ARTHRITIS & RHEUMATOLOGY
Vol. 69, No. 2, Feb ua y 2017, pp 362–375
DOI 10.1002/a .39856
V
C2016 The Au ho s. A h i is & Rheuma ology published by Wiley Pe iodicals, Inc.
on behal o he Ame ican College o Rheuma ology. This is an open access a icle unde
he e ms o he C ea i e Commons A ibu ion-NonComme cial License, which pe mi s
use, dis ibu ion and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly
ci ed and is no used o comme cial pu poses.
E icacy and Sa e y o Ep a uzumab in Mode a ely
o Se e ely Ac i e Sys emic Lupus E y hema osus
Resul s F om Two Phase III Randomized, Double-Blind,
Placebo-Con olled T ials
Megan E. B. Clowse,
1
Daniel J. Wallace,
2
Richa d A. Fu ie,
3
Michelle A. Pe i,
4
Ma ilyn C. Pike,
5
Pio Leszczy
nski,
6
C. Michael Neuwel ,
7
Ka h yn Hobbs,
8
Mau o Keise man,
9
Liliana Duca,
10
Kenne h C. Kalunian,
11
Ca inel Gala eanu,
12
Sabine Bonga d ,
13
Ch is ian S ach,
13
Ca olyn Beaudo ,
14
B ian Kilgallen,
14
and Ca oline Go don,
15
on behal o he EMBODY In es iga o G oup
Objec i e. Ep a uzumab, a monoclonal an ibody
ha a ge s CD22, modula es B cell signaling wi hou
subs an ial educ ions in he numbe o B cells. The
aim o his s udy was o epo he esul s o 2 phase III
mul icen e andomized, double-blind, placebo-
con olled ials, he EMBODY 1 and EMBODY 2
ials, assessing he e icacy and sa e y o ep a uzumab
in pa ien s wi h mode a ely o se e ely ac i e sys emic
lupus e y hema osus (SLE).
Me hods. Pa ien s me
‡
4 o he Ame ican Col-
lege o Rheuma ology e ised classi ica ion c i e ia o
SLE, we e posi i e o an inuclea an ibodies and/o
an i–double-s anded DNA an ibodies, had an SLE Dis-
ease Ac i i y Index 2000 (SLEDAI-2K) sco e o
‡
6
(inc eased disease ac i i y), had B i ish Isles Lupus
Assessmen G oup 2004 index (BILAG-2004) sco es o
g ade A (se e e disease ac i i y) in
‡
1 body sys em o
ClinicalT ials.go iden i ie s: NCT01262365; NCT01261793.
Suppo ed by UCB Pha ma.
1
Megan E. B. Clowse, MD, MPH: Duke Uni e si y Medical
Cen e , Du ham, No h Ca olina;
2
Daniel J. Wallace, MD: Ceda s-
Sinai Medical Cen e , Los Angeles, Cali o nia;
3
Richa d A. Fu ie,
MD: No hwell Heal h, New Yo k, New Yo k;
4
Michelle A. Pe i,
MD: Johns Hopkins Uni e si y School o Medicine, Bal imo e, Ma y-
land;
5
Ma ilyn C. Pike, MD: MedPha m Consul ing, Camb idge, Mas-
sachuse s;
6
Pio Leszczy
nski, MD: Poznan Uni e si y o Medical
Sciences, Poznan, Poland;
7
C. Michael Neuwel , MD: Alameda
Coun y Heal h Sys em, Oakland, Cali o nia;
8
Ka h yn Hobbs, MD:
Den e A h i is Clinic, Den e , Colo ado;
9
Mau o Keise man, MD:
Pon i ical Ca holic Uni e si y, Po o Aleg e, B azil;
10
Liliana Duca,
MD: Clinica Neomed, B aso , Romania;
11
Kenne h C. Kalunian, MD:
Uni e si y o Cali o nia San Diego School o Medicine, La Jolla;
12
Ca inel Gala eanu, MD: UCB Pha ma, B ussels, Belgium;
13
Sabine
Bonga d , MSc, Ch is ian S ach, MD: UCB Pha ma, Monheim, Ge -
many;
14
Ca olyn Beaudo , B ian Kilgallen, MSc: UCB Pha ma,
Raleigh, No h Ca olina;
15
Ca oline Go don, MD, FRCP: Uni e si y
o Bi mingham and Uni e si y Hospi al Bi mingham NHS Founda ion
T us , Bi mingham, UK. Membe s o he EMBODY In es iga o
G oup a e lis ed in Appendix A.
D . Clowse has ecei ed consul ing ees om UCB Pha ma
(mo e han $10,000). D . Wallace has ecei ed consul ing ees om
EMD Se ono, GlaxoSmi hKline, Lilly, and UCB Pha ma (less han
$10,000 each). D . Fu ie has ecei ed consul ing ees om UCB
Pha ma (mo e han $10,000). D . Pe i has ecei ed consul ing ees
om GlaxoSmi hKline, MedImmune, Me ck Se ono, Pa exel, and
UCB Pha ma (less han $10,000 each) and esea ch suppo om
UCB Pha ma. D . Pike has ecei ed consul ing ees om UCB
Pha ma (less han $10,000). D . Leszczy
nski has ecei ed consul ing
ees and speaking ees om Abb ie, Biogen, MSD, Roche, and P ize
(less han $10,000 each) and esea ch suppo om UCB Pha ma. D .
Neuwel has ecei ed consul ing ees om GlaxoSmi hKline, Human
Genome Sciences, and UCB Pha ma (less han $10,000 each). D .
Kalunian has ecei ed consul ing ees om B is ol-Mye s Squibb,
Genen ech, Biogen Idec, An he a, MedImmune, No o No disk,
Zymogene ics, Me ck Se ono, and UCB Pha ma (less han $10,000
each) and esea ch suppo om Genen ech, Biogen Idec, Cephalon,
Cyp ess, MedImmune, No o No disk, and UCB Pha ma. D . Go don
has ecei ed consul ing ees om B is ol-Mye s Squibb, Me ck
Se ono, and UCB Pha ma (less han $10,000 each) and esea ch sup-
po om UCB Pha ma.
Add ess co espondence o Megan E. B. Clowse, MD, MPH,
Duke Uni e si y Medical Cen e , Du ham, NC 27710. E-mail: megan.
[email p o ec ed].
