ARTICLE OPEN ACCESS

Impact of treatment on cellularimmunophenotype in MSA cross-sectional study

Maria Cellerino MD Federico Ivaldi Matteo Pardini MD PhD Gianluca Rotta PhD Gemma Vila BSc

Priscilla Backer-Koduah MSc Tone Berge PhD Alice Laroni MD PhD Caterina Lapucci MD

Giovanni Novi MD Giacomo Boffa MD Elvira Sbragia MD Serena Palmeri MD Susanna Asseyer MD

EinarHoslashgestoslashlMD Cristina Campi PhDMichele Piana PhDMatilde IngleseMD PhD Friedemann PaulMD

Hanne F HarboMD PhD Pablo VillosladaMD PhD Nicole Kerlero de Rosbo PhD and AntonioUccelli MD

Neurol Neuroimmunol Neuroinflamm 20207e693 doi101212NXI0000000000000693

Correspondence

Dr Uccelli

auccellineurologiaunigeit

AbstractObjectiveTo establish cytometry profiles associated with disease stages and immunotherapy in MS

MethodsDemographicclinical data and peripheral blood samples were collected from 227 patients withMS and 82 sex- and age-matched healthy controls (HCs) enrolled in a cross-sectional study at 4European MS centers (Spain Italy Germany and Norway) Flow cytometry of isolated pe-ripheral blood mononuclear cells was performed in each center using specifically preparedantibody-cocktail Lyotubes data analysis was centralized at the Genoa center Differences inimmune cell subsets were assessed between groups of untreated patients with relapsing-remitting or progressive MS (RRMS or PMS) and HCs and between groups of patients withRRMS taking 6 commonly used disease-modifying drugs

ResultsIn untreated patients with MS significantly higher frequencies of Th17 cells in the RRMSpopulation compared withHC and lower frequencies of B-memoryB-regulatory cells as well ashigher percentages of B-mature cells in patients with PMS compared with HCs emergedOverall the greatest deviation in immunophenotype in MS was observed by treatment ratherthan disease course with the strongest impact found in fingolimod-treated patients Fingolimodinduced a decrease in total CD4+ T cells and in B-mature and B-memory cells and increases inCD4+ and CD8+ T-regulatory and B-regulatory cells

ConclusionsOur highly standardized multisite cytomics data provide further understanding of treatmentimpact on MS immunophenotype and could pave the way toward monitoring immune cells tohelp clinical management of MS individuals

These 2 senior authors contributed equally to this manuscript

From the Department of Neurosciences Rehabilitation Ophthalmology Genetics Maternal and Child Health (MC FI MP AL CL GN GB ES SP MI NKdR) and Centerof Excellence for Biomedical Research (AU) University of Genoa Italy BD Biosciences Italy (GR) Milan Institut drsquoInvestigacions Biomediques August Pi Sunyer (GV PV)Barcelona Spain Charite Universitaetsmedizin Berlin and Max Delbrueck Center for Molecular Medicine (PB-K SA FP) Germany Department of Research Innovation andEducation (TB) Neuroscience Research Unit Oslo University Hospital Department of Mechanical Electronics and Chemical Engineering (TB) Oslo Metropolitan University NorwayUniversity of Oslo (EH HFH) and Oslo University Hospital (HFH) Norway Department of Mathematics and Padova Neuroscience Center (CC) University of Padua ItalyDepartment of Mathematics (MP) University of Genoa Italy and IRCCS Ospedale Policlinico San Martino (AL MP MI AU) Genoa Italy

Go to NeurologyorgNN for full disclosures Funding information is provided at the end of the article

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 2020 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology 1

Individual interactions between inflammatory demyelinatingand degenerative processes underlying MS determine a largeheterogeneity in clinical course and treatment response12e2

(linkslwwcomNXIA199) In addition to clinical and im-aging data characterization of immune cell alterations oc-curring during the disease and in response to treatmentprovides help in better understanding MS pathogenesis andthe mechanism of action of drugs3ndash6

Changes in immune cell subsets in peripheral blood mono-nuclear cells (PBMCs) have been proposed as surrogate bio-markers of activity to monitor treatment response in MS andhelp in treatment decision7ndash9 Immunophenotyping studies ofpatients with MS have focused on alterations in compositionand function of T- or B-cell subsets10ndash12 but comprehensivestudies are lacking and present discrepancies13 Although a fewstudies comparing the effect of different disease-modifyingtreatments (DMTs) on immune cells are available re-producibility of reliable findings is lacking possibly due totechnical issues including the use of different flow cytometryparameters and monoclonal antibodies defining cell sub-populations13 Other limitations are represented by immunesystem variations including time-dependent changes in-terindividual modifications and heritable and nonheritableinfluences such as microbial and environmental factors14 andby small groups of patients with different disease characteristics

To overcome technical difficulties large multisite studies arepreferable but concerns have been raised about the accuracyand consistency of sample processing A strict standardizationof assays is indeed essential to obtain accurate measurement ofvariations in the immunologic profile Although several in-ternational consortia have been created to standardize immu-nophenotyping by flow cytometry1516 multicentric studies arestill lacking

Here we present cytomics data obtained through a cross-sectional study within the European Sys4MS project whichuses a systemsmedicine approach combining integrative omicsimaging and clinical data to develop algorithms that can beused in clinical practice toward prognosis and treatmentmanagement We have set up a reliable methodology for highlystandardized multisite PBMC processing and flow cytometryacquisition and analysis and applied such methodology to es-tablish cytometry profiles possibly associated with diseasecourse andor response to treatment in a relatively large cohortof patients We observed differences in both T- and B-cell

subsets related to disease phase in untreated patients moni-toring of the effect on immune cells induced by DMTs pointedto fingolimod (FINGO) as the drug with the greatest impact onrelevant subpopulations in particular effector and regulatory Tand B cells

MethodsStudy populationA total of 227 patients with MS (180 relapsing-remitting MS[RRMS] and 47 progressive MS [PMS] including 25 sec-ondary and 22 primary PMS) and 82 sex- and age-matchedhealthy controls (HCs) were prospectively enrolled betweenOctober 2016 and August 2017 at 4 European MS centers(44 patients and 12 HCs from Hospital Clinic of BarcelonaSpain 51 patients and 25 HCs from Oslo University HospitalNorway 41 patients and 22 HCs from Charite-Uni-versitaetsmedizin Berlin Germany 91 patients and 23 HCsfrom Ospedale Policlinico San Martino Genoa Italy) Allpatients underwent collection of demographic data clinicalhistory peripheral blood samples and assessment of the Ex-panded Disability Status Scale (EDSS) score Demographicclinical data regarding the global population are reported intable 1 See supplemental material (linkslwwcomNXIA199)for inclusion criteria

