ARTICLE OPEN ACCESS

Profiling individual clinical responses by high-frequency serum neurofilament assessment inMSKatja Akgun MD Nicole Kretschmann Rocco Haase Undine Proschmann MD Hagen H Kitzler MD

Heinz Reichmann MD PhD and Tjalf Ziemssen MD PhD

Neurol Neuroimmunol Neuroinflamm 20196e555 doi101212NXI0000000000000555

Correspondence

Dr Ziemssen

TjalfZiemssen

uniklinikum-dresdende

AbstractObjectiveTo evaluate individual neurofilament light chain (NfL) variation over the time of disease courseand the potential of NfL measurement to predict treatment response in patients with MS

MethodsWe investigated 15 patients with MS after immune reconstitution treatment with alemtuzumab(ATZ) Monthly serum NfL (sNFL) measurements were correlated with Expanded DisabilityStatus Scale (EDSS) MRI and relapse activity over an observational period of up to 102months

ResultsBefore ATZ sNfL was significantly increased in correlation with previous relapseMRI activityAfter ATZ sNfL decreased quickly within the first 6 months In patients classified as NEDA-3sNfL declined and persisted at an individual low steady-state level of lt8 pgmL During follow-up 34 sNfL peaks with a gt20 fold increase could be detected which were associated withclinical orMRI disease activity Even patient-reported relapse-suspicious symptoms which havenot been confirmed because relapses were accompanied by sNfL increase proposing sNfLassessment as a marker for relapse activity sNfL started to increase earliest 5 months beforepeaked at clinical onset and recovered within 4ndash5 months sNfL presented at higher levels inactive patients requiring ATZ retreatment compared with responder patients During 2documented pregnancies sNfL was at a low level whereas a postpartum transient sNfL increasewas seen without any signs of activity

ConclusionsThis study applied a long-term high-frequency sNfL assessment in an ATZ-treated cohortallowing a holistic profiling on the individual level and highlighted that sNfL can eminentlycomplement the individual clinical and MRI monitoring in clinical practice

From the Center of Clinical Neuroscience (KA NK RH UP TZ) University Hospital Dresden Germany Department of Neuroradiology (HHK) University Hospital DresdenGermany and Department of Neurology (HR) University Hospital Dresden Germany

Funding information and disclosures are provided at the end of the article Full disclosure form information provided by the authors is available with the full text of this article atNeurologyorgNN

The Article Processing Charge was funded by the authors

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 40 (CC BY-NC-ND) which permits downloadingand sharing the work provided it is properly cited The work cannot be changed in any way or used commercially without permission from the journal

Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology 1

Alemtuzumab (ATZ Lemtrada Sanofi) is a humanized mono-clonal antibody and represents one of the most effective treat-ments for active relapsing-remitting MS12 ATZ might producea durable therapeutic response as a consequence of a permanentrebalancing of the immune system3 It has introduced the im-mune reconstitution therapy as a new therapeutic inductiontherapy-like concept in MS4 The definition of clinical responseprofiles after ATZ dosing is crucial to the identification of patientswho need retreatment or switching to other therapy Until nowprimarily clinical characteristics such as relapse activity disabilityand repeated MRI analysis have been used to define responseprofiles to MS treatment56 However this approach has severallimitations because even high-frequency clinical and MRI as-sessment is not sensitive enough to detect all neuroinflammatoryand neurodegenerative activity and additional biomarkers areneeded to complement clinical- and MRI-derived information7

In particular reliable quantification and longitudinal follow-up evaluation of neuroaxonal damage would be important forassessing MS disease activity monitoring treatment responsefacilitating treatment development and determining prog-nosis The neurofilament proteins that represent one of themain cytoskeletal constitutes in neurons show promise in this

context because their levels rise upon neuroaxonal damagenot only in the CSF but also in blood indicating neuroaxonalinjury independent of causal pathways8ndash10 Different studiespresented that neurofilament light chain (NfL) is increased inthe CSF of patients with clinically isolated syndrome and MSand correlates with MRI activity disability and relapseactivity11ndash16 Third-generation (electrochemiluminescence)and particularly fourth-generation (single-molecule array[SIMOA]) assays enable the reliable measurement of neu-rofilaments throughout the range of concentrations found inblood samples Previous studies suggest serum (s)NfL asa potential predictive marker for MS disease outcome1317ndash19

Nevertheless data and studies are missing that clarify howsNfL measurements develop over time at relapse or re-mission and whether sNfL level may be able to predicttreatment response

In this study we analyzed individual long-term series of sNfLlevels in patients with highly active relapsing-remitting MSafter immune reconstitution treatment with ATZ using high-frequency sNfL assessment over a long observational periodof up to 102 months By this approach we aim to observesNfL changes in the MS disease course and stratify the

GlossaryATZ = alemtuzumab cMRI = cerebral MRI EDSS = Expanded Disability Status Scale Gd+ = gadolinium enhancing IRT =immune reconstitution therapy NEDA = no evidence of disease activity NfL = neurofilament light chain SIMOA = singlemolecule array technology sNfL = serum neurofilament light chain SS = steady state

2 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 NeurologyorgNN

potential to predict disease activity vs remission and treatmentresponse in patients with MS in clinical practice

