS/MS and controls; LFLC: lambda free light chains; IQR: interquartile range. Values are given as n (%) or as median (IQR). take possible blood-CSF diffusion into account. The three combinations were as follows: KFLC with FLC indices were defined as [Q FLC] × [serum albu- OCB, LFLC with OCB, and KFLC with LFLC. We min/CSF albumin]. defined the outcome of the combination as followed: when one of the measurements is positive, the combi- nation is positive, when both are negative, the combi- nation is negative. Statistics Differences in demographics, clinical characteristics, To compare the sensitivity, specificity, and accuracy FLC concentrations, and FLC indices were tested via between the two different diagnostic tools (OCB and the Mann–Whitney U test for comparison of two FLC), the McNemar test was used (SPSS 22.0 (IBM groups of non-normally distributed data. For compar- Crop., Armonk, NY, USA)). The positive and negative ison of more than two groups, Kruskal–Wallis test predictive values (PPVs and NPVs) were compared with post hoc Dunn’s multiple comparison test was using the R package DTcompair. The p values < 0.05 applied. For normally distributed continuous varia- were considered statistically significant. bles, an independent-samples t test or one-way analy- sis of variance (ANOVA) with post hoc Bonferroni correction was applied. For binary variables, a chi- Results square test was performed. These statistical analyzes Paired CSF and serum samples from a total of 745 were performed in SPSS 22.0 (IBM Crop., Armonk, patients were included in this multicenter study (see NY, USA). Supplemental Figure 1 for the flowchart of patient selection). The patient groups (n = 526) consisted of Gaussian mixture modeling was used to define cut- 242 CIS and 284 MS patients. The control group offs for abnormal FLC indices using the R statistical (n = 219) consisted of 76 NINDC, 67 INDC, 49 SC, software program version 3.2.1 mixtools package. and 27 HC. The control group was older First, the number of distributions that best described (mean = 42 ± 12 years) than the CIS/MS patient group the data was determined with the R boot.comp func- (mean = 35 ± 10 years, p < 0.001). There was no sig- tion. Next, we defined a data-driven cut-off as the nificant difference in sex distribution between the two point where the lines of two fitted normal distribu- groups (p = 0.20). Only 7.7% of the control group had a tions crossed each other. The main analyses included positive OCB status. The diagnoses of these patients all subjects. Data were log-transformed because FLC are as follows: Tolosa hunt (n = 2), meningitis (n = 1), indices were not normally distributed. Based on the neurosarcoidosis (n = 2), lupus (n = 1), transient global defined cut-off, subjects were classified as positive or amnesia (n = 1), stroke (n = 1), post-infectious myelitis negative for kappa or lambda FLC as binary result, (n = 1), myelitis (n = 1), encephalitis (n = 1), epilepsy similar as the available results for OCB status. As (n = 1), ADEM (n = 1), neurodegeneration (n = 1), extra comparisons, we combined the different tools to INDC, exact diagnosis unknown (n = 2), and neuritis compare with the single measurement OCB. The optica (n = 1). journals.sagepub.com/home/msj 915 Multiple Sclerosis Journal 26(8) Figure 1. CSF KFLC and LFLC indices of CIS, MS, CIS/MS and controls. (a) Levels of KFLC indices, (b) levels of LFLC indices. Horizontal bars in the scatter dot plot represent the median. CSF: cerebrospinal fluid; KFLC: kappa free light chains; LFLC: lambda free light chains; CIS: clinically isolated syndrome; MS: multiple sclerosis. *p ⩽ 0.001. Figure 2. Cerebrospinal fluid (a) KFLC and (b) LFLC cut-off values based on mixture modeling. The red (low values) and green (high values) lines are the individual components of the estimated mixture distributions, the dotted line is the combined estimated mixture distribution. The cut-off is defined as the point where the red and green lines cross. For a detailed list including the diagnoses of non- See Figure 1(a) for the KFLC indices per subgroup inflammatory neurological diseases and inflamma- and Figure 1(b) for the LFLC indices per subgroup. tory neurological diseases included in the groups, see KFLC and