ARTICLE OPEN ACCESS CLASS OF EVIDENCE

Cladribine vs other drugs in MSMerging randomized trial with real-life data

Alessio Signori PhD Francesco Sacca MD Roberta Lanzillo MD PhD Giorgia Teresa Maniscalco MD

Elisabetta Signoriello MD PhD Anna Maria Repice MD Pietro Annovazzi MD Damiano Baroncini MD

Marinella Clerico MD PhD Eleonora Binello MD Raffaella Cerqua MD Giorgia Mataluni MD PhD

Paola Perini MD PhD Simona Bonavita MD Luigi Lavorgna MD PhD Ignazio Roberto Zarbo MD PhD

Alice Laroni MD PhD Lorena Pareja-Gutierrez MD Sara La Gioia MD Barbara Frigeni MD

Valeria Barcella MD Jessica Frau MD Eleonora Cocco MD Giuseppe Fenu MD PhD

Valentina Torri Clerici MD Arianna Sartori MD Sarah Rasia MD Cinzia Cordioli MD

Maria Laura Stromillo MD PhD Alessia Di Sapio MD Simona Pontecorvo MD PhD Roberta Grasso MD

Stefania Barone MD Caterina Barrila MD Cinzia Valeria Russo MD Sabrina Esposito MD

Domenico Ippolito MD Doriana Landi MD PhD Andrea Visconti MD and Maria Pia Sormani PhD

Neurol Neuroimmunol Neuroinflamm 20207e878 doi101212NXI0000000000000878

Correspondence

Dr Sormani

mariapiasormaniunigeit

AbstractObjectiveCladribine tablets were tested against placebo in randomized controlled trials (RCTs) In thisstudy the effectiveness of cladribine vs other approved drugs in patients with relapsing-remitting MS (RRMS) was compared by matching RCT to observational data

MethodsData from the pivotal trial assessing cladribine tablets vs placebo (CLARITY) were propensityscore matched to data from the Italian multicenter database i-MuST This database included3150 patients diagnosed between 2010 and 2018 at 24 ItalianMS centers who started a disease-modifying drug The annualized relapse rate (ARR) over 2 years from treatment start and the24-week confirmed disability progression were compared between patients treated with cla-dribine and other approved drugs (interferon glatiramer acetate fingolimod natalizumab anddimethyl fumarate) with comparisons with placebo as a reference Treatment effects wereestimated by the inverse probability weighting negative binomial regression model for ARR andCox model for disability progression The treatment effect has also been evaluated according tobaseline disease activity

ResultsAll weighted baseline characteristics were well balanced between groups All drugs tested had aneffect vs placebo close to that detected in the RCT Patients treated with cladribine had asignificantly lower ARR compared with interferon (relapse ratio [RR] = 048 p lt 0001)glatiramer acetate (RR = 049 p lt 0001) and dimethyl fumarate (RR = 06 p = 0001) asimilar ARR to that with fingolimod (RR = 074 p = 024) and a significantly higher ARR than

MORE ONLINE

Class of EvidenceCriteria for ratingtherapeutic and diagnosticstudies

NPuborgcoe

From the Department of Health Sciences (A Signori MPS) Section of Biostatistics University of Genoa Department of Neurosciences (FS RL CVR) Reproductive Sciences andOdontostomatology Multiple Sclerosis Center Federico II University Naples Neurological Clinic and Multiple Sclerosis Center of ldquoAORN ACardarellirdquo (GTM) Naples Centro diSclerosi Multipla (ES) II Clinica Neurologica Universita della Campania ldquoLuigi VanvitellirdquoNapoli 2nd Neurology Unit and CRRSM (Regional Referral Multiple Sclerosis Center) (AMR)Careggi University Hospital University of FlorenceMultiple Sclerosis Study Center (PA DB) ASST ValleOlona POdi Gallarate (VA) Clinical and Biological Sciences Department (MC)Neurology Unit University of Torino San Luigi Gonzaga Hospital Orbassano Centro SM (EB) Dipartimento di Neuroscienze Ospedale Universitario Citta della Salute e della Scienzadi Torino Neurological Clinic (RC) Department of Experimental and Clinical Medicine Marche Polytechnic University Ancona Policlinic Tor Vergata (GM DL) Rome The MultipleSclerosis Center of the Veneto Region (PP) Department of Neurosciences University of Padua Department of Medical Surgical Neurological Metabolic and Aging Sciences (SBonavita LL SE DI) University of Campania Luigi Vanvitelli Naples Department of Clinical and Experimental Medicine (IRZ) University of Sassari Department NeuroscienceRehabilitation Ophthalmology Genetics Maternal and Child Health (AL) Center of Excellence for Biomedical Research (CEBR) University of Genova Neuroimmunology andNeuromuscular Diseases Unit (LP-G VTC) IRCCS Foundation Carlo Besta Neurological Institute Milan Centro Sclerosi Multipla ASST Papa Giovanni XXIII di Bergamo (SLG BFVB) Department of Medical Science and Public Health (JF EC GF) University of Cagliari Neurology Clinic (A Sartori) Department of Medical Surgical and Health SciencesUniversity of Trieste Multiple Sclerosis Center (SR CC) ASST Spedali Civili PO di Montichiari (BS) Department of Medicine Surgery and Neuroscience (MLS) University of Siena2nd Neurology Unit and CReSM (Regional Referral Multiple Sclerosis Center) (ADS) AOU San Luigi Gonzaga Orbassano Torino Regina Montis Regalis Hospital (ADS) MondovıDepartment of Neurology and Psychiatry (SP) Sapienza University Rome Neurologia Universitaria OORR (RG) Foggia Institute of Neurology (S Barone) University Magna Graeciaof Catanzaro Department of Neurology (CB) Valduce Hospital Como Merck Serono SpA (AV) Rome and IRCCS Ospedale Policlinico San Martino (AL MPS) Genova Italy