MethodsPatients collection of clinical data andblood samplingIn this pilot study we included 15 patients with relapsing-remitting MS with clinical- and MRI-confirmed highly activedisease course (figures 1ndash5 table) All patients were treatedbased on a standardized infusion scheme as described andused in CARE-MS 1 and 2 clinical trials20ndash22 During the firstinfusion course 12 mg ATZ was infused for 5 consecutivedays During the second course 12 months later ATZ wasadministered for 3 consecutive days Serum samples wereobtained before the first ATZ infusion (baseline month 0)and then monthly during follow-up for up to 102 monthsClinical parameters including relapse activity and ExpandedDisability Status Scale (EDSS) were recorded at monthlyrespectively three monthly visits A cerebral (c)MRI wasperformed every 12 months or more frequently if needed

ATZ retreatment with an additional course of 12 mg ATZgiven for 3 consecutive days was considered when clinicaldisease activity including relapses appeared orMRI progression

was confirmed by new gadolinium-enhancing (Gd+) lesionsandor 2 new T2 lesions Definition of no evidence of diseaseactivity (NEDA) is a composition of main outcome measure intreated patients with relapsing-remitting MS23 Criteria forNEDA-3 status were defined as (1) absence of relapses (2)absence of disability worsening defined by confirmed EDSSprogression (ge10 point increase if the EDSS baseline score waslt55 ge05 point increase if the EDSS baseline score was ge55)and (3) absence ofMRI progression (no evidence for newGd+lesions andor new T2 lesions)

Study approvalThe experiments and the study were approved by the in-stitutional review board of the University Hospital ofDresden All patients gave their written informed consent

sNfL measurement using single-moleculeanalysisSerum samples were collected monthly and directly storedat minus20degC until after preparation because NfL protein isstable during freezing process2425 sNfL measurement wasperformed using the Advantage NF-Light Singleplex Kit andprepared as defined in the manufacturerrsquos instructions(Quartered Lexington MA Datasheet Quanterix SIMOANF-Light Advantage Kit) The SIMOA Human AdvantageNfL assay is a digital immunoassay for the quantitative

Table Clinical parameters and sNfL levels before starting ATZ

Patient Sex

Age atstartingATZ

EDSS atstartingATZ

LastrelapsebeforestartingATZ [mo]

New T2lesions inMRI atstartingATZ [no]

New Gd +lesions inMRI atstarting ATZ[no]

No oflesions inMRI atstartingATZ

sNfLbeforestartingATZ [pgmL]

IndividualsNfL atsteadystate [pgmL plusmn SD]

No ofmowhensNfL steadysate wasreached afterATZ

1 f 39 3 1 0 2 gt20 357 538 plusmn 117 7

2 f 24 35 4 4 3 gt30 195 461 plusmn 114 2

3 f 25 3 5 0 0 gt10 442 37 plusmn 08 mdash

4 f 44 25 10 4 5 gt20 689 556 plusmn 156 mdash

5 m 32 25 10 2 0 gt20 124 527 plusmn 071 5

6 m 31 35 4 7 5 gt20 291 522 plusmn 117 10

7 f 21 15 5 2 0 gt30 327 236 plusmn 048 mdash

8 f 28 15 10 2 0 gt15 413 396 plusmn 152 mdash

9 m 25 15 6 2 2 gt15 481 383 plusmn 113 mdash

10 f 31 4 2 17 15 gt40 235 563 plusmn 17 6

11 f 45 3 4 1 1 gt10 931 556 plusmn 104 3

12 f 27 15 15 0 0 gt10 137 471 plusmn 259 mdash

13 f 21 2 14 0 0 gt10 632 397 plusmn 085 mdash

14 f 29 3 3 1 0 gt15 522 469 plusmn 082 mdash

15 m 36 1 6 0 0 gt20 244 435 plusmn 114 mdash

Abbreviations ATZ = alemtuzumab Gd+ = gadolinium enhancing no = number sNfL = serum neurofilament light chainsNfL before ATZ in bold markmdashthe significant increased sNfL compared to steady state

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 3

determination of NfL and was ran on a Simoa HD-1 in-strument (Quanterix)26 The antibodies and calibratorsused in the assays have been developed by Uman Diag-nostics (Sweden) and described previously in their plate-based enzyme-linked immunosorbent assay27 Previousreports proved that the newly developed SIMOA techniqueis the most sensitive to detect NfL even in lowest

concentrations28 Calibrators (8 calibrator points) and di-luted serum samples (dilution 14) were measured induplicates Sample dilution was calculated using the in-strument The lower threshold of quantification was 0775pgmL Both the mean intraassay coefficient of variation ofduplicates and the mean interassay coefficient of variationwere lt10

Figure 1 Clinical data and variation of sNfL levels of patients 1ndash3 with ATZ-treated MS

Clinical data and sNfL variation since starting ATZ and during follow-up are presented Clinical data including relapse activity MRI activity and pretreatmentsare shown for the period before starting ATZ Since ATZ initiation a standardized monitoring scheme was performed by evaluation of relapse activity everymonths 3monthly EDSS and cerebralMRI every 12months Patient 1 to 3 (A toC) did receive only the first and the secondATZ infusion cycles (Patients 1ndash3 A-C) After the first and the second standard ATZ infusion courses patients presented NEDA-3 status and low level of sNfL in the follow-up ATZ = alemtuzumabcMRI = cerebral magnetic resonance imaging Gd+ = gadolinium enhancing sNfL = serum neurofilament light chain