Standard protocol approvals registrationsand patient consentsThe Sys4MS project was approved by the institutional reviewboard of University of Genoa University of Oslo Charite-Universitaetsmedizin and Hospital Clinic of BarcelonaPatients provided signed informed consent before their en-rollment on the study according to the Declaration of Helsinki

Standardization of blood processing and flowcytometry acquisition and analysisThe multicentric study was set up to ensure the highest stan-dardization at all levels and consequently reduce possible biasstandard operating procedures were followed by all 4 centersfor PBMC isolation and flow cytometry acquisition of data useof Lyotubes (BD Biosciences Milan Italy Cat No 625148)from a single batch by all centers use of identical type of flowcytometer equipment and software in all centers and central-ized analysis of the data from all centers See supplementalmaterial (linkslwwcomNXIA199) and figure e-1 (linkslwwcomNXIA200) for details of the antibody panels of theLyotubes and relevant flow cytometry procedures and analysis

GlossaryALEM = alemtuzumabANCOVA = analysis of covariance ANOVA = analysis of varianceDMF = dimethyl fumarateDMT =disease-modifying treatment EDSS = Expanded Disability Status Scale FINGO = fingolimod GA = glatiramer acetateHC =healthy control IFN = interferon-β NTZ = natalizumab PBMC = peripheral blood mononuclear cell PC = principalcomponent PMS = progressive MS RRMS = relapsing-remitting MS RTXOCRE = rituximabocrelizumab TERI =teriflunomide t-SNE = t-distributed Stochastic Neighbor Embedding

2 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 NeurologyorgNN

StatisticsAnalyses were performed using SPSS 220 GraphPad Prism 80and R 350 Differences in CD3CD4 subpopulations in T-regvs T-eff tubes and between centers were assessed by the t testand analysis of variance (ANOVA) respectively For all sub-group analyses demographic differences between groups wereanalyzed using the χ2 test Mann-WhitneyKruskal-Wallis testindependent samples t test and ANOVA where appropriateComparisons of cell frequencies between untreated patientswith RRMS untreated patients with PMS and HCs as well asacross groups of patients with RRMS treated with differentdrugs untreated patients with RRMS and HCs were assessedby analysis of covariance (ANCOVA) adjusting for age and sexWhen comparing untreated patients with RRMS vs untreatedpatients with PMS disease duration and EDSS score wereadded to the covariates listed above A p value le005 was con-sidered significant In addition significance levels after Bonfer-roni correction (to adjust for multiple testing) are providedPrincipal component (PC) analysis was used to show the dis-tribution of the populations and reduce the dimensionality

considering all the variables together Data were visualized usingt-distributed Stochastic Neighbor Embedding (t-SNE)17

a nonlinear data reduction algorithm that enables the visuali-zation of high-dimensional data represented as 2-dimensionalmaps which preserve the original spacing of the data sets

Data availabilityCodified data are available through the MultipleMS EU pro-ject and database (multiplemseu) on registration

ResultsStrict standardization permitted reliableresults from a multicentric cohortTo verify that there was no significant drift with time in thebehavior of the flow cytometry equipment or in the batch ofLyotubes used in each center we performed a linear regressionanalysis As shown in figure e-2 (linkslwwcomNXIA201)the coefficients of determination r2 indicated a very poor cor-relation between time (x-axis) and tube content (y-axis) withthe regression line slopes being close to 0 These data confirmthat there was very little variation in the behavior of theequipment and Lyotubes in the different centers with time

Because the same CD3+CD4+ broad cell population couldbe identified in both T-reg and T-eff tubes we assessed thedata obtained for this population in both tubes to monitorreproducibility between tubes in each center and across the 4centers (figure e-3 linkslwwcomNXIA202) There wereno differences in CD3+CD4+ T-cell frequencies in HCs orpatients with MS between the T-reg and T-eff tubes whenthe data from each center were assessed separately (figuree-3A) nor when T-reg and T-eff tubes were compared acrosscenters (figure e-3B) Altogether these data point to thehigh reliability of the data resulting from strict standardiza-tion of the multicentric study

Untreated patients with MS show differencesin T and B cells according to disease courseAs a prerequisite to exploring differences in terms of immunecells across the whole population we applied a PC analysisincluding all the lymphocyte subpopulations studied (figure 1)By plotting the first and second PCs we obtained a meaningfulrepresentation of the data variance whereby the first and sec-ond PCs explained the 344 and 138 of the variance re-spectively As we observed a stronger deviation in theimmunologic profile of patients with MS in relation to treat-ment than to disease phenotype (figure 1A) our initial analysisfocused on untreated patients and HCs to determine possibledifferences in patients with RRMS or PMS without the in-fluence of therapy Data are reported in table 2 As previouslyreported412 significantly higher frequencies of Th17 cells inthe untreated RRMS population compared with HC (p = 001)emerged Th17 cells did not significantly differ in the PMSgroup compared with HCs or with patients with RRMS Fre-quencies of Th1 classic cells were higher in patients with PMScompared with patients with RRMS (p = 001) but not

Table 1 Demographic and clinical characteristics ofglobal MS population and controls

HC(n = 82)

RRMS(n = 180)

PMS (n = 47)

SPMS n = 25PPMS n = 22

Female n () 60 (73) 127 (71) 37 (57)

Age mean(range) y

37 (22ndash71) 40 (19ndash60) 54 (32ndash68)

Disease durationmean (SD) y

mdash 91 (63) 188 (117)

EDSS scoremedian (range)

mdash 2 (0ndash65) 5 (2ndash8)

Treatment n[mean treatmentduration y]

GA mdash 28 [38 y] 2 [08 y]

INF mdash 30 [60 y] 2 [67 y]

DMF mdash 29 [17 y] 1 [22 y]

TERI mdash 15 [14 y] mdash

FINGO mdash 28 [26 y] 2 [5 y]

NTZ mdash 22 [36 y] 1 [93 y]

ALEM mdash 6 [15 y] 1 [16 y]

DAC mdash 2 [gt 05 y] mdash

RTXOCRE mdash mdash 7 [gt 05 y]

Untreated 82 20 31

Abbreviations ALEM = alemtuzumab DAC = daclizumab DMF = dimethylfumarate EDSS = Expanded Disability Status Scale FINGO = fingolimod GA= glatiramer acetate HC = healthy control IFN = interferon-β NTZ = nata-lizumab PMS = progressive MS PPMS = primary progressive MS RRMS =relapsing-remitting MS RTXOCRE = rituximabocrelizumab SPMS = sec-ondary progressive MS TERI = teriflunomide