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

The Article Processing Charge was funded by Merck

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

natalizumab (RR = 213 p = 0014) confirming results obtained by indirect treatment comparisons from RCTs (network meta-analyses) The relative effect of cladribine tablets 10 mg (cumulative dose 35 mgkg over 2 years) was higher in patients withhigh disease activity vs all treatments except fingolimod and natalizumab Effects on disability progression were largelynonsignificant probably due to lack of power for such analysis

ConclusionIn patients with RRMS cladribine tablets showed lower ARR compared with matched patients who started interferonglatiramer acetate or dimethyl fumarate was similar to fingolimod and was higher than natalizumab The beneficial effect ofcladribine tablets was generally amplified in the subgroup of patients with high disease activity

Classification of evidenceThis study provides Class III evidence that for patients with RRMS cladribine-treated patients had lower ARR compared withinterferon glatiramer acetate or dimethyl fumarate similar ARR compared with fingolimod and higher ARR compared withnatalizumab

Relapsing-remitting MS (RRMS) is characterized by periodicexacerbations of disease symptoms followed by periods ofremission in which symptoms are either partially or com-pletely absent Disease-modifying drugs (DMDs) are able todecrease the frequency and duration of relapses and some-times to slow the progression of the disease1

Cladribine tablet is hypothesized as an immune reconstitutiontherapy that targets specific subsets of the adaptive immunesystem2 As demonstrated in the CLARITY phase 3 ran-domized controlled study a short-course treatment (8ndash20 dy)with cladribine tablets provided a sustained treatment benefitfor patients with RRMS3 In particular the 2 tested doses ofcladribine (35 and 525 mgkg [35 mgkg is the approveddose]) were superior to placebo in suppressing relapse activityand increasing the probability of remaining relapse-free In theCLARITY extension the clinical benefits of 2 years of treat-ment with cladribine tablets were shown to be sustainedwithout the need for retreatment4

In this study we conducted a propensity score (PS)-matchedanalysis of observational data from the CLARITY trial data setand the Italian multicenter database i-MuST which includes3150 patients diagnosed between 2010 and 2018 at 24 ItalianMS centers who started a DMD

We aim to fill in the gap in the current RRMS treatmentlandscape as no direct head-to-head comparisons of cladribinewith other common immunotherapies have yet been con-ducted In particular we try to define the role of cladribinetablets in the context of the other available therapies for MS byan individual-level comparative efficacy analysis of cladribine

tablets vs other common immunotherapies such as interferonbeta (IFN β-1a and β-1b) glatiramer acetate (GA) fingolimod(FTY) natalizumab (NTZ) dimethyl fumarate (DMF) andteriflunomide (TERI) as first-line therapies for RRMS in Italy

MethodsPatientsThis is an observational retrospective study merging 2 data setsDetails on the study design of the CLARITY data set can befound in the article reporting the clinical trial results3 and detailsabout the i-MuST database collection can be found in previouspublications5 This study takes advantage of the availability ofthe placebo arm of the CLARITY study thus providing a ref-erence to which each treatment can be compared

Briefly in the CLARITY study 1326 patients were randomlyassigned in a 111 ratio to receive 1 of 2 doses of cladribinetablets (either 35 mg or 525 mgkg of body weight) or pla-cebo Patients randomized to the 35 mgkg cumulative dosereceived a first treatment course at weeks 1 and 5 of the firstyear and a second treatment course at weeks 1 and 5 of thesecond year Patients were eligible if they had received a di-agnosis of RRMS (according to the McDonald criteria 20016)had lesions consistent with MS on MRI had had at least 1relapse within 12 months before study entry and had a score ofle55 on the Expanded Disability Status Scale (EDSS)

The i-MuST data set is a multicenter retrospective databaseinvolving 24 Italian MS centers The data set used for thisanalysis included 3006 patients with RRMS From these

GlossaryARR = annualized relapse rateDMD = disease-modifying drugDMF = dimethyl fumarate EDSS = ExpandedDisability StatusScale FS = functional system FTY = fingolimod GA = glatiramer acetate HAD = high disease activity HR = hazard ratioIFN = interferon IPW = inverse probability weighting NMA = network meta-analysis NTZ = natalizumab PS = propensityscore RRMS = relapsing-remitting MS SC = subcutaneous TERI = teriflunomide

2 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 6 | November 2020 NeurologyorgNN

patients satisfying the same inclusionexclusion criteria asCLARITY were extracted Inclusion criteria were age over 16years diagnosis of RRMS (2001 International Panel Di-agnostic Criteria and the 2010 revision67) and initiating aDMD between January 2010 and June 2017 There were noexclusion criteria The EDSS score8 was collected for all pa-tients at therapy start at therapy switch (if any) every 6ndash12months depending on the center and at the last follow-up

Only treatment-naive patients were selected in the CLARITYdata set because all the patients enrolled in i-MuST were attheir first therapy As a consequence the study population wasdefined as all enrolled patients in CLARITY who did notreceive any prior DMD and all the patients enrolled ini-MuST who satisfied the inclusion criteria of CLARITY

The following variables were common to the 2 data sets andwere merged age at treatment start disease duration sinceonset baseline EDSS score presence of gadolinium-enhancing(active) lesions at baseline sex relapses in the previous year firsttherapy and date of first therapy start First relapse time to firstrelapse time to a progression event confirmed at 3 months andtime to a progression event confirmed at 6 months were alsoextracted fromCLARITY and calculated in the i-MuST data set