Submi ed o publica ion Ap il 19, 2016; accep ed in e ised
o m Augus 23, 2016.
362
g ade B (mode a e disease ac i i y) in
‡
2 body sys ems
(in he mucocu aneous, musculoskele al, o ca dio espi-
a o y domains), and we e ecei ing s anda d he apy,
including manda o y ea men wi h co icos e oids (5–
60 mg/day). BILAG-2004 g ade A sco es in he enal and
cen al ne ous sys em domains we e excluded. Pa ien s
we e andomized 1:1:1 o ecei e ei he placebo,
ep a uzumab 600 mg e e y week, o ep a uzumab
1,200 mg e e y o he week, wi h in usions deli e ed o
he i s 4 weeks o each 12-week dosing cycle, o 4
cycles. Pa ien s ac oss all 3 ea men g oups also con-
inued wi h hei s anda d he apy. The p ima y end
poin was he esponse a e a week 48 acco ding o he
BILAG-based Combined Lupus Assessmen (BICLA)
de ini ion, equi ing imp o emen in he BILAG-2004
sco e, no wo sening in he BILAG-2004 sco e, SLEDAI-
2K sco e, o physician’s global assessmen o disease
ac i i y, and no disallowed changes in concomi an medi-
ca ions. Pa ien s who discon inued he s udy medica ion
we e classi ied as non esponde s.
Resul s. In he EMBODY 1 and EMBODY 2 i-
als o ep a uzumab, 793 pa ien s and 791 pa ien s,
espec i ely, we e andomized, 786 (99.1%) and 788
(99.6%), espec i ely, ecei ed s udy medica ion, and
528 (66.6%) and 533 (67.4%), espec i ely, comple ed
he s udy. The e was no s a is ically signi ican di e -
ence in he p ima y end poin be ween he g oups, wi h
he week 48 BICLA esponse a es being simila
be ween he ep a uzumab g oups and he placebo g oup
( esponse a es anging om 33.5% o 39.8%). No new
sa e y signals we e iden i ied.
Conclusion. In pa ien s wi h mode a e o
se e ely ac i e SLE, ea men wi h ep a uzumab 1
s anda d he apy did no esul in imp o emen s in
esponse a es o e ha obse ed in he placebo 1
s anda d he apy g oup.
Sys emic lupus e y hema osus (SLE) is a ch onic
mul isys em au oimmune disease (1) ha mos e-
quen ly a ec s he musculoskele al, mucocu aneous,
hema ologic, and enal sys ems (2). The disease com-
monly ollows a elapsing– emi ing pa e n, wi h la es
o high disease ac i i y ollowed by empo a y educ-
ions in symp oms. The apeu ic op ions a e limi ed.
Co icos e oids, o en a high doses, o m he co ne -
s one o ea men . Thei long- e m use a high doses
(e.g., use o o al p ednisone a a dosage o 0.5–1.0 mg/
kg/day) is associa ed wi h signi ican complica ions,
which may ha e a subs an ial impac on a pa ien ’s heal h
and quali y o li e (3,4). Immunosupp essan s and an i-
mala ial d ugs a e equen ly included in he pa ien ’s
egimen, wi h he aim o educing disease ac i i y and
limi ing he long- e m o gan damage a ising ei he om
he disease i sel o om co icos e oid use.
Recen ad ances in he unde s anding o SLE
pa hogenesis and he cen al ole o B cells in he pa ho-
logicp ocesseso hediseaseha eled o head en o
biologic he apies o he managemen o lupus. One such
he apy is ep a uzumab, a humanized monoclonal an i-
body o he IgG1 class ha a ge s CD22 on B cells, pe -
u bing he B cell ecep o signaling complex and esul ing
in he modula ion o B cell ac i i y wi hou subs an ial
educ ions in he numbe o pe iphe al B cells (5,6).
Ep a uzumab has been e alua ed as a he apy
o SLE in 12 sponso ed clinical s udies. In he 2 phase
II/III double-blind, placebo-con olled ALLEVIATE
s udies (add essing he e icacy and sa e y o ep a-
uzumab in pa ien s wi h mode a e/se e e la ing SLE),
he doses o ep a uzumab used we e based on body su -
ace a ea, and clinical ou comes we e measu ed using
he B i ish Isles Lupus Assessmen G oup (BILAG)
imp o emen esponse. Pa ien s ecei ing a dose o
360 mg/m
2
had imp o emen s in he clinical signs and
symp oms o SLE (7) as well as imp o emen s in quali y
o li e measu es and educ ions in hei co icos e oid
dose (8). In he phase IIb EMBLEM s udy (add essing
he sa e y and e icacy o ep a uzumab in pa ien s wi h
se ologically posi i e ac i e SLE), ixed doses o
ep a uzumab we e in es iga ed. This double-blind, pla-
cebo-con olled, dose- anging and dose egimen– anging
s udy u ilized a composi e esponse index, he BILAG-
based Combined Lupus Assessmen (BICLA), which
emphasizes imp o emen based on changes in he
BILAG index, a measu e o disease ac i i y. The s udy
comp ised 1 dosing cycle, wi h he s udy d ug adminis-
e ed o e 4 weeks, and he p ima y end poin was
assessed a week 12. A posi i e ea men e ec , com-
pa ed o placebo, was seen in pa ien s ecei ing
ep a uzumab a a cumula i e dose o 2,400 mg. This
dose was he e o e ca ied o wa d in o he phase III
s udies, using he 2 di e en dosing egimens s udied in
he phase IIb s udy (5,7,9).
In he p esen epo , we p esen he esul s o
he 2 phase III mul icen e , andomized, placebo-
con olled, double-blind EMBODY 1 and EMBODY 2
ials, which aimed o demons a e he e icacy and
sa e y o ep a uzumab in he ea men o pa ien s wi h
mode a ely o se e ely ac i e SLE.