LFLC indices were significantly increased Supplemental Tables 2A and 2B. in CIS/MS compared to controls (both p < 0.001). In addition, FLC indices were higher in MS than in CIS (KFLC p = 0.07, LFLC p < 0.001). FLC concentrations and FLC indices Table 2 shows the levels and indices of the KFLC and LFLC in CIS/MS patients and controls. CSF KFLC FLC index cut-off and CSF LFLC concentrations were significantly In the total cohort (n = 745) (all diagnostic subgroups), increased in CSF of CIS/MS patients compared to a bimodal distribution of the log-transformed KFLC controls (both p < 0.001). In addition, KFLC and index values fitted the data best. This yielded a cut-off LFLC concentrations in CSF were higher in MS for the log-KFLC index of 1.89 (95% confidence patients than CIS patients (both p < 0.001). interval (CI) = 1.65–4.92) (Figure 2(a)), which corre- sponds to an KFLC index of 6.6 (95% CI = 5.2–138.1) KFLC and LFLC serum concentrations were signifi- on the original scale. cantly higher in the control group than CIS/MS (both, p = 0.001) but did not differ significantly between CIS For the LFLC index, a bimodal distribution yielded an and MS (KFLC p = 0.33, LFLC p = 1.00). optimal cut-off for the log-LFLC index of 1.9 (95% 916 journals.sagepub.com/home/msj CE Leurs, HAM Twaalfhoven et al. Table 3. Diagnostic sensitivity and specificity of the FLC indices compared to OCB. OCB KFLC p-value LFLC p-value CIS/MS vs controls Sensitivity 0.82 (95% 0.88 (95% < 0.001 0.66 (95% < 0.001 CI = 0.79–0.85) CI = 0.85–0.90) CI = 0.62–0.70) Specificity 0.92 (95% 0.83 (95% < 0.001 0.86 (95% 0.019 CI = 0.89–0.96) CI = 0.78–0.88) CI = 0.81–0.91) Accuracy 0.85 (95% 0.86 (95% 1.000 0.74 (95% < 0.001 CI = 0.82–0.88) CI = 0.84–0.89) CI = 0.70–0.77) PPV 0.96 (95% 0.92 (95% < 0.001 0.92 (95% 0.001 CI = 0.94–0.98) CI = 0.90–0.95) CI = 0.90–0.95) NPV 0.68 (95% 0.73 (95% 0.010 0.53 (95% < 0.001 CI = 0.63–0.73) CI = 0.68–0.79) CI = 0.48–0.58) CIS vs controls Sensitivity 0.77 (95% 0.81 (95% 0.150 0.60 (95% < 0.001 CI = 0.72–0.82) CI = 0.76–0.86) CI = 0.54–0.66) Specificity 0.92 (95% 0.83 (95% < 0.001 0.86 (95% 0.019 CI = 0.89–0.96) CI = 0.78–0.88) CI = 0.82–0.91) Accuracy 0.84 (95% 0.82 (95% 0.149 0.73 (95% < 0.001 CI = 0.80–0.88) CI = 0.78–0.85) CI = 0.68–0.77) PPV 0.92 (95% 0.84 (95% < 0.001 0.83 (95% 0.001 CI = 0.88–0.95) CI = 0.79–0.88) CI = 0.77–0.89) NPV 0.78 (95% 0.79 (95% 0.56 0.66 (95% < 0.001 CI = 0.73–0.83) CI = 0.74–0.84) CI = 0.61–0.72) MS vs controls Sensitivity 0.86 (95% 0.93 (95% < 0.001 0.75 (95% < 0.001 CI = 0.82–0.90) CI = 0.90–0.97) CI = 0.70–0.80) Specificity 0.92 (95% 0.83 (95% < 0.001 0.86 (95% 0.019 CI = 0.89–0.96) CI = 0.78–0.88) CI = 0.81–0.91) Accuracy 0.89 (95% 0.89 (95% 1.000 0.80 (95% < 0.001 CI = 0.86–0.92) CI = 0.86–0.92) CI = 0.77–0.84) PPV 0.94 (95% 0.88 (95% < 0.001 0.88 (95% 0.002 CI = 0.91–0.97) CI = 0.84–0.91) CI = 0.84–0.92) NPV 0.84 (95% 0.91 (95% 0.001 0.73 (95% < 0.001 CI = 0.79–0.88) CI = 0.86–0.95) CI = 0.68–0.78) FLC: free light chains; OCB: oligoclonal IgG bands; KFLC: kappa free light chains; LFLC: lambda free light chains; CIS: clinically isolated syndrome; MS: multiple sclerosis; PPV: positive predictive value; NPV: negative predictive value. To compare the sensitivity and specificity between the two different diagnostic tools (OCB and FLC), the McNemar test was used. CI = 1.5–3.1) (Figure 2(b)), which corresponds to an When identifying CIS patients from controls, the LFLC index of 6.9 (95% CI = 4.5–22.2) on the origi- sensitivities did not differ significantly between nal scale. KFLC and OCB (KFLC index = 0.81, OCB = 0.77; p = 0.15), but the specificity of OCB was significantly higher (KFLC index = 0.83, OCB = 0.92, p < 0.001). Diagnostic sensitivity and specificity of the Identifying MS patients from controls, the sensitivity indices compared to OCB of the KFLC index (0.93) was significantly higher KFLC index. When pooling CIS and MS as one than of OCB (0.86; p < 0.001) but the specificity group (CIS/MS), the sensitivity to identify CIS/ (0.83) was significantly lower (0.92; p < 0.001). MS from controls of the KFLC index (0.88) was significantly higher than of OCB (0.82; p < 0.001) The accuracies were similarly high for both biomark- at the cost of a significantly lower specificity ers