The study population was split into the following treatmentgroups placebo cladribine 35 mgkg cladribine 525 mgkgIFN β-1a (subcutaneous [SC]) IFN-1a (SC) IFN β-1b GAFTY NTZ DMF and TERI For treatment comparisons theIFNs were merged in a single IFN group and the 2 cladribinetablets arms in a single cladribine-treated group Thereforeunless differently specified the term IFN indicates the 3 IFNspooled together and the term cladribine tablets indicates the2 cladribine arms pooled together

Standard protocol approvals registrationsand patient consentsThe study was approved by the Ligurian Region ethical com-mittee (258REG2016) All the centers involved in the study askedfor written permission of the use of anonymized personal clinicaldata for research purposes and written informed consent wasobtained from all study patients included in this study CLARITYregistrationmdashClinicalTrialsgov number NCT00213135

OutcomesThe primary end point for the Class III comparative effec-tiveness analysis was annualized relapse rate (ARR) over 2years The secondary end point was time to onset of 24-weeksustained disability progression over 2 years Patients ini-MuST were censored at their treatment switch or at year 2whichever happened first

Definition of relapses in i-MuST followed clinical trial criteriachange in the EDSS score with an increase of ge05 points onthe total score or an increase of 1 point on 2 functional systems(FSs) or 2 points on 1 FS excluding changes involving bowelbladder or cerebral FS In CLARITY the definition of relapses

closest to those defined in i-MuST was identified to be that ofldquoqualifying relapsesrdquo and this variable was used for the analysis

The ARR for each treatment group was calculated as the totalnumber of relapses experienced in the group divided by thetotal number of patient-years on study

The time to onset of 24-week confirmed disability pro-gression is defined as the time from baseline to the firstdisability progression that is confirmed at the next visit ge24weeks after the initial disability progression Disability pro-gression was defined by 1 of the following an EDSS scoreincrease of ge15 points from a baseline EDSS score of 0 thatis sustained for ge12 weeks and an EDSS score increase ofge10 point from a baseline EDSS score of 10ndash55 (inclusive)for ge24 weeks

The date of the initial EDSS assessment at which the mini-mum increase in the EDSS score is met was the date of onsetof the progression The progression was defined as confirmedwhen this minimum EDSS change was present on the nextstudy visit occurring after 24 weeks or longer from the onsetof the progression EDSS assessments le30 days after aprotocol-defined relapse were not used for confirmation ofdisability progression If a patient met the defined criteria ofconfirmed progression and was also having a relapse thepatient was required to meet the defined minimum criteria atthe subsequent visit

Patients without a confirmed progression based on the aboverules were censored The censor date was the date of last EDSSassessment in the study Patients who withdrew from the studyafter the baseline visit but before the first clinical evaluationscheduled visit were censored at baseline

Statistical methodsBaseline was defined for the i-MuST database as the start oftherapy date and for the CLARITY database as the randomi-zation date Baseline disease and demographic characteristicswere summarized by group and overall before any adjustmentusing descriptive statistics and were compared between treat-ment groups using the standardized mean difference as calcu-lated according toCohen d effect size ACohen d effect size gt01denotes meaningful imbalance in the baseline covariates9ndash14

Tomitigate the baseline differences between compared groupsthe inverse probability weighting (IPW) approach based on PSwas used The weights correspond to the inverse of the con-ditional PS of receiving the treatment The PSwas calculated bymodeling on the baseline characteristics the probability of re-ceiving each treatment vs another one by a logistic regressionmodel applied to each treatment group in the i-MuST study vsthe overall CLARITY study population (since the assignmentto cladribine tablets or placebo was randomized) The logisticregression model had treatment group as the dependent vari-able (placebocladribine tablets vs other drug) and the fol-lowing baseline variables as independent covariates age

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

disease duration since onset baseline EDSS score sex andrelapses in the previous year

We used stabilized trimmed weights (weights that exceeded aspecified threshold are each set to that threshold15) to mitigatethe impact of extremely higher or lower weights on the vari-ability of the estimated treatment effect15 The threshold wasbased on the quantiles of the distribution of the weights (1st to5th and 95th to 99th percentiles)1516 For each treatment pairwe choose to trim the weights at the percentile giving thehighest reduction of standardized Cohen d differences

An IPW negative binomial regression model was used toassess differences between treatments on relapse rate over 2years The relapse count was used as a dependent variablethe treatment group as an independent variable and the logof follow-up duration as an offset variable

The semiparametric IPW Cox regression model was usedto assess treatment effect differences on time to disabilityprogression Cumulative probability to be progression-free during follow-up was calculated and graphicallydisplayed by mean of Kaplan-Meier survival curves Theestimates of treatment effect were expressed as hazardratios (HRs) and reported with 95 CIs and p values Stata(v14 StataCorp College Station TX) was used for thecomputation

Subgroup analysesA subgroup analysis assessed the relative treatment effect ofeach drug vs cladribine tablets in subgroups of patients de-fined according baseline disease activity The definition ofpatients with high disease activity (HDA) at baseline was theone used in a previous article4 as patients with ge2 relapsesduring the year before study entry whether on DMD Thedefinition including patients with previous relapses while ontreatment was not applied here because only naive patientswere included in the analyzed cohorts

All the analyses run on the overall population were rerun onsubgroups of patients with and without HDA at baseline Thesignificance of the difference in treatment effects betweensubgroups was assesses by interaction tests The results weredisplayed by forest plots

Sensitivity analysesThe following sensitivity analyses were run (1) analyses with aPS and IPW calculated including baseline MRI informationrun on a reduced sample (2) analyses including patientstreated with cladribine tablets 35 mgkg only (3) analysesadjusted by a 11 PS matching and a variable ratio le31 PSmatching and (4) analysis contrasting cladribine tablets vs eachIFN separately