PATIENTS AND METHODS
Pa ien s. Inclusion c i e ia. Eligible pa ien s we e age
$18 yea s and had a diagnosis o mode a ely o se e ely ac i e
SLE ha ul illed $4 o he 11 Ame ican College o Rheuma-
ology (ACR) e ised c i e ia o SLE (10) (i pa ien s we e
EFFICACY AND SAFETY OF EPRATUZUMAB IN SLE 363
posi i e o a neu ologic diso de , he diagnosis had o mee
$5 o 11 ACR c i e ia). All pa ien s had, a a minimum, dis-
ease ac i i y in he musculoskele al, mucocu aneous, o ca dio-
espi a o y body sys ems, as de ined by he 2004 e sion o he
BILAG index (BILAG-2004) (11). Pa ien s we e equi ed o
ha e a BILAG-2004 g ade A (se e e disease ac i i y) in $1o
hese body sys ems o a BILAG-2004 g ade B (mode a e dis-
ease ac i i y) in $2 o hese body sys ems. In addi ion, all
pa ien s had o ha e an SLE Disease Ac i i y Index 2000
(SLEDAI-2K) sco e o $6 (indica ing inc eased disease ac i -
i y) (12), and o be posi i e, a sc eening, o an inuclea an i-
bodies (ANAs; i e $1:80) and/o an i–double-s anded DNA
(an i-dsDNA) an ibodies (de ined as a posi i e esul om
ei he a mul iplex immunoassay o he Fa assay).
Pa ien s mus ha e been ecei ing co icos e oids a a
s able dosage o 5–60 mg/day (p ednisone o equi alen ) o a
leas 5 days (61 day) p io o baseline. An imala ials (hyd o-
xychlo oquine, chlo oquine, o quinac ine) and immunosup-
p essan s (aza hiop ine, mycophenola e mo e il, le lunomide,
o me ho exa e) we e no manda o y bu we e pe mi ed,
whe eas o he immunosupp essan s we e excluded. Pa ien s
ea ed wi h hese agen s mus ha e ecei ed hem a a s able
dose o a leas 28 days (61 day) p io o baseline.
Exclusion c i e ia. Pa ien s wi h se e e lupus neph i is
o se e e neu opsychia ic SLE a sc eening we e excluded.
Thus, BILAG-2004 g ade A sco es in hese body sys ems
( enal and neu opsychia ic domains) we e no pe mi ed
(wi h he excep ion o pa ien s achie ing a BILAG-2004 neu-
opsychia ic g ade A because o he p esence o mononeu i is
[single o mul iple] and/o polyneu opa hy, p o ided ha his
was no new o wo sening a sc eening). Se um c ea inine le -
els o .2.5 mg/dl, a clinically signi ican inc ease in he se um
c ea inine le el wi hin 4 weeks p io o sc eening, o p o ein-
u ia le els o .3.5 gm/day we e also exclusion c i e ia.
O he exclusions included pa ien s wi h known an i-
phospholipid synd ome, hose who we e p egnan o b eas -
eeding, hose who had a p o oundly immunosupp essed s a e,
and hose wi h signi ican hema ologic abno mali ies, ac i e
in ec ions, a his o y o ch onic in ec ions, o a his o y o malig-
nancies o h omboembolic e en s. Signi ican hema ologic
abno mali ies (any labo a o y inding o a hemoglobin le el
,8.0 gm/dl, a whi e blood cell coun ,2,000/mm
3
, an absolu e
neu ophil coun ,1,500/mm
3
, o a pla ele coun ,30,000/
mm
3
) we e no allowed. Fu he mo e, pa ien s we e excluded i
hey had ecei ed o al an icoagulan s wi hin 12 weeks p io o
sc eening, cyclophosphamide wi hin 6 mon hs p io o sc een-
ing, o calcineu in inhibi o s wi hin 4 weeks p io o sc eening.
P e ious use o biologic he apies was allowed, subjec o an
app op ia e p o ocol-de ined washou pe iod be o e sc eening.
S udy design. The EMBODY 1 and EMBODY 2 i-
als we e iden ical phase III, mul icen e , andomized, double-
blind, placebo-con olled s udies, wi h he only di e ence
being he si es a which he s udies ook place (bo h s udies
included si es in No h Ame ica, La in Ame ica, Wes e n
Eu ope, Eas e n Eu ope, and he Middle Eas and India;
EMBODY 1 addi ionally included he Paci ic egion [Aus alia]
and he Fa Eas [Republic o Ko ea and Taiwan]; EMBODY
2 addi ionally included Sou h A ica). All pa ien s p o ided
hei w i en in o med consen , and he s udies ecei ed
app o al om he local ins i u ional e iew boa ds/independen
e hics commi ees.
The p ima y end poin was he esponde a e a week
48, acco ding o he BICLA composi e end poin (13), which
equi es imp o emen om baseline in he BILAG-2004
sco e, wi h no wo sening in he BILAG-2004 sco e, SLEDAI-
2K sco e, o physician’s global assessmen o disease ac i i y,
and no disallowed changes in concomi an medica ions (dis-
cussed in mo e de ail below).
The s udies consis ed o a 2-week sc eening pe iod,
ollowed by a 48-week double-blind ea men pe iod, and a 4-
week sa e y ollow-up (13 weeks o pa ien s discon inuing
ea men p io o week 48). The sample size was selec ed o
p o ide 90% powe o de ec a 15% highe esponse, based on
he p ima y end poin , in ep a uzumab- ea ed pa ien s com-
pa ed o placebo- ea ed pa ien s (5,7).
In each s udy, 780 pa ien s we e planned o andomi-
za ion, and pa ien s we e andomized 1:1:1 h ough an in e ac-
i e oice and web esponse sys em o 1 o 3 ea men a ms:
placebo, ep a uzumab 600 mg e e y week, o ep a uzumab
1,200 mg e e y o he week. In usions we e deli e ed o e a 4-
week dosing pe iod a he beginning o each 12-week ea -
men cycle, i.e., 600-mg in usions o ep a uzumab gi en a
weeks 0, 1, 2, and 3, o 1,200-mg in usions o ep a uzumab
gi en a weeks 0 and 2 (wi h in usions o placebo a weeks 1
and 3, o main ain blinding), o in usions o placebo a weeks
0, 1, 2, and 3 (Figu e 1). This dosing pa e n was epea ed
e e y 12 weeks o 4 cycles, wi h a inal assessmen a week 48.
Assessmen s we e pe o med a baseline, and hen a weeks 4,
8, and 12 o each cycle. Pa ien s who ei he emained in he
s udy o comple ion a week 48 o wi hd ew a week 16 o
la e , due o lack o e icacy, we e allowed o en oll in he
open-label ex ension s udy (NCT01408576). The s udy d ug
was gi en in conjunc ion wi h he pa ien s’ exis ing s anda d
he apy (all concomi an medica ions a e desc ibed below).