for all three comparisons. The PPVs were higher (KFLC index = 0.83, OCB = 0.92; p < 0.001) (see for OCB (in all three comparisons p < 0.001). The Table 3). NPV was the same for both markers (p = 0.56) in CIS journals.sagepub.com/home/msj 917 Multiple Sclerosis Journal 26(8) versus controls, but when pooling CIS/MS and in MS No relevant differences were seen when we exclude alone, the NPVs were slightly higher for KFLC patients with the CIS diagnosis according to (p = 0.010, p < 0.001) (see Table 3). McDonald 2005 criteria. LFLC index. When pooling CIS and MS in one group When excluding INDC from the control group, higher (CIS/MS), the sensitivity to identify CIS/MS from con- specificities were seen for OCB, KFLC, and LFLC trols of the LFLC index (0.66) was significantly lower when comparing CIS/MS with the controls. However, than that of OCB (0.82; p < 0.001). The specificity to the p value did not change relevantly when comparing discriminate CIS/MS from controls was also signifi- the new specificities of OCB, KFLC, and LFLC (data cantly lower (0.86) than for OCD (0.92; p = 0.019). not shown). When identifying CIS patients from controls, the sensitivities did differ; the LFLC index was signifi- Discussion cantly lower than OCB (LFLC index = 0.60; Our study indicates that the KFLC index is a valid test OCB = 0.77; p < 0.001). The specificity of LFLC in for diagnosing CIS/MS. Compared to OCB, the CIS was lower than OCB (LFLC index = 0.86; KFLC index is more sensitive at the cost of a lower OCB = 0.92; p = 0.019). Identifying MS patients from specificity. This trade-off resulted in a higher NPV for controls, the sensitivity of the LFLC index (0.75) was the KFLC index compared to OCB, but a lower PPV. significantly lower than OCB (0.86; p < 0.001). The In addition, our results indicate that the LFLC index is specificity (0.86) was also significantly lower than not a valid test for diagnosing CIS/MS. OCB (0.92; p = 0.019). The accuracies and NPV were significantly lower for LFLC than OCB for all three Our sensitivity and specificity for the KFLC index in comparisons. The PPV was significantly lower for CIS/MS were lower than a few much smaller previous LFLC than OCB when comparing MS with controls, studies,18,23,26 which reported sensitivity in the range and in the other comparisons, the PPV of LFLC was of 0.93–0.95 and specificity in the range of 0.91–1.00. similar to OCB (see Table 3). The study of Desplat-Jégo et al.9 showed a lower sen- sitivity of 0.70 and a lower specificity of 0.82 for the KFLC index in MS patients. The more recent study of Combination of the different tools compared to Vasilj et al.27 showed a lower sensitivity of 0.71, how- single measurement ever a higher specificity of 0.98. The results of the Three combinations were made, KFLC index–OCB, comparison of KFLC and OCB are in line with results LFLC index–OCB, and KFLC index–LFLC index, of a recent multicenter study,14 where a cut-off of 5.9 and all were compared with the analysis of OCB was employed to validate KFLC in CSF as a diagnos- alone. Sensitivity and specificity were calculated for tic biomarker in 60 CIS patients, 60 MS patients com- all the subgroups compared to the control group. The pared to 60 OND, reporting a higher sensitivity of the sensitivity of the combination OCB with the KFLC KFLC index (0.78) compared to OCB (0.72) for diag- index increased to 0.88, and the specificity of this nosis of CIS. In MS, the sensitivity of the KFLC and combination decreased to 0.83. The same results were OCB was comparable (0.93 vs 0.93). However, the obtained for the combinations of LFLC index–OCB specificity (0.95) in CIS and MS was higher compared and KFLC index–LFLC index compared to single to our study. Another paper used the same 5.9 cut-off; OCB, showing a slightly higher sensitivity (LFLC this resulted in a sensitivity and specificity of 0.96 and index–OCB = 0.87, KFLC–LFLC index = 0.87) and a 0.98, respectively, in MS patients.21 lower specificity (0.83 and 0.80, respectively). This discrepancy in the crude sensitivity and specific- ity of the KFLCs may be due to the more heterogene- Sensitivities and specificities in alternative ous control group in our study compared to the subgroups previous studies, due to pooling of the