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 3

Figure 1 Principal component analysis showing population distribution according to phenotype and treatment

Representation of the data variance (PC1 explaining the 344 and PC2 138 of the variance respectively) The weights of PC1 and PC2 are reported in thesupplemental material file (linkslwwcomNXIA199) This figure represents the same scattered plot where (A) shows population distribution according totreatment (left panel) and to disease phenotype (right panel) and (B) shows the 95 confidence ellipses for different drugs and in particular immuno-modulatoryfirst-line-treated (GA IFN DMF and TERI) patients and FINGO-treated patients (upper left panel) antimigratory-treated (NTZ) patients andFINGO-treated patients (upper right panel) antindashCD20-treated (RTXOCRE) patients and FINGO-treated patients (lower left panel) and antindashCD52-treated(ALEM) patients and FINGO-treated patients (lower right panel) ALEM = alemtuzumab DAC = daclizumab DMF = dimethyl fumarate FINGO = fingolimod GA= glatiramer acetate HC = healthy control IFN = interferon-β NTZ = natalizumabOCRE = ocrelizumab PMS =progressiveMS RRMS = relapsing-remittingMSRTX = rituximab TERI = teriflunomide UNT = untreated patients

4 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 NeurologyorgNN

compared with HCs No difference in the Th1Th17 cellsemerged between groups Among the CD4+ cells we alsoobserved increased frequencies in CD4+CD25+CD127minus

(CD4+ T-reg) in patients with PMS compared with patientswith RRMS (p = 0004) or HCs (p = 0006) Frequency ofCD4+ T-reg cells was similar between RRMS andHC None ofthe CD8+ T-cell populations differed with disease phenotype

Analysis of total CD19+ cells showed an increased frequency inpatients with RRMS compared with patients with PMS(p = 0038) However further investigation showed robustimmunologic changes in B-cell subpopulations in patients withPMS In particular lower frequencies in CD19+CD24high

cells both CD38low (B-memory) or CD38high (B-reg) andhigher values of CD19+CD24lowCD38low (B-mature) wereevident in patients with PMS compared with HCs (p = 0004p = 0009 and p = 001 respectively) no significant differencesin these subpopulations were observed between untreatedpatients with PMS and untreated patients with RRMS None ofthe differences described above survived multiple testingcomparison (none of the p values reached Bonferroni correctedfor 16 variables across 3 groups p lt 0001)

Impactof treatmenton immunecell frequenciesAfter assessing the immunologic differences related to diseasecourse we investigated the effect of treatment Overall the

Table 2 Differences in terms of cell subpopulations in untreated patients with MS and HCs

Population characteristicsHC(n = 82)

U-RRMS(n = 20)

U-PMS(n = 31) p Valuesa p Valuesb p Valuesc

Female n () 60 (73) 16 (80) 18 (58) 03 01 01

Mean age y 37 42 56 006 lt00001 lt00001

Mean disease duration y (SD) mdash 105 (64) 203 (117) mdash mdash 0001

EDSS score median mdash 2 5 mdash mdash lt00001

Cell subpopulationCell frequencies (SD) p Valuesa p Valuesb p Valuesc

Total CD3+ 701 (114) 712 (142) 672 (23) 07 05 04

CD3+CD4+ 435 (94) 434 (117) 428 (21) 03 06 02

CD3+CD8+ 211 (70) 218 (106) 199 (16) 06 08 09

CD4+CD25+CD1272 (CD4+T-reg) 48 (12) 47 (21) 64 (04) 09 0006 0004

CD3+CD8+CD282CD1272 (CD8+T-reg) 229 (167) 259 (36) 267 (36) 04 06 04

CD3+CD4+CCR62CD1612CXCR3+ (Th1) 90 (57) 71 (56) 121 (12) 01 008 001

CD3+CD4+CCR6+CD161+CXCR32

CCR4+ (Th17)06 (05) 11 (06) 08 (01) 001 07 04

CD3+CD4+CCR6+CD161+CxCR3highCCR4low (Th1Th17)

20 (18) 22 (04) 21 (11) 03 06 03

Total CD19+ 61 (03) 81 (61) 68 (06) 0054 03 0038

CD19+CD24highCD38low (B-memory) 198 (91) 205 (98) 123 (18) 04 0004 02

CD19+CD24lowCD38low (B-mature) 442 (138) 430 (143) 556 (32) 05 001 02

CD19+CD24highCD38high (B-reg) 62 (47) 61 (45) 35 (09) 05 0009 03

CD19+CD242CD38high (B-plasma) 33 (31) 32 (23) 29 (05) 05 08 08

CD19+CD24highCD382

(B-memory-atypical)195 (89) 205 (89) 109 (102) 06 01 06

CD16+CD56 low (CD56 dim) 717 (74) 626 (127) 688 (80) 07 09 08

CD16+CD56 high (CD 56 bright) 29 (21) 32 (28) 26 (02) 02 04 08

Abbreviations EDSS = Expanded Disability Status Scale HC = healthy control U-PMS = untreated progressive MS U-RRMS = untreated relapsing-remitting MSa p Values for the HC vs U-RRMS comparison using the independent samples t-test (age) χ2 test (sex) and analysis of covariance adjusted for sex and age(cell frequencies)b p Values for the HC vs U-PMS comparison using the independent samples t-test (age) χ2 test (sex) and analysis of covariance adjusted for sex and age(cell frequencies)c p Values for the U-RRMS vs U-PMS comparison using the independent samples t-test (age) χ2 test (sex) Mann-Whitney (disease duration and EDSS score)and analysis of covariance adjusted for sex age disease duration and EDSS score (cell frequencies)Significant differences between the groups are reported in bold

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 5

strongest impact on immune cells seemed attributable toFINGO (figure 1) Indeed PC analysis showed a segregationof FINGO-treated patients compared with all the others evenwhen they were grouped according to the drugsrsquo mechanismof action (immunomodulatory first-line drugs glatirameracetate (GA)interferon-β (IFN)dimethyl fumarate(DMF)teriflunomide (TERI) antimigratory drug (otherthan FINGO) natalizumab (NTZ) anti-CD20 rituximabocrelizumab (RTXOCRE) and anti-CD52 alemtuzumab(ALEM) monoclonal antibodies figure 1B)

At the level of single immune cell populations we assesseddifferences induced by treatments only in patients with RRMSas they represented a more homogeneous population (samedisease course no differences in terms of the mean diseaseduration or EDSS score table 3) Because only a few patientswith RRMS were taking ALEM (n = 6) or daclizumab (n = 2)we excluded these patients from this analysis Thus we onlyconsidered patients with RRMS who were under treatmentwith the 6 DMTs most commonly used in these cohorts (GAIFN DMF TERI FINGO and NTZ) untreated RRMS andHC The frequencies of the different cell subpopulations foreach group and ANCOVA results are reported in table 3 Apost hoc analysis confirmed FINGO being the drug with thestrongest impact on different cell subpopulations as describedbelow and shown in figure 2 (depicted by t-SNE algorithm)Bonferroni correction for multiple comparison was applied forthe 16 variables across the 8 groups (p lt 00004)

DMTs and T cells downregulation of CD3+CD4+ cellsand upregulation of CD4+ andCD8+T-reg cells in FINGO-treated patientsData are presented in table 3 and Figure 2 AndashC FINGO-treated patients had lower frequencies in total CD4+ T cells (plt 00001 for FINGO vs each other group) no effect of FINGOon total CD8+ T cells emerged CD8+ T-cell frequencies wereslightly reduced in DMF-treated compared with IFN-treatedpatients (p = 003) and HCs (p = 002) but this difference didnot survive the multiple testing comparison We observeda clear increase in both CD4+ (CD4+CD25+CD127minus) andCD8+ (CD8+CD28minusCD127minus) T-reg cell frequencies inFINGO-treated patients compared with each other group(p lt 00001) No effect was observed in the FINGO-treatedpopulation in terms of Th1 Th17 andTh117 cells Th17 cellswere lower in HCs compared with IFN-treated patients(p = 003) whereas Th1Th17 cells were higher in IFN-treatedvs DMF-treated patients (p = 001) but these differences didnot survive multiple comparisons

DMTs and B cells upregulation of B-reg anddownregulation of B-memory and B-mature cellfrequencies in FINGO-treated patientsData are presented in table 3 and figure 2D Patients treatedwith anti-CD20 were excluded from this analysis because theyhave almost no B cells We observed reduced frequencies oftotal CD19+ cells in FINGO-treated compared with untreatedIFN- NTZ- or TERI-treated patients (p = 001 p lt 00001

p = 00003 and p = 0003 respectively) Only FINGO vs IFNand NTZ survived multiple testing comparison Frequencies ofCD19+CD24highCD38high B-reg cells were higher inFINGO-treated patients (p lt 00001 compared with each othergroup) in DMF-treated patients compared with HCs(p = 00002) untreated (p = 0007) or NTZ-treated (p = 001)patients and in the IFN-treated group compared with HCs(p lt 00001) untreated (p = 0001) or NTZ- and GA-treated(p = 0001 and p = 0002 respectively) patients Comparisonbetween DMF- and IFN-treated populations survived multipletesting comparison only vs HC Frequencies of CD19+-

CD24highCD38low B-memory cells were lower in theFINGO-treated patients vs HCs (p lt 00001) and vs untreated(p lt 00001) and NTZ-treated (p = 00014) patients the latternot surviving multiple testing comparison Nominal signifi-cance was also reached in the DMF-treated group with lowerfrequencies of B-memory cells vs untreated patients and HCs(p = 0006 and p = 0001 respectively) A significant increase inCD19+CD24highCD38minus B-memory-atypical cells in NTZ-treated patients compared with FINGO- IFN- TERI- andDMF-treated patients (p lt 00001 for all the comparisons)emerged nominal significance was also observed in theNTZ vsGA comparison (p = 0002) The FINGO-treated group alsoshowed reduced frequencies of CD19+CD24lowCD38lowmature B cells compared with DMF- (p = 00001) and TERI-treated (p lt 00001) patients reaching nominal significance inthe comparison with HCs (p = 0004) NTZ- (p = 002) GA-(p = 00004) and IFN-treated (p = 001) patients LastCD19+CD24minusCD38high plasma cells were higher in theFINGO-treated group compared with DMF- (p = 002) IFN-(p = 0001) NTZ- (p = 00002) treated patients or HCs(p = 001) with only the comparison with NTZ survivingmultiple testing comparison

No evident impact of treatment in terms of NK cells eitherCD56dim or CD56bright was observed (table 3 and figure2E) Nominal significance was reached in the comparison ofCD56bright NK cells for FINGO-treated (16) vs IFN-treated (49 p = 0001) or TERI-treated (49 p = 0001)patients but none of the comparisons survived multiple testinganalysis

DiscussionCharacterization of immunologic changes occurring duringMSand in response to DMTs could affect the choice of the mostappropriate therapy7 The impact of MS drugs on the immunesystem differs in terms of rapidity selectivity magnitude anddurability resulting in short- and long-term effects differentlyaffecting immune cell subsets The gold standard of therapy forautoimmune diseases such as MS is the eradication of autor-eactive cells and restoration of immune tolerance resulting inthe control of effectorpathogenicautoreactive cells by regu-latory networks Characterization of the impact of DMTs onthe immune system not only helps to better understand theirmechanisms of action but has also given insights into the

6 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 NeurologyorgNN

Table 3 Cell frequencies in the relapsing-remitting MS population and HCs

Population characteristics HC (n = 82) U-RRMS (n = 20) GA (n = 28) IFN (n = 30) DMF (n = 29) TERI (n = 15) FINGO (n = 28) NTZ (n = 22) p Valuesa

Female n () 60 (73) 16 (80) 18 (64) 19 (63) 20 (69) 10 (66) 21 (75) 18 (81) 07

Age mean y 37 42 41 42 41 38 39 38 02

Disease duration mean (SD) y mdash 105 (64) 89 (62) 97 (58) 85 (67) 76 (62) 78 (48) 109 (81) 05

EDSS score median mdash 2 15 15 2 2 25 25 006

Cell subpopulationCell frequencies (SD) p Valuesb

Total CD3+ 701 (114) 712 (142) 717 (109) 681 (116) 621 (173) 752 (57) 407 (186) 633 (107) lt00001c

CD3+CD4+ 435 (94) 434 (117) 484 (111) 436 (106) 452 (175) 459 (116) 103 (103)d 396 (87) lt00001c

CD3+CD8+ 211 (70) 218 (106) 192 (61) 213 (91) 151 (68) 225 (35) 181 (94) 213 (72) 001d

CD4+CD25+CD1272 (CD4+T-reg) 48 (12) 47 (21) 58 (23) 706 (28) 58 (41) 51 (23) 106 (57) 51 (15) lt00001c

CD3+CD8+CD282CD1272

(CD8+T-reg)229 (167) 259 (36) 272 (148) 226 (184) 249 (199) 289 (151) 545 (240) 218 (135) lt00001c