All pa ien s mus ha e been app o ed o andomiza-
ion by an ex e nal cen al e iewe o de e mine whe he ade-
qua e disease ac i i y was p esen . Randomiza ion was
s a i ied by geog aphic egion (Eas e n Eu ope, Wes e n
Eu ope, Middle Eas and India, Fa Eas , No h Ame ica,
La in Ame ica, and he Paci ic) and by disease se e i y a
baseline. Disease se e i y was de e mined using he Sys emic
Lupus In e na ional Collabo a ing Clinics/ACR Damage
Index (SDI) (14), ca ego ized as ollows: 1) an SDI sco e o 0
and a BILAG-2004 g ade A in ,2 body sys ems, 2) an SDI
sco e o .0 o a BILAG-2004 g ade A in $2 body sys ems, o
3) an SDI sco e o .0 and a BILAG-2004 g ade A in $2 body
sys ems.
Concomi an medica ions. A baseline, all pa ien s
mus ha e been ecei ing o al co icos e oids a a dosage o 5–
60 mg/day (p ednisone o equi alen ). Be ween weeks 0 and 8,
empo a y inc eases in he dose o co icos e oids up o a max-
imum o 25% abo e baseline le els we e pe mi ed a he dis-
c e ion o he in es iga o . Pa ien s we e classi ied as
non esponde s i hei co icos e oid dose emained abo e he
baseline le el a e week 8. F om week 4, ape ing o o al co -
icos e oids was encou aged, a a a e o 5 mg e e y 2 weeks o
a a ge dose o #7.5 mg/day. Co icos e oid dose changes
we e o be a oided wi hin 4 weeks o he week 24 and week 48
assessmen s.
Pa ien s ecei ing immunosupp essan s and/o an ima-
la ial agen s had o con inue hei s able baseline dose
h oughou he s udy, unless he e was suspec ed oxici y.
364 CLOWSE ET AL
Pa ien s changing hei dose o s a ing a new immunosup-
p essan o an imala ial we e conside ed non esponde s.
S udy a iables. The p ima y e icacy a iable was
he imp o emen esponse a week 48 based on he BICLA
de ini ion (13), which equi ed all o he ollowing c i e ia:
imp o emen in he BILAG-2004 sco e (imp o emen in all
BILAG-2004 g ade A sco es a s udy en y o g ades B, C, o
D a ollow-up; imp o emen in all BILAG-2004 g ade B
sco es a s udy en y o g ades C o D a ollow-up); no wo s-
ening in he BILAG-2004 sco e (no new BILAG-2004 g ade A
sco es; no mo e han 1 new BILAG-2004 g ade B sco e); no
wo sening (no inc ease) in he SLEDAI-2K o al sco e, as com-
pa ed o ha a s udy en y; no wo sening (,10-mm inc ease
on a 100-mm isual analog scale) in he physician’s global
assessmen o disease ac i i y, as compa ed o ha a s udy
en y; and no disallowed changes in concomi an medica ions.
The BILAG-2004 body sys em sco es and SLEDAI-2K da a
we e e i ied by a cen al e iew and adjudica ion commi ee o
ensu e consis en applica ion o he assessmen s.
Seconda y e icacy a iables we e he BICLA esponse
a es a weeks 12, 24, and 36, and he change om baseline in
daily co icos e oid doses a weeks 24 and 48.
Explo a o y e icacy a iables included each compo-
nen o he BICLA composi e end poin , he BICLA esponse
a ime poin s o he han hose included in he p ima y and
seconda y end poin s, and changes om baseline in he
BILAG-2004 o al sco e (sco es we e con e ed om g ades
A, B, C, D, and E o sco es o 12, 8, 1, 0, and 0, espec i ely)
(15), changes om baseline in he SLEDAI-2K o al sco e,
changes om baseline in he pa ien ’s and physician’s global
assessmen s o disease ac i i y (on 100-mm isual analog
scales), and he a e age change om baseline in co icos e oid
dose (calcula ed as he ime-weigh ed a ea unde he cu e
minus baseline, o baseline o week 48).
Analyses o pa ien - epo ed ou come measu es in-
cluded he p opo ions o pa ien s who achie ed a minimal
clinically impo an di e ence (MCID) om baseline in he
36-i em Sho Fo m (SF-36) Heal h Su ey physical and men-
al componen summa y sco es (de ined as a $2.5-poin
imp o emen ) (16), mean changes om baseline in each o he
LupusQoL heal h- ela ed quali y o li e domains (17), mean
changes om baseline in he Func ional Assessmen o
Ch onic Illness The apy–Fa igue (FACIT-F) scale (18), and
he p opo ions o pa ien s achie ing an MCID ($4-poin
inc ease) in he FACIT-F scale.
Time o a new la e was also de e mined, wi h a new
la e de ined as he de elopmen o a BILAG-2004 g ade A in
$1 body sys em om a p e ious g ade o B, C, D, o E a
baseline, o he de elopmen o a concu en BILAG-2004
g ade B in $2 body sys ems om a g ade o C, D, o E a base-
line (sys ems la ing a baseline we e no included). Fla es
we e based only on i ems ha we e new and we e con i med
by he cen al e iew and adjudica ion commi ee.
Pha macodynamic and immunologic a iables included
le els o CD191B cells, CD31T cells, CD191CD221B cells,
immunoglobulins, au oan ibodies (an i-dsDNA), ex ac able
nuclea an igen an ibodies, and complemen C3 and C4 p o eins.
S a is ical analysis. E icacy a iables we e analyzed
using he ull analysis se , consis ing o andomized pa ien s
who ecei ed a leas one pa ial dose o s udy medica ion.
One o he EMBODY 1 s udy si es was ound, by an audi , o
ha e ailed o conduc he s udy in line wi h applicable
egula ions, In e na ional Con e ence on Ha monisa ion/
Figu e 1. A, Design o he EMBODY s udies on e icacy and sa e y o ep a uzumab monoclonal an ibody (Emab) ea men in pa ien s wi h
mode a ely o se e ely ac i e sys emic lupus e y hema osus (SLE). B, Disposi ion o pa ien s in he EMBODY 1 and EMBODY 2 ials. * 5One
si e in EMBODY 1 was emo ed om he s udy owing o sa e y conce ns and a lack o compliance wi h he s udy p o ocol; all 45 pa ien s
en olled a his si e we e emo ed om he ull analysis se , bu e ained in he sa e y se . QOW 5e e y o he week; ST 5s anda d he apy;
QW 5e e y week; BICLA 5B i ish Isles Lupus Assessmen G oup–based Combined Lupus Assessmen ; PBO 5placebo.