CIS/MS group Including patients with the diagnosis CIS (according or the inclusion of not only clinical definite MS to McDonald 2005 criteria) as RRMS patients resulted patients. For example, similar as for the OCB, KFLCs in lower sensitivities of OCB, KFLC, and LFLC that can be elevated in inflammatory controls,28 and thus were seen when comparing CIS patients with con- specificity will be lower when included. Nevertheless, trols. However, the p value did not change relevantly this is a very relevant control group in differential when comparing the new sensitivities for OCB, diagnosis of MS. The unprecedented large number of KFLC, and LFLC. No relevant differences were seen patients and the large heterogeneous control group in when comparing MS to control group. this study gave us a reflection of the real-life clinical 918 journals.sagepub.com/home/msj CE Leurs, HAM Twaalfhoven et al. situation, thus avoiding spectrum bias, and allowed us MS centers (participating in the BioMSeu consortium) to give a more representative sensitivity and specific- using standardized protocols.33 Moreover, inter-labo- ity for OCB, KFLC, and LFLC indices. ratory agreement is reported to be good for OCB, for example an inter-laboratory agreement of kappa > 0.8 Noticeable is that the sensitivity and specificity of between 19 participating laboratories in Spain was OCB to discriminate MS patients from controls were observed.34 lower in our study than previously reported.10,29,30 However, a meta-analysis published in 201331 (13,467 One more important note is that the best set up for the patients) showed that the diagnostic specificity of study would have been if the test population should be OCB diminished if other inflammatory etiologies suspected MS cases and not already diagnosed with were considered. Therefore, the lower specificities in MS. Still, as provided in Supplemental Table 2B, we our study may also be due to inclusion of various con- included various INDC and quite some patients ini- trol groups. tially suspected for demyelinating disease. The cut-off in this study was calculated using a data- Alongside the sensitivity and specificity results of the driven Gaussian mixture modeling approach. We indices, we found significantly increased FLC con- chose a different approach compared to other studies centrations and quotients in CSF of CIS/MS patients that applied, for example, receiver operating charac- compared to the control groups. However, our main teristic (ROC) curve analysis and area under the curve focus in this study was in the FLC indices and not in (AUC) values,9,10,18,23,27,32 because we reasoned that the concentrations of the FLC. To control for blood- the cut-off should be defined by biological levels CSF barrier function, we used indices instead of (data driven) and not based on clinical diagnosis, concentrations. which is an imperfect golden standard. We deter- mined a cut-off of 6.6 for abnormal KFLC indices and Combination of different markers (KFLC index– 6.9 for abnormal LFLC indices. Our cut-off for KFLC OCB, LFLC index–OCB, and KFLC index–LFLC is in line with a previous multicenter study showing a index) compared to the single measurement OCB, KFLC index cut-off of 5.9.10 This almost comparable showed that the combination KFLC index–OCB com- cut-off for the KFLC index in two multicenter studies pared to single OCB gave a slightly higher sensitivity supports its robustness and implies that it can be used (0.88). However, the specificity became lower (0.83). as an universal cut-off. The same results were seen in the combination LFLC index–OCB and KFLC index–LFLC index compared There are some limitations in this study. One limita- to single OCB. By definition, the sensitivity become tion was that not all patients were diagnosed based on higher and the specificity lower when you decide the same MS criteria; most patients by McDonald beforehand that the combination test will be positive 2010 (84%) but a few with McDonald 2005, which when the test is positive in one of the two. may have influenced the diagnosis of CIS patients par- ticularly. CIS patients diagnosed before 2010 may For clinical practice, the KFLC index is more accu- very well be MS patients according to McDonald rate in excluding CIS/MS