Data availabilityCLARITY and i-MuST data would be available on reasonablerequest

ResultsDescriptive analysisA total of 2204 patients from i-MuST and 945 patients fromCLARITY were included in the final analysis Supplementaryfigure e-1 (linkslwwcomNXIA298) presents the flowchartfor selecting patients from the i-MuST andCLARITY data sets

In the i-MuST data set a total of 1168 patients were treatedwith IFN 402 with GA 113 with FTY 149 with NTZ 295with DMF and 77 with TERI In the CLARITY data set 305were in the placebo arm 322 received the cladribine 35 mgkg dose and 318 received the cladribine 525 mgkg dose

The unweighted characteristics of all patients included in theanalysis according to the treatment group in the i-MuST da-tabase and in CLARITY are presented in table e-1 (linkslwwcomNXIA298) In CLARITY patients were randomizedamong treatment arms therefore data are presented here forthe overall CLARITY study according to the IPW approachused The baseline characteristics detailed for each treatmentarm are reported in table e-2

Patients included in i-MuSTwere generally younger (excludingpatients treated with GA who were in the same age range andpatients treated with TERI who were older) than patientsincluded in CLARITY EDSS levels were generally well bal-anced (small standardized differences in range of 020ndash037)between CLARITY patients and patients treated with FTYNTZ and TERI whereas patients treated with IFN GA andDMF had lower EDSS baseline scores Disease duration wasgenerally lower in i-MuST compared with CLARITY patientswith mild to moderate differences Patients enrolled in i-MuST(especially FTY NTZ and DMF) showed a higher frequencyof active lesions at baseline compared with trial data Follow-upduration was lower in i-MuST for patients treated with FTYDMF and TERI The mean time between visits with the EDSSwas 157 days with amedian of 109 days Baseline characteristicsfor each treatment and for CLARITY study after IPW arepresented in table 1 The 3 arms of CLARITY are shownseparately and compared with each treatment in tables e-3 toe-8 (linkslwwcomNXIA298)

The weighted characteristics were well balanced between eachtreatment group from the i-MuST and CLARITY data sets Aresidual unbalance persisted for the EDSS in the IFN and GAarms (with standardized difference in range of 025ndash034)

Comparative effectiveness analyses

Annualized relapse rateThe IPW-adjusted ARR for each treatment arm is reported intable 2 Treatment with cladribine tablets was associated witha statistically significant reduction in ARR of 52 vs IFN 51vs GA and 40 vs DMF The reduction of ARR vs FTY wasnot statistically significant (p = 024) NTZ was superior tocladribine tablets and the ARR ratio of cladribine vs NTZ was

4 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 6 | November 2020 NeurologyorgNN

Table 1 Inverse probability-weighted demographic and clinical characteristics

Panel A IFN GA and FTY vs CLARITY study

CLARITY(n = 945)

IFN (n =1168)

Standardized meandifference IFN vs CLARITY

CLARITY(n = 945) GA (n = 402)

Standardized meandifference GA vs CLARITY

CLARITY(n = 945) FTY (n = 113)

Standardized meandifference FTY vs CLARITY

Age y mean (SD) 373 (101) 366 (107) 0071 388 (102) 395 (118) 007 384 (103) 381 (116) 0025

Females n () 629 (665) 768 (657) 0018 629 (665) 269 (669) 0008 614 (65) 69 (611) 008

EDSS score mean(SD) median (IQR)

246 (125) 2(15ndash35)

211 (117) 2(1ndash3)

029 271 (128)25 (2ndash35)

228 (132) 2(1ndash3)

033 286 (128)25 (2ndash4)

281 (123)25 (2ndash35)

004

Disease durationmean(SD) median (IQR)

246 (445)052 (02ndash26)

275 (468)07 (02ndash29)

0062 338 (505)12 (03ndash46)

310 (557)100 (017ndash3)

0052 408 (536)195 (05ndash58)

390 (611)179 (03ndash56)

003

ARR in previousyearmean (SD)

132 (056) 133 (058) 0013 133 (058) 132 (056) 0024 135 (06) 136 (057) 001

Active lesions n () 306 (324) 4351023(425)

021 290 (307) 146376(387)

017 294 (311) 4890 (533) 046

Panel B NTZ DMF and TERI vs CLARITY study

CLARITY(n = 945) NTZ (n = 149)

Standardized meandifference NTZ vsCLARITY

CLARITY(n = 945)

DMF(n = 295)

Standardized meandifference DMF vsCLARITY

CLARITY(n = 945) TERI (n = 77)

Standardized meandifference TERI vsCLARITY

Age y mean (SD) 379 (104) 366 (109) 012 383 (103) 381 (102) 0015 391 (104) 393 (118) 0019

Females n () 609 (644) 84 (564) 017 617 (653) 197 (668) 003 614 (65) 45 (584) 013

EDSS score mean (SD)median (IQR)

287 (128) 3(2ndash4)

274 (110) 25(2ndash35)

010 268 (130)25 (15ndash35)

238 (123)25 (15ndash3)

023 287 (128) 3(2ndash4)

256 (134) 2(15ndash35)

024

Disease durationmean (SD) median(IQR)

387 (519)18 (05ndash53)

328 (510)095(017ndash458)

011 384 (512)179(051ndash55)

359 (561)076 (03ndash44)

0045 423 (550)206(057ndash597)

376 (588)105(054ndash398)

008

ARR in previous yearmean (SD)

136 (061) 142 (066) 0089 132 (057) 129 (053) 0068 133 (058) 132 (057) 0017

Active lesions n () 299 (317) 89128 (695) 082 293 (31) 134276(486)