4 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 NeurologyorgNN

Statistical analysisAn individual steady state (SS) of sNfL levels during stabledisease was defined in each patient as mean (SS) plusmn SD of allsNfL measurements in the selected period of NEDA (noconfirmed relapses no confirmed EDSS progression de-fined by increase ge10 for minimum of 6 months and no

MRI progression) A clinically significant increase in sNfLwas defined as gt ([mean [SS] + [3 times SD]) in each patientLongitudinal data were analyzed by generalized linearmixed models with gamma distribution and log linkfunction because of the right-skewed distribution patternfor the evaluation of sNfL variation over the time CIs and

Figure 2 Clinical data and variation of sNfL levels of patients 4ndash6 with ATZ-treated MS

Clinical data and sNfL variation since starting ATZ and during follow-up are presented Clinical data including relapse activity MRI activity and pretreatmentsare shown for the period before starting ATZ Since ATZ initiation a standardized monitoring scheme was performed by evaluation of relapse activity everymonths 3monthly EDSS and cerebral MRI every 12months Patient 4 to 6 did receive only first and second ATZ infusion cycle (patients 4 and 5) After the firstand the second standard ATZ infusion courses patients presented NEDA-3 status and low level of sNfL in the follow-up In patient 4 and 5 sNfL significantlyincreased at selected time intervals At these events the patients presented acute neurologic symptoms that were not classified as acute relapses by thetreating neurologist No corticosteroid treatment was performed and symptoms resolved within weeks spontaneously (patient 6) After ATZ no clinicalrelapses appeared but ongoing disease activity was confirmed by MRI progression which was paralleled by sNfL increase ATZ = alemtuzumab cMRI =cerebral magnetic resonance imaging Gd+ = gadolinium enhancing NEDA = no evidence of disease activity sNfL = serum neurofilament light chain

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 5

p values were originated from that model For the analysesof pre-relapse phases doses of ATZ were included asa random factor Bonferroni correction for pairwise testswas used Values of p lt 005 were considered significantThe length of the comparable time segment (CTS) forcomparisons between subjects with and without relapseswas set to plusmn12 months for respective analyses The start of the

CTS will be the respective number of months prior to the firstrelapse of a subject or for relapse-free subjects under stableconditions after the second ATZ cycle (defined as 12 monthsafter the second ATZ cycle or 24 months after the initialtreatment) CTS will be individually censored in the case ofother events of disease activity in the predefined segmentFigures present means and 95 CIs Correlation was

Figure 3 Clinical data and variation of sNfL levels of patients 7ndash9 with ATZ-treated MS

Clinical data and sNfL variation since starting ATZ and during follow-up are presented Clinical data including relapse activity MRI activity and pretreatmentsare shown for the period before starting ATZ Since ATZ initiation a standardized monitoring scheme was performed by evaluation of relapse activity everymonths 3monthly EDSS and cerebralMRI every 12months (patient 7) Patient 7 did receive only the first and the secondATZ infusion cycles After 14monthsthe patientmoved andwas lost to follow-up Based on return of clinical andMRI disease activity ATZ retreatment was performed in patients 8 to 15 (patients8ndash9) After ATZ initiation patients stabilized rapidly in clinical symptoms and NEDA-3 status was reached Years after first ATZ infusions upcoming diseaseactivity was presented confirmed by new Gd+ lesions that was paralleled by the increase in sNfL ATZ = alemtuzumab cMRI = cerebral magnetic resonanceimaging Gd+ = gadolinium enhancing NEDA = no evidence of disease activity sNfL = serum neurofilament light chain

6 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 NeurologyorgNN

calculated using Spearman correlation test Sensitivity andspecificity of significant sNfL peaks were calculated for theperiod after the first and the second ATZ treatment Starting atmonth 24 sNfL levels were evaluated at 3 monthly intervalsand rated as normalnegative peak for levels (le [mean [SS] +[3 times SD]) and increasedpositive peak for levels (gt [mean

[SS] + [3 times SD]) In total 210 of the 3 monthly intervals wereevaluated while the presence of relapse activity confirmedEDSS progression defined by increasege10 for aminimumof 6months and MRI progression was assessed as well to definethe presence of significant disease activity vs stable diseasecourse Statistical analyses were performed using the IBM

Figure 4 Clinical data and variation of sNfL levels of patients 10ndash12 with ATZ-treated MS

Clinical data and sNfL variation since starting ATZ and during follow-up are presented Clinical data including relapse activity MRI activity and pretreatmentsare shown for the period before starting ATZ Since ATZ initiation a standardized monitoring scheme was performed by evaluation of relapse activity everymonths 3monthly EDSS and cerebralMRI every 12months Based on return of clinical andMRI disease activity ATZ retreatmentwas performed in patients 8to 15 (patient 10) After ATZ initiation patients stabilized rapidly in clinical symptoms andNEDA-3 statuswas reached Years after first ATZ infusions upcomingdisease activity was presented confirmed by new Gd+ lesions that was paralleled by the increase in sNfL (Patients 11ndash12) Early after starting ATZ patientspresented upcoming or ongoing disease activity paralleled by sNfL increase ATZ = alemtuzumab cMRI = cerebral magnetic resonance imaging Gd+ =gadolinium enhancing NEDA = no evidence of disease activity sNfL = serum neurofilament light chain