compared to OCB but for 2010, because in the 2005 criteria, MS diagnosis was ruling in a diagnosis of CIS/MS, analysis of OCB more stringent. None of the patients were diagnosed appears to be more accurate. If we replace OCB by by the new 2017 criteria, because of the retrospective KFLC in diagnostic practice, there is a slightly higher setup, and thus imaging information was not collected. chance that a patient with a diagnosis different from We address this problem by pooling all CIS and MS MS will get the diagnosis of MS and maybe unneces- patients. Another reason for pooling CIS and MS is sarily exposed to potential negative side effects of that we did not have the data to test CIS converting to early treatment. Since the KFLC index is more sensi- MS versus non-converting CIS, because we did not tive at the cost of a lower specificity, we should stress have follow-up data. We performed several sensitivity that replacement of OCB by the KFLC index is not analyses in CIS or MS patients separately, and by optimal to arrive at high diagnostic certainty. reclassifying and excluding specific clinical groups However, with the higher sensitivity of KFLC, an ear- (Table 3). In these analyses, similar results were lier treatment start may be considered. Whether it is observed, suggesting that our results are robust for the an option to start treatment based on the KFLC result total population. Another limitation is that we did not and clinical/MRI findings according to the novel repeat the OCB analysis per patient centrally, but McDonald criteria or whether the treatment may be relied on the original local outcomes. However, we adapted after a first-year evaluation is subject of fur- received the samples and OCB status from expertise ther studies and discussions. journals.sagepub.com/home/msj 919 Multiple Sclerosis Journal 26(8) In conclusion, this study indicates that the KFLC and was supported by the Republic of Serbia Ministry index is a valid tool in the diagnostic process of MS. of Education, Science and Technological Development Since this marker is measured by a faster and rater- grant support (No. 175031). M.P. participated in independent analytical procedure, it should be consid- Advisory Board Meetings of Bayer Schering and ered as a potential cost-effective replacement of the Genzyme, and received travel or speaker honoraria OCB, especially when CSF analysis will regain a from Almirall, Bayer Schering, Biogen Idec, Genzyme, more prominent role in the 2017 revisions of the Merck Serono, Novartis, Sanofi-Aventis, and Teva McDonald criteria. Neurosciences. J.K.’s institution (University Hospital Basel) received and used, exclusively for research Acknowledgements support, consulting fees from Biogen, Novartis, Reagents for this study were provided by the Binding Protagen AG, Roche, Teva; speaker fees from the Site®. The authors would like to thank G. Oral and Swiss MS Society, Biogen, Novartis, Roche, Genzyme; M. Atik for helping A. Altintas collecting clinical data travel expenses from Merck Serono, Novartis, and samples. Roche; and grants from ECTRIMS Research Fellowship Program, University of Basel, Swiss MS Author Contributions Society, Swiss National Research Foundation C.E.L. contributed to study design, collecting all (320030_160221), Bayer AG, Biogen, Genzyme, data, statistical analysis, interpretation of the data, Merck, Novartis, Roche. L.M.V. received payment for and drafting the manuscript. H.A.M.T. contributed to lecturing, travel expenses, or research grants from KFLC and LFLC data analyses, interpretation of the Sanofi-Genzyme, Merck, Biogen, Novartis, Teva, data, and revising the manuscript for intellectual con- Binding Site. J.C.A.-C. received payment for lectur- tent. B.I.L.-W. contributed to statistical analysis of the ing, travel expenses, or research grants from Bayer, data (Biostatistician). B.M.J.U., J.K., C.B., and C.T. Roche, Sanofi-Genzyme, Merck, Biogen, Novartis, contributed to study design, interpretation of the data, Teva. R. A.