CD3+CD4+CCR62CD1612

CXCR3+ (Th1)90 (57) 71 (56) 92 (68) 112 (54) 76 (63) 74 (53) 149 (285) 142 (48) 005

CD3+CD4+CCR6+CD161+

CXCR32CCR4+ (Th17)06 (05) 11 (06) 08 (06) 12 (08) 104 (11) 07 (06) 11 (08) 07 (06) 001e

CD3+CD4+CCR6+CD161+

CxCR3highCCR4low (Th1Th17)20 (18) 22 (04) 23 (21) 23 (21) 11 (15) 14 (159) 18 (25) 23 (11) 004f

Total CD19+ 61 (03) 81 (61) 61 (28) 94 (52) 72 (50) 91 (43) 40 (27) 92 (48) lt00001g

CD19+CD24highCD38low(B-memory)

198 (91) 205 (98) 123 (18) 151 (100) 114 (84) 146 (47) 92 (58) 194 (71) lt00000h

CD19+CD24lowCD38low(B-mature)

442 (138) 430 (143) 498 (171) 463 (155) 508 (185) 5803 (84) 317 (1403) 360 (118) lt00000i

CD19+CD24highCD38high(B-reg)

62 (47) 61 (45) 69 (43) 152 (107) 142 (71) 83 (25) 267 (164) 62 (32) lt00001cj

CD19+CD242CD38high(B-plasma)

33 (31) 32 (23) 31 (27) 18 (13) 27 (35) 22 (26) 705 (125) 08 (07) lt00001k

CD19+CD24highCD382

(B-memory-atypical)132 (68) 135 (73) 107 (89) 81 (75) 76 (61) 65 (29) 84 (75) 184 (94) lt00001l

CD16+ CD56 low(CD56 dim)

717 (74) 626 (127) 638 (157) 459 (199) 581 (233) 522 (132) 680 (193) 579 (146) 03

Continued

Neurolo

gyorgN

NNeu

rologyN

euroimmunology

ampNeuroinflam

mation

|Volum

e7N

umber

3|

May

20207

immunopathogenesis of MS3 A clear example comes from theuse of B-cell depleting drugs which demonstrated beyondantibody-dependent functions unexpected antibody-independent roles for B cells18 Accordingly studies address-ing comprehensively the impact of DMTs on large cohorts ofpatients with MS and taking advantage of robust and re-producible methodological approaches are of pivotalimportance

Altered levels andor functional abnormalities in T or B cellshave already been reported in patients with MS1219ndash21

However previous studies are generally characterized byseveral limitations including the focus on particular restrictedimmune cell subsets small groups of patients mainly treatedwith a single DMT1122ndash31 different disease characteristicslack of inclusion of controls and difficult standardization ofsample processing5131516 In this study we have undertakena large multicentric analysis of the immunologic profile of Tand B cells in a cohort of 227 patients with MS at differentdisease stages treated with different DMTs and 82 HCs Weshow that our highly standardized approach provided reliableresults compensating the paucity of reproducible findings inlarge and multicentric cohorts516

Using this standardized reliable methodology we have com-pared single-cell subpopulations between groups of patientswith untreated RRMS or PMS and HCs (to evaluate immu-nologic features at different MS phases without the confoundof immunomodulatory therapy exposure) and assessed dif-ferences induced by 6 commonly used DMTs with differentmechanism of action

Although the strongest immune deviation in patients with MSappeared to be induced by treatments rather than by diseasecourse we observed some differences in single immune cellsubpopulations also in untreated patients In particular un-treated patients with RRMS showed higher frequencies ofTh17 cells compared with HCs Although elevated Th17 cellshave been reported across the whole spectrum of MS pheno-types being particularly indicative of an active inflammatoryprocess1232 we only observed differences between RRMS andHC but not between PMS and HC or RRMS this discrep-ancy32 could be related to the smaller number of patients withPMS included in our study and to the fact that our analyseswere corrected for both age and sex as well as for diseaseduration and EDSS score when comparing RRMS and PMSWe also observed increased frequencies of CD4+T-reg cells inpatients with PMS compared with patients with RRMS andHCs CD4+CD25+CD127lowndashFoxP3 have been reported tobe lowest in RRMS but to recover in the later secondary PMSphase433 Although we did not evaluate FoxP3 expression as-sociated with CD4+ T-reg cells the monitoring ofCD4+CD25+CD127minus T cells is equivalent with the analysis ofCD4+CD25+CD127lowndashFoxP3 T cells4 and our resultssuggest that CD4+ T-reg cell frequencies increase in the pro-gressive phases of the disease but remain stable in the initialrelapsing phase In untreated patients with PMS a deviation inTa

ble

3Cellfrequen

cies

intherelapsing-remittingMSpopulationan

dHCs(con

tinue

d)

Cellsu

bpopulation

Cellfrequencies

(SD)

pValuesb

CD16

+CD56

high

(CD

56brigh

t)29(21)

32(28)

31(21)

49(39)

34(21)

49(35)

16(14)

37(30)

000

1m

AbbreviationsDMF=dim

ethyl

fumarate

EDSS

=Ex

pan

ded

Disab

ility

StatusSc

ale

FINGO

=fingo

limodG

A=glatiram

erac

etate

HC=hea

lthyco

ntrolIFN

=interferon-βN

TZ=natalizumab

TER

I=teriflunomide

U-RRMS=

untrea

tedrelapsing-remittingMS

apValues

forco

mparisonsac

ross

the8groupsforse

x(χ

2)a

ndag

e(analysisofva

rian

ce)o

r7groupsofpatients

(HCex

cluded

)fordisea

sedurationan

dED

SSscore

(Kru

skal-W

allis)

bpValues

forco

mparisonsac

ross

the8groups(analysisofco

varian

cea

djusted

forag

ean

dse

x)S

ignifican

tdifference

sarereported

inbolds

tatistically

sign

ifican

tpost

hocan

alysisdetailsaredes

cribed

below

cplt000

01in

theco

mparisonbetwee

nFINGO-treated

patients

andalltheother

groupsse

parately

dp=002

fortheDMFvs

HCco

mparisonp

=003

fortheDMFvs

IFN

comparison

ep=002

fortheHCvs

IFN

comparisonp

=004

fortheHCvs

patients

withU-RRMSco

mparison

fp=001

fortheIFN

vsDMFco

mparison

gplt000

01fortheFINGO

vsIFNp

=000

03fortheFINGO

vsNTZ

comparisonp

=000

3fortheFINGO

vsTE

RIc

omparisonp

=001

fortheFINGO

vsU-RRMSco

mparison

hp=000

6fortheDMFvs

U-RRMSco

mparisonp

=000

1fortheDMFvs

HCco

mparisonp

lt000

01fortheFINGO

vsU-RRMSan

dvs

HCco

mparisonp

=000

14forFINGO

vsNTZ

comparison

ip=000

01fortheFINGOvs

DMFco

mparisonp

lt000

01forFINGOvs

TERIcomparisonp

=000

4fortheFINGOvs

HCco

mparisonp

=002

fortheFINGOvs

NTZ

comparisonp

=000

04fortheFINGOvs

GAco

mparisonp

=001

forFINGO

vsIFN

comparison

jp=000

7fortheDMFvs

U-RRMSco

mparisonp

=000

02forDMFvs

HCco

mparisonp

=001

fortheDMFvs

NTZ

comparisonp

=002

fortheDMFvs

HCco

mparisonp

=003

fortheDMFvs

IFN

comparison

kp=002

fortheFINGO

vsHCorvs

DMFco

mparisonp

=000

1fortheFINGO

vsIFN

comparisonp

=000

02fortheFINGO

vsNTZ

comparison

lplt000

01fortheNTZ

vsFINGOIFN

TER

Ian

dDMFco

mparisonp

=000

2fortheNTZ

vsGAco

mparison

mp=000

1fortheFINGO

vsIFN

andTE

RIc

omparison

8 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 NeurologyorgNN

Figure 2 T-distributed Stochastic Neighbor Embedding (t-SNE) representation of the immunologic profiles of healthycontrols and patients with relapsing-remitting MS under different disease-modifying treatments

Representations as depicted by t-SNE algorithm (n = 600000 cells) of (A) total CD4+ and CD8+ T cells (B) T-reg cells (C) T-eff cells (D) CD19+ B cells and (E) NKcells DMF = dimethyl fumarate FINGO = fingolimod GA = glatiramer acetate HC = healthy control IFN = interferon-β NTZ = natalizumab TERI = teri-flunomide UNT = untreated patients

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 9

the frequencies of B-cell subpopulations was also observedcompared with HCs with a decrease of B-memory and B-regcells and an increase of the B-mature arm This is particularlyinteresting as the first DMT affecting disability progression inPMS is a B cellndashdepleting agent (ocrelizumab)34

When we compared the effect of treatments on immune cellsof patients included in this study the strongest impact wasobserved in the FINGO-treated patients Our results confirmand strengthen those of a previous study10 where FINGOhad been shown to induce the greatest changes in the pe-ripheral immune system compared with other drugs in thestudy In line with previous findings we observed a decrease intotal CD4+ T cells in patients treated with FINGO significantdecreases were also observed in these patients for total CD19+

B cells and more particularly mature and memory B cells Wealso observed a clear increase in both CD4+ and CD8+ T-regcells as well as in B-reg cells in FINGO-treated patients Thesefindings support previous findings suggesting a role forFINGO in restoring regulatory networks possibly impaired inMS27283536

Contradictory data on the effect of FINGO treatment onTh17 cells have been reported9272836 Thus opposite resultswere obtained in some longitudinal studies with FINGOupregulating9 or downregulating27 this T-cell subset Ina cross-sectional study a decrease in Th17 cells was ob-served28 In contrast on par with another study36 we did notobserve any overall difference in Th17 cell frequency inFINGO-treated patients This could be due to the use ofdifferent andor fewer surface markers used to define thisT-cell subset9272836

Other DMTs had an impact on immune cell profile Thus asalready reported CD8+ T cells3137 and B-reg and B-memorycells1125 were affected by DMF treatment The limited use inour cohort of patients of DMTs that more recently enteredthe market in Europe did not allow us to obtain enoughinformation on B cellndashdepleting agents and on cladribine

The mode of action of FINGO on immune cells is believed tobe antimigratory related to their expression of sphingosine-1-phosphate receptors whereby FINGO treatment wouldprevent their egress from lymphoid tissue and their migrationinto the CNS38 Of interest NTZ a recombinant humanizedanti-α4-integrin antibody preventing immune cells to crossthe blood-brain barrier affected immune cell levels especiallyatypical memory B cells albeit to a lesser extent than FINGOIt should be noted that in contrast to most other studies ourdata were obtained through comparison between severalDMTs at a cross-sectional level rather than through a longi-tudinal study of the effect of 1 drug In this context the modeof action of FINGO might not be strictly accounted for by itsantimigratory effect Recent studies suggest other possibleadjunctive mechanisms whereby FINGO would suppresslymphopoiesis39 and through blocking of the S1P1 receptorson T cells could also lead to the sequestration of T cells in the

bone marrow40 thereby reducing circulating deleteriousT cells and consequently neuroinflammation

Altogether the results of our highly standardized study whichcompares the immunophenotype changes in patients withMSundergoing several widely used treatments rather than lon-gitudinal monitoring of cell subtypes in patients under a singletreatment type strengthen and add to previous findings Theyalso suggest that immunomonitoring by flow cytometry as anadjunct to clinical and imaging data could deepen ourknowledge about the mechanism of action of DMTs Itappears particularly interesting for FINGO whose impact onthe immune system suggests that its effects might go beyondthe well-known antimigratory effect Further highly stan-dardized studies that include longitudinal data and encompasslarger cohorts of patients treated with each DMT couldprovide proof of concept toward the additional use ofimmunophenotype monitoring in treatment management

AcknowledgmentThis work was supported by the European Commission(ERACOSYSMED ERA-Net program Sys4MS project id43) Instituto de Salud Carlos III Spain (AC1500052) theItalian Ministry of Health (WFR-PER-2013-02361136) theGerman Ministry of Science (Deutsches Teilprojekt BldquoForderkennzeichen 031L0083Brdquo) and the NorwegianResearch Council (project 257955) The authors thank ECarpi I Mo and F Kropf for their technical andor help withpatients at Genoa and Oslo centers and all the participatingpatients with MS and healthy individuals

Study fundingThis work was supported by the European Commission(ERACOSYSMED ERA-Net program Sys4MS project id43) Instituto de Salud Carlos III Spain (AC1500052) theItalian Ministry of Health (WFR-PER-2013-02361136) theGerman Ministry of Science (Deutsches Teilprojekt BldquoForderkennzeichen 031L0083Brdquo) and the Norwegian Re-search Council (project 257955)

DisclosureM Cellerino and F Ivaldi report no disclosures M Pardinireceived research support from Novartis and Nutricia andhonoraria fromMerk andNovartis G Rotta is an employee ofBD Biosciences Italy G Vila reports no disclosures PBacker-Koduah is funded by the DFG Excellence grant to FP(DFG exc 257) and is a junior scholar of the Einstein foun-dation T Berge has received unrestricted research grantsfrom Biogen and Sanofi-Genzyme A Laroni received grantsfrom FISM and Italian Ministry of Health and receivedhonoraria or consultation fees from Biogen Roche TevaMerck Genzyme and Novartis C Lapucci G Novi G BoffaE Sbragia and S Palmeri report no disclosures S Asseyerreceived a conference grant from Celgene E Hoslashgestoslashl re-ceived honoraria for lecturing from Merck and Sanofi-Genzyme C Campi and M Piana report no disclosures MInglese received research grants from the NIH DOD NMSS

10 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 NeurologyorgNN

FISM and Teva Neuroscience and received honoraria orconsultation fees from Roche Merck Genzyme and BiogenF Paul received honoraria and research support from AlexionBayer Biogen Chugai Merck Serono Novartis GenzymeMedImmune Shire and Teva and serves on scientific advi-sory boards of Alexion MedImmune and Novartis He hasreceived funding from Deutsche Forschungsgemeinschaft(DFG Exc 257) Bundesministerium fur Bildung und For-schung (Competence Network Multiple Sclerosis) theGuthy-Jackson Charitable Foundation EU Framework Pro-gram 7 and the National Multiple Sclerosis Society of theUnited States HF Harbo received travel support honorariafor advice or lecturing from Biogen Idec Sanofi-GenzymeMerck Novartis Roche and Teva and unrestricted researchgrants from Novartis and Biogen P Villoslada received con-sultancy fees and holds stocks from Bionure Farma SL SpiralTherapeutics Inc Health Engineering SL QMenta Inc andAttune Neurosciences Inc N Kerlero de Rosbo reports nodisclosures A Uccelli received grants and contracts fromFISM Novartis Biogen Merck Fondazione Cariplo andItalian Ministry of Health and received honoraria or consul-tation fees from Biogen Roche Teva Merck Genzyme andNovartis Go to NeurologyorgNN for full disclosures

Publication historyReceived by Neurology Neuroimmunology amp NeuroinflammationAugust 30 2019 Accepted in final form January 5 2020

References1 Thompson AJ Baranzini SE Geurts J Hemmer B Ciccarelli O Multiple sclerosis

Lancet 20183911622ndash16362 Krieger SC Cook K De Nino S Fletcher M The topographical model of multiple

sclerosis a dynamic visualization of disease course Neurol Neuroimmunol Neuro-inflamm 20163e279 doi 101212NXI0000000000000279

3 Martin R Sospedra M Rosito M Engelhardt B Current multiple sclerosis treatmentshave improved our understanding of MS autoimmune pathogenesis Eur J Immunol2016462078ndash2090

4 Jones AP Kermode AG Lucas RM Carroll WM Nolan D Hart PH Circulatingimmune cells in multiple sclerosis Clin Exp Immunol 2017187193ndash203

5 Willis JC Lord GM Immune biomarkers the promises and pitfalls of personalizedmedicine Nat Rev Immunol 201515323ndash329

6 Kalincik T Manouchehrinia A Sobisek L et al Towards personalized therapy for multiplesclerosis prediction of individual treatment response Brain 20171402426ndash2443

7 Pardo G Jones DE The sequence of disease-modifying therapies in relapsing multiplesclerosis safety and immunologic considerations J Neurol 20172642351ndash2374

8 Jaye DL Bray RA Gebel HM Harris WA Waller EK Translational applications offlow cytometry in clinical practice J Immunol 20121884715ndash4719

9 Teniente-Serra A Grau-Lopez L Mansilla MJ et al Multiparametric flow cytometricanalysis of whole blood reveals changes in minor lymphocyte subpopulations ofmultiple sclerosis patients Autoimmunity 201649219ndash228

Appendix Authors

Name Location Role Contribution

MariaCellerinoMD

University of Genoa Author Acquisition of clinicaldata analyzed the datainterpreted the dataand drafted themanuscript forintellectual content

FedericoIvaldi

University of Genoa Author Acquisition of flowcytometry data andanalyzed the data

MatteoPardini MDPhD

University of Genoaand IRCCS OspedalePoliclinico SanMartino

Author Acquisition of clinicaldata analyzed the dataand interpreted thedata

GianlucaRotta PhD

BD Biosciences Italy Author Optimization ofLyotubes andstandardization of flowcytometry analysis

Gemma VilaBSc

IDIBAPS Author Acquisition of flowcytometry data

PriscillaBacker-KoduahMSc

ChariteUniversitaetsmedizinBerlin

Author Acquisition of flowcytometry data

SusannaAsseyer MD

ChariteUniversitaetsmedizinBerlin

Author Acquisition of clinicaldata

Tone BergePhD

University of Oslo Author Acquisition of flowcytometry data

EinarHoslashgestoslashlMD

University of Oslo Author Acquisition of clinicaldata

Appendix (continued)

Name Location Role Contribution

Alice LaroniMD PhD

University of Genoaand IRCCS OspedalePoliclinico SanMartino

Author Acquisition of clinicaldata

CaterinaLapucci MD

University of Genoa Author Acquisition of clinicaldata

GiovanniNovi MD

University of Genoa Author Acquisition of clinicaldata

GiacomoBoffa MD

University of Genoa Author Acquisition of clinicaldata

ElviraSbragia MD

University of Genoa Author Acquisition of clinicaldata

SerenaPalmeri MD

University of Genoa Author Acquisition of clinicaldata

CristinaCampi PhD

University of Padova Author Analyzed the data

MichelePiana PhD

University of Genoa Author Analyzed the data

MatildeInglese MDPhD

University of Genoaand IRCCS OspedalePoliclinico SanMartino

Author Revised the manuscriptfor intellectual content

FriedemannPaul MD

ChariteUniversitaetsmedizinBerlin

Author Revised the manuscriptfor intellectual content

Hanne FHarbo MDPhD

University of Oslo andOslo UniversityHospital

Author Revised the manuscriptfor intellectual content

PabloVillosladaMD PhD

IDIBAPS Author Contributed to thedesign of the study andrevised the manuscriptfor intellectual content

NicoleKerlero deRosbo PhD

University of Genoa Author Designed andconceptualized thestudy interpreted thedata and revisedthe manuscript forintellectual content

AntonioUccelli MD

University of Genoaand IRCCS OspedalePoliclinico SanMartino

Author(correspondingauthor)

Designed andconceptualized thestudy interpreted thedata and revisedthe manuscript forintellectual content

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 11

10 Dooley J Pauwels I Franckaert D et al Immunologic profiles of multiple sclerosistreatments reveal shared early B cell alterations Neurol Neuroimmunol Neuro-inflamm 20163e240 doi 101212NXI0000000000000240

11 Staun-Ram E Najjar E Volkowich AMiller A Dimethyl fumarate as a first- vs second-line therapy in MS focus on B cells Neurol Neuroimmunol Neuroinflamm 20185e508 doi 101212NXI0000000000000508

12 Durelli L Conti L Clerico M et al T-helper 17 cells expand in multiple sclerosis andare inhibited by interferon-beta Ann Neurol 200965499ndash509

13 Roep BO Buckner J Sawcer S Toes R Zipp F The problems and promises of researchinto human immunology and autoimmune disease Nat Med 20121848ndash53

14 Davis MM Brodin P Rebooting human immunology Annu Rev Immunol 201836843ndash864

15 FinakG LangweilerM JaimesM et al Standardizing flow cytometry immunophenotypinganalysis from the human immunoPhenotyping consortium Sci Rep 2016620686

16 Maecker HT McCoy JP Nussenblatt R Standardizing immunophenotyping for thehuman immunology project Nat Rev Immunol 201212191ndash200

17 van der Maaten L Hinton G Visualizing data using t-SNE J Mach Learn Res 200892579ndash2605

18 Li R Patterson KR Bar-Or A Reassessing B cell contributions in multiple sclerosisNat Immunol 201819696ndash707

19 Duddy M Niino M Adatia F et al Distinct effector cytokine profiles of memory andnaive human B cell subsets and implication in multiple sclerosis J Immunol 20071786092ndash6099

20 Viglietta V Baecher-Allan C Weiner HL Hafler DA Loss of functional suppressionby CD4+CD25+ regulatory T cells in patients with multiple sclerosis J Exp Med2004199971ndash979

21 Jelcic I Al Nimer F Wang J et al Memory B cells activate brain-homing autoreactiveCD4(+) T cells in multiple sclerosis Cell 201817585ndash100e23

22 Chiarini M Serana F Zanotti C et al Modulation of the central memory and Tr1-likeregulatory T cells in multiple sclerosis patients responsive to interferon-beta therapyMult Scler 201218788ndash798

23 Ireland SJ Guzman AA OrsquoBrien DE et al The effect of glatiramer acetate therapy onfunctional properties of B cells from patients with relapsing-remitting multiple scle-rosis JAMA Neurol 2014711421ndash1428

24 Klotz L Eschborn M Lindner M et al Teriflunomide treatment for multiple sclerosismodulates T cell mitochondrial respiration with affinity-dependent effects Sci TranslMed 201911

25 Lundy SK Wu Q Wang Q et al Dimethyl fumarate treatment of relapsing-remittingmultiple sclerosis influences B-cell subsets Neurol Neuroimmunol Neuroinflamm20163e211 doi 101212NXI0000000000000211

26 Stuve O Marra CM Jerome KR et al Immune surveillance in multiple sclerosispatients treated with natalizumab Ann Neurol 200659743ndash747

27 Serpero LD Filaci G Parodi A et al Fingolimod modulates peripheral effector andregulatory T cells in MS patients J Neuroimmune Pharm 201381106ndash1113

28 Mehling M Lindberg R Raulf F et al Th17 central memory T cells are reduced byFTY720 in patients with multiple sclerosis Neurology 201075403ndash410

29 Baker D Herrod SS Alvarez-Gonzalez C Zalewski L Albor C Schmierer K Bothcladribine and alemtuzumab may effect MS via B-cell depletion Neurol Neuro-immunol Neuroinflamm 20174e360 doi 101212NXI0000000000000360

30 Gross CC Ahmetspahic D Ruck T et al Alemtuzumab treatment alters circulatinginnate immune cells in multiple sclerosis Neurol Neuroimmunol Neuroinflamm20163e289 doi 101212NXI0000000000000289

31 Spencer CM Crabtree-Hartman EC Lehmann-Horn K Cree BA Zamvil SS Re-duction of CD8(+) T lymphocytes in multiple sclerosis patients treated with dimethylfumarate Neurol Neuroimmunol Neuroinflamm 20152e76 doi 101212NXI0000000000000076

32 Romme Christensen J Bornsen L Ratzer R et al Systemic inflammation in pro-gressive multiple sclerosis involves follicular T-helper Th17- and activated B-cells andcorrelates with progression PLoS One 20138e57820

33 Venken K Hellings N Broekmans T Hensen K Rummens JL Stinissen P Naturalnaive CD4+CD25+CD127low regulatory T cell (Treg) development and functionare disturbed in multiple sclerosis patients recovery of memory Treg homeostasisduring disease progression J Immunol 20081806411ndash6420

34 Montalban X Hauser SL Kappos L et al Ocrelizumab versus placebo in primaryprogressive multiple sclerosis N Engl J Med 2017376209ndash220

35 Grutzke B Hucke S Gross CC et al Fingolimod treatment promotes regulatoryphenotype and function of B cells Ann Clin Transl Neurol 20152119ndash130

36 Sato DK Nakashima I Bar-Or A et al Changes in Th17 and regulatory T cellsafter fingolimod initiation to treat multiple sclerosis J Neuroimmunol 201426895ndash98

37 Fleischer V Friedrich M Rezk A et al Treatment response to dimethyl fumarate ischaracterized by disproportionate CD8+ T cell reduction in MS Mult Scler 201824632ndash641

38 Massberg S von Andrian UH Fingolimod and sphingosine-1-phosphatendashmodifiersof lymphocyte migration N Engl J Med 20063551088ndash1091

39 Blaho VA Galvani S Engelbrecht E et al HDL-bound sphingosine-1-phosphaterestrains lymphopoiesis and neuroinflammation Nature 2015523342ndash346

40 Chongsathidkiet P Jackson C Koyama S et al Sequestration of T cells in bonemarrow in the setting of glioblastoma and other intracranial tumors Nat Med 2018241459ndash1468

12 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 NeurologyorgNN

DOI 101212NXI000000000000069320207 Neurol Neuroimmunol Neuroinflamm

Maria Cellerino Federico Ivaldi Matteo Pardini et al Impact of treatment on cellular immunophenotype in MS A cross-sectional study

This information is current as of March 5 2020

ServicesUpdated Information amp

httpnnneurologyorgcontent73e693fullhtmlincluding high resolution figures can be found at

References httpnnneurologyorgcontent73e693fullhtmlref-list-1

This article cites 39 articles 11 of which you can access for free at

Citations httpnnneurologyorgcontent73e693fullhtmlotherarticles

This article has been cited by 2 HighWire-hosted articles

Subspecialty Collections

httpnnneurologyorgcgicollectionmultiple_sclerosisMultiple sclerosisfollowing 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 2020 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