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 7

SPSS software for Windows (version 250 IBM CorporationArmonk NY)

Data availabilityK Akgun and T Ziemssen have full access to all dataset andtake full responsibility for the integrity of the data and accu-racy of the data analysis Data will be shared on request fromany qualified investigator

ResultssNfL baseline levels before ATZ initiationIndividual disease course and variation in sNfL are depicted(figures 1ndash5) Immediately before starting ATZ 6 of 15patients presented with significantly increased sNfL levels(2159 plusmn 1048 pgmL) (mean plusmn 95 CI) (table figure e-1AlinkslwwcomNXIA108) This was in line with MRI scans

Figure 5 Clinical data and variation of sNfL levels of patients 13ndash15 with ATZ-treated MS

Clinical data and sNfL variation since starting ATZ and during follow-up are presented Clinical data including relapse activity MRI activity and pretreatments areshown for the periodbefore starting ATZ Since ATZ initiation a standardizedmonitoring schemewas performedbyevaluation of relapse activity everymonths 3monthly EDSS and cerebral MRI every 12 months Based on return of clinical and MRI disease activity ATZ retreatment was performed in patients 8 to 15(patients 13ndash15) Early after starting ATZ patients presented upcoming or ongoing disease activity thatwas paralleled by the increase in sNfL ATZ = alemtuzumabcMRI = cerebral magnetic resonance imaging Gd+ = gadolinium enhancing NEDA = no evidence of disease activity sNfL = serum neurofilament light chain

8 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 NeurologyorgNN

one week before demonstrating progression of the number ofT2 lesions in all and new Gd+ lesions in 5 of these patientscompared to the previous MRI 12 months before The lastclinically confirmed relapse before starting ATZ occurred onaverage 42 months ago (range 1ndash10 months) In contrast 9patients presented with sNfL levels unchanged to their SSlevels (table) Here 4 patients presented an unchanged MRIlesion load 3 patients presented new T2 lesions and 2patients showed new Gd+ lesions in MRI compared to MRI12 months before In these patients last relapse activity wasdocumented a much longer time period ago about 82months (range 3ndash15 months) There was a strong negativecorrelation of sNfL increase and time interval of last relapseactivity (r = minus06018 p lt 005)

After starting ATZ initially elevated sNfL levels constantlydecreased and reached individual SS on average within 53months (range 2ndash10 months) (table figure e-1A linkslwwcomNXIA108) At the second ATZ infusion course 2patients presented a significant increase in sNfL levels thatdeclined after 2 respectively 6 months (figures 2 and 5 patient4 and 15) Interestingly after ATZ initiation 5 of the 15patients presented a transient increase in sNfL at month 2ndash3after the first ATZ infusion and 3 of the 15 patients after thesecond ATZ infusion course that declined in the followingmonths (figures 1ndash3 patient 3 4 5 7 and 10)

Only a limited number of blood samples were available for 4patients receiving interferon-beta therapy in the treatmentperiod before ATZ initiation (figure 5 patient 12ndash15) All 4patients presented ongoing disease activity reflected by clin-ical relapses and MRI progression Increase and high intra-individual variation of sNfL levels were also seen in the clinicalactive period before starting ATZ with the highest sNfL levelof up to 842 pgmL (figure 5 patient 12)

Variation of sNfL in patients with NEDA-3status after ATZ treatmentAfter receiving the first and the second ATZ standard infusioncourses all patients responded to treatment documented bydecreased clinical and MRI activity Seven of 15 patients afterthe first ATZ infusion course presented and one more patientafter the second ATZ infusion course presented confirmedEDSS improvement ge05 for at least 6 months In our cohort4 patients met the criteria for the NEDA-3 status even duringentire 102 months of long-term follow-up (figures 1 and 2patient 1ndash4) Others reached NEDA-3 status after ATZtreatment as well but presented reappearance of disease ac-tivity during the follow-up of 5 and 7 years that necessitatedATZ retreatment (figure 1 patient 8ndash10) When NEDA-3was confirmed sNfL levels reached an individual low SS levelof lt 8 pgmL (table figures 1ndash5)

sNFLmdashreal-time marker for upcoming diseaseactivity in patients with ATZ-treated MSAlthough some patients met the NEDA-3 status for a longterm after starting ATZ others presented with a decreased but

still ongoing or returning disease activity months or years afterthe first ATZ infusions After the initial ATZ use and decreaseof sNfL levels all patients reached individual SS levels of sNfLduring the stable disease period Evaluating sNfL levels duringfollow-up repeated significant sNfL peaks could be detectedat different time points in some patients After the first and thesecond ATZ treatment courses a total of 34 sNfL peaks couldbe identified in our patients (figure e-2A linkslwwcomNXIA108) Twenty-seven sNfL peaks were associated withconfirmed clinical or MRI disease activity (figure e-2A linkslwwcomNXIA108) In contrast 146 of 152 of quarterlyintervals without any sNfL peak were associated with stabledisease (figure e-2B linkslwwcomNXIA108) Regardingrelapses individual sNfL levels started to increase 5 monthsbefore the first symptoms of relapse appeared (estimatedmeans p lt 005 at month 5 4 and 2 before onset of relapse)and were significantly elevated at the onset of clinical relapseactivity (p lt 0001) (figure e-3A linkslwwcomNXIA108)At acute relapse activity sNfL levels peaked to up to 819 pgmL After clinical onset of relapse sNfL further increased andrecovered in the following months 95 CI of sNfL afterrelapse approximated to 95 CI of NEDA-3 status patientsearliest 3-month follow-up (figure e-3A linkslwwcomNXIA108)