-L. reported grants and personal fees from and revising the manuscript for intellectual content. MercK Serono, personal fees, and non-financial sup- All other authors helped by selecting and collecting port from Biogen Idec, grants, personal fees, and non- samples and clinical information. All other authors financial support from Novartis Pharmaceuticals S.A., helped by revising the manuscript for intellectual grants and personal fees from Genzyme, non-financial content. support from Teva Pharma, S.L., non-financial support from Roche. H.H. has participated in meetings spon- Declaration of Conflicting Interests sored by and received speaker honoraria or travel The author(s) declared no potential conflicts of inter- funding from Bayer Schering, Biogen, Merck Serono, est with respect to the research, authorship, and/or and Novartis, and received honoraria for acting as con- publication of this article. sultant for Teva Pharmaceuticals Europe. F.D. has par- ticipated in meetings sponsored by or received Disclosures honoraria for acting as an advisor/speaker for Biogen C.E. L. reports a grant from Stichting MS research. Idec, Genzyme-Sanofi, Merck, Novartis Pharma, H.A.M. T., B.I.L.-W., M.C.,T.B., C.R., L.M.W., D. G., Roche, and TEVA ratiopharm. E.T. has participated in E.S., and P.E.H.J., and D.L.P. report no disclosures. meetings sponsored by or received honoraria for act- V.V.P. received travel grants from Biogen, Bayer ing as an advisor/speaker for Biogen Idec, Genzyme- Schering, Genzyme, Merck, Teva, Sanofi and Roche. Sanofi, Merck, Novartis, Roche, and TEVA. M.C. has His institution receives honoraria for consultancy and received compensation for consulting services and lectures from Biogen, Bayer Schering, Sanofi, Merck, speaking honoraria from Bayer Schering Pharma, Roche, Teva and Novartis Pharma as well as research Merk Serono, Biogen Idec, Teva Pharmaceuticals, grants from Novartis Pharma, Bayer Schering, Sanofi, Sanofi-Aventis, Genzyme, and Novartis. X.M. has and Roche. I.D. received lecture fees and/or travel received speaking honoraria and travel expenses for grants from Merck Serono, Rosche, Bayer, Medis, scientific meetings, has been a steering committee Teva, and Boehringer Ingelheim; received honoraria member of clinical trials or participated in advisory for acting as an advisor for Bayer; and was supported boards of clinical trials in the past years with Bayer by the Ministry of Education, Science and Schering Pharma, Biogen Idec, EMD Merck Serono, Technological Development of the Republic of Serbia Genentech, Genzyme, Novartis, Sanofi-Aventis, (Grant No. 175031). J.D. received honoraria for con- Teva Phramaceuticals, Almirall and Roche. L.V.gave sulting services, travel expenses for scientific meet- invited lectures and was the chair of symposia ings, and speaking honoraria from Bayer Schering by Biogen, Genzyme, Teva, Roche, Merck, and Pharma, Merck Serono, Medis, Rosche, and Actavis Novartis. A.A. has received personal fees from 920 journals.sagepub.com/home/msj CE Leurs, HAM Twaalfhoven et al. received honoraria for giving educational presenta- and consulting fees from Merck Serono, Biogen Idec, tions on multiple sclerosis and neuroimmunology at Teva, Genzyme, and Novartis. C.B. reported a grant several national congresses or symposia from Teva from the Swiss MS Society. C.T. served on the advi- Turkey, Merck Serono, Biogen Idec-Gen Pharma of sory board of Fujirebio and Roche, received research Turkey, Novartis, Bayer, and Sanofi-Genzyme. She consumables from Euroimmun, IBL, Fujirebio, has received travel and registration coverage for Invitrogen and Mesoscale Discovery, performed con- attending several national and international congresses tract research for IBL, Shire, Boehringer, Roche and or symposia from Teva Turkey, Merck Serono, Biogen Probiodrug, and received grants from the European Idec-Gen Pharma of Turkey, Bayer, Sanofi-Genzyme. Commission, the Dutch Research Council (ZonMW), K.R. has received honoraria for consulting services Association of Frontotemporal Dementia/Alzheimer’s and lecturing from Bayer Schering Pharma, Merk Drug Discovery Foundation, ISAO and the Serono, Biogen Idec, Teva Pharmaceuticals, Sanofi- Alzheimer’s Drug Discovery Foundation. Aventis, Genzyme, and Novartis. J. L.F. has no fund- ing to support the presented work, but has served on Funding scientific advisory boards for and received funding for The author(s) received no financial support for the travel related to these activities as well as honoraria research, authorship, and/or publication of this article. from Biogen Idec, Merck Serono, Sanofi-Aventis, Teva, Novartis, and Almirall. She has received speaker Supplemental Material honoraria from Biogen Idec, Teva, and Novartis. She Supplemental material for this article is available has served as advisor on preclinical development for online. Takeda. G.P.-J. has no funding to support the presented work, but has received support from Biogen Idec for a ORCID iDs study in optic neuritis and has received support for M Castellazzi https://orcid.org/0000-0001-6555 travel expenses to conferences from Genzyme, -6075 Novartis, Merck Serono, Roche, Teva, and Biogen. R Alvarez-Lafuente https://orcid.org/0000-0002 M.K. has received funding for travel and speaker hon- -3132-1486 oraria from Bayer Schering Pharma, Novartis, F Deisenhammer https://orcid.org/0000-0003 Genzyme, Merck, Shire, Biogen Idec, and Teva -4541-8841 Pharmaceutical Industries. M.M.V. received funding M Khalil https://orcid.org/0000-0002-5350-3328 from the Austrian Federal Ministry of Science, MM Voortman https://orcid.org/0000-0001-7357 Research and Economics and was trained within the -2868 frame of the PhD Program Molecular Medicine of the A Saiz https://orcid.org/0000-0002-5793-8791 Medical University of Graz. F.F. served on scientific advisory boards for Biogen Idec, Genzyme, Merck, Novartis, Roche and Teva Pharmaceutical Industries References Ltd.; serves on the editorial boards of Multiple 1. Thompson AJ, Banwell BL, Barkhof F, et al. Sclerosis, the Polish Journal of Neurology and Diagnosis of multiple sclerosis: 2017 revisions of Neurosurgery, Neurology and the Swiss Archives of the McDonald criteria. Lancet Neurol 2017; 17(2): Neurology and Psychiatry; provides services for 162–173. Actelion and Parexel and has received speaker hono- raria and support from Merck, Roche and Teva 2. Polman CH, Reingold SC, Banwell B, et al. Diagnostic criteria for multiple sclerosis: 2010 Pharmaceutical Industries Ltd. A.S. has received com- revisions to the McDonald criteria. Ann Neurol 2011; pensation for consulting services and speaker hono- 69(2): 292–302. raria from Bayer Schering, Merck Serono, Biogen Idec, Sanofi-Aventis, Teva Pharmaceutical Industries 3. Arrambide G, Tintoré M, Espejo C, et al. The added Ltd, and Novartis. M.J.V. has received travel grants value of oligoclonal bands in the multiple sclerosis from the Binding Site and consumables from the diagnostic criteria. ECTRIMS 2017; 2017: 202450. Binding site, Siemens and Euroimmun. L.V. has 4. Kuhle J, Disanto G, Dobson R, et al. Conversion from received speaking honoraria and travel expenses for clinically isolated syndrome to multiple sclerosis: scientific meetings, has been a steering committee A large multicentre study. Mult Scler 2015; 21(8): member of clinical trials or participated in advisory 1013–1024. boards by Biogen, Genzyme, Novartis, Merck, Roche. 5. Tintore M, Rovira A, Rio J, et al. Do oligoclonal B.M.J.U. has received personal compensation for con- bands add information to MRI in first attacks of sulting from Biogen Idec, Genzyme, Merck Serono, multiple sclerosis. Neurology 2008; 70(13 Pt. 2): Novartis, Roche, and TEVA. J.K. has accepted speaker 1079–1083. journals.sagepub.com/home/msj 921 Multiple Sclerosis Journal 26(8) 6. Kaplan B, Golderman S, Yahalom G, et al. Free light immunoglobulin free light chains in serum and urine. chain monomer-dimer patterns in the diagnosis of Clin Chem 2001; 47(4): 673–680. multiple sclerosis. J Immunol Methods 2013; 390(1- 20. Presslauer S, Milosavljevic D, Huebl W, et al. Kappa 2): 74–80. free light chains: Diagnostic and prognostic relevance 7. Kaplan B, Aizenbud BM, Golderman S, et al. Free in MS and CIS. PLoS ONE 2014; 9(2): e89945. light chain monomers in the diagnosis of multiple 21. Voortman MM, Stojakovic T, Pirpamer L, et al. sclerosis. J Neuroimmunol 2010; 229(1-2): 263–271. Prognostic value of free light chains lambda and 8. Senel M, Tumani H, Lauda F, et al. Cerebrospinal kappa in early multiple sclerosis. Mult Scler 2016; 23: fluid immunoglobulin kappa light chain in clinically 1496–1505. isolated syndrome and multiple sclerosis. PLoS ONE 22. Passerini G, Dalla Costa G, Sangalli F, et al. Free 2014; 9(4): e88680. light chains and intrathecal B cells activity in multiple 9. Desplat-Jego S, Feuillet L, Pelletier J, et al. sclerosis: A prospective study and meta-analysis. Quantification of immunoglobulin free light chains in Mult Scler Int 2016; 2016: 2303857. cerebrospinal fluid by nephelometry. J Clin Immunol 23. Pieri M, Storto M, Pignalosa S, et al. KFLC Index 2005; 25(4): 338–345. utility in multiple sclerosis diagnosis: Further 10. Presslauer S, Milosavljevic D, Brücke T, et al. confirmation. J Neuroimmunol 2017; 309: 31–33. Elevated levels of kappa free light chains in CSF 24. Teunissen C, Menge T, Altintas A, et al. Consensus support the diagnosis of multiple sclerosis. J Neurol definitions and application guidelines for control 2008; 255(10): 1508–1514. groups in cerebrospinal fluid biomarker studies in 11. Rudick RA, Peter DR, Bidlack JM, et al. Multiple multiple sclerosis. Mult Scler 2013; 19(13): 1802– sclerosis: Free light chains in cerebrospinal fluid. 1809. Neurology 1985; 35: 1443–1449. 25. Polman CH, Reingold SC, Edan G, et al. Diagnostic 12. Bracco F, Gallo P, Menna R, et al. Free light chains criteria for multiple sclerosis: 2005 revisions to in the CSF in multiple sclerosis. Journal of Neurology the “McDonald Criteria.” Ann Neurol 2005; 58(6): 1987; 234: 303–307. 840–846. 13. DeCarli C, Menegus MA and Rudick RA. Free light 26. Susse M, Hannich M, Petersmann A, et al. Kappa free chains in multiple sclerosis and infections of the CNS. light chains in cerebrospinal fluid to identify patients Neurology 1987; 37(8): 1334–1338. with oligoclonal bands. Eur J Neurol 2018; 25(9): 1134–1139. 14. Presslauer S, Milosavljevic D, Huebl W, et al. Validation of kappa free light chains as a diagnostic 27. Vasilj M, Kes VB, Vrkic N, et al. Relevance of KFLC biomarker in multiple sclerosis and clinically isolated quantification to differentiate clinically isolated syndrome: A multicenter study. Mult Scler 2016; 22: syndrome from multiple sclerosis at clinical onset. 102–110. Clin Neurol Neurosurg 2018; 174: 220–229. 15. Solling K. Free light chains of immunoglobulins 28. Van der Heijden M, Kraneveld A and Redegeld F. in normal serum and urine determined by Free immunoglobulin light chains as target in the radioimmunoassay. Scand J Clin Lab Invest 1975; treatment of chronic inflammatory diseases. Eur J 35(5): 407–412. Pharmacol 2006; 533(1-3): 319–326. 16. Rudick RA, Pallant A, Bidlack JM, et al. Free kappa 29. Villar LM, Masjuan J, Sadaba MC, et al. Early light chains in multiple sclerosis spinal fluid. Ann differential diagnosis of multiple sclerosis using a Neurol 1986; 20(1): 63–69. new oligoclonal band test. Arch Neurol 2005; 62(4): 574–577. 17. Makshakov G, Nazarov V, Kochetova O, et al. Diagnostic and prognostic value of the cerebrospinal 30. Petzold A. Markers for different glial cell responses fluid concentration of immunoglobulin free light in multiple sclerosis: Clinical and pathological chains in clinically isolated syndrome with conversion correlations. Brain 2002; 125(Pt. 7): 1462–1473. to multiple sclerosis. PLoS ONE 2015; 10(11): 31. Petzold A. Intrathecal oligoclonal IgG synthesis in e0143375. multiple sclerosis. J Neuroimmunol 2013; 262: 1–10. 18. Duranti F, Pieri M, Centonze D, et al. Determination 32. Valencia-Vera E, Martinez-Escribano Garcia- of κFLC and κ Index in cerebrospinal fluid: A valid Ripoll A, Enguix A, et al. Application of kappa free alternative to assess intrathecal immunoglobulin light chains in cerebrospinal fluid as a biomarker synthesis. J Neuroimmunol 2013; 263: 116–120. in multiple sclerosis diagnosis: Development of a 19. Bradwell AR, Carr-Smith HD, Mead GP, et al. diagnosis algorithm. Clin Chem Lab Med 2018; Highly sensitive, automated immunoassay for 56(4): 609–613. 922 journals.sagepub.com/home/msj CE Leurs, HAM Twaalfhoven et al. 33. Gnanapavan S, Hegen H, Khalil M, et al. Guidelines 34. Abraira V, Alvarez-Cermeno JC, Arroyo R, et al. Visit SAGE journals online for uniform reporting of body fluid biomarker studies Utility of oligoclonal IgG band detection for MS journals.sagepub.com/ home/msj in neurologic disorders. Neurology 2014; 83(13): diagnosis in daily clinical practice. J Immunol 1210–1216. Methods 2011; 371(1–2): 170–173. SAGE journals journals.sagepub.com/home/msj 923