036 287 (304) 1972 (264) 0088

Abbreviations ARR = annualized relapse rate DMF = dimethyl fumarate EDSS = Expanded Disability Status Scale FTY = fingolimod GA = glatiramer acetate HDA high disease activity IFN = interferon β IQR = interquartilerange NTZ = natalizumab TERI = teriflunomide

Neurolo

gyorgN

NNeu

rologyN

euroim

munology

ampNeuroinflam

mation

|Volum

e7N

umber

6|

Novem

ber2020

5

Table 2 ARR after IPW adjustment in all the analyzed treatment groups

IPW clinical variables IPW clinical + MRI variables

ARR (95 CI)

ARR ratio (95 CI)

ARR (95 CI)

ARR ratio (95 CI)

IFN vs placebo Cladribine vs IFN IFN vs placebo Cladribine vs IFN

Placebo 0321 (0275ndash0367) 076 (064ndash091) p = 0002 048 (041ndash057) p lt 0001 0319 (0273ndash0365) 079 (066ndash095) p = 0011 046 (038ndash055) p lt 0001

Cladribine 0118 (0101ndash0134) 0116 (0099ndash0132)

IFN 0246 (0222ndash0269) 0252 (0226ndash0279)

GA vs placebo Cladribine vs GA GA vs placebo Cladribine vs GA

Placebo 0318 (0254ndash0382) 078 (057ndash108) p = 013 049 (036ndash068) p lt 0001 0318 (0254ndash0383) 079 (057ndash109) p = 014 049 (036ndash068) p lt 0001

Cladribine 0123 (0099ndash0147) 0122 (0098ndash0147)

GA 0249 (0187ndash0311) 0250 (0187ndash0313)

FTY vs placebo Cladribine vs FTY FTY vs placebo Cladribine vs FTY

Placebo 0320 (0257ndash0383) 054 (032ndash089) p = 0016 074 (045ndash122) p = 024 0321 (0257ndash0385) 047 (028ndash079) p = 0005 084 (050ndash141) p = 051

Cladribine 0127 (0103ndash0150) 0127 (0103ndash0150)

FTY 0171 (0091ndash0252) 0151 (0078ndash0223)

NTZ vs placebo Cladribine vs NTZ NTZ vs placebo Cladribine vs NTZ

Placebo 0326 (0261ndash0391) 018 (010ndash034) p lt 0001 213 (117ndash388) p = 0014 0329 (0265ndash0394) 017 (009ndash032) p lt 0001 228 (123ndash424) p = 0009

Cladribine 0128 (0104ndash0152) 0128 (0104ndash0152)

NTZ 0060 (0026ndash0094) 0056 (0023ndash0089)

DMF vs placebo Cladribine vs DMF DMF vs placebo Cladribine vs DMF

Placebo 0312 (0246ndash0377) 065 (044ndash097) p = 0036 060 (041ndash089) p = 0001 0322 (0256ndash0387) 064 (043ndash094) p = 0024 063 (043ndash093) p = 002

Cladribine 0123 (0099ndash0147) 0129 (0104ndash0154)

DMF 0204 (0135ndash0272) 0204 (0136ndash0273)

Abbreviations ARR = annualized relapse rate DMF = dimethyl fumarate FTY = fingolimod GA = glatiramer acetate IFN = interferon β IPW = inverse probability weighting NTZ = natalizumabResults are presented after IPW adjustment with clinical variables (left columns) and with clinical variable + MRI variables (right columns)

6NeurologyN

euroim

muno

logyampNeuroinflam

mation

|Vo

lume7N

umber

6|

Novem

ber

2020NeurologyorgN

N

213 (p = 0014) (table 2) Cladribine tablets compared withTERI was not analyzed because the TERI arm was too small(n = 77) to drawmeaningful conclusions The treatment effectsvs placebo were close to those observed in randomized con-trolled trials and estimated in previous network meta-analyses(table e-9 and figure e-2 linkslwwcomNXIA298)17ndash19 withlargely overlapping CIs for all the drugs In this study NTZshowed a reduction in the ARR vs placebo of 82 (with verylarge CIs)mdashhigher than that observed in the AFFIRM study(ARR ratio vs placebo = 68) and reestimated in the networkmeta-analyses but with overlapping CIs Similar results wereobtained for the analysis including baseline MRI activity in thePS calculation on the subgroup of patients with available MRIinformation at baseline (table 2 second column) Forest plotsfor these 2 analyses are presented in figure 1 Similar resultswere obtained when considering the 2 different doses of cla-dribine tablets separately (table e-10)

Our results of the effects of cladribine tablets vs other drugs onARR were compared with those reported by a previous net-work meta-analysis (NMA) (figure e-3 linkslwwcomNXIA298)20 All the cladribine effects vs the other DMDs werewithin the CIs of those estimated in the NMA20

Disability progressionTheHRs for comparisons of disability progression are reportedin table 3 and displayed in figure e-4 (linkslwwcomNXIA298) A small proportion of patients had missing data onfollow-up EDSS Excluded patients in each group were 40(342) for IFN 23 (572) for GA 5 (44) for FTY 15(10) for NTZ and 33 (112) for DMF A significant dif-ference in favor of cladribine tablets on time to disability pro-gression was observed only for GA (minus36 p = 0045)

The survival curves for progression-free probability of eachtreatment vs cladribine tablets 35 mgkg are shown in figure

e-5 (linkslwwcomNXIA298) whereas the HRs for com-parisons of disability progression considering the 2 differentdoses of cladribine tablets separately are reported in table e-11