Six patients came to an unscheduled consultation reportingacute worsening and symptoms that were not confirmed asa relapse by our neurologists presenting with an sNfL peak asdemonstrated by later analysis (figure e-2A linkslwwcomNXIA108) These symptoms covered acute worsening ofdysesthesia paresthesia fatigue or dizziness and resolvedspontaneously within weeks Comparable to the confirmedrelapses sNfL started increasing 5 months before peaked atclinical onset and within 3 months of follow-up (p lt 0013)and subsequently decreased after 4 months (figure e-3B linkslwwcomNXIA108)

Linking sNfL withMRI activity an increase in sNfL levels wasmostly associated with a progression of the T2 lesion load andnew Gd+ lesions as observed in 24 of 34 cases (figure e-2AlinkslwwcomNXIA108) Yearly MRI scans demonstratedthat an increase in sNfL was usually detectable already monthsbefore MRI could confirm new subclinical activity of the past12 months In particular in patients with new T2 lesions butwithout Gd+ lesions the peak and increase in sNfL was seenon average 6ndash1 months before MRI scan was taken (figuree-3C linkslwwcomNXIA108) In contrast the pre-sentation of Gd+ lesions was paralleled by an sNfL increasearound the time point of MRI scans reflecting the associationof acute disease activity between sNfL peak and Gd+ lesion intime that abated within the following 4ndash6 months (figuree-3D linkslwwcomNXIA108)

The sNfL pattern associated with disease activity was com-parable with clinical and MRI activity In contrast to ourNEDA-3 patients 95 CI of sNfL levels of active patientsincreased 5 months before the clinical event peaked within 3

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 9

months and decreased at about 5 months of follow-up (figuree-3E linkslwwcomNXIA108) Looking at our data quar-terly assessment of sNfL levels should be the minimum fre-quency of sNfL analysis which seems to be necessary todetect all real-time variations in patients with MS during long-term observation

For 7 sNfL peaks no signs of disease activity measured bycMRI and relapse activity or relapse-associated EDSS increasecould be documented (figure e-2A linkslwwcomNXIA108) In our cohort 4 pregnancies were documented afterATZ initiation During pregnancies sNfL was at low and stablelevels (figures 1 3 and 5 patient 3 8 and 13) After birtha transient sNfL increase was seen in 2 patients without anysigns of clinical or MRI activity (figures 1 and 5 patient 3 and13) and data for sNfL levels after birth are missing after 2pregnancies (figures 3 and 5 patient 8 and 13) Fivemore sNfLpeaks appeared at the period of stable disease that principallymet the NEDA-3 status However conclusive comparable dataon spinal MRI follow-up are missing in our study

In addition to sNfL peaks we searched the sNfL follow-upmeasurements for time intervals of 3 months with normalsNfL levels (le [mean [SS] + [3 times SD]) after ATZ initiation Intotal 210 sNfL quarterly intervals were analyzed For 146intervals stable disease without evidence of clinical or MRIprogression was documented as well as sNfL was at thenormal level (figure e-2B linkslwwcomNXIA108) Foronly 6 intervals normal sNfL was paralleled by a classifiedrelapse (1) disability progression (2) or MRI progression(3) For the remaining 58 quarterly intervals an increase insNfL was associated with clinical orand MRI activity

Evaluation of quarterly time intervals with and without sNfLpeak after ATZ initiation was used to calculate sensitivity andspecificity An sNfL peak could indicate acute clinical andorMRI disease activity in 8947 of the cases (sensitivity) LowsNfL levels were associated without clinical or MRI diseaseactivity in 9542 (specificity)

sNfL after ATZ retreatmentBased on the reappearance of disease activity ATZ retreat-ment was initiated in 8 patients (figures 3ndash5 patient 8ndash15) Atleast 10 ATZ retreatments (7 patients received 1 ATZ retreatand 1 patient received 3 ATZ retreats) had to be performed inour cohort During the period before ATZ retreatment whichwas characterized by upcoming disease activity as demon-strated by relapse or MRI progression sNfL levels showeda prolonged increase (figure e-1B linkslwwcomNXIA108) In 6 patients ATZ retreat led to a rapid decrease inclinical disease activity and stable disease since retreat in long-term follow-up sNfL decreased and reached a low SS level onaverage of 42 months (figure e-1B linkslwwcomNXIA108) Although clinical and MRI activity decreased afterATZ retreatment in 2 patients it did not led to a completelystable disease which was reflected again by increased sNfLlevels (figure 5 patient 12 15)