EFFICACY AND SAFETY OF EPRATUZUMAB IN SLE 365
Good Clinical P ac ice Guidelines, and he s udy p o ocol,
and consequen ly all 45 pa ien s en olled a ha si e we e
excluded om he ull analysis se and he e icacy analyses.
These pa ien s we e e ained in he sa e y se , o p o ide com-
ple e sa e y da a.
Fo de e mina ion o he p ima y end poin , BICLA
esponse a es we e calcula ed using logis ic eg ession, wi h
pa ien s who discon inued p io o week 48 o did no ha e a
week 48 assessmen being classi ied as non esponde s. In
hose cases in which single componen s o he BICLA
(BILAG-2004 sco es, SLEDAI-2K sco es, o physician’s
global assessmen o disease ac i i y) we e missing, he alue
was impu ed om he p e ious isi alue. Pa ien s wi h
disallowed changes in concomi an medica ions we e classi ied
as non esponde s om ha ime poin o wa d.
In he p ima y analysis, P alues o pai wise compa i-
sons be ween each ac i e ea men g oup and he placebo
g oup we e gene a ed using a logis ic eg ession model,
including ac o s o pooled geog aphic egion and disease
se e i y a baseline. The Hochbe g me hod was used o adjus
o he compa ison o he 2 doses o ep a uzumab o placebo.
The esponse a e based on he p ima y a iable was also ana-
lyzed in an explo a o y manne o subg oups o pa ien s,
including hose de ined by geog aphic egion, body mass
index, baseline disease se e i y, lupus-associa ed labo a o y
pa ame e s, and concomi an medica ion use a baseline. Sub-
g oup analyses we e no adjus ed o mul iplici y. Odds a ios
and co esponding 95% con idence in e als we e calcula ed
o each ep a uzumab ea men g oup compa ed o placebo,
using logis ic eg ession models ha included ac o s o ea -
men , pooled egion, and baseline disease s a us, as well as o
he subg oup being analyzed and he ea men -by-subg oup
in e ac ion.
Fi e key seconda y e icacy a iables we e es ed o s a-
is ical signi icance acco ding o a hie a chical es ing p ocedu e,
wi h Hochbe g adjus men o mul iplici y wi hin each s ep o
he p ocedu e. BICLA esponse a es a weeks 36, 24, and 12
we e calcula ed and analyzed using he same me hods as hose
used o esponse a es a week 48. The key seconda y end
poin s o he co icos e oid dose we e o de ed ca ego ical end
poin s (pa ien s we e ca ego ized acco ding o hei change in
daily co icos e oid dose, de ined as a .50% dec ease, 0–50%
dec ease, no change, o an inc ease in dose o missing da a),
analyzed using nonpa ame ic ank analysis o co a iance,
wi h missing alues impu ed as he wo s possible ou come. All
o he seconda y and explo a o y e icacy a iables we e summa-
ized using desc ip i e s a is ics, wi h missing da a impu ed using
las obse a ion ca ied o wa d o con inuous a iables o
non esponde impu a ion o dicho omous a iables.
The SLE Responde Index (SRI) a week 48, which
was o iginally ca ego ized as a $4-poin (SRI-4), $6-poin
(SRI-6), o $8-poin (SRI-8) educ ion in sco e on he Sa e y
o Es ogens in Lupus E y hema osus Na ional Assessmen
e sion o he SLEDAI (SELENA-SLEDAI) (19), we e ana-
lyzed pos hoc, wi h modi ica ions om he o iginal SRI end
poin de ini ion o include he a iables used in he EMBODY
s udies, i.e., imp o emen in he SLEDAI-2K sco es, BILAG-
2004 sco es, and he physician’s global assessmen o disease
ac i i y (ins ead o he SRI-4 calcula ed om he SELENA-
SLEDAI, he classic BILAG index, and he physician’s global
assessmen o disease ac i i y). To be conside ed a esponde
in ou SRI analyses, pa ien s mus ha e achie ed all o he ol-
lowing: imp o emen in he SLEDAI-2K sco e o a leas 4, 6,
o 8 poin s (SRI-4, SRI-6, and SRI-8, espec i ely) om s udy
en y; no wo sening in he BILAG-2004 o al sco e (no new
BILAG-2004 g ade A sco es, no mo e han 1 new BILAG-
2004 g ade B sco e); no wo sening (,10-mm inc ease on a
100-mm isual analog scale) in he physician’s global assess-
men o disease ac i i y, compa ed o ha a s udy en y; and
no disallowed changes in concomi an medica ions a any ime
poin om baseline. Missing da a we e impu ed using modi-
ied non esponde impu a ion, as was used o he p ima y
end poin .
Sa e y and immunologic a iables we e analyzed using
he sa e y se . These a iables we e summa ized wi h desc ip-
i e s a is ics.
RESULTS
Disposi ion o he pa ien s and baseline cha ac-
e is ics. In he EMBODY 1 ial, 1,257 pa ien s we e
sc eened, and 793 we e andomized. In he EMBODY 2
ial, 1,194 pa ien s we e sc eened, and 791 we e andom-
ized. The majo i y o pa ien s ailed sc eening because
hey we e deemed ineligible on he basis o he inclusion
and exclusion c i e ia (367 o 464 sc een ailu es in he
EMBODY 1 ial, and 315 o 403 sc een ailu es in he
EMBODY 2 ial). O he easons o exclusion included
occu ence o ad e se e en s, loss o ollow-up, and wi h-
d awal o consen du ing he sc eening pe iod.
O e all, 265 pa ien s (33.4%) in EMBODY 1
and 258 (32.6%) in EMBODY 2 discon inued om he
s udies, mos commonly due o lack o e icacy (112
pa ien s in EMBODY 1, and 113 pa ien s in EMBODY
2). One s udy si e in he EMBODY 1 ial was emo ed
due o p o ocol iola ions, and all 45 pa ien s andom-
ized a ha si e we e emo ed om he ull analysis se .