Figure 1 ARR ratio of cladribine vs other treatments

ARR = annualized relapse rate DMF = dimethyl fumarate FTY = fingolimod GA = glatiramer acetate IFN = interferon β NTZ = natalizumab

Table 3 Risk of 24-week disability progression ofcladribine vs other drugs

IPW clinicalvariables

IPW clinical + MRIvariables

HR (95 CI) HR (95 CI)

IFN (n = 1128) vsplacebo

072 (055ndash094)p = 0017

075 (057ndash099)p = 0046

Cladribine vs IFN 086 (068ndash109)p = 022

081 (062ndash105)p = 011

GA (n = 379) vsplacebo

093 (058ndash147)p = 074

100 (063ndash159)p = 099

Cladribine vs GA 064 (041ndash099)p = 0045

058 (038ndash090)p = 0015

FTY (n = 108) vsplacebo

061 (030ndash127)p = 019

065 (032ndash130)p = 023

Cladribine vs FTY 089 (044ndash183)p = 076

084 (042ndash167)p = 062

NTZ (n = 134) vsplacebo

045 (022ndash092)p = 0028

035 (016ndash078)p = 001

Cladribine vs NTZ 120 (059ndash242)p = 061

151 (069ndash333)p = 031

DMF (n = 262) vsplacebo

054 (029ndash100)p = 0051

068 (038ndash119)p = 018

Cladribine vs DMF 107 (059ndash194)p = 083

080 (046ndash139)p = 043

Abbreviations DMF = dimethyl fumarate FTY = fingolimod GA = glatirameracetate HR = hazard ratio IFN = interferon β IPW = inverse probabilityweighting NTZ = natalizumab

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 6 | November 2020 7

Subgroup analysesThe effect of cladribine tablets vs other drugs was tested insubgroups defined according to baseline disease activityThe weighted baseline characteristics for each treatmentby subgroup (HDA vs non-HDA) are presented in tablese-12 and e-13 (linkslwwcomNXIA298) The samecharacteristics are reported in tables e-14 to e-18 with the3 arms of CLARITY presented separately Results on ARRare shown in figure 2 The effect of cladribine tablets wasgenerally amplified in the HDA subgroup and the samewas true for cladribine 35 mg alone (table e-19) Thesubgroup analysis run on time to disability progression ispresented in tables e-20 and e-21 The same trend for ahigher efficacy of cladribine tablets is observable in theHDA subgroup

Sensitivity analyses

PS matching 11Baseline characteristics of patients after 11 PS matching arereported in table e-22 (linkslwwcomNXIA298) A goodbalance with standardized differences lt010 for almost allcharacteristics and treatments was observed The ARR ratiobetween cladribine tablets and single treatment arms after 11PS matching (table e-23) confirmed in general the resultsobserved with the IPW approach

PS matching 31The characteristics of matched samples after up to 31 PSmatching are reported in table e-24 (linkslwwcomNXIA298) Here also a good balance was observed between the

Figure 2 Relapse rate ratio of cladribine vs each other treatment according to HDA subgroups

ARR = annualized relapse rate CLAD = cladribine DMD = disease-modifying drug DMF = dimethyl fumarate FTY = fingolimod GA = glatiramer acetate HDAhigh disease activity IFN = interferon β NTZ = natalizumab RR = relapse ratio

8 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 6 | November 2020 NeurologyorgNN

compared groups The ARR ratio between cladribine tabletsand single treatment arms after 31 PS matching (table e-25)confirmed the results observed with the IPW approach

Stratification of IFN armsThe IPW-weighted baseline characteristics of the 3 IFN armssplit are reported in table e-26 (linkslwwcomNXIA298)Good balance with standardized differences lt010 was ob-served for almost all characteristics across the 3 arms comparedwith CLARITY Theweighted ARR ratio (table e-27 linkslwwcomNXIA298) showed a higher effect of cladribine tablets inIFN-1b and IFN-1a SC compared with IFN-1a IM

DiscussionThis study presents a novel approach matching individual pa-tient data from the CLARITY randomized controlled phase IIItrial with those included in a large multicenter observationalstudy in newly diagnosed patients The presence of the placeboarm from CLARITY allowed us to compare as a first step theeffectiveness of each drug from the observational study vs pla-cebo with the effect reported in randomized clinical trials Theeffect of each treatment vs placebo was close to that observed inthe clinical trials except for NTZ where a larger effect in thedrug vs placebo was seen (ARR ratio 82 vs 68 reported inthe AFFIRM trial21) even if the large CIs of the estimateobtained in this study limit the interpretation of such result

The results from comparative effectiveness showed a superi-ority of cladribine tablets vs IFN GA and DMF on the relapserate in the first 2 years No significant differences were ob-served with FTY whereas NTZ was shown to be superior

Similar results were observed in a previous observational study22

on comparative effectiveness of cladribine conducted in a cohortof patients from the MSBase International Registry A smallcohort of patients treated with cladribine vs FTY NTZ and IFNshowed a superiority of cladribine vs IFN and a superiority ofNTZ on cladribine on hazard of first relapse On the sameoutcome no significant differences between cladribine and FTYwere detected No comparisons vs GA or DMFwere performed

A recent NMA20 compared the effects of various DMDs onARR and confirmed disability progression in active and highlyactive RRMS including 41 studies Cladribine was ranked infourth position on a scale of efficacy for ARR reduction afteralemtuzumab NTZ and ocrelizumab In the same NMA asignificant reduction of ARR with cladribine estimated around40 vs GA and up to 48 vs IFN was reported and was closeto that observed in our study When compared with DMF theARR reduction due to cladribine was not significant and lowerthan that observed in our study (22 95 CI 43ndash7) inNMA vs 36 in our study our data are therefore consistentwith those reported in NMA) The ARR under cladribine wascomparable to FTY and higher than the ARR under NTZ evenif the difference was not significant