DiscussionThe evaluation of sNfL is a promising opportunity to evaluateMS disease activity While measurement of NfL in CSF hasalready demonstrated a significant correlation with in-flammatory disease activity12ndash14 fourth-generation immuno-assays allow the implementation of sNfL measurement frompatients` blood as ldquoeasy-to-measurerdquo marker to reflect acuteneuroaxonal damage Of particular interest is the proof thatsNfL levels can be used to separate not only patients with MSfrom healthy controls but also to separate patients with MSwith and without persistent clinical and subclinical MRI dis-ease activity192930 In previous studies increased sNfL levelsin patients with MS have been independently associated withdisability and relapse status and suggested to be a predictivemarker for future relapses and disability worsening1719 Fur-ther data have proved that patients on disease-modifyingtreatment had lower sNfL levels than untreated patients29 Inthis line another study found that patients who switched frominjectable therapies to fingolimod had significantly lower sNfLlevels than staying on injectables31 Associations betweendisease activity and treatment-related reductions in sNfLlevels were confirmed by another study in a large independentcohort of patients with relapsing-remitting MS11 A longitu-dinal observational study demonstrated that patients withincreased sNfL levels at baseline independent of other clinicaland MRI variables experienced significantly more brain andspinal cord volume loss over 2 and 5 years of follow-up17

Nevertheless detailed data on high-frequency measurementsof sNfL levels in patients with MS are still missing This couldassist their implementation into clinical practice by profilingindividual sNfL dynamics during active vs stable disease Thisis especially interesting for patients treated with immune re-constitution therapies as ATZ Different response profiles toATZ have been described so far32 which only were defined byclinical characteristics as relapse activity and disability as wellas repeated MRI analysis56

In this pilot study we analyzed the individual variation ofsNfL levels using monthly measurements in patients withrelapsing-remitting MS over a long period Our 15 patientswere treated with ATZ which mediates its long-term efficacyby depleting T and B lymphocyte subsets and immune-reconstitution during follow-up The definition of treatmentresponse and the resulting potential need for retreatment areimportant aspects in ATZ therapy However careful evalua-tion of clinical and MRI response profiles in ATZ-treatedpatients combined with high-frequency measurement of sNfLlevels is a straightforward approach to learn more about sNfLin the follow-up of patients with MS

In parallel with the highly active disease course which was thereason to treat with ATZ increased sNfL levels were docu-mented before starting ATZ which declined after ATZ in-fusion within the following months Even high-frequencysNfL assessment could prove the positive long-term treat-ment effect of only 2 ATZ courses mostly During stable

10 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 NeurologyorgNN

disease a low individual SS level of sNfL without significantvariations was found In the period after ATZ treatment ourdata indicate that sNfL increase is clearly associated with thereturn of clinical and MRI disease activity In particular re-lapse activity and development of new Gd+ lesions wereparalleled by an increase in sNfL The peak in sNfL levels wasdirectly linked with the onset of relapse and acute MRI ac-tivity High-frequency evaluation of sNfL could demonstratethat the increase in sNfL already starts several months beforerelapse activity is clinically visible Nevertheless the samplesize in our pilot study group is too small to draw final con-clusions on the exact time period when sNfL peak is predictivefor disease activity Interestingly patient-reported symptomsthat have not been classified as clinical relapse by the treatingneurologists have been sometimes associated with an sNfLincrease proposing acute sNfL assessment as a parameter toidentify real relapse activity in patients with MS Although thenumber of patients is limited in this pilot evaluation the es-timation on sensitivity and specificity proposed high quality inthe classification of active vs stable disease using sNfL meas-ures Previous studies have evaluated the relationship of sNfLlevels with clinical and MRI data at the group level111929

Only a few reports are available that assess sNfL levels duringfollow-up but using low-frequency assessment periods ofa year1131 Our high-frequency sNfL assessment is capable todocument the dynamics of sNfL peaks as increases for a shorttime period of 3ndash6 months which are associated with diseaseactivity Our data demonstrate that at the minimum quarterlysNfL measurements would be sufficient to detect all real-timevariations during long-term observation A monthly assess-ment of sNfL does not seem to be necessary as performed inour high-frequency study

During our observation few sNfL peaks could be seen inpatients without evidence of MS disease activity as measuredby relapses disability progression or MRI activity UsuallysNfL levels were at their lowest level at the period when theNEDA-3 status was confirmed However only cerebral MRIwas performed in our study whereas conclusive longitudinaldata on spinal MRI were not available Other studies alreadyconfirmed the correlation between increased sNfL levels andGd+ spinal lesions as well as brain and spinal cordatrophy111718 which were not evaluated in our study aswell In our cohort there were no other reported events astrauma episodes stroke anesthesia during surgery or met-abolic diseases that were supposed to be associated withsNfL increase33ndash36 However we are not able to completelyexclude such factors In MS sNfL peaks can indicate sub-clinical disease activity even when clinical parameters andcerebral MRI seem to be stable Further studies are necessaryto investigate the relevance of additional MRI parameters toconfirm active vs stable disease

Several studies highlighted the relevance of sNfL levels to re-flect ongoing neuroinflammation as well as neurodegenerationin patients with MS837 In our cohort only patients with highlyactive relapsing-remitting MS disease course were included

Instability and increase in sNfL level were clearly associatedwith relapses and occurrence of new T2 and Gd+ lesions incerebral MRI that are typically linked to the inflammatorycharacteristics of MS pathology Our data represent sNfL asa real-time marker of current or upcoming inflammatory dis-ease activity as well as treatment response on an individual levelrather than only a predictive parameter for disease outcome inpatients with relapsing-remitting MS at the group level Col-lection of real-world data and observational studies providinglongitudinal information on drug profile and different out-comes in real life are essential to derive more and importantimplications for sNfL use in clinical practice38ndash40