Despi e his, discon inua ions we e balanced ac oss he
ea men g oups and ac oss he s udies ( ange o dis-
con inua ions ac oss g oups 30.9–35.5% in EMBODY
1, and 30.8–34.7% in EMBODY 2) (Figu e 1).
Pa ien cha ac e is ics a baseline we e also bal-
anced be ween he 2 s udies (Table 1). Mo e han 90%
o pa ien s we e emale, and he mean age was 42.1
yea s in EMBODY 1 and 41.0 yea s in EMBODY 2.
Time since diagnosis anged om 0 yea s o 43 yea s,
wi h a median o 6.3 yea s in EMBODY 1 and 5.1 yea s
in EMBODY 2. The mean 6SD daily co icos e oid
dosage was 11.2 68.8 mg/day in EMBODY 1 and
13.0 69.6 mg/day in EMBODY 2, wi h 45.1% o
pa ien s in EMBODY 1 and 36.0% o pa ien s in
EMBODY 2 ecei ing a dosage o #7.5 mg/day. The
p opo ion o pa ien s ecei ing an imala ials a baseline
was sligh ly lowe in EMBODY 2 han in EMBODY 1
(72.5% o pa ien s in EMBODY 1 e sus 63.6% o
pa ien s in EMBODY 2).
366 CLOWSE ET AL

Table 1. Pa ien demog aphics and disease cha ac e is ics a baseline in he EMBODY 1 and EMBODY 2 ials*
EMBODY 1 EMBODY 2
Placebo
1s anda d
he apy
(n 5249)
Emab
(1,200 mg QOW)
1s anda d he apy
(n 5244)
Emab
(600 mg QW)
1s anda d he apy
(n 5248)
Placebo
1s anda d
he apy
(n 5263)
Emab
(1,200 mg QOW)
1s anda d he apy
(n 5261)
Emab
(600 mg QW)
1s anda d he apy
(n 5264)
Age, mean 6SD yea s 41.2 612.8 42.2 611.7 42.2 611.4 41.1 611.8 40.8 611.5 41.2 612.7
Female 237 (95.2) 228 (93.4) 226 (91.1) 245 (93.2) 247 (94.6) 245 (92.8)
Race
Asian 26 (10.4) 22 (9.0) 18 (7.3) 7 (2.7) 7 (2.7) 12 (4.5)
Black/A ican Ame ican 26 (10.4) 32 (13.1) 33 (13.3) 25 (9.5) 29 (11.1) 34 (12.9)
Whi e 187 (75.1) 178 (73.0) 188 (75.8) 204 (77.6) 198 (75.9) 193 (73.1)
Hispanic/La ino 50 (20.1) 52 (21.3) 44 (17.7) 56 (21.3) 50 (19.2) 50 (18.9)
Yea s since diagnosis, median ( ange) 5.8 (0–36) 7.3 (0–34) 6.1 (0–43) 5.7 (0–37) 5.0 (0–29) 4.8 (0–42)
Physician- epo ed measu e
SLEDAI-2K o al sco e, mean 6SD 10.7 64.1 9.9 63.7 10.2 63.6 10.1 63.6 10.1 63.8 10.2 63.6
PhGA, mean 6SD 55.5 612.9 55.7 614.3 56.5 614.9 56.2 614.4 57.2 614.0 57.3 615.6
Pa ien s wi h $1 BILAG-2004 g ade A 139 (55.8) 142 (58.2) 147 (59.3) 157 (59.7) 148 (56.7) 161 (61.0)
BILAG-2004 o al sco e, mean 6SD† 20.0 65.5 19.8 65.9 19.6 65.6 21.0 66.7 21.3 66.6 21.0 66.7
Pa ien - epo ed measu e
P GA, mean 6SD 58.3 619.2 58.5 619.1 58.1 620.2 58.6 619.6 60.2 619.0 59.1 618.9
SF-36, mean 6SD
PCS sco e 35.0 69.2 34.2 610.1 34.2 68.0 34.6 69.7 34.9 68.8 35.2 69.5
MCS sco e 38.3 611.8 38.0 611.3 37.8 613.4 37.9 612.0 37.9 612.6 38.0 611.9
FACIT-F sco e, mean 6SD 24.4 612.0 24.1 611.7 23.6 610.7 24.3 611.5 24.1 611.4 25.3 611.4
Concomi an medica ion
Immunosupp essan 116 (46.6) 123 (50.4) 112 (45.2) 121 (46.0) 113 (43.3) 129 (48.9)
Aza hiop ine 53 (21.3) 52 (21.3) 41 (16.5) 48 (18.3) 51 (19.5) 56 (21.2)
Le lunomide 1 (0.4) 3 (1.2) 7 (2.8) 3 (1.1) 6 (2.3) 6 (2.3)
Me ho exa e 41 (16.5) 49 (20.1) 41 (16.5) 40 (15.2) 39 (14.9) 45 (17.0)
Mycophenola e 22 (8.8) 29 (11.4) 30 (12.1) 38 (14.4) 32 (12.3) 34 (12.9)
An imala ial 175 (70.3) 181 (74.2) 181 (73.0) 162 (61.6) 160 (61.3) 179 (67.8)
Co icos e oid 248 (99.6) 241 (98.8) 247 (99.6) 256 (97.3) 257 (98.5) 263 (99.6)
0–#7.5 mg/day 120 (48.2) 112 (45.9) 102 (41.1) 97 (36.9) 88 (33.7) 99 (37.5)
7.5–#20 mg/day 117 (47.0) 105 (43.0) 131 (52.8) 137 (52.1) 142 (54.4) 135 (51.1)
.20 mg/day 11 (4.4) 24 (9.8) 14 (5.6) 22 (8.4) 27 (10.3) 29 (11.0)
Labo a o y pa ame e
ANAs .1:80 216 (86.7) 212 (86.9) 218 (87.9) 231 (87.8) 232 (88.9) 233 (88.3)
An i-dsDNA posi i e 133 (53.4) 126 (51.6) 131 (52.8) 143 (54.4) 126 (48.3) 134 (50.8)
Low complemen le els 115 (46.2) 104 (42.6) 110 (44.4) 125 (47.5) 122 (46.7) 134 (50.8)
* Excep whe e indica ed o he wise, alues a e he numbe (%) o pa ien s. Emab 5ep a uzumab monoclonal an ibody; QOW 5e e y o he week; QW 5e e y week; SLEDAI-
2K 5Sys emic Lupus E y hema osus Disease Ac i i y Index 2000; PhGA 5physician’s global assessmen o disease ac i i y (on 0–100-mm isual analog scale); P GA 5pa ien ’s
global assessmen o disease ac i i y (on 0–100-mm isual analog scale); SF-36 536-i em Sho Fo m; PCS 5physical componen summa y; MCS 5men al componen summa y;
FACIT-F 5Func ional Assessmen o Ch onic Illness The apy–Fa igue; ANAs 5an inuclea an ibodies; an i-dsDNA 5an i–double-s anded DNA.
† The B i ish Isles Lupus Assessmen G oup 2004 index (BILAG-2004) was calcula ed as ollows: A 512, B 58, C 51, D 50, and E 50.
EFFICACY AND SAFETY OF EPRATUZUMAB IN SLE 367
A baseline, 428 pa ien s (57.8%) in EMBODY 1
and 466 pa ien s (59.1%) in EMBODY 2 had a BILAG-
2004 g ade A in $1 body sys em, while 371 pa ien s
(50.1%) in EMBODY 1 and 411 pa ien s (52.2%) in
EMBODY 2 had a BILAG-2004 g ade B in $2 body
sys ems. Mos pa ien s had mode a e- o-se e e disease
ac i i y in he musculoskele al and mucocu aneous body
sys ems: o he musculoskele al sys em, BILAG-2004
g ades A o B we e p esen in 92.2% o EMBODY 1
pa ien s and 93.5% o EMBODY 2 pa ien s; o he
mucocu aneous sys em, BILAG-2004 g ades A o B
we e p esen in 80.3% o EMBODY 1 pa ien s and
83.0% o EMBODY 2 pa ien s. O he pa ien cha ac e -
is ics we e also simila ac oss ea men g oups (Table 1).
E icacy. The p ima y end poin was no me in
ei he s udy. A week 48, imp o emen s in disease ac i -
i y, as measu ed by he BICLA esponse a es, occu ed
in simila p opo ions o pa ien s ac oss ea men
g oups; no signi ican di e ences we e seen in he
p opo ion o esponde s be ween pa ien s ecei ing
placebo 1s anda d he apy and hose ecei ing ei he
dose o ep a uzumab 1s anda d he apy (Figu es 2A
and B). In he EMBODY 1 s udy, BICLA esponse
a es we e 34.1% in he placebo 1s anda d he apy
g oup, 39.8% in he ep a uzumab 1,200 mg e e y o he
week 1s anda d he apy g oup (P50.175 e sus
placebo), and 37.5% in he ep a uzumab 600 mg
e e y week 1s anda d he apy g oup (P50.442 e sus
placebo). In he EMBODY 2 s udy, BICLA esponse a es
we e 33.5% in he placebo 1s anda d he apy g oup,
34.1% in he ep a uzumab 1,200 mg e e y o he
week 1s anda d he apy g oup (P50.899 e sus placebo),
and 35.2% in he ep a uzumab 600 mg e e y week 1
s anda d he apy g oup (P50.716 e sus placebo).
In he EMBODY 1 ial, 87 pa ien s who ecei ed
placebo (34.9%) and 160 pa ien s who ecei ed
ep a uzumab (32.5%) did no achie e a ea men
esponse a week 48, which was a ibu ed o ea ly wi h-
d awal o missing da a. Mo eo e , 54 pa ien s in he pla-
cebo g oup (21.7%) and 134 pa ien s in he ep a uzumab
g oups (27.2%) we e classi ied as non esponde s as a
esul o disallowed changes in concomi an medica ions
(p edominan ly, disallowed inc eases in he dose o co -
icos e oids; pa ien s may ha e had mo e han one eason
o a non esponse). In he EMBODY 2 ial, 84 pa ien s
who ecei ed placebo (31.9%) and 166 pa ien s who
ecei ed ep a uzumab (31.6%) had missing da a o wi h-
d ew ea ly om he s udy. Mo eo e , 60 pa ien s in he
placebo g oup (22.7%) and 119 pa ien s in he
ep a uzumab g oups (22.7%) had p ohibi ed medica ion
changes (again, p ima ily inc eases in he dose o co -
icos e oids, wi h mo e han one eason o a non e-
sponse being possible).
Rapid imp o emen s om baseline we e ini ially
seen in bo h he placebo g oup and he ep a uzumab
dosing g oups in bo h s udies. A se e al ime poin s,
Figu e 2. BICLA esponde a es by ea men g oup in he EMBODY 1 ial (A) and EMBODY 2 ial (B), and week 48 change om baseline in
daily co icos e oid (CS) dose in EMBODY 1 (C) and EMBODY 2 (D), as well as a e age change om baseline (E). See Figu e 1 o o he de ini ions.
368 CLOWSE ET AL
Table 2. E icacy ou comes a week 48*
EMBODY 1 EMBODY 2
Placebo
1s anda d
he apy
(n 5249)
Emab
(1,200 mg QOW)
1s anda d
he apy
(n 5244)
Emab
(600 mg QW)
1s anda d
he apy
(n 5248)
Placebo
1s anda d
he apy
(n 5263)
Emab
(1,200 mg
QOW)
1s anda d
he apy
(n 5261)
Emab
(600 mg QW)
1s anda d
he apy
(n 5264)
Clinical ou comes
BICLA esponse a e 85 (34.1) 97 (39.8) 93 (37.5) 88 (33.5) 89 (34.1) 93 (35.2)
BILAG-2004 index
Imp o emen and no wo sening 103 (41.4) 126 (51.6) 120 (48.4) 115 (43.7) 104 (39.8) 110 (41.7)
A e age CFB in o al sco e, LS mean 6SD 28.6 66.3 29.8 66.5 29.1 66.4 28.1 66.7 29.1 67.6 29.6 66.6
SLEDAI-2K sco e
A e age CFB in o al sco e, LS mean 6SD 23.6 64.8 23.8 64.2 23.6 64.5 23.3 64.2 23.5 64.6 23.9 64.3
No wo sening 228 (91.6) 221 (90.6) 228 (91.9) 234 (89.0) 230 (88.1) 242 (91.7)
PhGA
A e age CFB, LS mean 6SD mm 223.8 622.2 225.6 622.4 221.9 624.2 222.3 623.9 223.7 622.6 226.0 623.2
No wo sening 228 (1.6) 229 (93.9) 223 (90.3) 228 (87.0) 236 (91.1) 247 (94.3)
Concomi an medica ion use, no disallowed changes 195 (78.3) 175 (71.7) 183 (73.8) 203 (77.2) 198 (75.9) 208 (78.8)
SRI-4 esponse a e 89 (35.7) 93 (38.1) 83 (33.5) 85 (32.3) 91 (34.9) 96 (36.4)
Pa ien - epo ed ou comes
P GA, a e age CFB, LS mean 6SD mm 215.5 627.1 217.1 628.1 215.7 628.1 213.7 628.0 216.8 627.6 217.5 629.3
SF-36 sco e
MCID in PCS 119 (48.6) 129 (54.0) 125 (50.4) 117 (45.0) 125 (48.8) 134 (51.7)
MCID in MCS 116 (47.3) 116 (48.5) 102 (41.1) 121 (46.5) 115 (44.9) 106 (40.9)
S abiliza ion/lack o de e io a ion 53 (21.6) 70 (29.3) 65 (26.2) 63 (24.2) 62 (24.2) 67 (25.9)
FACIT-F sco e
LS mean 6SD 4.0 610.1 4.2 69.9 3.8 610.4 2.6 69.3 2.6 610.1 2.6 610.0
MCID 110 (44.5) 110 (45.8) 105 (43.0) 110 (42.6) 113 (44.1) 116 (44.8)
* Excep whe e indica ed o he wise, alues a e he numbe (%) o pa ien s. BICLA 5B i ish Isles Lupus Assessmen G oup–based Combined Lupus Assessmen ; CFB 5change
om baseline; LS 5leas squa es; SRI-4 5Sys emic Lupus E y hema osus Disease Ac i i y Index Responde Index 4-poin imp o emen ; MCID 5minimal clinically impo an
di e ence (see Table 1 o o he de ini ions).
EFFICACY AND SAFETY OF EPRATUZUMAB IN SLE 369
Table 3. Immunologic ou comes a week 48*
EMBODY 1 EMBODY 2
Placebo
1s anda d
he apy
(n 5249)
Emab
(1,200 mg QOW)
1s anda d
he apy
(n 5244)
Emab
(600 mg QW)
1s anda d
he apy
(n 5248)
Placebo
1s anda d
he apy
(n 5263)
Emab
(1,200 mg QOW)
1s anda d
he apy (n 5261)
Emab
(600 mg QW)
1s anda d
he apy
(n 5264)
CD191B cells, % CFB,
median ( ange) cells/
m
l
29.7
(293, 783)
(n 5175)
231.5
(2100, 1,800)
(n 5179)
236.5
(294, 24,500)
(n 5169)
210.2
(2100, 1,367)
(n 5178)
232.7
(297, 24,500)
(n 5173)
237.7
(296, 646)
(n 5184)
CD31T cells, % CFB,
median ( ange) cells/
m
l
23.0
(285, 1,106)
(n 5175)
5.3
(278, 974)
(n 5180)
23.2
(265, 359)
(n 5169)
0.9
(283, 856)
(n 5178)
3.8
(280, 301)
(n 5173)
23.2
(285, 1,364)
(n 5184)
% o CD221WBCs,
CFB, mean 6SD†
21.4 62.7
(n 52)
260.8 636.4
(n 52)
– 2.7 66.2
(n 510)
256.6 631.7
(n 516)
261.8 624.6
(n 518)
IgA, % CFB, median ( ange)
gm/li e
3.0
(242.7, 113.2)
(n 5175)
6.5
(239.5, 64.4)
(n 5178)
4.1
(279.8, 536.0)
(n 5168)
3.6
(256.2, 59.5)
(n 5175)
7.9
(261.1, 858.3)
(n 5172)
4.3
(279.2, 119.2)
(n 5183)
IgG, % CFB, median ( ange)
gm/li e
0.9
(288.7, 143.5)
(n 5175)
0.8
(253.5, 94.1)
(n 5178)
2.3
(248.8, 128.1)
(n 5168)
2.2
(251.9, 79.6)
(n 5175)
3.6
(242.5, 857.6)
(n 5172)
1.6
(277.2, 65.2)
(n 5183)
IgM, % CFB, median ( ange)
gm/li e ,
0.8
(253.0, 98.2)
(n 5174)
218.7
(265.4, 142.7)
(n 5178)
222.0
(285.4, 185.7)
(n 5168)
0
(275.8, 200.0)
(n 5175)
217.7
(283.8, 712.5)
(n 5172)
220.0
(270.0, 53.3)
(n 5183)
Reduc ion in an i-dsDNA,
no. (%)‡
11 (6.4)
(n 547)
12 (6.7)
(n 547)
7 (4.2)
(n 538)
17 (9.7)
(n 544)
14 (8.2)
(n 545)
11 (6.0)
(n 539)
ANA shi om posi i e
o nega i e, no. (%)§
8 (4.9)
(n 5162)
3 (1.7)
(n 5173)
5 (3.1)
(n 5163)
7 (4.1)
(n 5169)
12 (7.4)
(n 5162)
8 (4.5)
(n 5176)
C3 no maliza ion, no. (%)¶ 11 (22.4)
(n 549)
11 (24.4)
(n 545)
8 (20.5)
(n 539)
12 (21.4)
(n 556)
12 (24.5)
(n 549)
11 (20.8)
(n 553)
C4 no maliza ion, no. (%)¶ 14 (21.9)
(n 564)
12 (21.8)
(n 555)
6 (12.5)
(n 548)
5 (7.7)
(n 565)
15 (22.4)
(n 567)
15 (18.8)
(n 580)
* CFB 5change om baseline; WBCs 5whi e blood cells (see Table 1 o o he de ini ions).
† CD191CD221cells, as a pe cen age o CD191cells, we e measu ed in a subse o pa ien s.
‡ Reduc ion in an i-dsDNA is de ined as e e ing o nega i e in only pa ien s es ing posi i e ia s anda d assay a baseline.
§ Posi i e ANA alues a e de ined as hose abo e he lowe limi o quan i ica ion, and nega i e alues a e hose below he lowe limi o quan i ica ion. Only pa ien s es ing
posi i e a baseline a e shown.
¶ Only pa ien s wi h low baseline le els a e shown.
370 CLOWSE ET AL