In the NMA20 cladribine was statistically similar to the otherDMDs on confirmed disability progression In our study thesuperiority of cladribine was detected only vs GA

Cladribine is approved and indicated mainly for the treatmentof patients with highly active RRMS Having individual pa-tient data we were able to perform a subgroup analysis whichindicated a higher efficacy of cladribine vs other DMDs inpatients with HDA with ARR reduction close to 70 vs IFNGA and DMF in this specific subgroup of patients

Study limitationsThis study carries all the well-known limitations of non-randomized comparisons with the advantage of a placebo armgiving a benchmark for comparisons23 In addition the samplesize for all the comparisons on disability progression was lowgiving estimates with large CIs To mitigate the potential biasdue to the lack of randomization we applied an IPW in themain analysis as well as several sensitivity analyses with differentmatching algorithms

With our comparative results based on a wide national data-base (i-MuST) and the CLARITY data set we simulated ascenario representing the complexity of management of pa-tients with MS which confirms the effectiveness of cladribinetablets in treating RRMS Our approach let us overcome thelack of direct head-to-head comparisons of cladribine tablets toother common DMDs We confirmed the superiority of cla-dribine tablets in reducing relapse rate during the first 2 yearscompared with its major competitors such as IFN GA andDMF NTZ data were still superior as given in the resultsFinally having individual patient data we were able to showthe higher efficacy of cladribine in the subgroup of patientswith HDA

Study fundingThis study was sponsored by Merck Serono SpA RomeItaly an affiliate of Merck KGaA Darmstadt Germany

DisclosureA Signori F Sacca R Lanzillo GT Maniscalco E Signo-riello A M Repice P Annovazzi D Baroncini M Clerico EBinello R Cerqua G Mataluni P Perini S Bonavita LLavorgna I R Zarbo A Laroni L Pareja-Gutierrez S LaGioia B Frigeni V Barcella J Frau E Cocco G Fenu V TClerici A Sartori S Rasia C Cordioli M L Stromillo A DiSapio S Pontecorvo R Grasso S Barone C Barrila C VRusso S Esposito D Ippolito and D Landi have nothing todisclose A Visconti is an employee at Merck Serono Italy MP Sormani received consulting fees from Merck KGaANovartis Biogen Roche Sanofi Teva Celgene MedDayGeNeuro and Immunic Go to NeurologyorgNN for fulldisclosures

Publication historyReceived by Neurology Neuroimmunology amp NeuroinflammationMay 27 2020 Accepted in final form July 16 2020

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 6 | November 2020 9

References1 Thompson AJ Baranzini SE Geurts J et al Multiple sclerosis Lancet 2018391

1622ndash16362 Baker D Pryce G Herrod SS Schmierer K Potential mechanisms of action related to

the efficacy and safety of cladribine Mult Scler Relat Disord 201930176ndash1863 Giovannoni G Comi G Cook S et al CLARITY Study Group A placebo-controlled

trial of oral cladribine for relapsing multiple sclerosis N Engl J Med 2010362416ndash426

4 Giovannoni G Soelberg Sorensen P Cook S et al Safety and efficacy of cladribinetablets in patients with relapsing-remitting multiple sclerosis results from the ran-domized extension trial of the CLARITY study Mult Scler 2018241594ndash1604

5 Sacca F Lanzillo R Signori A et al Determinants of therapy switch in multiplesclerosis treatment-naive patients a real-life study Mult Scler 2019251263ndash1272

Appendix Authors

Name Location Contribution

Alessio SignoriPhD

University of Genoa Designed andconceptualizedthe study analyzed thedata and drafted themanuscript forintellectual content

Francesco SaccaMD

University Federico IINaples

Major role in theacquisition of data

Roberta LanzilloMD PhD

University Federico IINaples

Major role in theacquisition of data

Giorgia TeresaManiscalco MD

AORN CardarelliNaples

Major role in theacquisition of data

ElisabettaSignoriello MDPhD

University LuigiVanvitelli Naples

Major role in theacquisition of data

Anna MariaRepice MD

Careggi UniversityHospital Florence

Major role in theacquisition of data

Pietro AnnovazziMDDamianoBaroncini MD

ASST Valle Olona POGallarate

Major role in theacquisition of data

Marinella ClericoMD PhD

University of Turin Major role in theacquisition of data

Eleonora BinelloMD

Turin Major role in theacquisition of data

Raffaella CerquaMD

Marche PolytechnicUniversity Ancona

Major role in theacquisition of data

Giorgia MataluniMD PhD

Policlinic Tor VergataRome

Major role in theacquisition of data

Paola Perini MDPhD

University of Padua Major role in theacquisition of data

Simona BonavitaMD

University LuigiVanvitelli

Major role in theacquisition of data

Luigi LavorgnaMD PhD

University LuigiVanvitelli

Major role in theacquisition of data

Ignazio RobertoZarbo PhD

University of Sassari Major role in theacquisition of data

Alice Laroni MDPhD

University of Genoa Major role in theacquisition of data

Lorena Pareja-Gutierrez MD

IRCCS Besta Milan Major role in theacquisition of data

Sara La Gioia MD Centro SM PapaGiovanni XXIIIBergamo

Major role in theacquisition of data

Barbara FrigeniMD

Centro SM PapaGiovanni XXIIIBergamo

Major role in theacquisition of data

Valeria BarcellaMD

Centro SM PapaGiovanni XXIIIBergamo

Major role in theacquisition of data

Jessica Frau MD University of Cagliari Major role in theacquisition of data

Eleonora CoccoMD

University of Cagliari Major role in theacquisition of data

Appendix (continued)

Name Location Contribution

Giuseppe FenuMD PhD

University of Cagliari Major role in theacquisition of data

Valentina TorriClerici MD

IRCCS Besta Milan Major role in theacquisition of data

Arianna SartoriMD

University of Trieste Major role in theacquisition of data

Sarah Rasia MD Spedali Civili Brescia Major role in theacquisition of data

Cinzia CordioliMD

Spedali Civili Brescia Major role in theacquisition of data

Maria LauraStromillo MDPhD

University of Siena Major role in theacquisition of data

Alessia Di SapioMD

Regina MontisHospital Mondovı

Major role in theacquisition of data

SimonaPontecorvo MDPhD

Sapienza UniversityRome

Major role in theacquisition of data

Roberta GrassoMD

NeurologiaUniversitaria Foggia

Major role in theacquisition of data

Stefania BaroneMD

University ofCatanzaro

Major role in theacquisition of data

Caterina BarrilaMD

Valduce HospitalComo

Major role in theacquisition of data

Cinzia ValeriaRusso MD

University Federico IINaples

Major role in theacquisition of data

Sabrina EspositoMD

University LuigiVanvitelli Naples

Major role in theacquisition of data

DomenicoIppolito MD

University LuigiVanvitelli Naples

Major role in theacquisition of data

Doriana LandiMD PhD

Policlinic Tor VergataRome

Major role in theacquisition of data

Andrea ViscontiMD

Merck Serono SpARome

Drafted the manuscriptfor intellectual content

Maria PiaSormani PhD

University of Genoa Designed andconceptualizedthe study analyzed thedata drafted themanuscriptfor intellectual contentand supervised thestudy

10 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 6 | November 2020 NeurologyorgNN

6 McDonald WI Compston A Edan G et al Recommended diagnostic criteria formultiple sclerosis guidelines from the International Panel on the diagnosis of multiplesclerosis Ann Neurol 200150121ndash127

7 Polman CH Reingold SC Banwell B et al Diagnostic criteria for multiple sclerosis2010 revisions to the McDonald criteria Ann Neurol 201169292ndash302

8 Kurtzke JF Rating neurologic impairment in multiple sclerosis an Expanded Dis-ability Status Scale (EDSS) Neurology 1983331444ndash1452

9 Cohen J Statistical Power Analysis for the Behavioral Sciences 2nd ed Hillsdale NJLawrence Erlbaum Associates Publishers 1988

10 Normand ST Landrum MB Guadagnoli E et al Validating recommendations forcoronary angiography following acute myocardial infarction in the elderly a matchedanalysis using propensity scores J Clin Epidemiol 200154387ndash398

11 Rosenbaum PR Rubin DB The central role of the propensity score in observationalstudies for causal effects Biometrika 19837041ndash55

12 Rosenbaum PR Rubin DB Reducing bias in observational studies using sub-classification on the propensity score J Am Stat Assoc 198479516ndash524

13 Austin PC An introduction to propensity-score methods for reducing the effects ofconfounding in observational studies Multivariate Behav Res 201146399ndash424

14 Rosenbaum PR Model-based direct adjustment J Am Stat Assoc 198782387ndash39415 Austin PC Stuart EA Moving towards best practice when using inverse probability of

treatment weighting (IPTW) using the propensity score to estimate causal treatmenteffects in observational studies Stat Med 2015343661ndash3679

16 Cole SR HernanMA Constructing inverse probability weights for marginal structuralmodels Am J Epidemiol 2008168656ndash664

17 Lucchetta RC Tonin FS Borba HHL et al Disease-modifying therapies for relapsingndashremitting multiple sclerosis a network meta-analysis CNS Drugs 201832813ndash826

18 Hamidi V Couto E Ringerike T Klemp M A multiple treatment comparison ofeleven disease-modifying drugs used for multiple sclerosis J Clin Med Res 20181088ndash105

19 Fogarty E Schmitz S Tubridy N Walsh C Barry M Comparative efficacy of disease-modifying therapies for patients with relapsing remitting multiple sclerosis systematicreview and network meta-analysis Mult Scler Relat Disord 2016923ndash30

20 Siddiqui MK Khurana IS Budhia S et al Systematic literature review andnetwork meta-analysis of cladribine tablets versus alternative disease-modifyingtreatment for relapsing-remitting multiple sclerosis Curr Med Res Opin 2018341361ndash1371

21 Hutchinson M Kappos L Calabresi P et al The efficacy of natalizumab in patientswith relapsing multiple sclerosis subgroup analyses of AFFIRM and SENTINELJ Neurol 2009256405ndash415

22 Kalincik T Jokubaitis V Spelman T et al Cladribine versus fingolimod natalizumaband interferon β for multiple sclerosis Mult Scler 2018241617ndash1626

23 Trojano M Tintore M Montalban X et al Treatment decisions in multiplesclerosismdashinsights from real-world observational studies Nat Rev Neurol 201713105ndash118

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 6 | November 2020 11

DOI 101212NXI000000000000087820207 Neurol Neuroimmunol Neuroinflamm

Alessio Signori Francesco Saccagrave Roberta Lanzillo et al Cladribine vs other drugs in MS Merging randomized trial with real-life data

This information is current as of August 14 2020

ServicesUpdated Information amp

httpnnneurologyorgcontent76e878fullhtmlincluding high resolution figures can be found at

References httpnnneurologyorgcontent76e878fullhtmlref-list-1

This article cites 22 articles 0 of which you can access for free at

Subspecialty Collections

httpnnneurologyorgcgicollectionmultiple_sclerosisMultiple sclerosis

httpnnneurologyorgcgicollectionclass_iiiClass IIIfollowing 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