Here we present long-term high-frequency sNfL assessmentsin an ATZ-treated cohort for up to 85 years of follow-upallowing a holistic clinical MRI and sNfL responder evalua-tion on the individual patient level sNfL can complement theindividual clinical and MRI monitoring after immune re-constitution treatment with ATZ More long-term data re-garding individual sNfL patient profiles have to be generatedfor different treatment options before sNfL could becomea key parameter for decision-making in clinical practice

Author contributionsStudy concept and design K Akgun and T Ziemssen Ac-quisition of data K Akgun N Kretschmann U ProschmannHH Kitzler Analysis and interpretation of data K Akgun RHaase Drafting of the manuscript K Akgun and T ZiemssenCritical revision of the manuscript for important intellectualcontent N Kretschmann U Proschmann H Kitzler HReichmann Statistical analysis R Haase K Akgun Studysupervision T Ziemssen

AcknowledgmentThe authors thank M Marggraf and N Dartsch for technicalassistance

Study fundingNot applicable

DisclosureK Akgun received honoraria from Sanofi for consulting andspeaking service T Ziemssen received personal compen-sation from Sanofi for consulting services and financialsupport for research activities from Sanofi H H Kitzlerreceived honoraria from Sanofi for speaking serviceN Kretschmann R Haase U Proschmann andH Reichmann declare no competing interest Disclosuresavailable NeurologyorgNN

Publication historyReceived by Neurology Neuroimmunology amp NeuroinflammationDecember 19 2018 Accepted in final form January 28 2019

References1 Kalincik T Brown JWL Robertson N et al Treatment effectiveness of alemtuzumab

compared with natalizumab fingolimod and interferon beta in relapsing-remittingmultiple sclerosis a cohort study Lancet Neurol 201716271ndash281

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 11

2 Hassoun LEJ Thomas K Ziemssen T Hands on Alemtuzumab-experience fromclinical practice whom and how to treat Mult Scler Demyelinating Disord 2016110

3 Ziemssen T Thomas K Alemtuzumab in the long-term treatment of relapsing-remitting multiple sclerosis an update on the clinical trial evidence and data from thereal world Ther Adv Neurol Disord 201710343ndash359

4 DrsquoAmico E Ziemssen T Cottone S Induction therapy for the management of earlyrelapsing forms of multiple sclerosis a critical opinion Expert Opin Pharmacother2017181553ndash1556

5 Ziemssen T Thomas K Treatment optimization in multiple sclerosis how do weapply emerging evidence Expert Rev Clin Immunol 201713509ndash511

6 Ziemssen T Derfuss T de Stefano N et al Optimizing treatment success in multiplesclerosis J Neurol 20162631053ndash1065

7 Ziemssen T Kern R Thomas KMultiple sclerosis clinical profiling and data collectionas prerequisite for personalized medicine approach BMC Neurol 201616124

8 Khalil M Teunissen CE Otto M et al Neurofilaments as biomarkers in neurologicaldisorders Nat Rev Neurol 201814577ndash589

9 Bergman J Dring A Zetterberg H et al Neurofilament light in CSF and serum isa sensitive marker for axonal white matter injury in MS Neurol NeuroimmunolNeuroinflamm 20163e271 doi 101212NXI0000000000000271

10 Teunissen CE Khalil M Neurofilaments as biomarkers in multiple sclerosis MultScler 201218552ndash556

11 Disanto G Barro C Benkert P et al Serum Neurofilament light a biomarker ofneuronal damage in multiple sclerosis Ann Neurol 201781857ndash870

12 Hakansson I Tisell A Cassel P et al Neurofilament light chain in cerebrospinal fluidand prediction of disease activity in clinically isolated syndrome and relapsing-remitting multiple sclerosis Eur J Neurol 201724703ndash712

13 Bhan A Jacobsen C Myhr KM Dalen I Lode K Farbu E Neurofilaments and 10-yearfollow-up in multiple sclerosis Mult Scler 2018241301ndash1307

14 van der Vuurst de Vries RMWong YYMMescheriakova JY et al High neurofilamentlevels are associated with clinically definite multiple sclerosis in children and adultswith clinically isolated syndrome Mult Scler Epub 2018 May 1

15 Petzold A The prognostic value of CSF neurofilaments in multiple sclerosis at 15-yearfollow-up J Neurol Neurosurg Psychiatry 2015861388ndash1390

16 Disanto G Adiutori R Dobson R et al Serum neurofilament light chain levels areincreased in patients with a clinically isolated syndrome J Neurol Neurosurg Psy-chiatry 201687126ndash129

17 Barro C Benkert P Disanto G et al Serum neurofilament as a predictor of diseaseworsening and brain and spinal cord atrophy in multiple sclerosis Brain 20181412382ndash2391

18 Kuhle J Nourbakhsh B Grant D et al Serum neurofilament is associated withprogression of brain atrophy and disability in early MS Neurology 201788826ndash831

19 Siller N Kuhle J Muthuraman M et al Serum neurofilament light chain is a bio-marker of acute and chronic neuronal damage in early multiple sclerosis Mult SclerEpub 2018 Mar 1

20 Caon C Namey M Meyer C et al Prevention and management of infusion-associated reactions in the comparison of alemtuzumab and rebif((R)) efficacy inmultiple sclerosis (CARE-MS) program Int J MS Care 201517191ndash198

21 Cohen JA Coles AJ Arnold DL et al Alemtuzumab versus interferon beta 1a as first-line treatment for patients with relapsing-remitting multiple sclerosis a randomisedcontrolled phase 3 trial Lancet 20123801819ndash1828

22 Coles AJ Twyman CL Arnold DL et al Alemtuzumab for patients with relapsingmultiple sclerosis after disease-modifying therapy a randomised controlled phase 3trial Lancet 20123801829ndash1839

23 Giovannoni G Tomic D Bright JR Havrdova E No evident disease activity the useof combined assessments in the management of patients with multiple sclerosis MultScler 2017231179ndash1187

24 Kuhle J Plattner K Bestwick JP et al A comparative study of CSF neurofilament lightand heavy chain protein in MS Mult Scler 2013191597ndash1603

25 Keshavan A Heslegrave A Zetterberg H Schott JM Stability of blood-based bio-markers of Alzheimerrsquos disease over multiple freeze-thaw cycles Alzheimers Dement(Amst) 201810448ndash451

26 Wilson DH Rissin DM Kan CW et al The Simoa HD-1 analyzer a novel fullyautomated digital immunoassay analyzer with single-molecule sensitivity and multi-plexing J Lab Autom 201621533ndash547

27 Petzold A Altintas A Andreoni L et al Neurofilament ELISA validation J ImmunolMethods 201035223ndash31

28 Kuhle J Barro C Andreasson U et al Comparison of three analytical platforms forquantification of the neurofilament light chain in blood samples ELISA electro-chemiluminescence immunoassay and Simoa Clin Chem Lab Med 2016541655ndash1661

29 Novakova L Zetterberg H Sundstrom P et al Monitoring disease activity in multiplesclerosis using serum neurofilament light protein Neurology 2017892230ndash2237

30 Kuhle J Barro C Disanto G et al Serum neurofilament light chain in early relapsingremitting MS is increased and correlates with CSF levels and with MRI measures ofdisease severity Mult Scler 2016221550ndash1559

31 Piehl F Kockum I Khademi M et al Plasma neurofilament light chain levels inpatients with MS switching from injectable therapies to fingolimod Mult Scler 2018241046ndash1054

32 Wiendl H Calabresi PA Meuth SG Defining response profiles after alemtuzumabrare paradoxical disease exacerbation Neurology 201890309ndash311

33 Evered L Silbert B Scott DA Zetterberg H Blennow K Association of changes inplasma neurofilament light and tau levels with anesthesia and surgery results from theCAPACITY and ARCADIAN studies JAMA Neurol 201875542ndash547

34 Kuhle J Gaiottino J Leppert D et al Serum neurofilament light chain is a biomarkerof human spinal cord injury severity and outcome J Neurol Neurosurg Psychiatry201586273ndash279

35 Shahim P Zetterberg H Tegner Y Blennow K Serum neurofilament light as a bio-marker formild traumatic brain injury in contact sports Neurology 2017881788ndash1794

36 Pujol-Calderon F Portelius E Zetterberg H Blennow K Rosengren LE Hoglund KNeurofilament changes in serum and cerebrospinal fluid after acute ischemic strokeNeurosci Lett 201869858ndash63

37 Khalil M Are neurofilaments valuable biomarkers for long-term disease prognosti-cation in MS Mult Scler 2018241270ndash1271

38 Ziemssen T Hillert J Butzkueven H The importance of collecting structured clinicalinformation on multiple sclerosis BMC Med 20161481

39 Ziemssen T Engelmann U Jahn S et al Rationale design and methods of a non-interventional study to establish safety effectiveness quality of life cognition health-related and work capacity data on Alemtuzumab in multiple sclerosis patients inGermany (TREAT-MS) BMC Neurol 201616109

40 Haase R Wunderlich M Dillenseger A Kern R Akgun K Ziemssen T Improvingmultiple sclerosis management and collecting safety information in the real world theMSDS3D software approach Expert Opin Drug Saf 201817369ndash378

12 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 3 | May 2019 NeurologyorgNN

DOI 101212NXI000000000000055520196 Neurol Neuroimmunol Neuroinflamm

Katja Akguumln Nicole Kretschmann Rocco Haase et al assessment in MS

Profiling individual clinical responses by high-frequency serum neurofilament

This information is current as of April 8 2019

ServicesUpdated Information amp

httpnnneurologyorgcontent63e555fullhtmlincluding high resolution figures can be found at

References httpnnneurologyorgcontent63e555fullhtmlref-list-1

This article cites 38 articles 3 of which you can access for free at

Citations httpnnneurologyorgcontent63e555fullhtmlotherarticles

This article has been cited by 1 HighWire-hosted articles

Subspecialty Collections

httpnnneurologyorgcgicollectionoutcome_researchOutcome research

httpnnneurologyorgcgicollectionmultiple_sclerosisMultiple sclerosis

httpnnneurologyorgcgicollectiondiagnostic_test_assessment_Diagnostic test assessmentfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

Academy of Neurology All rights reserved Online ISSN 2332-7812Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the AmericanPublished since April 2014 it is an open-access online-only continuous publication journal Copyright

is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm